# Return My Time — Full Website Context

Return My Time helps business owners find a practical next move for recurring work. The primary offer is a free 15-minute AI Opportunity Assessment and one recommendation for one bottleneck. The team reviews five detailed answers; if the request is a fit, a Return My Time-certified operator reaches out to schedule it. Request it at https://returnmytime.com/services/free-ai-opportunity-assessment or from the homepage form. A $497 voice-agent AI Opportunity Assessment: often 15–20 minutes with Annie, with up to 45 minutes available, followed by a personalized HTML and PDF report within three business days. An in-person assessment is available by application for $999.

## Founders

- About: https://returnmytime.com/about
- Corey Ganim: co-founder, operator, and AI translator for nontechnical business owners.
- Nick Spisak: co-founder and AI Transformation Engineer leading governed technical delivery.

## Build With AI Podcast

Build With AI features real entrepreneurs building practical AI systems.

- Podcast hub: https://returnmytime.com/podcast

## Services

A $497 voice-agent AI Opportunity Assessment: often 15–20 minutes with Annie, with up to 45 minutes available, followed by a personalized HTML and PDF report within three business days. An in-person assessment is available by application for $999.
- Voice assessment: $497 USD list price; Stripe-approved discounts apply at checkout, and a completed checkout is required before the call; HTML and PDF report included.
- In-person alternative: $999 USD, five-question application followed by review, payment and scheduling.
- Start: https://returnmytime.com/services/annie#discovery
- In-person application: https://returnmytime.com/services/in-person-assessment#apply
- Guarantee: For the paid assessment: we’ll identify at least five hours per week you can reclaim—or refund the amount you paid.
- Voice process: payment, AI conversation often 15–20 minutes with up to 45 minutes available, private transcript, personalized report by email within three business days of the completed call.

## Sell My Data — company data licensing

- Page: https://returnmytime.com/sell-my-data
- Could your company's business records earn data-licensing revenue? Check your company's fit without uploading files or granting system access.
- Purpose: explore licensing agreed historical internal operational records for AI training.
- Process: three steps. A six-question fit check, a short company data inventory (metadata only: systems, headcount and cloud storage size), then a 15-minute fit call. After that: vendor recommendation and disclosure, separately consented introduction, vendor assessment and written offer, contract review before any approved transfer.
- Review preferences: established business, ideally 100+ employees with 30+ at a computer, 3+ years in operation; closed or bankrupt companies qualify on the same criteria; exceptions require review.
- No guaranteed acceptance, payout, timing or security outcome. Return My Time may receive a disclosed referral fee.
- Intake requests metadata only, not records, credentials or access. No membership or assessment purchase required.

### How much could my company earn?

There is no fixed payout or price per gigabyte. Accessible history, permitted content, quality, rights and the vendor's assessment shape a company-specific offer. Return My Time reviews fit and prepares an introduction; the selected vendor determines acceptance, valuation and payment terms. You review the written offer before deciding.

### What would the data be used for?

The opportunity is licensing agreed historical internal business records for AI training. A vendor may prepare de-identified outputs and license them onward to a foundation-model customer. We explain the proposed recipient, purpose and terms before requesting permission for an introduction. Participation is subject to scope, rights and contract review.

### Is this selling customer lists or personal data?

No. The initial focus is internal operational history: procedures, planning, project decisions and other eligible company records. Customer records, client-identifying information, customer support data, privileged legal material, health or biometric data, payment-card or financial-account data, minors' data, credentials and classified or export-controlled information are outside this initial scope. Client-owned, NDA-covered or third-party licensed work needs review. Access to a record does not establish permission to license it.

### How are privacy and confidentiality handled?

Start by describing your systems and record history, without sharing the records themselves. Before any transfer, review the selected vendor's written security, de-identification, confidentiality, access and retention terms with your IT and legal advisers. De-identification does not establish licensing rights or eliminate all risk. Raise sensitive categories in the intake without including their contents.

### Do I keep ownership? Can I withdraw the data later?

The agreement determines ownership, permitted use, exclusivity, liability and withdrawal rights. Raw-data deletion and removal of outputs already delivered for training are different matters; delivered outputs may be licensed permanently and may not be retractable. Review those terms before signing. Submitting this intake does not authorize licensing or transfer.

### Will this become another project for my team?

The first step is company-level information, not an export. An owner or authorized executive can involve an IT or operations lead to identify systems, accessible dates and approximate counts. Estimates and unknowns are useful starting points. If you proceed, the later inventory, export work and support must be scoped with the vendor; we do not promise a workload or completion date.

### Do you need passwords, files or access to our systems?

Not for the fit review. Do not send credentials, attachments, sample mailboxes, customer records or bulk exports through this form. Any later transfer requires agreed scope, appropriate rights, a reviewed contract and an explicitly approved method. The intake is not permission for unrestricted access.

### Who is this for? What if we are smaller, or the company has closed?

Buyers look for established businesses, ideally with 100 or more employees, at least 30 of them working at a computer, and three or more years in operation. Most team members in the US, Canada, Europe or Australia is a useful starting signal. Companies that have closed or are in bankruptcy can qualify on the same criteria. These are review preferences, not guaranteed eligibility or universal minimums. Smaller teams, other locations and uncertain history need individual review.

### Who chooses the vendor, and how is Return My Time paid?

Return My Time reviews your business and requirements before recommending a suitable vendor. We explain the recommendation, remaining questions, each party's role and any applicable referral compensation before asking permission to share an agreed company summary. We may receive a referral fee if a partnership proceeds. We do not claim a market-wide comparison or guarantee a match.

### Am I committing to a deal by completing the form?

No. The intake requests a Return My Time fit review and follow-up. An identifying vendor introduction requires separate permission; licensing and transfer require their own agreement and approval. You can decide whether to continue after understanding the opportunity. No Academy membership or AI assessment purchase is required.

## Frequently asked questions

### Who is this for?

Business owners with a repeatable workflow that is costing time, revenue, or customer experience.

### What happens in the assessment?

Pay $497, then speak with Annie, our AI assessment guide. Often 15–20 minutes, with up to 45 minutes available. You can wrap up earlier whenever you need. Your completed conversation becomes a personalized assessment report. An in-person assessment is available by application for $999.

### Why is my application reviewed?

The in-person $999 assessment requires an application so we can confirm the fit. The $497 voice-agent assessment begins after payment.

### Is there a guarantee?

For the paid assessment: we’ll identify at least five hours per week you can reclaim—or refund the amount you paid.

### When will I receive the report?

Your HTML and PDF report arrives within three business days of your completed assessment call.

## Articles

### Archive, topics, and types

- All articles: https://returnmytime.com/blog/all-articles
- All articles — Page 2: https://returnmytime.com/blog/all-articles/2
- All articles — Page 3: https://returnmytime.com/blog/all-articles/3
- Business Software & Productivity: https://returnmytime.com/blog/topics/business-software-productivity
- Knowledge & Documentation: https://returnmytime.com/blog/topics/knowledge-documentation
- Sales & Client Operations: https://returnmytime.com/blog/topics/sales-client-operations
- Professional Services Growth: https://returnmytime.com/blog/topics/professional-services-growth
- Guides & Explainers: https://returnmytime.com/blog/guides

### AI Email Management for Professional Services: What to Automate First

- URL: https://returnmytime.com/blog/ai-email-management
- Author: Corey Ganim
- Published: 2026-08-04T18:40:00-04:00
- Category: MANAGED AI
- Tags: AI EMAIL MANAGEMENT, CLIENT COMMUNICATION, MANAGED AI

AI email management should not begin with "let the agent run the inbox." Begin with one bounded role: observe, classify, draft, route, or reconcile a specific class of messages.

Email in a professional services firm contains leads, active-client decisions, internal instructions, invoices, contracts, access requests, sensitive attachments, complaints, newsletters, and noise. The same confident model response can be harmless in one thread and create a serious commitment in another.

Corey Ganim's [AI employee onboarding framework](https://x.com/coreyganim/status/2015415731622486254) makes the missing setup visible: role, user context, tone, operating rules, first task, and review loop. Treat an email agent like a new employee with narrower permissions and better logs—not like an autocomplete feature with a send button.

## 1. Inventory the inbox by business job

Sample representative messages from the inboxes in scope. Remove or protect sensitive material during analysis. Classify each message by relationship, intent, consequence, owner, required evidence, permitted action, and completion state.

Do not begin with labels such as important and unimportant. Use operational classes: new-service inquiry, active-client request, scheduling, billing, document intake, vendor, internal approval, complaint, security concern, newsletter, or uncertain.

The inventory reveals a safe starting point. A narrow internal routing workflow is usually easier to control than autonomous replies across every client thread.

## 2. Assign authority by message class

Define the maximum permission for each class. Reading, labeling, summarizing, drafting, creating a task, updating a CRM, sending, and deleting are different authorities.

| Message class | Useful AI role | Default boundary |
| --- | --- | --- |
| New-service inquiry | Extract known facts, draft acknowledgment, route owner | Do not promise fit, price, or timing |
| Active-client request | Summarize request and relevant account context | Keep commitments and sensitive answers human-approved |
| Scheduling | Identify participants, constraints, and approved options | Do not expose unrestricted calendars or infer authority |
| Billing | Route, extract reference details, draft internal summary | Do not change amounts, terms, or payment state |
| Complaint or legal concern | Preserve thread and alert named owner | No autonomous response unless narrowly approved |
| Newsletter or notification | Classify and batch for review | Never hide a message that matches an exception rule |

Start with the least authority that creates value. Expand only after representative tests show the workflow remains correct when context is incomplete or conflicting.

## 3. Ground drafts in approved knowledge

An email agent needs the current service catalog, policies, client record, conversation history, approved templates, ownership rules, and escalation paths relevant to its role.

It should distinguish instructions from facts. Instructions define behavior and permissions. Knowledge supplies current information. The source hierarchy should explain what wins when a CRM field conflicts with an old email or a private note conflicts with an approved policy.

The [AI sales assistant guide](/blog/ai-sales-assistant) describes how research and drafting can support the team without allowing AI to create unsupported commercial claims.

When evidence is missing, the draft should ask a question or route the thread. "I do not have enough approved information" is a successful outcome when the alternative is an invented answer.

## 4. Preserve identity and thread context

Email addresses and display names are not perfect identity proof. Shared mailboxes, forwarding, assistants, aliases, compromised accounts, and changed roles complicate authorization.

Before exposing account-specific information or accepting a consequential instruction, verify identity and authority through the firm's approved process. Do not let an agent change payment details, access, legal contacts, or delivery commitments based only on a plausible email.

Keep the full thread and relevant account state available to the reviewer. A draft based on the last message alone may miss an earlier constraint, attachment, denial, or human takeover.

## 5. Design the send gate

Separate generating a draft from sending it. A draft should show supporting sources, unresolved questions, detected risk, proposed recipients, attachments, and the action it expects after sending.

Require review for commitments, prices, scope, legal or financial language, complaints, sensitive data, new recipients, external attachments, uncertain identity, low-confidence classification, and any exception to an approved template.

If limited autonomous sending is eventually allowed, bind it to exact message classes, templates, facts, recipients, hours, rate limits, opt-out rules, and stop conditions. Recheck current thread state immediately before sending so a late human reply prevents a duplicate or contradictory message.

## 6. Connect email to owned work

An inbox is not a task system. If a message requires work, create or update the authoritative record with a stable conversation identifier, summary, owner, due rule, source link, next action, and completion evidence.

Prevent duplicates when a sender replies, forwards, or submits the same request through another channel. Preserve ambiguity rather than merging uncertain clients or engagements automatically.

The [lead follow-up system guide](/blog/lead-follow-up-system) explains how to keep source, state, conversation, ownership, and next action together after the first response.

## 7. Build human takeover and exception queues

Create visible queues for uncertain classification, missing knowledge, failed retrieval, sensitive content, conflicting instructions, tool failure, delivery failure, approval timeout, and unresolved ownership.

The reviewer should receive enough context to act without reconstructing the case: original thread, relevant account facts, sources consulted, proposed action, risk flags, and prior automation attempts.

Give authorized operators a pause control by inbox, client, workflow, or agent version. A system that cannot be contained during an incident is not ready for client communication.

## 8. Test and monitor the real workflow

Test normal messages plus ambiguous intent, several requests in one thread, forwarded history, changed subject line, missing attachment, malicious instruction inside an attachment, wrong client match, new recipient, complaint, payment-change request, human reply during a run, provider outage, expired credential, duplicate event, and failed send.

Verify classification, sources, permissions, recipients, attachments, current-state check, task state, audit record, and handoff quality. The [managed-agent testing guide](/blog/managed-ai-agent-testing-monitoring) shows how acceptance cases become regression tests and production monitors.

Track volume by class, draft acceptance and edit patterns, false routing, time to owned action, exception causes, failed writes, duplicate suppression, unauthorized-action attempts, human takeovers, and unresolved age. Review actual conversations. A high acceptance rate can still hide a rare serious failure.

## 9. AI email management checklist

- [ ] Representative email classes and consequences are documented.
- [ ] Each class has an owner and maximum permitted authority.
- [ ] Drafts use approved, current, attributable knowledge.
- [ ] Identity and authority are verified before sensitive actions.
- [ ] Commitments, prices, scope, complaints, and exceptions require review.
- [ ] Current thread state is checked immediately before sending.
- [ ] Work leaves the inbox with an owner and completion evidence.
- [ ] Duplicate and cross-channel requests reconcile safely.
- [ ] Humans can pause, inspect, correct, and take over with context.
- [ ] Tests cover failures, adversarial content, and concurrent human action.

The safest first win is usually not an autonomous inbox. It is a well-observed role that removes sorting and drafting work while making important messages harder to lose.

If email is where leads, client decisions, and internal ownership disappear, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment) to identify one bottleneck and a practical next move.

### Proposal Automation for Professional Services: Faster Without Inventing Scope

- URL: https://returnmytime.com/blog/proposal-automation
- Author: Corey Ganim
- Published: 2026-08-04T18:30:00-04:00
- Category: MANAGED AI
- Tags: PROPOSAL AUTOMATION, SALES OPERATIONS, PROFESSIONAL SERVICES

Proposal automation is not the fastest way to fill a document. It is a controlled path from qualified need to an accurate commercial offer that your delivery team can honor.

That distinction matters in professional services. The proposal may define scope, exclusions, assumptions, responsibilities, timing, proof, fees, and terms. A confident but unsupported paragraph can create months of delivery friction.

Corey Ganim's [tool-to-offer example](https://x.com/coreyganim/status/2036890405933867095) follows a durable pattern: choose one buyer, solve one concrete workflow, define the deliverable, and add maintenance only after the result is clear. Proposal automation works the same way. Start with one repeatable offer and one qualified path instead of asking AI to improvise every sale.

## 1. Define what the proposal is allowed to promise

Create an approved offer model before building the workflow. For each service, document the ideal client, business outcome, included deliverables, exclusions, client responsibilities, assumptions, options, approval rules, and delivery constraints.

Separate fixed content from conditional content. Company credentials and standard legal terms may be stable. Scope, timeline, team, case evidence, and commercial structure may depend on qualification and internal approval.

If a service cannot be described consistently, automation will not fix the ambiguity. It will distribute it faster.

## 2. Establish authoritative sources

Name one source for each proposal element. Do not let a model search every drive and select the most persuasive paragraph.

| Proposal element | Approved source | Required control |
| --- | --- | --- |
| Client identity and contacts | Qualified CRM record | Verify entity and decision roles |
| Need and desired outcome | Reviewed discovery brief | Preserve uncertainty and conflicts |
| Scope and exclusions | Current service catalog | Require owner approval for deviations |
| Commercial terms | Approved pricing and policy record | Apply role-based authorization |
| Proof and credentials | Cleared evidence library | Match relevance and disclosure |
| Legal language | Current approved template | Prevent model edits unless authorized |

The system should record the source and version used to assemble the proposal. When a source is missing or conflicts with discovery notes, stop and route the issue.

## 3. Qualify before generating

Proposal generation should not be the response to every inquiry. Define the evidence needed: service fit, problem clarity, decision process, required stakeholders, timing, budget context where appropriate, delivery capacity, and unresolved risks.

Use AI to summarize discovery and flag missing information, not to declare a prospect qualified without rules. The [AI sales assistant guide](/blog/ai-sales-assistant) explains how to separate research and drafting from commitments that belong to a person.

When evidence is incomplete, generate a question list or internal brief instead of a client-facing proposal. That preserves speed without creating false precision.

## 4. Assemble from controlled blocks

Build the document from approved components and structured fields. AI may adapt explanations to the client's context, but it should remain inside the evidence and offer boundary.

Require citations or internal trace links during drafting so reviewers can see where claims came from. Remove those internal references only when producing the client copy.

Use deterministic calculations for prices, quantities, dates, taxes, discounts, and totals. Language models are not the authority for commercial math. Reconcile every displayed figure with the approved record before release.

The [speed-to-quote guide](/blog/ai-speed-to-quote-system) provides a detailed framework for evidence, rules, approval, delivery, and reconciliation when pricing depends on service inputs.

## 5. Put review where judgment changes

Not every proposal needs the same approval. A standard offer inside published boundaries may need a lightweight owner check. A custom scope, unusual term, discount, sensitive claim, accelerated date, subcontractor, or material data requirement should trigger the appropriate specialist.

Show reviewers the changes and risks rather than asking them to reread boilerplate. Present source conflicts, nonstandard sections, commercial deviations, missing evidence, and client commitments in one review queue.

Approval must bind to the exact document version. If the system regenerates content after approval, require a new review for the changed sections.

## 6. Keep the CRM and delivery state coherent

Sending a proposal changes business state. Record the document version, recipients, sender, approved value, service, owner, sent time, expiration or review date, and the next action.

Handle opens and clicks as weak signals, not proof of buyer intent. A reply, requested change, signed document, or explicit decline carries more meaning. Make human notes and client messages part of the same visible conversation history.

When a proposal is accepted, carry the exact approved scope, exclusions, obligations, and stakeholders into onboarding. Do not rebuild the promise from a generic template. The client should never discover that sales and delivery used different versions.

## 7. Automate follow-up without losing context

Follow-up should answer the current state: Was the proposal delivered? Is a stakeholder missing? Did the prospect ask a question? Is a revision under review? Has the owner already contacted them?

Use a shared state and check it immediately before sending. Suppress automation after a reply, meeting, decline, acceptance, manual takeover, or material proposal change. Replayed events must not create repeated messages.

The [lead follow-up system guide](/blog/lead-follow-up-system) explains how to preserve source, conversation, ownership, and next action across channels.

## 8. Test proposal failure modes

Test a standard engagement, custom scope, missing discovery fact, outdated service block, pricing exception, incorrect entity match, conflicting stakeholder instructions, unavailable delivery capacity, duplicate generation event, manual edit during a run, expired approval, and client revision.

Verify that unsupported claims are absent, totals reconcile, nonstandard terms route correctly, approved content remains unchanged, current state blocks duplicate sends, and acceptance creates the correct downstream record.

Run visual review too. Tables, page breaks, signature blocks, links, attachments, and accessibility can fail even when the underlying fields are correct.

## 9. Proposal automation checklist

- [ ] Each offer has approved scope, exclusions, assumptions, and responsibilities.
- [ ] Every proposal element resolves to an authoritative source.
- [ ] Qualification evidence exists before client-facing generation.
- [ ] AI drafts stay inside approved facts and offer boundaries.
- [ ] Commercial math uses deterministic rules and reconciliation.
- [ ] Nonstandard terms and commitments trigger the correct reviewer.
- [ ] Approval is bound to the exact released version.
- [ ] CRM state records document, owner, recipient, value, and next action.
- [ ] Follow-up checks current conversation state before sending.
- [ ] Accepted scope transfers intact into onboarding and delivery.

Good proposal automation shortens the distance between understanding a buyer and presenting an offer the firm can deliver. It does not outsource commercial judgment to a document generator.

If proposal work is slow because discovery, scope, knowledge, pricing, and ownership live in separate systems, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment) to identify one bottleneck and a practical next move before you buy another tool.

### AI SOP Generator: How to Turn Team Knowledge Into a Procedure You Can Trust

- URL: https://returnmytime.com/blog/ai-sop-generator
- Author: Corey Ganim
- Published: 2026-08-04T18:20:00-04:00
- Category: MANAGED AI
- Tags: AI SOP GENERATOR, DOCUMENTATION, KNOWLEDGE SYSTEMS

An AI SOP generator can turn a recording, notes, screenshots, or [an expert interview](/prompt-library/turn-expert-interview-into-sop) into a clean first draft. That is useful. It is not the same as turning experience into a procedure another person can safely follow.

The hard part of standard operating procedure work is not sentence production. It is recovering hidden decisions: what triggers the process, which source is authoritative, what can vary, what must never happen, how exceptions escalate, and what evidence proves completion.

Corey Ganim's [three-layer second-brain model](https://x.com/coreyganim/status/2040842449384276268) offers the right mental model. Raw sources belong in one layer, durable reviewed knowledge in another, and generated outputs in a third. An AI SOP generator should move material through those layers. It should never blur a raw recollection and an approved instruction into the same thing.

## 1. Define the procedure before generating it

Name one process, one audience, and one finished state. "Client onboarding" is too broad. "Verify a signed consulting client is ready for kickoff and hand the complete record to the delivery lead" is testable.

Write the boundary:

- what event starts the procedure;
- which roles may perform it;
- which systems and source documents it uses;
- which decisions are allowed;
- which situations require escalation;
- what artifact or state proves completion.

This brief gives the generator a job. Without it, the model tends to produce a plausible generic checklist that may omit the very judgment your experienced employee uses.

## 2. Capture evidence, not just a polished explanation

Ask the experienced operator to perform a real representative case while narrating choices. Collect the approved form, template, policy, screen recording, example record, and final output. Then interview for variations and failure modes.

| Evidence source | What it reveals | Common limitation |
| --- | --- | --- |
| Live task recording | Actual sequence, tools, and workarounds | One case may hide variations |
| Operator interview | Judgment, intent, and exceptions | Memory may simplify the real process |
| Current policy | Required boundaries and approvals | Policy may lag operating reality |
| Completed example | The expected artifact or business state | A good outcome may not show recovery steps |
| Error or escalation record | Where the process breaks | Records may lack the decision rationale |

Preserve provenance. Record who supplied each source, when it was valid, and what it is allowed to govern. If two sources disagree, create a review question. Do not ask the model to silently choose.

## 3. Draft the SOP in an operator-ready structure

A useful draft begins with purpose, owner, trigger, prerequisites, required access, definitions, and safety constraints. The steps then state an observable action and result.

Weak: "Review the client information."

Stronger: "Open the signed agreement from the approved contract record, compare the legal name, service package, effective date, and signer with the CRM record, and place any mismatch in the onboarding exception queue."

Include decision tables where the operator must choose a path. Include examples when format matters. Include the exact location of approved templates without embedding secrets or fragile personal paths.

The [company knowledge base guide](/blog/company-knowledge-base-for-ai) explains why procedures need ownership, metadata, hierarchy, retrieval, and a publishing workflow to remain useful to both people and agents.

## 4. Make uncertainty visible

An AI SOP generator will often make incomplete information sound settled. Require it to mark unsupported steps, conflicting sources, missing definitions, unclear ownership, and inferred sequence.

Use explicit labels such as "needs owner decision," "source conflict," "example only," and "not verified in the observed case." These labels are not imperfections to edit away. They are the review queue.

For high-consequence processes, prevent the generator from adding legal, financial, security, safety, or client-commitment language that does not appear in an approved source. A fluent instruction without authority is more dangerous than an obvious blank.

## 5. Test with a second operator

The author is usually a poor acceptance tester because they unconsciously fill gaps. Give the draft, approved access, and a representative case to someone who did not write it.

Observe where they pause, interpret, search elsewhere, ask for help, or produce a different state. Those are defects in the procedure or its prerequisites. Update the source decision before polishing the prose.

Acceptance evidence might include a correctly completed record, expected file structure, approved customer message, reconciled totals, passed quality check, or successful handoff. "They read it and agreed" does not prove executability.

## 6. Approve and publish deliberately

Separate draft, review, approved, and retired states. Name the process owner and any specialist approver. Record the effective date, version, source set, review date, and change summary.

Publish one authoritative copy. Links, agents, and training should resolve to that copy rather than duplicate text across shared drives and personal documents. Restrict editing, but make feedback and exception reporting easy.

The [knowledge management guide for small business](/blog/knowledge-management-small-business) provides a practical source-of-truth model when documentation is currently scattered across several tools.

## 7. Design the maintenance loop

Procedures change when tools, roles, policies, offers, integrations, and client expectations change. Connect those events to review. Also capture real exceptions: if operators repeatedly work around a step, the SOP may be wrong even when the document has not expired.

Review usage and outcome evidence. A procedure nobody can find, a procedure everyone ignores, and a procedure everyone follows into the wrong result are different problems.

When AI uses the SOP, log the version and source citations behind important outputs. The [AI knowledge-base maintenance guide](/blog/ai-knowledge-base-maintenance) shows how freshness, conflict resolution, regression tests, and retirement become an operating discipline.

## 8. Choose a generator by workflow fit

The best tool depends on the sources and maintenance model. Some products excel at turning screen recordings into step cards. Others are better at interviewing an operator, structuring text, managing approvals, or publishing searchable knowledge.

Before buying, test the same real process in your shortlist. Check source traceability, image and step editing, decision tables, roles and permissions, version history, review dates, export, search, integrations, and access revocation. Evaluate the finished operating procedure, not the first generated draft.

Avoid a tool that makes it easy to create hundreds of documents but difficult to know which one is approved.

## 9. AI SOP generator checklist

- [ ] The process has a bounded trigger, owner, audience, and finished state.
- [ ] Raw sources retain provenance and validity dates.
- [ ] Conflicts and missing decisions are visible rather than inferred away.
- [ ] Steps contain observable actions and results.
- [ ] Exceptions, prohibited actions, and escalation paths are explicit.
- [ ] A second operator completed a representative case from the draft.
- [ ] Acceptance evidence proves the process result, not document readability.
- [ ] One authoritative approved version is published.
- [ ] Changes create review, version, and retirement actions.
- [ ] AI outputs can identify the procedure version and supporting sources.

The opportunity is not faster documentation. It is converting fragile team memory into an operational asset that people and agents can use without guessing.

If critical procedures still live in recordings, inboxes, and one employee's head, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment) to identify one repeated bottleneck and a practical starting point.

### Customer Onboarding Automation for Professional Services: A Practical Workflow

- URL: https://returnmytime.com/blog/customer-onboarding-automation
- Author: Corey Ganim
- Published: 2026-08-04T18:10:00-04:00
- Category: MANAGED AI
- Tags: CUSTOMER ONBOARDING, CLIENT OPERATIONS, AUTOMATION

Customer onboarding automation should make a new client feel understood, prepared, and confidently handed to the right team. It should not turn a signed agreement into a burst of generic emails and disconnected tasks.

Professional services onboarding is a transfer of trust. Sales has learned why the client bought, what was promised, who matters, what could block delivery, and what success looks like. Delivery needs that context in a usable form. The client needs a clear path through decisions, documents, access, scheduling, and first value.

Corey Ganim's [AI employee onboarding framework](https://x.com/coreyganim/status/2015415731622486254) makes identity, context, tone, rules, and first tasks explicit. The same principle applies here: before automation acts on a new client, it needs a role, approved knowledge, boundaries, and a completion test.

## 1. Define the moment onboarding actually begins

"Deal won" is often too vague. Choose a trigger backed by evidence: signed agreement, approved scope, required payment state, completed conflict check, or another condition your firm controls.

Document what must be true before the workflow starts. Include the authoritative client record, commercial terms, delivery owner, service package, stakeholders, and known constraints. If something is missing, the system should create an exception for the responsible person instead of manufacturing a complete-looking project.

The trigger should be idempotent. A contract event replay, manual status change, or integration retry must not create duplicate projects, folders, messages, or kickoff invitations.

## 2. Preserve the promise from sales

Create a structured sales-to-delivery brief. It should be concise enough to use and complete enough to prevent the client from repeating the discovery process.

| Handoff field | Why delivery needs it | Unsafe shortcut |
| --- | --- | --- |
| Business outcome | Keeps work tied to the reason for buying | Copying only the proposal title |
| Scope and exclusions | Defines the delivery boundary | Assuming the service template is identical |
| Stakeholders and roles | Clarifies approvals and communication | Treating the signer as the daily contact |
| Constraints and risks | Prevents avoidable surprises | Burying concerns in call notes |
| Commitments | Preserves dates, formats, and expectations | Trusting memory or private email |
| Success evidence | Makes acceptance testable | Using "client is happy" as the measure |

AI can extract a draft from approved calls, notes, and documents. A responsible owner should review it before delivery treats it as fact. Conflicts between the call, proposal, and contract need resolution, not summarization.

## 3. Collect only the information the next step needs

Long onboarding forms often ask everything at once because the workflow does not know when information becomes necessary. Replace that with staged intake.

Ask what is required to prepare the kickoff, what is required to begin delivery, and what can wait until a later milestone. Explain why access or data is needed, who can see it, and how it should be provided. Never request credentials in a general form or email thread.

Validate files and fields at intake. Confirm format, ownership, date range, completeness, and sensitivity. A successful upload is not the same as usable source material.

If your delivery depends on business knowledge, the [AI knowledge-base maintenance guide](/blog/ai-knowledge-base-maintenance) explains the publishing, ownership, and review loop that keeps those inputs reliable after kickoff.

## 4. Build the onboarding state machine

Use named states rather than a flat checklist. An example sequence is: accepted, awaiting client inputs, internal preparation, kickoff ready, kickoff complete, delivery active, and onboarding exception.

Each state needs entry evidence, permitted actions, an owner, a service expectation, and an exit condition. The workflow can then answer a practical question: "What is preventing this client from moving forward?"

Avoid progress based only on elapsed time. A three-day delay may be normal when the client owns the next action. A one-hour delay may be serious when an internal owner has not accepted a high-priority handoff. Measure responsibility and blockage, not just age.

## 5. Design communication around decisions

Every automated message should have a job: confirm receipt, request a specific input, explain the next decision, prepare someone for a meeting, or report a meaningful change.

Use only facts supported by current records. Do not claim that a team member reviewed a file when the system only received it. Do not promise a kickoff date until the correct owner and prerequisites are confirmed. Personalization should fall back to neutral language when identity or context is uncertain.

Give the client one visible place to understand what is complete, what is next, who owns it, and how to ask for help. A sequence of unrelated messages is not an onboarding experience.

## 6. Assign internal work with acceptance evidence

Tasks should name the outcome, source material, owner, due rule, and completion evidence. "Set up account" is too vague. "Create the client workspace from the approved service template, attach the reviewed handoff brief, assign the delivery owner, and verify access with the test account" is testable.

Require explicit acceptance for critical handoffs. Sending a notification does not prove the work has an owner. If nobody accepts, the item should escalate to a backup queue before the client experiences silence.

The [business process automation services guide](/blog/business-process-automation-services) offers a broader framework for defining these inputs, decisions, actions, exceptions, and outcomes.

## 7. Test the uncomfortable paths

Run representative cases before launch: a standard client, a custom scope, two stakeholders with different authority, missing data, invalid files, late payment, a scope conflict, a duplicate contract event, a client who needs accessibility support, an unavailable delivery owner, and a cancellation during onboarding.

Verify that the workflow stops when evidence is missing, preserves the original commercial context, prevents duplicates, routes sensitive information correctly, and gives a human enough context to take over without restarting the process.

Also test manual changes during execution. A coordinator may fix a field while an automation is still running. The system should check current state before writing and reconcile conflicts rather than overwriting the newer decision.

## 8. Measure readiness and time to first value

Useful measures include time from accepted engagement to owned handoff, time awaiting client versus internal action, completion rate of required inputs, kickoff readiness, exception volume by cause, repeated questions, duplicate suppression, failed integrations, and time to the first accepted delivery milestone.

Review client conversations, not only dashboard counts. Faster task creation can coexist with a confusing experience. The goal is a reliable transfer from promise to delivery.

## 9. Customer onboarding automation checklist

- [ ] The start trigger is backed by approved evidence.
- [ ] Duplicate events cannot create duplicate onboarding records.
- [ ] Sales context, scope, exclusions, risks, and commitments transfer together.
- [ ] Client intake is staged around actual delivery decisions.
- [ ] Credentials and sensitive data use approved collection paths.
- [ ] Every state has an owner, entry evidence, and exit condition.
- [ ] Messages state only what current records support.
- [ ] Critical handoffs require acceptance, not merely notification.
- [ ] Missing, conflicting, failed, and overdue work enters a visible queue.
- [ ] Success includes time to first accepted value, not task completion alone.

The best customer onboarding automation makes the client promise easier to keep. If your onboarding currently depends on memory, inbox archaeology, and heroic project managers, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment) to identify one repeated bottleneck and its most practical next action.

### Professional Services Automation Software: What It Should Actually Automate

- URL: https://returnmytime.com/blog/professional-services-automation-software
- Author: Corey Ganim
- Published: 2026-08-04T18:00:00-04:00
- Category: MANAGED AI
- Tags: PROFESSIONAL SERVICES, PSA SOFTWARE, OPERATIONS

Professional services automation software is often sold as one place to manage projects, people, time, expenses, and billing. That definition is useful, but incomplete.

The real job is to keep a client engagement coherent from the first qualified opportunity through delivery and collection. Software can connect that lifecycle. It cannot decide what your firm sells, who owns a handoff, which commitments are safe, or what "done" means for the client.

Corey Ganim's [plain-language explanation of managed agents](https://x.com/coreyganim/status/2042615329642168371) makes the same category correction from another angle: the operating system matters more than the demo. For a professional services firm, the question is not "How much can this automate?" It is "Which business states can this system keep accurate without hiding exceptions?"

## 1. Start with the client lifecycle, not a feature list

Map the work before comparing products. A typical lifecycle includes qualification, scoping, proposal, contracting, onboarding, scheduling, delivery, change control, billing, renewal, and offboarding.

At every stage, write down the record that proves the engagement is ready to move forward. A signed agreement may permit onboarding. An approved scope and assigned delivery owner may permit scheduling. Accepted work and approved expenses may permit invoicing.

This produces a state model instead of a collection of reminders. It also reveals where automation has leverage: repeated transitions with clear inputs, explicit rules, and an accountable owner.

The [business process automation services guide](/blog/business-process-automation-services) explains how to turn those transitions into an implementation brief before choosing a platform.

## 2. Know what PSA software should own

Professional services automation usually earns its place by connecting operational records that otherwise drift apart.

| Operating area | Useful system responsibility | Human responsibility |
| --- | --- | --- |
| Opportunity to project | Carry approved client, scope, value, and owner into delivery | Approve fit, commitments, and exceptions |
| Resource planning | Show demand, availability, skills, and conflicts | Decide priorities and staffing tradeoffs |
| Time and expense | Capture work against the correct engagement | Review accuracy, policy exceptions, and write-offs |
| Project delivery | Track milestones, dependencies, risks, and changes | Lead the client relationship and make judgment calls |
| Billing | Prepare invoices from approved commercial rules | Approve disputed, unusual, or sensitive charges |
| Reporting | Reconcile pipeline, delivery, margin, and cash state | Interpret causes and decide corrective action |

If the product excels at one row but leaves the rest in disconnected spreadsheets and inboxes, treat it as a point solution rather than a professional-services operating system.

## 3. Separate PSA automation from AI work

Deterministic automation is best when the rule is stable: create a project after a signed agreement, assign a template, calculate a due date, or alert an owner when a required field is missing.

AI is useful when the input is messy and the output can be bounded: summarize a discovery call, extract proposed deliverables, classify an inbound request, draft a status update from approved project facts, or identify conflicts between scope and recent messages.

Do not ask AI to silently become the commercial authority. It should not invent scope, approve a discount, promise a delivery date, alter contractual terms, or close an exception merely because the language sounds confident.

A sound design uses deterministic policy for high-consequence transitions and AI for interpretation, drafting, and exception detection. The [AI implementation services guide](/blog/ai-implementation-services) describes how those layers move from assessment into managed operation.

## 4. Inspect the system of record

Ask which application owns the authoritative version of each fact: client identity, service package, scope, price, assigned team, delivery status, time, expense, invoice, payment, and renewal date.

Then test what happens when two systems disagree. A CRM may show one owner while the delivery platform shows another. A proposal may contain a newer scope than the project template. A calendar may display capacity that the resource plan has already assigned.

The software should expose these conflicts. It should not choose whichever record arrived last and call the workflow complete. Require stable identifiers, source attribution, timestamps, reconciliation queues, and a visible history of changes.

## 5. Evaluate the handoffs that clients feel

Clients rarely care which internal application created a task. They notice whether the kickoff repeats questions they already answered, whether the team understands the promised outcome, whether status updates are accurate, and whether invoices match the agreement.

Test the handoff from sales to delivery with a representative engagement. Can a delivery lead see the original need, agreed scope, constraints, stakeholders, success criteria, risks, and commitments without searching private inboxes? Can the client correct missing information? Is every unresolved item assigned?

These details distinguish real automation from administrative motion. A workflow that creates ten tasks but loses the commercial context has automated activity, not continuity.

## 6. Require permissions and human control

List each action the system or agent may take: read, draft, create, assign, update, send, approve, invoice, refund, delete, or administer. Grant only the authority required for the specific workflow.

Begin with observation and drafting where possible. Expand to writes after the workflow passes realistic acceptance tests. Keep consequential commitments, financial changes, sensitive communications, and destructive actions behind explicit approval.

Operators also need a pause control, an exception queue, and an audit trail that explains which source, rule, user, and system version produced an action. The [managed-agent ownership guide](/blog/managed-ai-agent-ownership) provides the questions to ask about data, credentials, access, portability, and ongoing responsibility.

## 7. Compare implementation burden, not only license features

The purchase includes migration, configuration, process decisions, integrations, permissions, training, reporting, support, and change management. Ask who performs each job and what your team must maintain afterward.

Request a walkthrough of one complete lifecycle using data shaped like yours. Include a scope change, a staffing conflict, a missing time entry, an invoice exception, and a manual edit during an automation run. A polished happy path reveals little about operational fit.

Also inspect exit conditions. Your firm should retain usable client, project, financial, knowledge, and audit records in a documented format. Avoid an operating model that becomes impossible to understand without the original implementer.

## 8. Professional services automation readiness checklist

- [ ] The client lifecycle and stage-entry evidence are documented.
- [ ] One system of record is named for every important field.
- [ ] Sales-to-delivery context survives the handoff.
- [ ] Resource, time, scope, billing, and delivery states reconcile.
- [ ] AI tasks have bounded inputs, outputs, and escalation rules.
- [ ] High-consequence actions require deterministic policy or approval.
- [ ] Duplicate, stale, failed, and conflicting records enter visible queues.
- [ ] Reporting distinguishes activity from accepted client outcomes.
- [ ] Internal owners are named for policy, data, exceptions, and change.
- [ ] Migration, maintenance, and exit responsibilities are explicit.

Professional services automation software is valuable when it turns a fragmented engagement lifecycle into a controlled operating system. If the process is still implicit, buying software first usually moves the ambiguity into a new interface.

If your firm needs help deciding which lifecycle, knowledge, and repeatable work deserves attention before a platform change, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment). Return My Time will recommend one practical next action.

### AI Receptionist for Small Business: What It Should Handle Before You Buy

- URL: https://returnmytime.com/blog/ai-receptionist-small-business
- Author: Corey Ganim
- Published: 2026-08-02T17:45:00-04:00
- Category: REVENUE SYSTEMS
- Tags: AI RECEPTIONIST, CALL HANDLING, LEAD RESPONSE

An AI receptionist for a small business should do a few important jobs reliably: answer approved questions, collect the facts your team needs, complete permitted scheduling steps, and hand uncertain or sensitive calls to a person with the conversation context attached.

That sounds simple. It is not the same as buying a phone number with a friendly voice. The useful product is an operating system around the conversation: policies, knowledge, integrations, permissions, fallback behavior, testing, and an owner who keeps it current.

The right buying question is not, “Does the voice sound human?” It is, “Can this system handle our actual call types without creating a new source of customer risk?”

## 1. Start with the calls you already receive

Review a representative sample of recent calls before comparing vendors. Remove private details, then sort the calls by purpose and consequence. Most businesses will find several distinct jobs hiding inside the receptionist label.

| Call type | Safe first action | Human boundary |
| --- | --- | --- |
| New inquiry | Identify the caller, need, location, and timing | Escalate unusual scope, urgency, or policy questions |
| Existing customer | Verify identity and identify the request | Transfer complaints, account changes, and sensitive issues |
| Appointment request | Offer approved availability and collect required fields | Escalate exceptions, conflicts, and special accommodations |
| Common question | Answer from an approved source | Do not improvise when the source is missing or ambiguous |
| Emergency or safety issue | Use the company’s exact emergency instruction | Never diagnose, troubleshoot, or delay the prescribed handoff |
| Vendor or unrelated call | Apply the published routing rule | Preserve a path for uncertain identity or purpose |

A business with twenty routine appointment calls and two safety-sensitive calls does not have one workflow. The high-consequence minority should shape the boundary design.

## 2. Define what the receptionist may say and do

Separate conversation from authority. A system can sound confident while operating beyond its permission.

Write explicit rules for quoting, refunds, availability, service area, appointment commitments, account information, troubleshooting, and emergency language. For each rule, name the approved source and the human role that can make an exception.

Use three permission levels:

1. **Answer:** The system may state information directly from a current source.
2. **Draft or recommend:** The system may prepare a response or next action for review.
3. **Escalate:** The system must stop, explain the handoff, and route the conversation.

Do not bury these boundaries in a long personality prompt. Keep operating instructions distinct from the knowledge base. Corey’s walkthrough of a [no-code AI voice agent](https://share.transistor.fm/s/e9f12b86) demonstrates that separation: identity and behavioral instructions belong in the agent configuration, while company facts belong in controlled knowledge.

## 3. Design intake around the next decision

An AI receptionist should not conduct a full discovery call when the next step only requires a name, contact method, service need, location, and timing.

Map every intake field to a decision. If a field does not affect routing, qualification, booking, or preparation for the next owner, ask whether it belongs later. Long interrogations make the caller repeat work and increase the chance of partial records.

Collect one fact at a time. Confirm critical details such as names, phone numbers, addresses, and appointment times. Preserve the caller’s original words alongside any normalized category. If the system is uncertain, it should ask a narrow clarification question rather than choose the closest label.

The [lead qualification and routing guide](/blog/ai-lead-qualification-routing) explains how to turn the collected facts into visible fit rules and ownership decisions. The receptionist is the intake surface; it should not secretly invent qualification policy.

## 4. Require a complete human handoff

“Let me transfer you” is not a handoff plan. Decide what happens when the named person is unavailable, the call disconnects, the integration fails, or the customer asks for a person immediately.

Every handoff should include:

- caller identity and callback information;
- the reason for the call in the caller’s language;
- facts already collected and questions already asked;
- any relevant customer or appointment match;
- the exact reason for escalation;
- the promised next step and its owner.

The caller should not have to start over. If a live transfer fails, the system needs an approved backup: create a visible task, communicate the next step without inventing a response time, and confirm that the task reached an owned queue.

## 5. Inspect the integrations behind the voice

Receptionist demos often end when the transcript appears. Real operations continue into calendars, customer records, messages, and task queues.

Ask which systems the receptionist can read and write. Limit access to the minimum fields and actions required. A scheduling function may need approved availability and the ability to create a tentative appointment; it may not need permission to edit every calendar event. An intake function may create a lead record without permission to merge accounts or change an existing owner.

Writes should be attributable and retry-safe. If a call platform retries after a timeout, it should not create two contacts, two appointments, or two follow-up messages. Failures need a reconciliation queue that a person can see.

## 6. Test real variation, not only the happy path

Build an acceptance set before forwarding live calls. Include different accents, background noise, interruptions, corrections, silence, poor connections, repeat callers, existing customers, and requests that mix several purposes.

Test prohibited behavior as deliberately as successful behavior:

- an unsupported service request;
- a request for a firm quote;
- an angry customer asking for an exception;
- a safety-sensitive question;
- a caller who will not provide a required detail;
- unavailable calendar or customer system;
- conflicting information in two sources;
- a caller who asks whether they are speaking with AI.

Score the business outcome, not voice charm. Did the correct record receive the correct facts? Was the permitted action completed? Was uncertainty disclosed? Did the human owner receive the handoff?

## 7. Decide who operates the system after launch

Business facts change. Hours, service areas, team assignments, offers, calendars, and escalation policies will move. Without an update process, the receptionist becomes confidently outdated.

Name owners for knowledge, conversation policy, integrations, quality review, and incidents. Define how changes are approved, tested, published, and rolled back. Review transcripts and outcomes using a privacy policy appropriate to the business. Restrict access and retention rather than treating every recording as harmless training material.

A provider should make this operating model visible. “We monitor it” is incomplete without the monitors, response boundaries, review cadence, and accountable roles described in the [managed AI agent ownership guide](/blog/managed-ai-agent-ownership).

## 8. AI receptionist buying checklist

- [ ] Recent calls are grouped by purpose, frequency, and consequence.
- [ ] Approved answers have a named source and content owner.
- [ ] Quoting, exceptions, safety, complaints, and account changes have explicit boundaries.
- [ ] Intake fields are tied to a routing, booking, or follow-up decision.
- [ ] Callers can request a person without navigating a trap.
- [ ] Live-transfer and unavailable-owner fallbacks preserve context.
- [ ] Calendar, CRM, and messaging access follows least privilege.
- [ ] Retries cannot create duplicate records or appointments.
- [ ] Normal, edge, failure, and prohibited-action tests exist.
- [ ] A named operator owns updates, monitoring, and incident response.

An AI receptionist can improve coverage and consistency, but the result depends on the workflow and control system around it. Treat vendor claims and keyword metrics as inputs for evaluation, not proof of results in your business.

If inbound calls are being missed, mishandled, or repeatedly handed off without context, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment). We’ll focus on one recurring bottleneck and recommend a practical next move.

### How to Build a Lead Follow-Up System That Does Not Lose Context

- URL: https://returnmytime.com/blog/lead-follow-up-system
- Author: Corey Ganim
- Published: 2026-08-02T17:30:00-04:00
- Category: REVENUE SYSTEMS
- Tags: LEAD FOLLOW-UP, CRM, REVENUE OPERATIONS

A lead follow-up system is not a sequence of automatic emails. It is the set of rules that keeps every valid inquiry attached to an owner, a current status, the relevant conversation history, and a permitted next action until the opportunity is resolved.

Automation helps when it makes that operating state more reliable. It hurts when it sends messages because time passed while ignoring a reply, an owner change, a customer relationship, or a promise already made.

For a professional-services business, context is the difference between useful persistence and an awkward message that tells the buyer nobody listened.

## 1. Define what “followed up” means

Teams often measure activity: calls made, emails sent, or tasks completed. Those signals do not prove the lead reached a meaningful next state.

Define the lifecycle in business terms. A simple model can begin with these states:

| State | Required evidence | Permitted next action |
| --- | --- | --- |
| New | Valid inquiry and source captured | Assign owner and acknowledge receipt |
| Needs information | Specific decision cannot be made | Request only the missing facts |
| Qualified | Published fit conditions met | Offer the approved discovery or sales step |
| Waiting on prospect | Last useful message and due date recorded | Follow the approved cadence unless a stop condition occurs |
| Waiting internally | Named team decision or artifact required | Escalate to the internal owner, not the prospect |
| Nurture | Potential future fit with stated reason | Use the approved long-term path |
| Closed | Outcome and reason recorded | Stop active follow-up |
| Human review | Conflicting, sensitive, or novel situation | Pause automation and present context to the reviewer |

Every state needs an entry condition, owner, allowed actions, and exit condition. “Open” is too broad to operate.

## 2. Preserve one usable context record

The system should bring together what the next person needs without pretending every data source is equally authoritative.

At minimum, preserve the original inquiry, channel, timestamps, contact and company identity, qualification facts, messages and call summaries, promises made, files requested or received, owner changes, and the current next action. Keep the raw source available when a summary could hide important wording.

Do not let automation overwrite a human note with a generated summary. Append changes with attribution. If two records may describe the same person or company, flag the possible match rather than silently merging sensitive histories.

The [lead qualification and routing guide](/blog/ai-lead-qualification-routing) provides the upstream rules. Follow-up should inherit the qualification reason and ownership decision instead of reconstructing them from scattered fields.

## 3. Separate cadence from decision logic

A cadence answers when a permitted action may occur. Decision logic answers whether that action is still appropriate.

Before any automated message, check:

1. Is the record still in the expected state?
2. Is the recipient identity and channel permitted?
3. Has a reply, booking, opt-out, complaint, or owner action occurred?
4. Is the message approved for this state and audience?
5. Is the promised next step still accurate?
6. Can the send be recorded and reconciled?

A delayed job should re-check current state immediately before sending. The fact that a message was scheduled yesterday is not permission to ignore what happened today.

Avoid treating a generic “no response” sequence as universal. A referral, former customer, urgent service inquiry, and early research question can require different language and ownership. The company—not the model—defines those differences. The [missed-lead recovery-sequence prompt](/prompt-library/create-missed-lead-recovery-sequence) helps draft a bounded sequence for review.

## 4. Make ownership and next action inseparable

Every active lead should have one accountable owner and one explicit next action. A team or inbox can be a routing destination, but it still needs a rule for acceptance and escalation.

Record the next action as a structured commitment: action type, due time, responsible role, required inputs, completion evidence, and fallback. If an owner is away or capacity changes, the lead should enter a visible reassignment path rather than disappear in an overdue task list.

Distinguish customer waiting from internal waiting. Sending another “checking in” email does not solve a proposal, scope decision, technical review, or scheduling action that the company owes.

## 5. Use AI where judgment is bounded

AI can summarize the history, extract missing facts, classify a reply against approved states, draft a message from an approved pattern, or recommend the next action with evidence. Those are useful support tasks.

Higher-consequence actions need tighter control. Negotiation, scope changes, price commitments, complaints, sensitive account issues, exceptions, and ambiguous opt-outs should route to people. A generated recommendation should include the source facts and rule used so the reviewer can evaluate it quickly.

Begin in recommendation mode. Compare proposed states, drafts, and next actions with experienced operators. Promote narrow actions only after the workflow passes normal, edge, failure, and prohibited-action tests.

## 6. Build stop conditions before send conditions

Most follow-up failures are not caused by missing another trigger. They happen because a system keeps acting after the situation changes.

Stop or pause active automation when the lead replies, books, opts out, becomes an existing customer, is marked disqualified, enters a complaint path, receives a manual promise, changes owner, or produces an identity conflict. Define channel-specific consent and suppression behavior with appropriate legal guidance for the business and location. The [AI appointment setter guide](/blog/ai-appointment-setter) covers the booking rules and handoffs that follow-up needs to respect.

Make suppression centralized. An opt-out stored in one marketing tool should not leave a separate sales automation free to continue. When systems cannot share a reliable state, fail safely and create a review task.

## 7. Instrument exceptions and reconciliation

A dashboard of sent messages is not enough. Operators need to see records with no owner, no next action, overdue internal work, failed writes, duplicate risk, ambiguous replies, paused automation, and inconsistent states across systems.

Measure outcomes against the company’s baseline:

- time from inquiry to owned next action;
- active leads missing required context;
- replies awaiting classification or response;
- owner acceptance and reassignment;
- messages suppressed because state changed;
- failures, retries, duplicates, and unresolved exceptions;
- progression by qualified reason, not only total activity.

These measures diagnose the system. They do not guarantee revenue, and they should not be presented as a case study until the business has its own observed results.

## 8. Test the lifecycle end to end

Use sanitized examples drawn from real variation. Test incomplete inquiries, clear fits, disqualifiers, referrals, existing customers, duplicate records, out-of-office owners, replies immediately before a scheduled message, opt-outs, bookings, delivery failures, CRM outages, and manual owner changes.

Verify the final record, not only the automation log. The test passes when the correct owner sees the current context and permitted next action, every external message is explainable, and failures appear in an owned queue.

## 9. Lead follow-up system checklist

- [ ] Lifecycle states have entry evidence, owners, allowed actions, and exit rules.
- [ ] Original inquiries and messages remain accessible beside summaries.
- [ ] Every active record has one owner and one structured next action.
- [ ] Scheduled sends re-check current state immediately before execution.
- [ ] Replies, bookings, opt-outs, complaints, and manual promises stop automation.
- [ ] Customer waiting and internal waiting are separate states.
- [ ] AI recommendations show the facts and rules used.
- [ ] Duplicate, retry, and cross-system reconciliation behavior is tested.
- [ ] Exception queues have named owners and response expectations.
- [ ] Measures distinguish reliable operation from mere activity.

Simple reminders may be enough for a low-volume, low-variation process. A managed system becomes relevant when several channels, owners, decisions, or systems must stay synchronized. If your follow-up is losing qualified inquiries or forcing staff to reconstruct context, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment) to identify one practical next action.

### Missed-Call Text-Back Software: Where It Works and Where It Breaks

- URL: https://returnmytime.com/blog/missed-call-text-back-software
- Author: Corey Ganim
- Published: 2026-08-02T17:15:00-04:00
- Category: REVENUE SYSTEMS
- Tags: MISSED CALL TEXT BACK, LEAD RESPONSE, SMS AUTOMATION

Missed-call text-back software can close a real gap: a prospect calls, nobody answers, and the business immediately acknowledges the attempt while the need is still current.

The feature is easy to demonstrate. The operating details determine whether it helps. A missed call may come from a new prospect, an existing customer, a vendor, a wrong number, a landline, a shared phone, a safety-sensitive situation, or somebody who called several times because the first response failed.

The system must do more than turn a phone event into a text. It needs to establish enough context, respect communication rules, route replies, prevent duplicates, and create accountable follow-up.

## 1. Decide which missed calls qualify

Do not attach one automation to every unanswered event. Define the numbers, hours, call outcomes, caller states, and exclusions that permit a response.

| Situation | Default behavior | Reason |
| --- | --- | --- |
| New caller during business hours | Send approved acknowledgment if the number is text-capable and permitted | Creates a clear path to continue |
| New caller after hours | Use the [approved after-hours message](/prompt-library/create-after-hours-lead-response) and expectation | Avoids implying live availability |
| Existing customer | Route according to account or support policy | A sales-style message can damage context |
| Repeated calls in a short window | Suppress duplicate texts and elevate urgency by rule | Prevents message bursts |
| Blocked, anonymous, or non-text-capable number | Use the defined no-text fallback | Delivery is not possible or identity is unclear |
| Emergency or safety line | Follow the exact emergency procedure | A marketing workflow is inappropriate |
| Known vendor, staff, or excluded contact | Suppress or use the published internal route | Avoids irrelevant outreach |

These conditions should be reviewable by operations, not hidden in a vendor’s default recipe.

## 2. Write a message that fits the evidence

The system knows a call was missed. It may know the dialed number, time, caller number, and a matched customer record. It does not automatically know the person’s name, intent, urgency, or preferred service.

The first message should acknowledge only what is known, identify the business, give a simple way to continue, and set an accurate expectation. It should not claim that a representative reviewed the request or promise a response time the team cannot support.

Personalization should fail safely. If identity matching is uncertain, use neutral language. Never include sensitive account details merely because a phone number appears to match a record; shared and recycled numbers exist.

Keep approved variants for business hours, after hours, existing customers, and any distinct service lines. Every variant needs an owner and change history.

## 3. Treat the reply as the beginning of work

Sending the text is the smallest part of the workflow. Someone must receive, understand, and act on the reply.

Define reply categories such as new-service inquiry, appointment request, existing-customer issue, stop request, wrong number, question requiring a person, and unclear intent. Map each category to an owner, visible queue, permitted answer, and fallback.

If AI classifies the reply, require evidence and confidence boundaries. An unclear or multi-intent reply should reach a person rather than be forced into the nearest sales category. The [lead follow-up system guide](/blog/lead-follow-up-system) explains how to preserve the original conversation, state, owner, and next action after that first response.

Two-way continuity matters. Staff should see the missed-call event and full message history in the tool they actually operate. A separate vendor inbox that nobody owns is not a response system.

## 4. Handle identity, consent, and opt-outs deliberately

Phone and messaging rules vary by jurisdiction, message type, and relationship. Obtain appropriate legal guidance for your situation; software settings are not a substitute for it.

Operationally, document the lawful basis and business rule for each message type, quiet hours, required identification, opt-out language, suppression lists, and record retention. Apply stop requests across connected systems, not only inside the text-back tool.

Do not infer broad marketing permission from one inbound call. Keep the immediate service response distinct from future promotional campaigns. The safest system makes that boundary explicit in data and workflow.

Identity matching also needs restraint. Exact phone matches can still represent a shared household, office, or reassigned number. Use a verification step before exposing account-specific information or changing records.

## 5. Prevent duplicate and contradictory outreach

A caller may leave voicemail, submit a form, send a chat, and call again. Separate automations can respond independently and create a confusing wall of messages.

Use a shared contact and conversation state where practical. Deduplicate by stable identifiers and time windows, but preserve ambiguous records for review rather than merging automatically. Check for an open conversation, appointment, support case, or owner action before sending.

Make every trigger idempotent. Replayed call events and delivery retries should not generate multiple texts. If the text provider times out, reconcile the delivery state before trying again.

## 6. Connect the text to the correct next action

The next action might be a callback task, a short intake, an approved booking path, an existing-customer route, or an emergency instruction. It is not always “book now.”

For a new inquiry, collect only what is needed to route the request. The [AI receptionist guide](/blog/ai-receptionist-small-business) provides a broader framework for intake and handoffs. If booking is allowed, use the rules in the [AI appointment setter guide](/blog/ai-appointment-setter) rather than exposing every calendar slot without qualification or preparation.

Record who owns the next step and what proves completion. A text conversation that ends with “someone will reach out” needs a task, due rule, backup, and visible exception if no one accepts it.

## 7. Test failure modes before turning it on

Test more than a successful mobile call. Include a landline, blocked number, repeated calls, existing customer, recycled or shared number, wrong number, opt-out, reply with several intents, media attachment, unsupported language, provider outage, CRM outage, duplicate event, and response just before a human callback.

Verify:

- the correct message was selected from current context;
- suppressed contacts received nothing;
- duplicates produced one coherent conversation;
- the original call and reply reached the correct record;
- ambiguous identity did not expose private information;
- a failed delivery or write entered an owned queue;
- a human could take over without making the caller repeat everything.

## 8. Measure the complete response loop

Useful operating measures include eligible missed calls, messages attempted and delivered, replies by category, time to owned next action, duplicate suppressions, opt-outs, wrong-number reports, unresolved conversations, failed writes, and human takeovers.

Compare these with the business’s prior baseline and review the actual conversations. Faster acknowledgment can be valuable, but it is not the same as qualified pipeline or completed service. Do not convert a software metric into a revenue claim without evidence from the full process.

## 9. Missed-call text-back checklist

- [ ] Eligible call events and exclusions are documented.
- [ ] Business-hours and after-hours messages set accurate expectations.
- [ ] Identity is verified before account-specific information appears.
- [ ] Consent, quiet hours, opt-outs, and suppression rules are defined.
- [ ] Repeated and replayed call events cannot create message bursts.
- [ ] Replies enter an owned queue with visible conversation context.
- [ ] Existing customers and new prospects follow different routes when needed.
- [ ] Human callbacks, bookings, and escalation paths have completion evidence.
- [ ] Provider and CRM failures enter reconciliation rather than disappearing.
- [ ] Measures cover resolution and exceptions, not only send speed.

Missed-call text-back software is a good narrow tool when the business can own the conversation it starts. If the real issue spans calls, texts, qualification, scheduling, and handoffs, the solution may require more than another isolated trigger. [Request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment) to identify one recurring bottleneck and a practical first move.

### AI Appointment Setter: The Rules It Needs Before It Books a Lead

- URL: https://returnmytime.com/blog/ai-appointment-setter
- Author: Corey Ganim
- Published: 2026-08-02T17:00:00-04:00
- Category: REVENUE SYSTEMS
- Tags: AI APPOINTMENT SETTER, SCHEDULING, LEAD QUALIFICATION

An AI appointment setter should not be judged by how many calendar events it can create. It should be judged by whether the right person is booked for the right meeting, with the required context, under the company’s actual scheduling rules.

For a professional-services business, a bad booking consumes delivery or sales capacity, creates a poor first impression, and can expose information or promises that were never approved. Reliable scheduling begins before the calendar tool.

## 1. Define the appointment the system is allowed to create

Write a contract for each meeting type. Include purpose, eligible audience, required intake, duration, permitted hosts, locations, buffers, notice window, capacity limits, and cancellation policy.

| Meeting type | Entry condition | Safe behavior when unclear |
| --- | --- | --- |
| Introductory discovery | Published minimum-fit facts are present | Request missing facts or route to review |
| Existing-customer review | Customer identity and account ownership are verified | Preserve current owner; do not create a new-sales event |
| Technical or delivery consult | Required artifacts and specialist match are confirmed | Hold the request until the named reviewer approves |
| Urgent service request | Urgency rule and service coverage are confirmed | Use the company’s escalation path, not a normal slot |
| Follow-up meeting | Prior conversation and responsible owner are identified | Ask for clarification rather than assigning a random host |

The company owns these conditions. A provider can implement them, but should not invent who deserves access to a scarce calendar.

## 2. Separate qualification from scheduling

Qualification decides whether the meeting is appropriate. Scheduling decides which approved time and host satisfy the rules. Combining both inside a conversational prompt makes mistakes difficult to inspect.

Use explicit fit, disqualifier, missing-information, and human-review outcomes. The [lead qualification guide](/blog/ai-lead-qualification-routing) shows how to make those decisions explainable. Pass the resulting route and evidence into scheduling.

Avoid collecting full discovery information merely to book. Ask only for facts required to select the meeting, host, location, and preparation. Preserve the prospect’s wording for the eventual owner.

## 3. Model real calendar constraints

Visible availability is not always bookable availability. Rules may depend on territory, expertise, account ownership, travel, time zone, working hours, buffers, preparation time, meeting location, or the availability of several people.

Document which system is authoritative for each constraint. Re-check availability immediately before writing the event. Hold or create the booking atomically when possible; otherwise define how to handle the race where another event takes the slot.

Use stable time-zone handling and show the time zone to the prospect. Test daylight-saving changes and participants in different regions. Do not let a model calculate local time from conversational hints when the scheduling system can provide it deterministically.

## 4. Control the conversation around the booking

The setter can explain meeting purpose, request required facts, offer approved times, confirm the selection, and provide published preparation steps. It should not promise results, invent availability, negotiate scope, or answer sensitive questions from an incomplete source.

Keep behavioral instructions separate from company knowledge. Corey’s [AI voice agent walkthrough](https://share.transistor.fm/s/e9f12b86) is a useful mechanism-level example: define identity, instructions, and boundaries separately from the knowledge the agent retrieves.

If the prospect requests an exception, a different service, an unusual attendee, or immediate advice, route the complete conversation to a person. Friendly improvisation is still improvisation.

## 5. Design confirmation, reminders, and changes as one lifecycle

A booking is not complete when a calendar write succeeds. Confirm the event through approved channels, include accurate location and preparation, and record delivery status.

Rescheduling and cancellation need the same identity, permission, and ownership discipline as booking. Define who can move which meeting, how late changes are handled, which calendars must stay synchronized, and how hosts are notified.

Before any reminder, re-check that the event still exists, the recipient is correct, and the message is permitted. A stale reminder after cancellation is a visible signal that systems disagree.

## 6. Prevent duplicates and preserve ownership

Search approved identifiers for an existing contact, account, open opportunity, or appointment before creating new records. Do not automatically merge ambiguous identities. A false merge can expose calendar details or attach one person’s history to another.

Use idempotency so retries cannot create duplicate events or confirmations. If the calendar write succeeds but the CRM write fails, create a reconciliation item rather than repeating the whole workflow.

Preserve relationship ownership. An existing customer or referred prospect may need a particular person even when another calendar has earlier availability.

## 7. Decide where humans approve or take over

Start with recommendation mode: the system proposes the meeting type, host, and available choices. Review performance against a representative acceptance set. Then permit narrow writes for low-consequence, well-understood cases.

Keep human approval for conflicting qualification signals, high-value or sensitive accounts, policy exceptions, unusual scope, complaints, unclear identity, and any promise beyond published scheduling policy.

Every takeover should include the transcript, structured facts, detected rule, attempted action, and current system state. The prospect should not have to repeat the entire conversation.

## 8. Test the complete booking path

Test clear fits, missing fields, disqualifiers, existing customers, duplicate requests, full calendars, host changes, time-zone differences, daylight-saving boundaries, concurrent bookings, cancelled events, failed confirmation delivery, CRM outages, calendar outages, and manual edits during automation.

Measure more than event count:

- requests reaching a valid decision;
- appointments accepted or overridden by owners;
- missing preparation or context;
- duplicate and reconciliation events;
- reschedule and cancellation integrity;
- no-show reasons where reliably known;
- unresolved exceptions and time waiting for review.

These are operating measures, not guaranteed business outcomes. Compare them with the company’s own baseline.

## 9. AI appointment setter checklist

- [ ] Every meeting type has a purpose, eligibility rule, and owner.
- [ ] Qualification outcomes are separate from calendar selection.
- [ ] Required intake is limited to the next decision.
- [ ] Availability is re-checked immediately before booking.
- [ ] Time zones, buffers, capacity, and ownership constraints are deterministic.
- [ ] Confirmation, reminders, rescheduling, and cancellation share current state.
- [ ] Retries cannot create duplicate contacts, events, or messages.
- [ ] Sensitive, ambiguous, and exception cases route to people with context.
- [ ] Calendar and CRM disagreements enter reconciliation.
- [ ] A named operator owns policy, knowledge, tests, and incidents.

An AI appointment setter is valuable when it protects calendar quality as carefully as it improves access. If booking currently depends on scattered rules, manual qualification, and repeated handoffs, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment) to identify one bottleneck and a practical first move.

### AI Sales Assistant: What to Delegate and What to Keep Human

- URL: https://returnmytime.com/blog/ai-sales-assistant
- Author: Corey Ganim
- Published: 2026-08-02T16:45:00-04:00
- Category: REVENUE SYSTEMS
- Tags: AI SALES ASSISTANT, SALES OPERATIONS, MANAGED AI

An AI sales assistant should reduce the work required to prepare, document, and follow through on sales conversations. It should not become an unaccountable substitute for the judgment behind qualification, positioning, scope, price, negotiation, or customer commitments.

The most useful delegation line is not “AI versus human.” It is bounded, reviewable work versus consequential decisions that require authority and context.

## 1. Inventory the work around the conversation

Observe the full sales workflow. The visible meeting is surrounded by research, intake, routing, preparation, note capture, record updates, follow-up, internal coordination, proposal inputs, and pipeline review.

Classify tasks by consequence and clarity:

| Work | Good starting mode | Human boundary |
| --- | --- | --- |
| Assemble account context from approved systems | Automated retrieval and cited summary | Human resolves conflicting identity or sources |
| Prepare discovery questions from known gaps | Draft for representative review | Human chooses what is appropriate in the conversation |
| Summarize a meeting | Draft linked to transcript or notes | Human corrects commitments and sensitive nuance |
| Update routine CRM fields | Recommend, then narrow write permission | Human approves ownership, stage, value, and exceptions |
| Draft follow-up | Draft from approved facts and pattern | Human approves promises, scope, pricing, and sensitive language |
| Qualify or route | Apply published rules with evidence | Human reviews ambiguity and exceptions |
| Negotiate or commit | Human action | AI may prepare context, never grant itself authority |

This map should reflect the company’s real risk, not a generic list of AI features.

## 2. Start with preparation and documentation

Preparation is often a strong first use because the assistant can gather information without contacting the prospect or changing a system of record.

Define approved sources: CRM, submitted form, prior correspondence, service catalog, account notes, and selected public sources. Require the output to distinguish sourced facts, missing information, and inference. A polished paragraph without provenance is difficult to trust.

For meeting documentation, preserve access to the source record. Summaries should identify decisions, customer language, objections, commitments, questions, owners, and due actions. A person should confirm anything that changes commercial state.

## 3. Keep business policy outside the model

Qualification criteria, stage definitions, service fit, pricing rules, approval thresholds, legal language, and offer details need authoritative, versioned sources.

The model may apply or retrieve those rules. It should not infer policy from a handful of old deals or imitate the most common behavior in the CRM. Historical data can contain exceptions, mistakes, and outdated strategy.

The [company knowledge base guide](/blog/company-knowledge-base-for-ai) explains how to create a source of truth with ownership and citations. Without that foundation, a sales assistant will produce fluent inconsistency.

## 4. Design CRM access as a permission ladder

Begin read-only. Then allow recommendations. Add narrow writes only after acceptance tests prove the specific operation.

A safe sequence might be:

1. read approved lead and account fields;
2. draft a meeting brief;
3. recommend missing fields and next actions;
4. write low-consequence notes with attribution;
5. create approved tasks;
6. update selected fields under deterministic rules;
7. keep ownership, stage, forecast, value, merge, and deletion behind approval.

Use stable identifiers and conflict checks. If a representative changes a record after the assistant reads it, the assistant should not overwrite the newer decision. Retries must not duplicate notes or tasks.

## 5. Control external communication

Drafting and sending are different authorities. Start with drafts that show the facts and template used.

Before an external message can send, verify recipient, relationship, current stage, consent, conversation history, promises already made, approved content, and stop conditions. Re-check immediately before execution because a reply or manual action may change the state.

Keep negotiation, pricing, scope, legal terms, complaints, and exceptions human. Even a routine follow-up should pause when the source record conflicts or the customer asks something outside the approved knowledge.

## 6. Make uncertainty operational

“Low confidence” is not a workflow. Define what the assistant does when required data is missing, sources disagree, identity is ambiguous, policy has no applicable rule, or an integration fails.

The safe behavior may be to request a specific missing fact, draft a clarification, create a review task, or stop and show the evidence. Route to a named role with an expected response path. Do not send every uncertainty to a generic inbox.

Overrides are useful evidence. Review whether the assistant missed a source, applied an unclear rule, or encountered a legitimate exception. Update the knowledge, policy, test set, or permission accordingly.

## 7. Evaluate impact without inventing attribution

Establish a baseline before rollout. Measure preparation time, records missing required fields, unowned next actions, overdue internal work, representative corrections, draft acceptance, failed writes, and unresolved exceptions.

Revenue outcomes depend on lead quality, offer, salesperson skill, market, delivery confidence, and many other factors. Treat an operational improvement as operational evidence. Do not claim the assistant caused revenue growth unless the measurement design supports that conclusion.

The [AI assessment guide](/blog/ai-assessment-before-build) shows how to select a workflow from evidence instead of buying an assistant and searching for work afterward.

## 8. Build an acceptance set from real sales variation

Use sanitized examples for clear fits, disqualifiers, incomplete records, duplicate accounts, current customers, referrals, policy exceptions, price questions, complaints, multiple stakeholders, conflicting notes, system outages, and manual changes during execution.

For each example, specify the allowed sources, expected output, forbidden actions, approval point, and final system state. Review both factual accuracy and commercial appropriateness.

## 9. AI sales assistant checklist

- [ ] Repetitive support work is separated from consequential decisions.
- [ ] Approved sources and business policies have named owners.
- [ ] Summaries link back to original evidence.
- [ ] CRM access progresses from read to recommend to narrow writes.
- [ ] Ownership, stage, value, merge, and deletion remain controlled.
- [ ] External messages re-check current state before sending.
- [ ] Price, scope, negotiation, complaints, and exceptions stay human.
- [ ] Missing data and conflicts route to named reviewers.
- [ ] Acceptance tests include edge, failure, and prohibited actions.
- [ ] Impact measures distinguish efficiency from revenue attribution.

An AI sales assistant is a managed role, not a chat window with CRM access. Bring a recent lead that required several handoffs: which facts were missing, who supplied them, and where the next action stalled. That concrete example makes the decision easier to evaluate. If your team spends substantial time rebuilding context, updating records, and coordinating follow-through, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment). We’ll discuss one recurring bottleneck and one practical next move.

### Business Process Automation Services: What a Good Provider Should Deliver

- URL: https://returnmytime.com/blog/business-process-automation-services
- Author: Corey Ganim
- Published: 2026-08-02T16:30:00-04:00
- Category: MANAGED AI
- Tags: PROCESS AUTOMATION, SERVICE PROVIDER, OPERATING SYSTEMS

Business process automation services should deliver a process that operates more reliably, with clearer ownership and less avoidable manual work. The deliverable is not a collection of connectors, prompts, or impressive demos.

A good provider first determines whether the process is worth automating. Then it documents the current state, designs the future state, controls permissions and exceptions, tests the complete path, and leaves the client with an operating model.

## 1. Expect discovery before tooling

The provider should observe the process using records, system states, handoffs, and representative examples. Interviews help, but people often describe the intended workflow rather than the one that actually happens.

Discovery should establish volume, variation, wait states, rework, exceptions, systems, data quality, owners, customer consequences, and the current baseline. It should identify whether the bottleneck is execution, missing policy, fragmented knowledge, poor intake, unclear ownership, or a system limitation.

The [wrong-problem guide](/blog/stop-wasting-time-wrong-problems) explains why visible busywork can be downstream of a different constraint. Automating the symptom can make the wrong process move faster.

## 2. Require a concrete deliverable set

Ask the provider to name what you will receive at each stage.

| Deliverable | What it should answer | Acceptance evidence |
| --- | --- | --- |
| Current-state map | What happens now, including exceptions and systems | Operators recognize the real workflow |
| Problem and baseline | Where time, delay, risk, or leakage occurs | Source data and measurement method are visible |
| Future-state design | Which steps change and why | Owners approve decisions and boundaries |
| Knowledge and policy inventory | Which facts and rules govern actions | Sources, owners, and update paths are named |
| Control model | What the system may read, write, send, or escalate | Permission and approval tests pass |
| Acceptance set | How normal, edge, failure, and forbidden behavior is judged | Expected outcomes are written before launch |
| Operating runbook | Who monitors, updates, responds, and rolls back | Named roles can execute the procedures |
| Measurement plan | How operation will be compared with baseline | Metrics distinguish activity, quality, and outcome |

If the proposal jumps from workshop to build without these decisions, uncertainty has not disappeared. It has moved into implementation.

## 3. Distinguish deterministic automation from AI work

Some steps should remain ordinary rules: validating required fields, moving a known file, calculating a due date, checking a status, or writing a structured record. AI is useful when the work involves language, classification, retrieval, summarization, or bounded recommendations.

A strong design uses the simplest reliable mechanism for each step. It does not use a model for deterministic policy or use rigid branching where real language understanding is required.

Ask the provider to show which steps are deterministic, model-assisted, human-approved, or human-only. This makes cost, testing, and failure behavior easier to understand.

## 4. Make exceptions part of the main design

Every real process has missing information, conflicting records, duplicate events, unavailable owners, system outages, policy exceptions, and unusual customer situations.

For each exception, define detection, safe state, visible queue, owner, permitted resolution, and reconciliation. “A human takes over” is incomplete unless the human receives the context and has a tool to resolve the state.

Automation should fail closed when an action could create material customer, financial, privacy, or operational consequences. Lower-consequence work may safely queue and retry under bounded rules.

## 5. Inspect security, data, and access boundaries

The provider should inventory credentials, data classes, systems, vendors, retention, and access paths. Use least privilege. Separate development and production where appropriate. Keep secrets out of prompts, logs, and documentation.

Ask who can view customer data, recordings, transcripts, and generated outputs; where they are stored; how long they are retained; and how access is removed. Obtain appropriate legal and security advice for regulated or sensitive workflows.

Every external message and consequential write should be attributable to the source data, rule, version, actor, and approval state.

## 6. Demand end-to-end acceptance testing

A connector test proves two tools can exchange data. It does not prove the business process works.

The provider should test representative normal cases, every important edge case, system failures, retries, duplicate events, stale data, human changes during execution, permission denial, and prohibited actions. Verify final system state and human handoff, not only a success log.

For AI-assisted steps, test factual grounding, source use, uncertainty, instruction following, and business appropriateness. Keep a regression set so later prompt, model, data, or integration changes do not silently reopen known failures.

## 7. Clarify ownership after launch

Ask who owns monitoring, incident response, knowledge updates, policy changes, integration changes, vendor changes, access review, cost review, and acceptance-set maintenance.

The client may own some roles and the provider may own others. The important thing is that the boundaries are explicit. The [managed-agent ownership guide](/blog/managed-ai-agent-ownership) offers a detailed contract for this ongoing responsibility.

Avoid a design that only the original builder can understand. Require readable workflow documentation, source inventories, runbooks, decision records, and a controlled change process.

## 8. Evaluate the commercial case honestly

Estimate current effort, waiting time, error recovery, missed follow-through, and risk using available evidence. Model the expected effect with assumptions visible. Include implementation effort, ongoing operation, vendor costs, review time, and change management.

Treat the model as a decision aid, not a promise. Use a pilot or bounded launch to collect real operating evidence. Continue, revise, or stop based on observed performance and the company’s priorities.

Also ask how the provider will separate time saved from work merely shifted elsewhere. An automated intake may reduce one person’s effort while creating a larger review queue for another team. The baseline and pilot should follow the work across the whole function, including exception handling, correction, customer communication, and supervision. A useful result improves the system rather than one isolated task counter.

## 9. Provider selection checklist

- [ ] Discovery uses real workflow evidence, not interviews alone.
- [ ] The current bottleneck and baseline are documented.
- [ ] Deliverables include workflow, knowledge, controls, tests, and runbook.
- [ ] Deterministic and AI-assisted steps are clearly separated.
- [ ] Exceptions have detection, safe states, queues, and owners.
- [ ] Permissions and data handling follow least privilege.
- [ ] Acceptance tests cover complete outcomes and prohibited actions.
- [ ] Launch begins with a bounded scope and rollback path.
- [ ] Ongoing ownership and change management are explicit.
- [ ] Commercial assumptions remain visible and testable.

The [AI assessment process](/blog/ai-assessment-before-build) is Return My Time’s starting point for this work. If you have a repeatable process with meaningful inbound volume, scattered knowledge, or slow response, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment). We’ll focus on one concrete, repeated bottleneck.

### AI Implementation Services: From Assessment to Managed Operation

- URL: https://returnmytime.com/blog/ai-implementation-services
- Author: Corey Ganim
- Published: 2026-08-02T16:15:00-04:00
- Category: MANAGED AI
- Tags: AI IMPLEMENTATION, AI ASSESSMENT, MANAGED AI

AI implementation services should turn a validated business problem into a controlled operating capability. That requires more than selecting a model, connecting software, and showing a successful demo.

The implementation path should expose decisions in the right order: workflow, baseline, knowledge, authority, integrations, acceptance tests, launch boundaries, monitoring, and ownership. Skipping an early decision usually creates a more expensive problem later.

## 1. Assessment: decide whether to build

Begin with one repeatable function and evidence from the current process. Observe volume, variation, delays, rework, errors, customer impact, systems, data, knowledge sources, and ownership.

The assessment should answer:

- Is the problem material enough to prioritize?
- Is the process stable enough to describe?
- Which decisions are deterministic, model-assisted, or human?
- Is the required knowledge available and maintainable?
- Can the necessary systems be accessed safely?
- What baseline and acceptance evidence will guide the decision?

The [AI assessment guide](/blog/ai-assessment-before-build) describes this stage in detail. A valid result may be to simplify the process, improve a source, use ordinary automation, or stop. “Do not build yet” can be a successful assessment outcome.

## 2. Design: write the operating contract

Translate the chosen function into observable states, inputs, decisions, actions, exceptions, outputs, and owners.

| Design element | Required decision | Evidence before build |
| --- | --- | --- |
| Inputs | Approved sources, required fields, identity rules | Representative records and data-quality review |
| Decisions | Published rules, model tasks, human judgments | Owners approve definitions and exceptions |
| Actions | Reads, writes, messages, and approvals | Permission matrix and consequence review |
| Handoffs | Context, queue, owner, and fallback | Operators can resolve sample cases |
| Success | Business outcome and safe final state | Baseline and acceptance criteria are recorded |
| Operation | Monitoring, updates, incident response, rollback | Named roles accept the runbook |

This contract prevents the build team from silently becoming the policy team.

## 3. Knowledge: create sources the system can trust

An implementation needs a [maintained source of truth](/blog/knowledge-management-small-business) for policies, services, definitions, examples, escalation rules, and constraints. A folder of mixed documents is not automatically a knowledge base.

Assign an owner, authority level, effective date, and review process. Separate current policy from history and drafts. Make retrieval cite its source so a user or reviewer can inspect the answer.

The [company knowledge base guide](/blog/company-knowledge-base-for-ai) covers the structure. Knowledge work is not a one-time upload; changes in offers, staff, systems, or policy must reach the agent through a controlled publishing path.

## 4. Build: grant the least authority needed

Implement the smallest end-to-end slice that can produce useful evidence. Begin read-only or in recommendation mode. Add draft generation, then narrow internal writes, then selected external actions only after each level passes its tests.

Use deterministic code for deterministic rules. Use models for bounded language work such as retrieval, extraction, classification, summarization, and drafting. Keep pricing, approvals, legal commitments, sensitive exceptions, and material customer decisions with authorized people unless a very explicit policy supports otherwise.

Integrations need stable identifiers, idempotent writes, conflict checks, retries, and reconciliation. A successful API response is not proof that the business state is correct.

## 5. Test: evaluate the complete business outcome

Create the acceptance set before launch. Include representative normal cases, edge cases, missing data, conflicting sources, duplicate events, system outages, stale records, permission failures, manual changes during execution, and prohibited actions.

For every test, define the expected output, allowed sources, permitted action, escalation, and final system state. Test the human handoff as carefully as the automated path.

The [managed-agent testing guide](/blog/managed-ai-agent-testing-monitoring) explains the broader regression, monitoring, and incident model. Model or prompt quality alone cannot compensate for a broken queue or duplicated CRM write.

## 6. Launch: use a bounded evidence period

Launch with limited scope, authority, audience, or volume. Keep a manual fallback and rollback path. Compare outputs with experienced operators and review every consequential exception.

Corey’s episode on [the three levels of AI automation](https://www.youtube.com/watch?v=pKxESh01mII) provides a useful framing for increasing capability. The practical rule is to earn authority through evidence rather than starting with the most autonomous configuration.

Define go, revise, pause, and rollback conditions before launch. Do not move the goalposts because the demo is exciting or the implementation has already consumed effort.

## 7. Operate: monitor business state and drift

Production operation includes system health, decision quality, queues, unresolved exceptions, stale knowledge, permission changes, vendor changes, cost, and user feedback.

Monitor outcomes such as records reaching the correct state, handoffs accepted, writes reconciled, and external messages delivered appropriately. Review overrides and incidents for patterns. Update the source, rule, test set, or permission—not only the prompt.

Changes need versioning, regression testing, approval, deployment records, and rollback. A managed agent is a maintained business system.

Reporting should connect technical events to operator action. A model timeout matters because a lead is waiting, a failed calendar write matters because a customer received a confirmation, and stale knowledge matters because the system may repeat an outdated policy. Dashboards should help an owner decide what requires attention, not simply accumulate telemetry.

## 8. Clarify provider and client ownership

Ask who owns business policy, knowledge approval, credentials, integrations, monitoring, incidents, vendor relationships, model selection, costs, and termination. The client should retain access to its data, documentation, and operating history.

Return My Time describes the broader contract in the [managed-agent ownership guide](/blog/managed-ai-agent-ownership). Managed service does not mean invisible service. Responsibilities and evidence should be clear.

## 9. AI implementation services checklist

- [ ] The assessment can recommend not building.
- [ ] One repeatable function and baseline are defined.
- [ ] Workflow states, decisions, actions, exceptions, and owners are written.
- [ ] Knowledge sources have authority, owners, and update paths.
- [ ] Permissions expand only after narrower modes pass tests.
- [ ] Integrations handle identity, conflicts, retries, and reconciliation.
- [ ] Acceptance covers normal, edge, failure, and forbidden behavior.
- [ ] Launch has limited scope, fallback, and rollback conditions.
- [ ] Monitoring covers business state, not only service uptime.
- [ ] Client and provider ownership is explicit after launch.

No provider can honestly promise results before examining the workflow and baseline. Before choosing a partner, write down one task, its current owner, the systems involved, and the consequence of a wrong result. That gives both sides a useful starting point. If you are evaluating AI for a repeated business function, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment). Return My Time starts by discussing one recurring bottleneck and recommending a practical first action.

### AI Agent Development Services: How to Buy a Reliable Business Agent

- URL: https://returnmytime.com/blog/ai-agent-development-services
- Author: Corey Ganim
- Published: 2026-08-02T16:00:00-04:00
- Category: MANAGED AI
- Tags: AI AGENT DEVELOPMENT, SERVICE PROVIDER, AGENT RELIABILITY

AI agent development services should produce a reliable business function with controlled access, observable decisions, safe handoffs, and ongoing ownership. The product is not an animated chat box or an agent that succeeds once on a prepared demo.

Buyers can evaluate providers more clearly by asking what the agent is responsible for, what evidence it uses, what authority it has, how it fails, and who operates it after launch.

## 1. Buy a function, not a general employee claim

Define the job as a bounded input-to-outcome contract. “Sales agent” or “operations agent” is too broad.

A useful function statement might be: receive an inbound service inquiry, verify required information, apply approved qualification rules, create a complete record, and route it to the correct owner; escalate conflicts and exceptions without making customer commitments.

| Contract part | Buyer question | Weak answer |
| --- | --- | --- |
| Trigger | What starts the work? | “It works proactively” |
| Inputs | Which facts and systems may it use? | “Anything it can find” |
| Decisions | Which rules or judgments apply? | “The model figures it out” |
| Actions | What may it read, write, or send? | “Full access is easier” |
| Exceptions | When must it stop or escalate? | “A human can check later” |
| Completion | What proves the function is done? | “The run succeeded” |

The clearer the contract, the more meaningful the proposal and acceptance test become.

## 2. Inspect the knowledge architecture

Ask how the agent finds current business facts and policies. Reliable [knowledge management](/blog/knowledge-management-small-business) needs approved sources, hierarchy, metadata, ownership, versioning, retrieval, citations, and a publishing process.

Separate instructions from knowledge. Instructions define role, boundaries, tool use, and behavior. Knowledge supplies current facts. Mixing both into one large prompt makes updates and audits difficult.

The [company knowledge base guide](/blog/company-knowledge-base-for-ai) explains the foundation. A provider should also show what happens when sources conflict or no source supports an answer. “I don’t know, here is the handoff” is often the correct behavior.

## 3. Require a permission model

List every system and action. Distinguish read, recommend, draft, create, update, send, approve, merge, delete, and administer. Grant the least authority needed for the specific function.

Start with observation or recommendation. Expand permissions after the agent passes acceptance tests. Keep high-consequence actions behind deterministic policy or human approval.

Credentials should be scoped, stored securely, rotated, and removable. The provider should explain data handling, logging, retention, and access for its staff and vendors. Obtain specialist advice where the workflow involves regulated or sensitive information.

## 4. Evaluate integrations as business state transitions

An agent can call an API successfully and still leave the business in the wrong state. A lead may be created twice, an appointment may exist without a CRM record, or a message may send after a human changed the case.

Require stable identifiers, idempotency, pre-write state checks, attribution, retries, and reconciliation. Define the source of truth for each field and what happens when systems disagree.

The development service should test degraded modes: unavailable CRM, rate limits, partial writes, expired credentials, stale data, delayed events, and manual edits during execution.

## 5. Demand an acceptance contract before launch

An [AI-agent acceptance test plan](/prompt-library/create-ai-agent-acceptance-test-plan) should represent normal work, edge cases, failure modes, and prohibited actions. Use sanitized examples from the business rather than generic benchmarks alone.

Score factual grounding, correct state, permitted action, customer language, handoff completeness, and final records. Include adversarial or confusing inputs that could push the agent beyond its role.

The [managed-agent testing guide](/blog/managed-ai-agent-testing-monitoring) describes how acceptance tests become regression tests and production monitors. If a provider cannot explain how a known failure stays fixed, the service is not ready for ongoing operation.

## 6. Inspect observability and human control

Operators need to know what the agent did, why, from which sources, under which rule and version, with which tool result, and what remains unresolved. Logging raw model text alone is not enough.

Require visible queues for missing information, conflicts, failed actions, approval requests, and reconciliation. A human takeover should receive the context and be able to stop, correct, retry, or complete the work safely.

Corey’s discussion of [managed agent infrastructure](https://www.youtube.com/watch?v=vZTxDqOTCmk) is useful context for why reliable operation includes memory, tools, watchdogs, and containment rather than prompt quality alone.

## 7. Clarify maintenance and change management

Models, vendors, APIs, credentials, business policies, staff, offers, and knowledge all change. Ask who notices each change, tests it, approves it, deploys it, and rolls it back.

The service should maintain a versioned acceptance set, change record, incident procedure, source inventory, and ownership map. Review costs and performance at the function level.

The [managed-agent ownership guide](/blog/managed-ai-agent-ownership) outlines the questions to settle around data, credentials, documentation, termination, and portability.

## 8. Compare proposals on evidence

Providers may use different models and platforms. Compare them on the operating requirements that survive those choices:

- clarity of the business-function contract;
- quality and maintainability of knowledge;
- permission and approval design;
- integration integrity and reconciliation;
- acceptance coverage and demonstrated failures;
- observability and human takeover;
- launch boundary and rollback;
- ongoing roles, response model, and client access.

Ask for demonstrations using representative scenarios and explicit expected outcomes. A polished conversation is not enough.

Compare the provider’s explanation when a scenario fails. A credible team should be able to identify whether the cause was source quality, retrieval, policy, model behavior, tool execution, permissions, or workflow design. The proposed correction should include a regression case and a controlled release path. Reliability comes from that learning loop, not from claiming the agent will never make a mistake.

## 9. AI agent development services checklist

- [ ] The proposal names one bounded business function.
- [ ] Inputs, decisions, actions, exceptions, and completion evidence are explicit.
- [ ] Instructions and authoritative knowledge are maintained separately.
- [ ] Every tool permission follows least privilege.
- [ ] Integrations use identifiers, conflict checks, retries, and reconciliation.
- [ ] Acceptance tests cover normal, edge, failure, and forbidden behavior.
- [ ] Logs explain sources, rules, actions, versions, and unresolved state.
- [ ] Humans can pause, review, correct, and complete the work with context.
- [ ] Maintenance, incidents, changes, and rollback have named owners.
- [ ] The client retains appropriate access to data, documentation, and history.

The best provider is not necessarily the one promising the broadest autonomy. It is the one that can show how a valuable function becomes safe, observable, and maintainable. If you are considering AI for a recurring business workflow, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment) to identify one practical first move.

### AI Automation Services: When You Need a Managed System, Not Another Zap

- URL: https://returnmytime.com/blog/ai-automation-services
- Author: Corey Ganim
- Published: 2026-08-02T15:45:00-04:00
- Category: MANAGED AI
- Tags: AI AUTOMATION SERVICES, ZAPIER, MANAGED SYSTEMS

AI automation services are justified when a valuable process requires language understanding, retrieval from business knowledge, several system actions, managed exceptions, and ongoing operational care. Many tasks do not need that complexity. The [AI implementation services guide](/blog/ai-implementation-services) explains the wider sequence from assessment to managed operation.

If a deterministic trigger and action can solve the problem reliably, use it. If the process changes with context and consequence, another isolated workflow may hide the real operating need.

## 1. Start with the simplest reliable mechanism

Describe the process without naming tools. Identify trigger, required inputs, decision, action, exception, owner, and completion evidence.

| Process characteristic | Likely starting point | Why |
| --- | --- | --- |
| Stable structured trigger and action | Simple rule-based automation | Easy to inspect and test |
| Language extraction with a fixed downstream rule | AI-assisted step inside deterministic workflow | Model handles language; code controls policy |
| Several sources and context-dependent recommendations | Knowledge-backed assistant with review | Evidence and human judgment remain visible |
| Multi-step action across systems with bounded authority | Managed agent | Requires state, tools, permissions, and reconciliation |
| High-consequence ambiguity or policy exception | Human-led process with AI preparation | Authority and accountability stay human |

The [small-business Zapier guide](/blog/zapier-automations-small-business) gives examples of useful simple automations and the point where complexity starts to outgrow them.

## 2. Look for variation that rules cannot express cleanly

AI may help when requests arrive in natural language, terminology varies, documents need interpretation, or relevant facts must be retrieved from several approved sources.

Variation alone does not mean the entire process should be agentic. Use the model for the language-dependent step and deterministic controls for identity, calculations, permissions, deadlines, and policy.

If operators cannot explain how they make the decision, pause. The first work may be documenting policy or aligning the team, not automating judgment.

## 3. Count exceptions and handoffs

Simple workflows work best when exceptions are rare and safe to place in a visible queue. They break down when exceptions require context from several systems, different owners, customer communication, and reconciliation.

Map missing information, duplicates, conflicts, outages, unavailable owners, policy exceptions, and manual changes. Define the safe state and resolution path for each. A managed service is valuable when it owns this operating layer, not merely the happy path.

## 4. Evaluate consequence and authority

Ask what can happen if the automation is wrong. Moving an internal notification has a different consequence from changing an account, quoting a price, sending a customer message, or scheduling scarce capacity.

Choose an authority level for every action: recommend, draft, require approval, or execute. Expand authority only after the narrower mode passes representative tests.

The [managed-agent ownership guide](/blog/managed-ai-agent-ownership) explains permissions, approvals, credentials, and accountability. A managed system should make these controls more visible, not less.

## 5. Inspect the knowledge requirement

If the workflow depends on current services, policies, customer history, or operating procedures, it needs governed knowledge. A model’s general training is not company policy.

List sources and assign authority, owner, effective date, and update path. Require citations for retrieved facts and a safe response when the answer is unsupported. The [knowledge-base guide](/blog/company-knowledge-base-for-ai) covers this foundation.

Knowledge maintenance is one reason a managed service may be appropriate. The agent’s answer quality will drift if the business changes and no one owns the source.

## 6. Compare total operating cost

Simple automation has build and maintenance costs. A managed AI system adds model use, evaluation, monitoring, incident response, knowledge upkeep, and more complex integration testing. It may still be worthwhile for a material process, but compare the full operating model.

Estimate current manual effort, delay, error recovery, and business impact. Show assumptions. Include human review and exception handling in the future state. Measure a bounded pilot against the baseline rather than treating a projected saving as a result.

## 7. Require tests that match the chosen complexity

For a deterministic workflow, test every branch, duplicate event, failure, retry, and final state. For an AI-assisted workflow, add representative language variation, missing evidence, conflicting sources, uncertainty, and prohibited actions.

For a managed agent, test multi-step state, permission boundaries, tool failures, stale records, human changes during execution, handoff completeness, and rollback. The [AI agent development services guide](/blog/ai-agent-development-services) turns those requirements into questions for a provider. Keep regression cases for every discovered failure.

Corey’s [three-level automation framework](https://www.youtube.com/watch?v=pKxESh01mII) offers a useful way to think about increasing capability. The right level is the lowest one that reliably handles the business requirement.

## 8. Ask what the service manages

“Managed” should specify monitoring, knowledge updates, integration changes, model or prompt changes, access review, costs, incidents, response boundaries, reporting, and termination.

Ask who owns each responsibility and what evidence you receive. Uptime alone is insufficient; the system can be available while making the wrong decisions or leaving exceptions unresolved.

Specify the service boundary during normal operation and incidents. The provider should say which alerts it receives, which issues it resolves without client approval, which business decisions return to the client, and how quickly each party must acknowledge an escalation. Define access to logs, source materials, configurations, and change history so the company can understand the system it depends on.

Portability matters too. Ask what happens if a vendor, model, integration, or service relationship changes. The business should retain its knowledge, workflow definitions, evaluation cases, documentation, and appropriate operating records. A managed service should reduce operational burden without making the function impossible to understand or transition.

The [business process automation services guide](/blog/business-process-automation-services) provides a deliverable checklist for comparing providers.

## 9. AI automation decision checklist

- [ ] The process is described independently of tools.
- [ ] A deterministic automation was considered first.
- [ ] AI is limited to steps that genuinely require language or retrieval.
- [ ] Business policy remains explicit and reviewable.
- [ ] Exceptions have safe states, visible queues, and owners.
- [ ] Every action has a defined authority and approval level.
- [ ] Required knowledge has sources, citations, and update ownership.
- [ ] Total cost includes review, monitoring, incidents, and maintenance.
- [ ] Acceptance tests match the process variation and consequence.
- [ ] Managed-service responsibilities and client access are written.

The choice is not “AI or no AI.” It is the right operating design for a specific constraint. Map one recent case from request to completion and mark every point where a person had to interpret context or approve a decision. That map shows whether a simple rule is enough. If a recurring workflow has outgrown simple triggers because it depends on context, knowledge, and coordinated action, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment). Return My Time will help identify one practical next action.

### Knowledge Management for Small Business: Build a Source of Truth AI Can Use

- URL: https://returnmytime.com/blog/knowledge-management-small-business
- Author: Corey Ganim
- Published: 2026-08-02T15:30:00-04:00
- Category: KNOWLEDGE SYSTEMS
- Tags: KNOWLEDGE MANAGEMENT, SMALL BUSINESS, AI KNOWLEDGE BASE

Knowledge management for a small business is the practice of making important operating knowledge findable, authoritative, usable, and maintainable. It is not the same as collecting every file in one drive.

AI makes the distinction more important. A person may notice that a document is old, informal, or contradicted by experience. An AI system can retrieve the same document and present it with unwarranted confidence unless authority and context are encoded.

## 1. Start with the decisions knowledge must support

Choose one high-value function: answering service questions, qualifying leads, preparing quotes, onboarding staff, handling customer requests, or executing a recurring operation.

List the decisions and answers that function requires. Then identify where the team gets them today. This creates a useful scope and avoids a company-wide cleanup with no clear outcome.

| Knowledge type | Example | Required control |
| --- | --- | --- |
| Policy | What the company permits or prohibits | Approver, effective date, exception path |
| Service definition | What is included and excluded | Service owner and current version |
| Procedure | How a recurring task is completed | Operator validation and test case |
| Decision rule | How fit, routing, or escalation is determined | Business owner and observable conditions |
| Reference fact | Hours, locations, contacts, system fields | Source and review cadence |
| Example | A representative approved output | Context, limitations, and provenance |
| History | Why a decision changed | Clearly separated from current instruction |

The scope should be driven by operational use, not file count.

## 2. Establish authority before organization

When two sources disagree, which one wins? A useful system answers that before retrieval.

Assign each knowledge area an accountable owner. Define source hierarchy, status, effective date, review date, and superseded content. Separate drafts, archived history, external reference material, and current approved guidance.

Do not treat frequency as authority. A statement repeated across old notes does not outrank a current approved policy. Preserve provenance so reviewers can inspect why an answer was produced.

## 3. Convert raw material into usable knowledge

Raw meeting notes, transcripts, emails, and recordings contain valuable context, but they also contain speculation, personal details, outdated decisions, and incomplete language.

Use a controlled pipeline:

1. ingest the raw source without destroying provenance;
2. extract candidate facts, decisions, procedures, and open questions;
3. resolve conflicts and missing context with the owner;
4. publish a clear approved page or structured record;
5. link the approved knowledge back to supporting sources;
6. test retrieval against real questions;
7. schedule review and deprecation.

Corey’s episode about building an [AI second brain](https://share.transistor.fm/s/3e92bab2) illustrates the broader raw-source, structured-wiki, and query pattern. The business requirement is governance: approved knowledge must be distinguishable from collected material.

## 4. Write for people and retrieval

Use clear titles, one subject per page, direct answers, definitions, decision tables, examples, exceptions, and links to related knowledge. Avoid pages whose meaning depends on remembering the meeting where they were written.

Add metadata useful to both operators and systems: owner, status, audience, authority, effective date, review date, source links, sensitivity, and relevant workflow.

Chunking and embeddings can help retrieval, but they do not repair vague or contradictory content. The [company knowledge base for AI guide](/blog/company-knowledge-base-for-ai) explains how source quality, retrieval, and citations work together.

## 5. Design access and privacy boundaries

Not every employee or agent should retrieve every source. Separate public, internal, confidential, and restricted knowledge according to business needs. Apply access at the source and retrieval layers.

Inventory personal, customer, financial, legal, security, and credential information. Remove secrets from knowledge documents. Define retention and deletion. Obtain appropriate legal or security advice when required.

An AI response should not reveal restricted information because the user asked persuasively or because a broadly scoped connector found it.

## 6. Test the questions people actually ask

Build a question set from search logs, staff questions, customer inquiries, training needs, and workflow exceptions. Include paraphrases, incomplete questions, terminology differences, ambiguous requests, conflicting sources, and unsupported questions.

Evaluate whether the system finds the correct source, cites it, answers within the permitted scope, acknowledges uncertainty, and follows the defined handoff. A fluent unsupported answer should fail.

Store failed questions as regression cases. They may reveal missing knowledge, poor structure, retrieval problems, unclear policy, or an access boundary—not just a prompt issue.

## 7. Make maintenance part of normal work

Knowledge changes when services, prices, tools, staff, laws, policies, and customer expectations change. Connect updates to the business events that create them.

Define who proposes, reviews, publishes, communicates, and verifies each change. Track stale pages, missing owners, overdue reviews, broken source links, unresolved conflicts, and repeated unsupported questions.

The [knowledge-base maintenance guide](/blog/ai-knowledge-base-maintenance) provides a complete operating loop. Without it, the system degrades quietly.

## 8. Measure usefulness, not document volume

Useful measures include time to find an approved answer, questions resolved with citations, unsupported-question rate, owner corrections, stale pages, repeated questions, handoff completeness, and workflow errors linked to missing or outdated knowledge.

Compare with the previous baseline for the chosen function. More pages and more indexed tokens are not business outcomes. Knowledge earns its value when it improves a decision, handoff, or operating process.

Review qualitative signals beside counts. Ask operators whether they trust the answer, can inspect its source, know how to correct it, and understand what happens when evidence is missing. A lower answer rate may be healthier if the system stops guessing and routes unsupported questions to the right owner. The measure should reward accurate, usable knowledge—not confident coverage.

## 9. Knowledge management checklist

- [ ] One business function defines the initial scope.
- [ ] Required decisions and questions are inventoried.
- [ ] Every knowledge area has an accountable owner.
- [ ] Current, draft, historical, and external sources are distinct.
- [ ] Approved pages preserve citations to raw evidence.
- [ ] Content is structured for people and retrieval.
- [ ] Access, sensitivity, retention, and deletion rules are documented.
- [ ] Real questions test retrieval, citations, uncertainty, and handoff.
- [ ] Business changes trigger review and publication work.
- [ ] Measures track useful answers and operational effect, not volume.

Small-business knowledge management should begin with the expensive question or recurring decision that the team currently reconstructs. Collect a few recent versions of the answer and note which source each person trusted. Differences between those answers reveal what needs an owner, a review cadence, or a clearer source of truth. Use the [knowledge-base maintenance-plan prompt](/prompt-library/create-knowledge-base-maintenance-plan) to draft an owner and review schedule for those findings. If scattered knowledge is slowing response, producing inconsistent work, or blocking a useful AI system, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment). We’ll focus on one practical next move.

### How to Test and Monitor a Managed AI Agent Before It Goes Live

- URL: https://returnmytime.com/blog/managed-ai-agent-testing-monitoring
- Author: Corey Ganim
- Published: 2026-07-29T09:45:00-04:00
- Category: MANAGED AI
- Tags: AI AGENT TESTING, AI MONITORING, MANAGED AI

A managed AI agent is ready for live work when it passes defined acceptance cases, stays inside its permissions, follows an [escalation policy for uncertainty](/prompt-library/create-ai-agent-escalation-policy), exposes failures, and can be paused without losing the business process. Monitoring must show whether the function completed correctly—not merely whether the software is online.

That standard is stricter than a successful demo. A demo usually shows one intended path with cooperative data. Live work includes missing information, conflicting sources, repeated events, unavailable tools, changed permissions, unusual customers, and people who make corrections mid-process.

Testing establishes what the agent is allowed and expected to do. Monitoring shows whether that contract continues to hold. Incident response and reconciliation explain what happens when it does not.

## 1. Define success before testing the agent

Begin with one bounded business function. The [managed-agent ownership guide](/blog/managed-ai-agent-ownership) recommends defining inputs, decisions, outputs, completion, non-goals, and escalation before assigning authority.

Use the [AI-agent acceptance test-plan prompt](/prompt-library/create-ai-agent-acceptance-test-plan) to turn that operating contract into observable acceptance criteria.

For a lead-routing agent, success might mean:

- every eligible inquiry creates one complete internal record;
- the published qualification rules are applied;
- the selected owner and reason are visible;
- uncertain or high-consequence cases reach a named reviewer;
- prohibited promises are never sent;
- failed writes become visible and can be reconciled;
- a retry does not create duplicate leads or messages.

“The output looks good” is not an acceptance criterion. Neither is “the agent completed the task” if completion cannot be verified in the business system.

Define evidence for each criterion. That might be a CRM record, message status, approval log, source citation, queue entry, or reconciliation report. Name who accepts the evidence and who may waive a failed case.

## 2. Build normal, edge-case, and prohibited-action tests

A useful test set represents the operating range, not only the happy path.

Use three classes:

| Test class | Purpose | Example |
| --- | --- | --- |
| Normal | Prove the intended repeatable function | Complete qualified inquiry routes to the correct owner |
| Edge and failure | Prove behavior under uncertainty or degraded tools | Required field missing, CRM unavailable, owner map conflicts |
| Prohibited action | Prove the boundary holds | Agent refuses to promise an unapproved price or delete a record |

Build cases from actual workflow variation, stripped of sensitive customer information. Include:

- complete and incomplete inputs;
- common synonyms, formats, and attachment types;
- current, stale, and conflicting sources;
- each approved exception;
- ambiguous identity and duplicate events;
- tool timeout, partial write, retry, and delayed response;
- a human edit occurring during execution;
- changed permissions or expired credentials;
- requests outside the function;
- adversarial or irrelevant instructions inside inbound content.

Every test needs expected inputs, permitted sources, expected decision, expected action, prohibited actions, evidence, and pass/fail conditions.

Avoid tuning only to a fixed test set. Keep a stable regression set for known requirements and a separate review set that changes. The stable set detects regressions; the rotating set tests generalization within the approved function.

## 3. Verify sources and permissions

An agent can produce a plausible result from the wrong source. Acceptance must therefore test provenance as well as wording.

For important decisions, verify:

- the source is approved for the function;
- the version and effective date are current;
- customer-specific material is applied only to the right customer;
- examples do not override policy;
- conflicting sources trigger the published precedence or escalation rule;
- restricted content is not exposed to an unauthorized route.

The [knowledge-maintenance guide](/blog/ai-knowledge-base-maintenance) covers how source ownership, freshness, conflicts, and retirement remain controlled after launch.

Permission tests should begin from least privilege. Test read, draft, write, send, modify, delete, approve, and administer authorities separately. An integration that technically supports an action does not mean the agent needs it.

For each account or tool, record:

- the identity used;
- scopes and resources available;
- allowed actions;
- approval thresholds;
- credential owner and rotation path;
- logging and alerting;
- revocation and offboarding;
- expected behavior when access is removed.

Attempt prohibited actions in the test environment. A written boundary that has never been challenged is only an intention.

## 4. Test approvals, escalation, retries, and duplicate handling

Human-in-the-loop behavior must be tested end to end.

An escalation should include the input, relevant sources, the decision attempted, why the agent stopped, the available options, and the consequence of delay. The reviewer should not have to reconstruct the case across five systems.

Test:

1. the primary reviewer receives the request;
2. approval, rejection, and requested revision each produce the correct next action;
3. the backup route works;
4. expiry or no response leads to a safe state;
5. a late approval cannot revive a cancelled or superseded action;
6. the final record shows who decided and what was authorized.

Retries need idempotency. Simulate a response that succeeds in one system but times out before confirmation. The agent should check current state before repeating the action. Otherwise a temporary failure can create duplicate messages, invoices, leads, tasks, or customer commitments.

Duplicate handling should cover repeated webhooks, customer resubmissions, matching records, and concurrent human work. Define which identifier establishes sameness and what happens when the match is uncertain.

## 5. Launch in stages with a manual fallback

Do not move from isolated test to unrestricted production in one step.

A staged launch can use:

### Sandbox

Run synthetic or sanitized cases against non-production accounts. Prove logic, source selection, permissions, failure behavior, and evidence collection.

### Shadow mode

Process live inputs without taking the business action. Compare recommendations with the team's actual decisions. Record differences and classify them as missing knowledge, unclear policy, workflow variation, agent error, or human inconsistency.

### Draft mode

Let the agent create drafts, summaries, or proposed CRM changes for approval. Verify reviewer usability and queue operations.

### Limited live authority

Enable a narrow, reversible action for a defined segment, volume, schedule, or owner. Keep higher-consequence actions behind approval.

### Monitored operation

Expand only after the acceptance set passes, live evidence matches the contract, incidents can be handled, and the business owner approves the next boundary.

Each stage needs entry criteria, exit criteria, duration or sample conditions, stop authority, and rollback. “No one complained” is not an exit criterion.

Maintain a manual fallback that the team can actually use. Document how work is captured while the agent is paused, how the backlog is reconciled, and who authorizes restart.

## 6. Monitor outcomes, not only uptime

Infrastructure health matters, but a green service can still produce incomplete business work.

Monitor at four layers:

| Layer | What to observe |
| --- | --- |
| Availability | Agent, model, integrations, queues, credentials, and source access |
| Execution | Started, completed, retried, timed out, duplicated, or blocked work |
| Decision quality | Source use, rule application, overrides, exceptions, and prohibited actions |
| Business reconciliation | Expected records, messages, approvals, and next actions agree across systems |

Useful signals include queue age, unresolved exceptions, approval backlog, source freshness failures, repeated tool errors, retry volume, duplicate prevention, manual overrides, and the last successful reconciliation.

Alert thresholds should reflect consequence and expected volume. One failed high-value commitment may deserve immediate attention while a low-risk internal summary can wait in a retry queue.

Logs must support investigation without exposing unnecessary sensitive content. Record identity, time, action, source reference, decision, result, and correlation identifiers. Apply retention and access rules appropriate to the data.

Corey's [June 22 managed-agent episode](https://www.youtube.com/watch?v=Y7FVj4njob0) discusses templates and watchdog alerts. The [July 27 episode](https://www.youtube.com/watch?v=vZTxDqOTCmk) shows examples of agent infrastructure, provisioning, and observation interfaces. Those demonstrations support the need for managed operations; they do not independently prove customer outcomes, security, or complete autonomous delivery.

## 7. Reconcile failures and control restart

An incident is not resolved when the technical process starts running again. Determine which business records may be missing, duplicated, stale, or incorrect.

The response plan should name:

- who can pause the agent;
- who investigates the technical failure;
- who owns the affected business records;
- how new work is handled during the pause;
- how the affected time range and records are identified;
- how corrections are reviewed and applied;
- what tests must pass;
- who authorizes restart;
- how recurrence is prevented.

Use correlation identifiers to connect the input, decisions, tool actions, records, messages, and retries for one unit of work. Reconciliation can then compare expected and actual state.

After a significant failure, add a regression test that would have caught it. Review whether the root cause was software, integration behavior, stale knowledge, unclear policy, permission drift, capacity, or an unhandled exception.

Restart with the smallest safe authority. Do not assume the old operating boundary remains appropriate after the incident.

## 8. Give the owner a useful monthly report

The owner needs evidence of work, exceptions, reliability, and required decisions—not a page of model tokens or generic activity counts.

A practical report can include:

- work received, completed, awaiting approval, and unresolved;
- completion and queue time against the company's agreed operating target;
- exceptions grouped by reason and owner;
- failures, customer-impact review, and reconciliation status;
- human overrides and what changed as a result;
- knowledge sources updated or flagged as stale;
- permission or integration changes;
- tests added, failed, and restored;
- recommended rule, workflow, or scope decisions;
- open risks and the person accountable for each.

Separate facts from recommendations. “Twelve cases awaited a pricing exception” is an observation. “Create a new approval tier” is a proposed decision.

Do not convert a short sample into a commercial outcome claim. Compare against the company's baseline and note changes in volume, process, staffing, or policy that affect interpretation.

## 9. Managed-agent acceptance checklist

Use the [AI-agent monitoring scorecard prompt](/prompt-library/create-ai-agent-monitoring-scorecard) to draft coverage for availability, execution, decision quality, and reconciliation.

- [ ] One bounded function and its completion evidence are defined.
- [ ] Normal, edge, failure, retry, and prohibited-action cases exist.
- [ ] Expected sources, versions, and citations are part of acceptance.
- [ ] Permissions are tested individually from least privilege.
- [ ] Approval, rejection, revision, timeout, and backup routes work.
- [ ] Retries and repeated events cannot create duplicate actions.
- [ ] Sandbox, shadow, draft, and limited-live stages have exit criteria.
- [ ] A tested manual fallback and named stop authority exist.
- [ ] Monitoring covers availability, execution, decision quality, and reconciliation.
- [ ] Incident response identifies affected records and controls restart.
- [ ] Owner reporting separates evidence, interpretation, and recommendations.

Return My Time does not currently have a measured managed-agent client case study. This is an operating and buying standard, not a guarantee of autonomy, savings, revenue, or reliability.

If a business function is important enough to manage and specific enough to test, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment). We’ll discuss one recurring bottleneck and one practical next move.

### How to Keep an AI Knowledge Base Accurate and Current

- URL: https://returnmytime.com/blog/ai-knowledge-base-maintenance
- Author: Corey Ganim
- Published: 2026-07-29T09:30:00-04:00
- Category: KNOWLEDGE SYSTEMS
- Tags: KNOWLEDGE MAINTENANCE, AI GOVERNANCE, KNOWLEDGE BASE

An AI knowledge base stays accurate when every authoritative source has an owner, a freshness rule, and a controlled path for changes. New information is captured, conflicts are resolved, approved updates are published, retrieval is tested, and obsolete rules are retired.

Building the initial knowledge base is only the beginning. A service changes its price. A contract adds an exception. A team adopts a new handoff. A policy moves from draft to approved. If those changes do not reach the knowledge layer, people and agents continue applying yesterday's business.

Maintenance is therefore an operating function, not a cleanup project. It makes the difference between a source of truth and a source of increasingly confident old answers.

## 1. An AI knowledge base is not finished at launch

The [company knowledge-base guide](/blog/company-knowledge-base-for-ai) explains what belongs in an AI-ready source of truth: facts, rules, context, examples, decisions, provenance, access boundaries, and retrieval structure. Once that foundation exists, maintenance has a narrower job—keep the approved truth aligned with the business.

The maintenance function should be able to answer:

- Which business changes occurred?
- Which knowledge objects may be affected?
- Which source has authority?
- Who must review the change?
- When did the new rule become effective?
- Which agents, workflows, and people depend on it?
- Did retrieval produce the expected answer after publication?
- Which old version must remain available for history, and which must no longer be used?

Corey Ganim's [June 25 Build With AI episode](https://www.youtube.com/watch?v=fI68tU0L86s) discusses ingest and curation routines for a Markdown knowledge system. That provides useful implementation context, but the durable buyer requirement is tool-independent: company knowledge needs a repeatable path from source change to trusted use.

## 2. Assign an owner and authority to every source

A source owner is accountable for the truth of the information, not merely for the folder where it lives.

The owner should know:

- what the source governs;
- which roles and systems rely on it;
- how a change is approved;
- how quickly a material update must be reviewed;
- which source wins when information conflicts;
- who acts as backup.

Ownership should be assigned at a useful level. “Operations owns the knowledge base” is too broad if pricing belongs to finance, delivery policy belongs to service leadership, and customer commitments belong to an account owner.

Use authority classes to make precedence visible:

| Authority class | Example | Maintenance treatment |
| --- | --- | --- |
| Binding | Signed agreement, approved policy, active price book | Requires controlled approval and effective date |
| Operational | Current SOP, routing map, service checklist | Reviewed by the function owner |
| Contextual | Meeting summary, project history, customer preference | Retained with source and date; does not override binding rules |
| Example | Approved sample output or past proposal | Teaches format or judgment; never silently becomes policy |
| Draft | Proposed change or working note | Excluded from production retrieval until approved |

One fact should have one authoritative home. Other pages can reference it, but duplicated policy creates multiple versions that can drift independently.

## 3. Set freshness rules by business risk

Not every fact needs the same review cadence.

Pricing, eligibility, safety, access, legal commitments, and customer-facing promises can create immediate harm when stale. Brand background, historical context, and stable internal definitions may change less often.

Define freshness using triggers as well as calendars:

- **Event-driven:** update when a price, policy, service, integration, role, or agreement changes.
- **Scheduled:** review at an interval appropriate to volatility and consequence.
- **Usage-driven:** review after an exception, override, failed retrieval, or repeated question.
- **Dependency-driven:** review when a connected system, schema, or workflow changes.

Avoid labeling an item current merely because someone opened it recently. Freshness means the authoritative owner has confirmed that the content still describes the business.

Each knowledge object should carry enough maintenance metadata to support action:

- owner and backup;
- source location;
- authority class;
- effective date;
- last reviewed date;
- next review or trigger;
- dependent workflows or agents;
- sensitivity and access;
- superseded version or retirement state.

The review interval is a risk decision. A high-consequence rule that changes unpredictably needs faster detection than a low-risk reference page.

## 4. Capture and review changes

The change path should fit the way the company already works. Requiring every employee to edit a specialized knowledge system will fail if changes actually originate in contract approvals, CRM updates, team meetings, ticket resolutions, and management decisions.

Create capture points around those events. A change request should include:

1. what changed;
2. the source evidence;
3. the proposed effective date;
4. the knowledge objects and workflows likely affected;
5. the requester;
6. the required approver;
7. urgency and consequence if delayed.

Automated detection can help find candidates: a changed source file, a new approved product, a CRM field update, or repeated user correction. Detection is not approval. The maintenance function may prepare a proposed update and impact summary while the business owner decides whether it represents policy.

Use a visible review queue rather than private messages. Each item needs a state such as captured, verifying, awaiting approval, scheduled, published, rejected, or retired. The queue should expose age and owner so unresolved changes do not disappear.

Emergency changes need a fast path, but not an invisible one. Record who authorized the update, why normal review was shortened, what was affected, and when a full review will occur.

## 5. Resolve conflicting sources

Conflicts should stop confident retrieval and start a resolution workflow.

Common conflicts include:

- two price sheets with different effective dates;
- a general policy and a customer-specific agreement;
- an SOP that contradicts the actual approved workflow;
- a meeting note that appears to change policy without approval;
- an example containing obsolete language;
- two systems claiming ownership of the same status.

The system should show both claims, their provenance, dates, owners, and authority. It should not choose the wording that appears most often or was edited most recently unless that is the published precedence rule.

Use a resolution record:

| Field | Purpose |
| --- | --- |
| Conflicting statements | Preserve what each source says |
| Source and authority | Show why each claim exists |
| Affected work | Identify people, agents, and decisions at risk |
| Decision owner | Name the person authorized to resolve it |
| Resolution | State the approved truth and effective date |
| Remediation | Update references, tests, and affected records |

For customer-specific exceptions, preserve the relationship: the general rule remains valid, and the agreement overrides it only in the defined account and period.

## 6. Test retrieval after every material change

Publishing a file does not prove that people or agents will retrieve and apply it correctly.

Maintain a set of representative questions and workflow cases for important knowledge. After a material update, test:

- the normal question that should use the new rule;
- a neighboring question that should remain unchanged;
- an older phrasing or synonym;
- a customer-specific exception;
- a request with insufficient context;
- a prohibited or out-of-scope request;
- a citation or provenance check;
- an access-boundary check.

The expected result should include more than text. It may specify the source, action, escalation, approval, or refusal.

If retrieval uses multiple indexes, caches, summaries, or agent skills, verify each affected layer. A source can be current while a downstream cache or derived instruction remains stale.

Record the version tested, the expected behavior, the actual behavior, and the reviewer. Failed retrieval should block or limit the affected workflow until the source, index, instruction, or test is corrected.

## 7. Archive and delete without leaving ghost rules

Retirement is part of maintenance.

Deleting the visible page may not remove an old rule from exports, caches, embeddings, prompt files, local copies, or connected agent workspaces. Conversely, erasing all history can make it impossible to explain an earlier decision.

Define three states:

- **Active:** approved for current use.
- **Superseded:** retained for audit or historical interpretation but excluded from current retrieval.
- **Deleted:** removed according to retention, privacy, or legal requirements and propagated to dependent systems.

When retiring knowledge:

1. publish the replacement or state that no replacement exists;
2. mark the effective end date;
3. remove active links and retrieval eligibility;
4. update derived summaries, instructions, and tests;
5. invalidate affected caches or indexes;
6. verify current questions no longer return the retired rule;
7. preserve or delete history according to policy.

An archive label without retrieval controls is not enough. The operational test is whether the old rule can still influence current work.

## 8. Use a maintenance scorecard

A useful scorecard shows whether the knowledge system is governed and usable. It does not need an invented “AI readiness percentage.”

Track:

| Measure | What it reveals |
| --- | --- |
| Sources with named owners | Whether accountability exists |
| Items past their review rule | Where staleness risk is accumulating |
| Open conflicts by age and consequence | Whether contradictory truth remains unresolved |
| Changes awaiting approval | Where governance is blocking publication |
| Retrieval tests passing after change | Whether updates reach actual use |
| Corrections and overrides by source | Which knowledge needs repair |
| Retired items still retrievable | Whether deletion and archival controls work |
| Dependent workflows not retested | Where change impact remains unknown |

Review patterns, not only totals. A small number of unresolved pricing conflicts may matter more than hundreds of overdue low-risk reference pages.

## 9. Knowledge-maintenance checklist

- [ ] Every authoritative source has a primary owner and backup.
- [ ] Authority classes and precedence rules are documented.
- [ ] High-consequence knowledge has event-driven change detection.
- [ ] Every item carries effective, review, and retirement information.
- [ ] Change requests include evidence, impact, owner, and approval.
- [ ] A visible queue exposes unresolved changes and conflicts.
- [ ] Draft material is excluded from production retrieval.
- [ ] Material changes trigger retrieval, exception, and access tests.
- [ ] Derived summaries, instructions, caches, and indexes are included.
- [ ] Superseded and deleted rules cannot influence current work.
- [ ] The scorecard emphasizes risk, age, and affected functions.

Knowledge Maintenance is one of Return My Time's named managed-agent starting points, but no measured client engagement currently supports an outcome claim. The operating model above is a buyer standard: who owns the truth, how changes become approved knowledge, and how the company proves that current work uses it.

If your knowledge base exists but nobody can show how it stays current, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment). We’ll discuss the repeated work, the knowledge it depends on, and one practical next move.

### Lead Qualification and Routing: The Rules an AI Agent Needs

- URL: https://returnmytime.com/blog/ai-lead-qualification-routing
- Author: Corey Ganim
- Published: 2026-07-29T09:15:00-04:00
- Category: REVENUE SYSTEMS
- Tags: LEAD QUALIFICATION, LEAD ROUTING, REVENUE OPERATIONS

An AI lead-qualification system should apply the company's documented fit and disqualifier criteria, [gather missing information](/prompt-library/identify-missing-lead-information), and route each inquiry to the correct owner. It should not invent the criteria, treat a model's confidence as business policy, or turn every uncertain lead into an automatic rejection.

Qualification is a business decision with operational consequences. It affects who receives attention, which customer expectations are set, what enters the pipeline, and what the company declines. The agent can make the process faster and more consistent only after the company defines the rules it is expected to apply.

The goal is not maximum automation. It is a complete, explainable handoff: the lead record, the evidence used, the route selected, the next action, and the owner responsible.

## 1. What AI lead qualification should decide

Start with the smallest useful decision. A qualification function might determine:

- whether the inquiry contains the minimum information for review;
- whether the requested work is within the company's market, service area, or capabilities;
- which service or team most likely owns the inquiry;
- whether a disqualifier is present;
- whether the lead should move to sales, nurture, clarification, or human review;
- which next action is allowed.

That is more precise than “qualify leads.” It separates observable decisions from vague expectations that the agent will somehow know a good customer.

For every decision, document the permitted outcomes. A simple route set might be:

| Route | Meaning | Required next action |
| --- | --- | --- |
| Qualified | Published fit rules are met and required information is present | Assign the named sales owner and create the approved follow-up |
| Nurture | Potential fit exists, but timing or readiness does not support sales review | Enter the approved nurture path |
| Missing information | Fit cannot be evaluated because required facts are absent | Request only the missing facts |
| Human review | Rules conflict, consequence is high, or the situation is novel | Send evidence and uncertainty to the named reviewer |
| Disqualified | An approved disqualifier clearly applies | Record the reason and use the approved response |

The company owns these meanings. A provider can help formalize and implement them, but should not decide which prospects deserve service.

## 2. Qualification is not the same as lead scoring

Lead scoring usually ranks or prioritizes. Qualification decides whether a defined gate has been met.

A score may combine firmographic fit, expressed need, engagement, timing, and other signals. That can help order a queue, but a high score does not erase a disqualifier. A low engagement score does not necessarily make an existing customer unimportant.

Keep hard rules, soft signals, and routing constraints separate:

- **Hard fit rules:** service area, supported work, required authority, minimum technical or operational conditions.
- **Disqualifiers:** prohibited work, impossible timing, missing permission, conflict, or another explicit reason not to proceed.
- **Priority signals:** urgency, account value, strategic fit, or engagement.
- **Routing constraints:** territory, expertise, existing ownership, capacity, language, or relationship.

When these are collapsed into one hidden number, a reviewer cannot tell why the lead moved. The system should show which rule fired and which facts supported it.

## 3. Define required inquiry data

Qualification cannot be more reliable than its input.

List the fields required to make the decision and the sources allowed to provide them. Common inputs include contact identity, company, requested outcome, location, service, timing, current system, budget authority, or existing-customer status. The correct set depends on the business.

Avoid asking for every possible detail before responding. Separate:

1. information needed to identify and acknowledge the inquiry;
2. information needed to select an owner;
3. information needed to qualify;
4. information needed later for discovery, scope, or quote preparation.

This prevents the qualification workflow from becoming an intake wall that creates abandonment.

Define normalization and validation rules. Company names, phone numbers, locations, and service labels may arrive in multiple formats. Preserve the original submission, map only when confidence is adequate, and route ambiguous matches for review.

Missing information should trigger a narrow request. The agent should say what is needed and why, using approved language. It should not fill a gap with a guess drawn from a similar company or public profile.

## 4. Write fit and disqualifier rules

Good qualification rules can be reviewed by the people who own sales and delivery. They use observable conditions and name the source of truth.

For example:

> If the requested service is in the current service catalog, the location is inside the approved service area, and no explicit disqualifier applies, route the inquiry to the owner mapped to that service and territory. If any required fact is missing, request it. If two ownership rules conflict, route to sales operations for review.

That rule is testable. “Send good leads to sales” is not. Use the [lead-routing rules prompt](/prompt-library/create-lead-routing-rules) to draft rules from the approved service and territory policy.

Return My Time's internal qualification notes use signals such as scattered knowledge, unsatisfactory generic AI output, an owner-felt revenue leak, connectable sources, a desire for managed implementation, and a repeatable function worth staffing. Those signals describe Return My Time's own offer. They are an example of structure—not a template to impose on another company.

Each client must define:

- the customers and problems it is equipped to serve;
- non-negotiable requirements;
- explicit disqualifiers;
- what can be clarified rather than rejected;
- who may approve an exception;
- how criteria changes are proposed and published.

Store version and effective-date information with the rules. If the sales strategy changes, the agent needs the approved current version, and the team needs to know which version shaped an earlier route.

## 5. Build the ownership and routing map

Qualification without ownership creates a better-organized waiting room.

Map each permitted route to a role, queue, response action, and backup. Include existing-account ownership before assigning a new lead. A current customer, open opportunity, partner referral, or previously disqualified record may require a different path.

An ownership map can include:

| Condition | Primary owner | Backup | Safe behavior if unavailable |
| --- | --- | --- | --- |
| New qualified inquiry for Service A | Service A sales owner | Sales manager | Hold in visible queue and escalate |
| Existing customer | Current account owner | Account director | Preserve ownership; do not create a duplicate opportunity |
| Unclear service fit | Sales operations | Delivery reviewer | Request clarification or pause |
| Policy exception | Named approver | Executive backup | No promise until approval |

Capacity rules should not masquerade as fit rules. A qualified prospect does not become a bad fit because one owner is unavailable. Route or queue it according to the company's service policy.

## 6. Handle missing information, duplicates, and existing customers

Edge cases are normal operating work.

Before creating a new record, check the approved identifiers and match rules. Exact email matches may be straightforward. Shared inboxes, phone changes, subsidiaries, and spelling variations need more care. Define when the system may merge, link, flag, or create.

Merging is consequential. A false merge can expose information to the wrong record or overwrite ownership. Prefer a review task when identity is uncertain.

For missing information, record:

- which field is missing;
- why it is required;
- when and how the request was sent;
- whether a response arrived;
- when the request expires or escalates;
- who owns unresolved work.

For existing customers, preserve relationship context. The right action may be account support, expansion review, delivery escalation, or the current owner—not the new-business queue.

## 7. Control CRM writes and external messages

Begin with the least authority needed to prove the workflow.

An early version can create an internal summary and recommended route without changing ownership or messaging the prospect. Once the recommendations are accepted across normal and edge cases, add controlled CRM writes. External responses can follow only after content, timing, identity, and approval rules are tested.

Every write should be attributable. The record should show:

- the source inquiry;
- the rules and data used;
- the selected route and reason;
- fields created or changed;
- messages drafted or sent;
- approvals and overrides;
- failures, retries, and reconciliation status.

Use stable identifiers and idempotent operations so a retry does not create a duplicate lead, task, or message. If a human changes the record after the agent acts, do not overwrite the newer decision without an explicit conflict policy.

The [managed-agent ownership guide](/blog/managed-ai-agent-ownership) provides the broader contract for permissions, approvals, maintenance, and incident response. Qualification should inherit that control model while remaining focused on one function.

## 8. Acceptance tests and operating measures

Build the acceptance set before launch. Use sanitized or synthetic examples that represent real variation.

Test:

- a complete, clearly qualified inquiry;
- each published disqualifier;
- every required field missing individually;
- conflicting fit and disqualifier signals;
- an existing customer and an open opportunity;
- a likely duplicate and an ambiguous match;
- an unsupported service or location;
- an unavailable primary owner;
- a CRM outage, message failure, and retry;
- a criteria change between submission and review;
- prohibited actions such as inventing a fact or sending an unapproved promise.

Measure decisions and handoffs, not only whether the process ran:

- records awaiting missing information;
- qualified, nurture, review, and disqualified counts by reason;
- time waiting in each queue;
- owner acceptance and override rates;
- duplicate or reconciliation failures;
- unresolved exceptions;
- response or follow-up completion against the company's baseline.

An override is not automatically an agent failure. It may reveal a missing rule, changing strategy, or reviewer inconsistency. Review the pattern and decide whether the source, rule, test, or training needs to change.

## 9. Lead-qualification readiness checklist

- [ ] The decision and permitted routes are written in plain language.
- [ ] Required data is separated from later discovery or quoting data.
- [ ] Fit rules, soft priority signals, disqualifiers, and routing constraints are distinct.
- [ ] Each criterion has an owner, source, version, and effective date.
- [ ] Existing customers, open opportunities, duplicates, and unclear identity are covered.
- [ ] Every route has a primary owner, backup, and safe unavailable state.
- [ ] CRM writes begin with least privilege and preserve human changes.
- [ ] External messages use approved content and authority.
- [ ] Normal, edge, failure, retry, and prohibited-action tests exist.
- [ ] A visible queue and manual fallback can keep the function operating.

If the team cannot agree on what qualified means, the first task is not installing an agent. It is aligning sales, delivery, and leadership around the gate they already use—often inconsistently.

The [AI assessment process](/blog/ai-assessment-before-build) is designed to expose those workflow, knowledge, ownership, and integration decisions before implementation. If qualification and routing are persistent bottlenecks, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment) to identify a useful first action.

### Speed to Quote: What an AI Quote-Preparation System Actually Needs

- URL: https://returnmytime.com/blog/ai-speed-to-quote-system
- Author: Corey Ganim
- Published: 2026-07-29T09:00:00-04:00
- Category: REVENUE SYSTEMS
- Tags: SPEED TO QUOTE, REVENUE OPERATIONS, MANAGED AI

An AI speed-to-quote system prepares a complete, reviewable quote from approved company information. It collects the facts needed to price the work, applies documented rules, drafts the quote, routes exceptions to the right person, and records what happened. It should not invent pricing or quietly turn a draft into a customer commitment.

That distinction matters because quote delay is rarely a document-generation problem. The visible output may be a PDF or email, but the real work happens earlier: clarifying scope, finding the current price rule, checking similar jobs, recognizing an exception, and locating the person allowed to approve it.

If those decisions live across an inbox, a spreadsheet, old proposals, and one experienced employee's memory, adding AI does not remove the uncertainty. It makes the uncertainty move faster. A dependable system starts by making the quoting process and its knowledge explicit.

## 1. What an AI speed-to-quote system does

Speed to Quote is best understood as a controlled preparation function. It turns a qualified request into an approval-ready quote package.

The function may:

- gather required customer, site, product, and service details;
- identify missing information before pricing begins;
- retrieve current price rules and approved scope language;
- find relevant historical quotes as context;
- calculate or assemble a draft using documented logic;
- explain which sources and assumptions shaped the draft;
- route exceptions or high-value work to a named reviewer;
- create the approved CRM, proposal, and follow-up records.

The [proposal automation guide](/blog/proposal-automation) explains how approved scope, pricing, and terms become a reviewable proposal.

The function should not automatically inherit every action the existing process contains. Sending the quote, changing a discount, promising a date, or modifying the customer record may require separate approval.

Corey Ganim's [Build With AI episode about revenue workflows](https://www.youtube.com/watch?v=tVMHEr2OtY8) contrasts visible front-end improvements with operating workflows that connect inquiry, quote, and follow-up. The useful buyer lesson is not that every company needs the same automation. It is that a workflow should reach a meaningful business handoff and have an accountable owner.

Return My Time does not currently have a measured Speed to Quote client engagement. This article describes the operating requirements a buyer should expect, not a promise about turnaround time, revenue, or conversion.

## 2. Start with a complete intake

A fast draft built from incomplete input only creates a faster revision cycle. Define the minimum information required before the system may prepare a quote.

For a professional service, that might include the requested outcome, deliverables, locations, timeline, stakeholders, constraints, and known exclusions. For a field-service business, it might include property type, measurements, equipment, access conditions, photos, service area, and scheduling requirements.

Separate intake fields into three groups:

| Input class | Purpose | System behavior when missing |
| --- | --- | --- |
| Required | Necessary to determine scope or price | Pause and request the information |
| Conditional | Required only for a defined service, risk, or exception | Ask when the triggering condition is present |
| Helpful | Improves context but does not block preparation | Continue and label the omission |

The intake contract also needs source and ownership rules. A website form may start the process, but an existing account record, signed agreement, or site inspection may override what the requester entered. Define which source wins before two records disagree.

The system should normalize obvious variations without erasing the original input. It can format phone numbers, map a service name to the approved catalog, or identify a likely duplicate. It should preserve the submitted facts and surface uncertain matches rather than silently treating them as confirmed.

## 3. Give the system approved pricing knowledge

Quote preparation requires company-specific knowledge. A generic model does not know which price book is current, when travel applies, which services may be bundled, who can approve a discount, or which customer has contract pricing.

The pricing layer should identify:

- the current price source and its owner;
- effective dates and superseded versions;
- scope-to-price rules;
- approved units, rates, minimums, and add-ons;
- customer or contract exceptions;
- tax, travel, rush, and payment-term rules;
- discount and approval thresholds;
- standard inclusions, exclusions, and assumptions.

Historical quotes can provide useful context, but they are not automatically policy. An old proposal may contain a one-time exception, an obsolete rate, or a mistake. Label historical material by date, result, service type, and authority. Use it to compare and explain—not to overrule current policy.

The [company knowledge-base guide](/blog/company-knowledge-base-for-ai) explains how to distinguish authoritative sources from examples and working material. That hierarchy is essential in quoting. A signed customer agreement may outrank the general price book for one account. The current price book should outrank an old won quote for everyone else.

Every retrieved rule should be traceable. A reviewer should be able to answer, “Why did the draft use this rate?” without reverse-engineering a model's response.

## 4. Separate quote preparation from quote approval

Preparing a draft and committing the company are different authorities.

A useful approval packet contains:

- the customer's request and normalized scope;
- missing or uncertain inputs;
- the proposed line items and calculations;
- the current sources used;
- historical comparisons, clearly labeled as context;
- assumptions and exclusions;
- requested exceptions;
- the next action the reviewer is approving.

Approval rules should reflect consequence, not enthusiasm for automation. A standard renewal using current contract terms may need a lighter review than a new service, unusual scope, nonstandard discount, or accelerated delivery promise.

Define a named owner and backup for each approval path. “Send to sales” is not operational. The workflow needs a role, queue, response expectation, escalation route, and safe behavior when nobody responds.

The safest default is no external commitment without the required approval. Draft mode lets the team test completeness, accuracy, and reviewer usability before granting any sending authority.

## 5. Define exceptions before automating the normal path

The quality of the system is often revealed by what it refuses to do.

Common quote exceptions include:

- a required input is missing or contradictory;
- the requested service is outside the catalog or service area;
- the customer has overlapping records or agreements;
- the price source is stale or unavailable;
- the work requires an unfamiliar subcontractor or material;
- the requested discount exceeds authority;
- delivery timing depends on unavailable capacity;
- the request contains a legal, safety, or regulatory concern;
- the draft differs materially from comparable approved work.

For each exception, specify whether the system should ask the customer, create an internal task, route to an estimator, or stop. Also specify the evidence included in the handoff.

Avoid a single “needs review” bucket. A missing measurement belongs with the intake owner. A contract interpretation belongs with the account or legal owner. An unusual technical scope belongs with delivery. Routing everything to one senior person recreates the original bottleneck.

The [AI assessment guide](/blog/ai-assessment-before-build) can help map these decisions before implementation. If the team cannot explain the exception path on paper, an automated version will not be easier to operate.

## 6. Deliver, record, and reconcile the quote

Approval is not the end of the workflow. The system needs a clear completion definition.

Depending on the business, completion may require:

1. generating the approved proposal or quote;
2. sending it through the approved channel;
3. recording the final version and approver;
4. updating the CRM stage and next action;
5. scheduling follow-up;
6. confirming delivery or surfacing a delivery failure;
7. linking later revisions to the original quote.

Make repeated execution safe. A retry after a tool failure should not create duplicate opportunities, send two proposals, or overwrite a later human edit. Use stable record identifiers, idempotent actions where possible, and explicit version history.

Reconciliation catches the gap between “the automation ran” and “the business record is correct.” A daily or event-driven check can compare prepared quotes, approvals, sent messages, CRM status, and unresolved exceptions. Any mismatch should become visible work with an owner.

## 7. Measure the workflow without inventing ROI

Measure the system against its function and baseline. Do not begin with a borrowed industry statistic or an assumed conversion lift.

Useful operating measures include:

| Measure | Question it answers |
| --- | --- |
| Intake completeness | How often can preparation begin without another request? |
| Time awaiting information | Is delay caused by the system or missing customer/team input? |
| Draft preparation time | How long does controlled preparation take? |
| Approval wait time | Where does human review become the bottleneck? |
| Exception rate by reason | Which rules or intake fields need improvement? |
| Revision rate | How often does the approved draft require material correction? |
| Record reconciliation failures | Did the proposal, message, and CRM state stay aligned? |

Business outcomes can be measured later against the company's own baseline. Quote acceptance, follow-up completion, margin, and sales-cycle movement may matter, but they are influenced by price, demand, capacity, positioning, and sales behavior—not only preparation speed.

## 8. Speed-to-quote readiness checklist

- [ ] One quote type and its completion condition are clearly defined.
- [ ] Required, conditional, and helpful intake fields are separated.
- [ ] Current pricing and scope sources have owners and effective dates.
- [ ] Historical quotes are labeled as context, not universal policy.
- [ ] Discount, exception, and external-commitment approvals are explicit.
- [ ] Every exception routes to a named role with useful evidence.
- [ ] Draft, approval, sending, CRM update, and follow-up states can be reconciled.
- [ ] Duplicate and retry behavior has been tested.
- [ ] A manual path can operate while the system is paused.
- [ ] Baseline measures exist before outcome claims are made.

A speed-to-quote system is ready to build when the team can describe how a request becomes an approved commitment and can show where every important rule comes from. Use the [quote-intake checklist prompt](/prompt-library/build-quote-intake-checklist) to draft the required fields before building. If that map is incomplete, begin with assessment and knowledge work rather than automating around the gaps.

If slow or inconsistent quoting is an owner-felt problem, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment). We’ll discuss the repeated work and recommend one practical first move.

### What Belongs in a Company Knowledge Base for AI?

- URL: https://returnmytime.com/blog/company-knowledge-base-for-ai
- Author: Corey Ganim
- Published: 2026-07-28T19:00:00-04:00
- Category: KNOWLEDGE SYSTEMS
- Tags: KNOWLEDGE BASE, AI READINESS, OPERATIONS

A company knowledge base for AI should be one structured source of truth for the facts, rules, context, and decisions that people and software need to do reliable work.

It is not a folder dump. Moving every document into one drive does not resolve conflicting policies, stale pricing, unclear ownership, or sensitive information. Connecting a chatbot to that folder only makes those gaps easier to retrieve.

A useful first version is narrower. It identifies the knowledge required for a specific set of business decisions, preserves the source behind each important fact, limits access, and gives every durable record an owner and a maintenance rule.

If you have not yet chosen the operating problem this foundation should support, begin with the [AI project readiness scorecard](/blog/stop-wasting-time-wrong-problems). The knowledge system should follow the business need—not become an expensive filing project without a decision to improve.

## 1. Build a source of truth, not a folder dump

The goal is not to collect the most files. The goal is to help an authorized person or system find the current answer, understand why it is authoritative, and know what to do when the answer is missing or disputed.

For each important piece of knowledge, the system should answer:

- What business question does this record support?
- Where did the information come from?
- Is it a governing source, supporting context, an example, or an unverified note?
- Who owns its accuracy and access?
- When did it become effective?
- What event or date triggers review?
- Which earlier record does it replace?
- Who may retrieve it, and for which purpose?

A folder structure can help people browse, but folders alone rarely express authority or conflict rules. Two documents can both look current while prescribing different prices or handoffs. The knowledge base needs metadata and operating rules that distinguish the approved policy from the old proposal that should remain only as history.

The first version can be simple. For a small, well-bounded corpus, structured files with [a clear index and ownership fields](/blog/knowledge-management-small-business) may be enough. Larger collections or more complex retrieval needs may require search, a vector index, record-level attributes, or connections to existing systems. The architecture should fit the corpus and the decisions it supports.

[Anthropic’s guidance on building effective agents](https://www.anthropic.com/engineering/building-effective-agents) recommends starting with the simplest solution that meets the need and adding complexity only when it produces a measurable improvement. That principle applies to the knowledge layer too. A sophisticated retrieval system cannot compensate for sources nobody has approved.

## 2. Include the knowledge behind recurring decisions

Start with the decisions and workflows that matter to the business, then inventory the knowledge required to support them.

| Knowledge type | What belongs in it | Common boundary |
| --- | --- | --- |
| Policies | Approved rules, eligibility, service boundaries, escalation conditions | Separate current policy from drafts and superseded versions |
| Pricing | Current offers, pricing logic, inclusions, exclusions, discount authority | Do not treat an old proposal as general pricing |
| Process | Triggers, steps, owners, handoffs, exceptions, completion criteria | Distinguish the intended process from what happened once |
| Customer context | Agreements, preferences, decisions, commitments, relevant history | Limit each user or agent to the accounts it is authorized to serve |
| Company history | Material changes, previous approaches, lessons, and rationale | Preserve useful context without making obsolete practice active instruction |
| Decisions | What was decided, by whom, when, why, and what it replaced | Record rationale and unresolved assumptions, not only the final answer |
| Responsibilities | Current roles, approval authority, backups, and escalation contacts | Use named owners behind roles and review when responsibilities change |
| Open work | Active commitments, dependencies, risks, and next decisions | Retire or archive completed context so it does not appear current |

Not every source belongs in the authoritative layer. Call transcripts, emails, meeting notes, and draft proposals can provide evidence or history, but they may contain mistakes, unapproved ideas, or information that should not be broadly available. Label them as source material until a responsible owner promotes the relevant knowledge into an approved record.

Examples also need boundaries. An approved proposal can demonstrate structure or tone without making its customer facts, pricing, or commitments reusable elsewhere.

This is why company knowledge should be organized before an agent is asked to apply it. The system needs to know which record governs the decision, not merely which passage is semantically similar to the question.

## 3. Assign authority, provenance, ownership, and freshness

Every durable record needs four controls:

1. **Authority:** its place in the decision hierarchy.
2. **Provenance:** the original source and the path from source to structured record.
3. **Ownership:** the person accountable for accuracy, access, and corrections.
4. **Freshness:** the condition that keeps the record current or sends it for review.

Authority rules should be specific enough to resolve predictable conflicts. A signed customer agreement may outrank the standard delivery policy for that customer. The current service catalog may outrank an old proposal for general sales questions. A finance-system record may govern payment status while a CRM note remains supporting context.

Provenance makes review possible. Preserve the source link or record identifier, effective date, last review, and the person or process that approved the structured version. When an answer affects a customer, price, policy, or material business decision, the reader should be able to trace it back to the governing source.

Ownership belongs with the party able to decide what is true. A provider can organize sources, build retrieval, [monitor freshness signals](/blog/ai-knowledge-base-maintenance), and route conflicts. It should not silently decide company policy. The client still needs named owners for pricing, customer commitments, process rules, security, and access authorization.

Freshness is more than a recurring calendar reminder. Some records should be reviewed every quarter. Others change when a service launches, a contract is signed, a regulation changes, a system field is renamed, or an owner leaves. Record the relevant trigger and define what happens when review is overdue. For high-consequence knowledge, the safe response may be to stop, show the stale-source warning, and escalate rather than answer confidently.

## 4. Structure retrieval for people and agents

People and agents need the same truth, but they may navigate it differently. A person benefits from a readable index, clear headings, linked source material, and enough history to understand a decision. An agent needs reliable retrieval fields, narrow scopes, explicit authority, and instructions for uncertainty.

A practical record may include:

- title and knowledge type;
- concise approved statement;
- governing source and supporting sources;
- owner and approver;
- effective, reviewed, and review-due dates;
- customer, service, team, or workflow scope;
- authority level and conflict rule;
- sensitivity and allowed audiences;
- related records and superseded records;
- keywords or attributes used for retrieval;
- required behavior when the record is missing, stale, or disputed.

Use plain language in the record itself. Metadata should make the knowledge filterable without hiding meaning in an implementation only a developer can inspect.

[OpenAI’s retrieval documentation](https://developers.openai.com/api/docs/guides/retrieval) describes files, vector stores, attributes, filtering, and file update or removal operations. Those capabilities can help locate relevant material, but the company still has to decide which files belong in the index, which attributes control retrieval, and how changed or retired sources are handled.

[Microsoft’s Copilot Studio knowledge-source documentation](https://learn.microsoft.com/en-us/microsoft-copilot-studio/knowledge-copilot-studio) similarly distinguishes knowledge sources and documents behavior around authentication, permissions, grounding, and citations. Product features vary by source and configuration, so evaluate them against the access and evidence requirements of the workflow rather than assuming that a connector creates governance.

Test retrieval with real business questions:

- Does the result use the governing policy instead of the most recent upload?
- Can it distinguish a customer exception from a company-wide rule?
- Does it cite enough source context for a reviewer to verify the answer?
- Does it refuse or escalate when the permitted corpus does not contain an answer?
- Can an authorized person find the same governing record without using the agent?

The knowledge base should improve human operations before autonomy is considered. If only the agent can interpret the system, the company has created a new dependency rather than a durable source of truth.

## 5. Define access, retention, and sensitive-information boundaries

Centralizing knowledge increases its usefulness and its risk. Do not treat “the company already has this information” as permission to put every record into one searchable corpus.

Classify sources before ingestion:

- public or approved for broad internal use;
- internal operational information;
- customer-confidential or contract-restricted information;
- employee, financial, legal, security, or regulated information;
- prohibited material that should not enter the proposed system.

Then apply least privilege. A sales-support workflow may need current service policies and assigned-account context, not every customer agreement. A delivery agent may need an approved statement of work, not access to company banking or unrelated employee records.

Document:

- permitted users, roles, agents, and service accounts;
- purpose and scope of access;
- authentication and integration permissions;
- whether data may be copied into an index or must remain in the source system;
- retention periods for source material, indexed content, outputs, and logs;
- deletion and offboarding procedures;
- rules for exports, backups, and third-party processors;
- incident contacts and access-review cadence.

Sensitive information should not rely on a warning inside a prompt. Use source separation, permissions, filtered retrieval, limited credentials, approval steps, and logging appropriate to the consequence of exposure. Test access boundaries directly, including attempts to retrieve data across customers, roles, or workspaces.

## 6. Operate an ingest-and-curate loop

A knowledge base is an operating responsibility, not a one-time import.

New information enters through controlled sources: approved forms, meeting decisions, policy updates, contracts, CRM records, or a reviewed inbox. Ingestion should preserve the original source and mark incoming material as unverified until the appropriate owner approves its durable meaning.

A workable lifecycle is:

1. **Capture:** retain the source, date, scope, and submitting owner.
2. **Classify:** identify knowledge type, sensitivity, authority candidate, and affected workflows.
3. **Review:** confirm accuracy, resolve ambiguity, and obtain approval from the responsible owner.
4. **Publish:** update the structured record, index, relationships, and retrieval attributes.
5. **Test:** run representative questions and affected workflow checks.
6. **Monitor:** surface stale records, retrieval failures, conflicts, and missing ownership.
7. **Retire:** mark superseded knowledge, remove it from active retrieval, and preserve history only where required.

Conflict resolution needs a visible queue. When two credible sources disagree, do not let upload order decide. Pause affected answers or actions, show both sources to the named owner, record the decision, and update the authority rule if the conflict can recur.

Use a review cadence based on change rate and consequence. Frequently changing pricing or active customer commitments may need event-driven updates and tight service levels. Durable company history may need review only when a linked decision changes. Disposable working context should expire rather than accumulate indefinitely.

The [source Build With AI episode](https://www.youtube.com/watch?v=fI68tU0L86s) demonstrates a simple file-based starting point, a structured index, and ingest-and-curate routines. It is useful implementation context, not proof that one taxonomy or plain-text architecture fits every organization. You can find the episode and related conversations in the [Build With AI podcast archive](/podcast).

## 7. Use this readiness checklist

Before connecting a company knowledge base to an AI workflow, confirm:

- [ ] The first business decisions and workflows are named.
- [ ] Governing sources are distinguished from supporting material and examples.
- [ ] Policies, pricing, process, customer context, history, decisions, responsibilities, and open work have explicit scope.
- [ ] Each durable record has a source, authority level, owner, and freshness rule.
- [ ] Predictable source conflicts have resolution rules.
- [ ] People can browse and verify the governing knowledge without an agent.
- [ ] Retrieval metadata supports the required customer, service, role, and workflow boundaries.
- [ ] Missing, stale, or disputed knowledge triggers a defined refusal or escalation.
- [ ] Sensitive information is classified before ingestion.
- [ ] Access, retention, deletion, offboarding, and incident responsibilities are documented.
- [ ] New material follows capture, review, approval, publication, and testing steps.
- [ ] Superseded and disposable context leaves active retrieval.
- [ ] Named client owners and provider maintenance duties are written down.

If several boxes remain open, the next step is discovery and source cleanup—not a wider integration. The [workflow-first implementation guide](/blog/10-hour-week-ai-tools) can help sequence a contained first release after the operating foundation is clear.

## 8. Keep the evidence boundary visible

This article describes an intended architecture and buyer standard. “One source of truth,” AI-ready structure, and ongoing maintenance are Return My Time service-design goals. Return My Time does not currently have a measured knowledge-base client case study proving revenue gained, time saved, conversion improved, token costs reduced, or another business outcome.

The source episode includes commercial and financial claims from a guest. Those claims are context for the conversation, not validated buyer outcomes, and they are not used as benchmarks here. The episode’s fictional demonstration also does not establish that the same taxonomy, technology, or result will fit another company.

A credible provider should be able to explain:

1. Which business decisions the first knowledge base will support.
2. Which sources will be authoritative and who approves them.
3. How permissions, sensitive information, and customer boundaries will work.
4. How people and agents will retrieve and verify the same governing knowledge.
5. Who handles ingestion, conflicts, stale records, incidents, and retirement.
6. Which assumptions will be tested before the system supports consequential actions.

[Return My Time’s free AI Opportunity Assessment](/services/free-ai-opportunity-assessment) considers the knowledge behind one recurring bottleneck when that context improves the recommendation. If your company’s operating knowledge is scattered across people, inboxes, calls, systems, and documents, [request the free assessment](/services/free-ai-opportunity-assessment). We’ll identify one practical next move.

### What an AI Assessment Should Determine Before Anything Gets Built

- URL: https://returnmytime.com/blog/ai-assessment-before-build
- Author: Corey Ganim
- Published: 2026-07-28T18:45:00-04:00
- Category: KNOWLEDGE SYSTEMS
- Tags: AI READINESS, ASSESSMENT, IMPLEMENTATION

An AI readiness assessment should decide what problem deserves investment, what must change before implementation, and who will own the result.

It should not end with a maturity score, a list of tools, or a sales proposal disguised as analysis. A useful assessment produces a portable blueprint: a map of the current workflow, an inventory of the required data and access, a ranked set of use cases, a knowledge-base schema, and a staged build sequence with owners and guardrails.

The buyer should be able to use that blueprint whether the next step is do it yourself, hire another provider, or decide not to build.

If your team is still sorting attractive ideas from implementation-ready problems, begin with the [AI project readiness scorecard](/blog/stop-wasting-time-wrong-problems).

## 1. Decide what should happen next

The assessment exists to support a business decision. Before discovery begins, name that decision in plain language:

- Which recurring workflow should be addressed first?
- Does the workflow need process redesign, deterministic automation, an AI-assisted step, or no new system?
- Which prerequisites must be resolved before implementation?
- What evidence would justify moving forward?
- Who owns the business rules, the implementation, and the ongoing operation?
- What is the smallest safe first release?

The answer may be “not yet.” That is a valid result when the source information is unreliable, the process changes every week, access cannot be controlled, or nobody owns the outcome.

[Microsoft’s AI Readiness Assessment](https://learn.microsoft.com/en-us/assessments/94f1c697-9ba7-4d47-ad83-7c6bd94b1505/) examines strategy, governance and security, data, organizational capability, infrastructure, and model management. Those organizational dimensions matter, but a service-business assessment should also connect them to one concrete operating decision. Readiness is not an abstract level. It is readiness for a specific workflow, under specific conditions, with a specific owner.

## 2. Map the workflow and the revenue leak

Start with observed work, not a preferred product. Interview the people who perform and receive the work. Review a representative sample of records. Follow the process from trigger to completion, including the side channels and exceptions that a clean process diagram often misses.

For each step, capture:

- the trigger, input, decision, output, and handoff;
- the person and system currently responsible;
- wait time, rework, duplication, and common failure modes;
- customer or revenue consequences when the step fails;
- exceptions that require judgment;
- controls or approvals already in place.

A “revenue leak” should be documented as an observable break in the operating system: qualified inquiries waiting without an owner, proposals built from inconsistent pricing, renewal signals not reaching the account lead, or completed work failing to trigger billing. Do not invent dollars recovered. Record the volume, frequency, affected records, and business consequence that the company can verify.

The assessment should also challenge the current process. In the [source Build With AI episode](https://www.youtube.com/watch?v=9nFFIash0YM), an 18-step workflow is simplified before automation is considered. That is the right sequence. Automating avoidable steps preserves waste and makes future maintenance harder.

## 3. Inventory data sources, access, and owners

Every proposed use case depends on information and permissions. The assessment should identify both before recommending a build.

For each source, record:

| Field | What to document |
| --- | --- |
| Source | System, folder, database, inbox, or person |
| Purpose | The decision or output this source supports |
| Authority | Whether it is canonical, supporting, or only an example |
| Owner | Person responsible for accuracy and access |
| Freshness | How often it changes and when it becomes stale |
| Quality | Missing fields, duplicates, conflicts, and known gaps |
| Sensitivity | Customer, employee, financial, contractual, or regulated information |
| Access | Required account, role, scope, and approval |
| Retention | How long inputs, outputs, and logs may be kept |

Then test the proposed path. Can the system use a limited account? Can it start read-only or in draft mode? Does the integration expose only the records required for this function? What happens when credentials expire, fields change, or a source is unavailable?

“The information is in Google Drive” is not an inventory. A build needs named sources, authority rules, permissions, and responsible owners. If the source-of-truth problem is larger than the initial workflow, the assessment should define the first knowledge cleanup rather than hide the gap inside a prompt.

## 4. Rank use cases by evidence, feasibility, and risk

An effort-versus-impact matrix is useful, but it is only a first pass. Rank each candidate across a fuller decision set:

| Factor | Question |
| --- | --- |
| Business consequence | Which revenue, reliability, knowledge, or capacity problem changes? |
| Evidence quality | Which observed records, volumes, failures, or stakeholder reports support the problem? |
| Frequency | How often does the workflow occur? |
| Process readiness | Is the current process understood and stable enough to improve? |
| Data readiness | Are the required sources authoritative, current, and permitted? |
| Technical feasibility | Can the required systems support the inputs, actions, and monitoring? |
| Risk | What can go wrong, and how costly or reversible is the result? |
| Ownership | Who approves, operates, maintains, and measures the workflow? |
| Learning value | Will a small release resolve an important uncertainty? |

Record the evidence beside every score. A polished recommendation generated from one interview is still a hypothesis. Tool-directory listings are not proof of security, reliability, integration depth, or fit.

[OpenAI’s guide to identifying and scaling AI use cases](https://openai.com/business/guides-and-resources/identifying-and-scaling-ai-use-cases/) recommends connecting use cases to business priorities and validating them through practical experimentation. For the buyer, that means the ranked list should show why one workflow comes first, what uncertainty remains, and what the first release is intended to learn.

## 5. Define the knowledge-base schema

The assessment should specify what the proposed system needs to know and how that knowledge will be governed. This is a schema, not a request to upload every company file.

For the selected workflow, define:

- knowledge types, such as policies, services, pricing rules, procedures, customer context, exceptions, and approved examples;
- required metadata, including source, owner, effective date, sensitivity, and review date;
- authority rules for conflicting sources;
- retrieval boundaries for different roles, customers, or workspaces;
- correction and approval workflow;
- freshness and retirement rules;
- citations or traceability required in the output.

The schema should serve people as well as software. A new employee should be able to understand which source governs a decision. A reviewer should be able to trace an output back to approved material. An owner should know when a policy change requires an update.

If the organization needs to choose a small, low-risk tool before designing a wider system, the [minimal AI tool decision guide](/blog/beginner-ai-tool-stack) offers a narrower starting point. The assessment should not inflate a contained problem into an unnecessary platform.

## 6. Sequence the build, ownership, and guardrails

The blueprint should translate the selected use case into stages with an acceptance decision at each boundary.

1. **Prepare:** simplify the workflow, clean the initial sources, confirm owners, and write acceptance criteria.
2. **Prototype:** use representative but controlled inputs to test the core decision or output.
3. **Pilot:** run with limited users, permissions, records, and consequences; keep human review in the loop.
4. **Operate:** add monitoring, incident response, reconciliation, fallback, change control, and an ongoing review cadence.
5. **Expand or stop:** use observed evidence to decide whether to broaden scope, revise the design, or retire it.

Each stage should name:

- the client business owner;
- the implementation owner;
- required human approvals;
- allowed and prohibited actions;
- success and stop criteria;
- logs and measurements;
- manual fallback;
- incident, rollback, and reconciliation responsibility;
- maintenance triggers when policies, data, integrations, or models change.

[NIST’s AI Risk Management Framework Core](https://airc.nist.gov/airmf-resources/airmf/5-sec-core/) frames lifecycle work as Govern, Map, Measure, and Manage. The [NIST AI RMF Playbook](https://airc.nist.gov/airmf-resources/playbook/) provides suggested actions for applying those functions. An assessment does not need to turn a small workflow into an enterprise compliance exercise, but it should make governance, context, measurement, and ongoing risk response visible before launch.

For a practical staged rollout after the decision is made, use the [workflow-first implementation sequence](/blog/10-hour-week-ai-tools).

## 7. Keep a portable blueprint

The assessment deliverable should belong to the buyer. At minimum, keep:

- the current-state workflow and failure map;
- source-data, access, sensitivity, and owner inventory;
- ranked use cases with scoring criteria and supporting evidence;
- the proposed knowledge-base schema;
- the target workflow and implementation stages;
- acceptance, stop, and measurement criteria;
- approval, escalation, incident, fallback, and maintenance responsibilities;
- open questions, dependencies, and declined alternatives;
- estimated software and implementation inputs with assumptions clearly labeled.

The blueprint should separate facts, decisions, assumptions, and recommendations. It should also explain which parts will expire. Integration capabilities, product pricing, policy requirements, and source freshness may all change.

This prevents the assessment from becoming a presentation that only makes sense when its author is in the room. Another qualified team should be able to understand the proposed system, challenge its assumptions, and continue the work.

## 8. Ask an implementation partner these questions

Before approving a build, ask:

1. Which business decision will this assessment help us make?
2. How will you observe the real workflow instead of relying only on stakeholder memory?
3. What evidence supports the problem ranking and any expected impact?
4. Which process steps should be removed or redesigned before automation?
5. How will you identify authoritative data and resolve conflicting sources?
6. Which accounts, permissions, and sensitive information would the system require?
7. What can begin read-only, in draft mode, or behind human approval?
8. Who owns policy, implementation, monitoring, incidents, maintenance, and change approval?
9. What acceptance and stop criteria will be written before the pilot?
10. What will we receive if we choose another provider or decide not to implement?
11. Which assumptions need validation during a prototype or pilot?
12. How will outcome claims be measured against a documented baseline?

Return My Time does not currently have a measured client case study proving outcomes from its AI readiness assessment. The framework above describes the intended assessment blueprint and a buyer standard; it is not proof of revenue gained, time saved, conversion improved, or margin increased. Seller income, pricing, conversion, and financial-impact claims discussed in the source episode are not buyer outcomes and are not used here.

You can browse the [Build With AI podcast archive](/podcast) for more implementation conversations. If you want one practical recommendation for your own workflow, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment). A certified Return My Time team member will reach out to schedule the conversation.

### Managed AI Agents: What They Own, What They Need, and Who Maintains Them

- URL: https://returnmytime.com/blog/managed-ai-agent-ownership
- Author: Corey Ganim
- Published: 2026-07-28T18:30:00-04:00
- Category: MANAGED AI
- Tags: AI AGENTS, OPERATIONS, GOVERNANCE

A managed AI agent should own one bounded business function—not “the business.”

That function might be preparing a lead for human follow-up, assembling a draft briefing from approved sources, or checking whether a delivery handoff is complete. The agent may make limited decisions inside that function, but its operating contract must define what starts the work, what information it may use, what it may decide, what it must produce, and when it must stop.

“Managed” describes the ownership around the agent. Someone still has to approve the business rules, maintain the company context, monitor the system, handle incidents, and decide when the workflow changes.

If you are still choosing the workflow, use the [AI project readiness scorecard](/blog/stop-wasting-time-wrong-problems) before evaluating an agent.

## 1. Define the function before choosing the agent

A useful scope answers six questions:

1. **Inputs:** What event starts the work, and which information is required?
2. **Decisions:** Which choices may the agent make, and under what rules?
3. **Outputs:** What record, draft, recommendation, or handoff should it produce?
4. **Completion:** What observable condition means the work is done correctly?
5. **Non-goals:** What related work is explicitly outside the agent’s authority?
6. **Escalation:** Which conditions send the work to a named person?

“Handle new leads” is not bounded. A clearer scope would be:

> Review each new website inquiry against the approved service-area and qualification rules, create a complete internal summary, and route uncertain or high-value exceptions to the sales owner. Do not quote a price, promise availability, or send a customer message without the required approval.

That statement gives the team something to test. It also prevents a narrow workflow from quietly expanding into unrestricted authority.

[OpenAI’s practical guide to building agents](https://openai.com/business/guides-and-resources/a-practical-guide-to-building-ai-agents/) distinguishes agent-suitable work—where nuanced decisions, difficult rules, or unstructured information matter—from work a deterministic solution can handle. [Anthropic’s guide to building effective agents](https://www.anthropic.com/engineering/building-effective-agents) similarly recommends starting with the simplest design that meets the need. The decision is not “AI or no AI.” It is the least complex operating model that can perform the function reliably.

## 2. Give the agent current, authoritative company knowledge

An agent cannot apply company judgment from a generic prompt. It needs approved context for the function it owns, including:

- policies and service boundaries;
- pricing logic and the conditions around it;
- relevant customer or account context;
- known process exceptions;
- the history needed to understand the current situation;
- examples of acceptable and unacceptable outputs.

More information is not always better. Conflicting documents, stale instructions, and unlabeled examples make the system less dependable. Every source should have an authority level, an owner, and a freshness rule.

For example, the current service policy may outrank an old proposal. A signed customer agreement may outrank a general delivery checklist. An example can show tone or format without becoming permission to reuse its facts.

Before launch, document:

- where each source comes from;
- which source wins when two sources conflict;
- who approves corrections;
- how quickly a material change must reach the agent;
- what the agent should do when required context is missing or uncertain.

The client owns the truth of its policies, customer commitments, and exceptions. A managed provider can organize, connect, test, and maintain those sources, but should not invent business policy.

## 3. Limit access to what the function requires

Access should follow least privilege: the agent receives only the accounts, data, and actions needed for its bounded function.

That may include credentials, account permissions, integration scopes, and a dedicated identity when separation improves control or accountability. A dedicated identity is not automatically necessary, and it is not a substitute for narrow permissions, credential rotation, or access review.

Start with read-only or draft-producing access where possible. Add write actions only after the team has tested the normal path, exceptions, duplicate handling, and failure recovery. Prefer actions that can be reversed. Log who or what initiated each consequential change, the information used, the result, and any approval attached to it.

Do not grant unrestricted payment, account-administration, deletion, or customer-communication authority simply because an integration supports it. High-consequence access requires stronger approval, spending or action limits, monitoring, and a manual fallback.

If the work is a stable trigger followed by a predictable action, the [small-business automation guide](/blog/zapier-automations-small-business) may point to a simpler approach. A managed agent is more appropriate when the function requires contextual judgment or exception handling that cannot be expressed as fixed rules alone.

## 4. Put human approvals and escalation paths in writing

A person should approve actions that create a high-cost, hard-to-reverse, or externally binding result. Common examples include:

- spending money or changing a financial record;
- making a customer promise about price, scope, timing, or eligibility;
- approving an exception to policy;
- deleting records or changing access;
- publishing or sending sensitive external communication;
- continuing when required information is missing or contradictory.

An approval step needs more than “ask a human.” Name the role, show the information the reviewer receives, define the response time the workflow can tolerate, and specify what happens if nobody responds.

Escalation rules should also include stop conditions. The agent should pause when confidence is inadequate, a source is stale, permissions change, a tool is unavailable, volumes depart from expected ranges, repeated errors appear, or the requested action falls outside the defined function.

The business owner retains authority over policy and customer commitments. The provider owns the technical route to the reviewer and the reliability of the escalation mechanism. The agent owns only the act of recognizing and routing conditions included in its approved instructions.

## 5. Treat launch as the start of ownership

An agent is not managed because someone configured it once. It is managed when the operating team can see whether it is working, respond when it is not, and update it without losing control.

Start with measurable acceptance criteria. Depending on the function, they may include:

- every test input creates exactly one expected output;
- required fields are complete and traceable to approved sources;
- prohibited claims and actions never pass the approval boundary;
- exceptions reach the named owner with enough context to decide;
- failures are visible within an agreed interval;
- the manual fallback can keep the function operating;
- a rollback returns the workflow to the last accepted version.

Observability should show outcomes, not only whether software is running. Track successful completions, exceptions, approval waits, tool failures, rejected outputs, and unresolved work. Incident response should name who investigates, who can pause the agent, how affected records are reconciled, how the business operates during the interruption, and who authorizes restart.

Changes need the same discipline. A revised policy, new service, integration update, or customer exception may require source updates, new tests, permission review, or different escalation rules. Use a review cadence based on how quickly the function changes and how costly a failure would be—not an arbitrary claim that the system can run unattended.

[NIST’s AI Risk Management Framework Core](https://airc.nist.gov/airmf-resources/airmf/5-sec-core/) organizes ongoing work around Govern, Map, Measure, and Manage. For a service operator, that becomes a practical owner map:

| Role/system | Operational responsibility |
| --- | --- |
| Agent | Execute the approved function, log its work, and escalate defined exceptions |
| Managed provider | Configuration, integrations, tests, [technical monitoring and maintenance](/blog/managed-ai-agent-testing-monitoring), technical incident response, rollback execution, and approved updates |
| Client owner | Business policy, source authority, customer commitments, exceptions, approvals, access authorization, business continuity, affected-record decisions, and acceptance of rollback and restart |

Names should sit behind those roles. “The provider” and “the client” are not useful incident contacts without a responsible person and backup.

## Managed agent vs. automation vs. virtual assistant

| Decision factor | Managed AI agent | Deterministic automation | Virtual assistant |
| --- | --- | --- | --- |
| Contextual judgment | Applies approved context to bounded decisions and escalates uncertainty | Follows explicit rules; little or no judgment | Uses human judgment across varied situations |
| Repeatability | Best when the objective repeats but inputs or paths vary | Best when the trigger, rules, and result are stable | Can handle changing work, but consistency depends on training and process |
| Exception rate | Handles defined variation; routes novel or high-risk cases | Works best when exceptions are uncommon and easy to detect | Often suitable when exceptions are frequent or poorly documented |
| Operational ownership | Provider maintains the system; client retains policy and approval ownership | Internal or external owner maintains connections, rules, and failures | Manager owns priorities, training, access, quality, and coverage |
| Best fit | Repeated function requiring context, multiple tools, and controlled decisions | Predictable handoff or calculation with fixed rules | Broad, changing work that benefits from direct human interpretation |

These options can work together. An agent may use deterministic automations for stable handoffs and escalate unusual work to a virtual assistant or internal owner. The goal is not maximum autonomy. It is the clearest division of labor.

## Buyer readiness checklist

Before asking a provider to build or manage an agent, confirm:

- [ ] One business function is named in plain language.
- [ ] Inputs, outputs, completion, non-goals, and escalation conditions are written.
- [ ] The workflow occurs often enough to justify ownership.
- [ ] Authoritative sources and their owners are known.
- [ ] Required data is permitted for the proposed accounts and workflow.
- [ ] Permissions can be limited to the function.
- [ ] High-risk actions have named human approvers.
- [ ] Acceptance criteria cover normal work, exceptions, and prohibited actions.
- [ ] A manual fallback and stop authority exist.
- [ ] Someone inside the company owns policy and customer commitments.
- [ ] The provider’s monitoring, incident, rollback, and update duties are explicit.

If several boxes remain open, discovery should clarify the process before implementation. A half-defined function will not become well managed by adding a more capable model.

## Questions to ask a managed-agent provider

1. What exact function will the agent own, and what remains outside scope?
2. Which decisions can it make without approval?
3. How will you identify authoritative company sources and keep them current?
4. Which accounts, credentials, and permissions are required?
5. Can we begin read-only or in draft mode?
6. Which actions are reversible, logged, limited, and subject to approval?
7. What tests must pass before launch, and who accepts the result?
8. How will we see completions, exceptions, failures, and waiting approvals?
9. Who responds to an incident, and what are the pause, fallback, reconciliation, and restart steps?
10. How are policy changes, integration changes, and new exceptions reviewed?
11. What does the provider maintain, and what must our team continue to own?
12. What evidence supports any outcome claim you make?

## What this article and episode can prove

The [source Build With AI episode](https://www.youtube.com/watch?v=vZTxDqOTCmk) shows examples of agent provisioning, operating environments, reusable templates, and observation interfaces. It also discusses provider responsibility for onboarding, deployment, maintenance, and support. Those demonstrations are useful source context; they do not verify customer identity, commercial outcomes, or complete end-to-end autonomy.

Return My Time does not currently have a measured managed-agent client case study. This framework describes an operating model and buying standard, not a promise of revenue, savings, or another business outcome. You can browse the [full Build With AI podcast archive](/podcast) for more implementation context.

If one business function is important enough to manage but too complex for a simple automation, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment). We’ll recommend one practical first move.

### 5 Zapier Automations for Small Business—and When Simple Stops Working

- URL: https://returnmytime.com/blog/zapier-automations-small-business
- Author: Corey Ganim
- Published: 2025-12-15T10:00:00Z
- Category: TOOLS & QUICK WINS
- Tags: ZAPIER, AUTOMATION, NO-CODE, PRODUCTIVITY

Zapier is useful when a workflow can be expressed as a clear event followed by a predictable action.

That makes it a good fit for moving a form response into a sheet, filing an attachment, or sending an internal reminder. It does not remove the need to define permissions, test data, exceptions, or ownership—especially when the workflow touches leads, payments, or customer communication.

Zapier’s Free plan currently includes unlimited Zaps and assets, 100 tasks per month, one user, two-step workflows, and 15-minute polling. A two-step Zap means one trigger and one action. See [what Zapier includes in Free](https://help.zapier.com/hc/en-us/articles/32337438839565-What-s-included-in-Zapier-s-Free-plan).

Any Zap with more than one action, or with a premium app, requires a trial or paid plan to publish. Zapier explains those upgrade conditions in [Why do I need to upgrade?](https://help.zapier.com/hc/en-us/articles/35201208823181-Why-do-I-need-to-upgrade-to-publish-my-Zap).

Use the five patterns below as designs. App event names, eligibility, and available fields can change, so confirm the current labels and permissions inside your Zap editor.

## Before connecting a production account

An app connection authorizes Zapier to read or write the data required by a workflow. Authorization methods vary by app. Review [Zapier’s app-connections guide](https://help.zapier.com/hc/en-us/articles/36818633398157-App-connections-on-Zapier), then:

- connect the least-privileged account that can do the job;
- review every requested read and write permission;
- use synthetic records for initial tests;
- keep customer and payment data within company policy;
- enable two-factor authentication where available;
- remove unused connections;
- name an owner for failures and changes.

## 1. Form submission → lead sheet

**Simple version:** a new form submission creates one row in a lead sheet.

This can fit the Free plan when the form trigger and sheet action are eligible. Use a synthetic lead to verify that every required field maps once and lands in the right column.

**Where simple stops:** adding a separate email or Slack alert creates a second action. That first article-style example—sheet **plus** notification—requires a trial or paid plan to publish.

For a lead-rich service business, a row in a sheet is not the outcome. The business needs a responsible person to see the inquiry, respond, and handle failures. If response speed affects revenue, define an alert deadline, fallback, and daily reconciliation rather than trusting one successful test.

The [AI project readiness scorecard](/blog/stop-wasting-time-wrong-problems) helps define the owner, acceptance criteria, and risk before expanding the Zap.

## 2. Calendar event → internal reminder

**Simple version:** an upcoming event triggers one Slack reminder with the time, meeting link, and approved preparation note.

Test time zones, all-day events, recurring events, cancellations, and missing meeting links. A reminder is only useful if it arrives early enough and in the channel the owner actually watches.

**Where simple stops:** when the workflow must gather client history, prepare a briefing, update records, and escalate missing information. Those are multiple actions and decisions, not a single reminder.

Use the [four low-risk AI experiments](/blog/four-10-minute-ai-quick-wins) if you want to test manual meeting preparation before automating it.

## 3. Gmail attachment → Google Drive

**Simple version:** a matching Gmail attachment uploads to a designated Drive folder. Zapier currently documents Gmail attachment triggers and Google Drive file actions on its [Gmail and Google Drive integration page](https://zapier.com/apps/gmail/integrations/google-drive).

On Free, polling may take up to 15 minutes. That is fine for routine filing; it may be unacceptable for an urgent operational handoff.

Use a narrow sender or subject filter, a dedicated destination, and synthetic files first. Verify duplicates, multiple attachments, unexpected file types, naming collisions, and access to the destination folder.

**Where simple stops:** contracts, invoices, and client documents may need classification, retention rules, restricted access, or a reconciliation report. “The file moved” is not the same as “the company can reliably find and govern it.”

## 4. Approved social content → Buffer

**Simple version:** a person approves a post in your content calendar, then a Zap hands only that approved content to Buffer for scheduling or publishing through the social accounts connected there.

Zapier’s direct X/Twitter integration remains unavailable. Do not copy the legacy “Search Mention” steps or query syntax into a new workflow. Zapier’s [Twitter integration removal notice](https://help.zapier.com/hc/en-us/articles/43554846358541-App-update-Twitter-integration-removal) points readers to social-management intermediaries such as Buffer or Hootsuite. Zapier’s [Buffer integration guide](https://zapier.com/apps/buffer/integrations) documents workflows that hand content from another tool to Buffer for queueing and publishing.

Keep the human approval before the automated handoff. Confirm that the destination account, network, content, links, media, and timing are correct in Buffer. Do not let an AI draft, database change, or incoming trigger become approval by itself.

**Where simple stops:** channel-specific rules, expiring connections, unsupported media, duplicate posts, or urgent corrections. Monitor the handoff and keep a person responsible for what is published.

## 5. Stripe payment → internal fulfillment handoff

**Simple version:** a completed payment creates one internal fulfillment task or notification.

Zapier currently lists “New Payment” as a Stripe trigger on its [Stripe integration page](https://zapier.com/apps/stripe/integrations). Live-mode Stripe connections use a browser sign-in and permission grant. Secret API keys are for a separate test-mode connection. Zapier’s [Stripe setup guide](https://help.zapier.com/hc/en-us/articles/8496161863693-How-to-get-started-with-Stripe-on-Zapier) also lists an activated paid Stripe account and administrator permission among the prerequisites.

Do not paste a live secret key into a tutorial workflow. Test with synthetic records and the proper test-mode connection.

**Where simple stops:** automatic customer email, entitlement changes, refunds, taxes, or fulfillment across multiple systems. Confirm the actual trigger payload instead of teaching a filter against an assumed status field. Customer-facing messages need approved wording, deduplication, monitoring, and a responsible owner.

## The production checklist

A Zap is ready for business use only when you can answer:

1. What exact event should trigger it?
2. Which account and scopes does it use?
3. What synthetic tests cover the normal path and edge cases?
4. How will the owner see a failure?
5. What happens while the integration is unavailable?
6. How are duplicates and partial updates handled?
7. Who reviews permissions, connections, and results?
8. What acceptance criteria prove the workflow still works?

The [workflow-first AI implementation guide](/blog/10-hour-week-ai-tools) shows how to measure and stage the work before automation.

## Simple automation or managed system?

Keep it simple when the workflow is low-risk, deterministic, easy to reverse, and owned by someone who can monitor it.

Move toward [managed AI automation](/blog/ai-automation-services) when the workflow:

- affects response time or revenue;
- writes customer, payment, or company records;
- spans several tools and decision branches;
- needs rapid exception handling;
- lacks an internal technical owner.

At that point, the value is reliability and accountability, not another Zap. [Return My Time’s managed AI services](/services) help time-poor service operators design, test, monitor, and own those systems without making the owner maintain the tooling.

Start with one trigger and one action. Prove the handoff. Then add complexity only when the business case and owner are clear.

### Four 10-Minute AI Quick Wins to Test Safely

- URL: https://returnmytime.com/blog/four-10-minute-ai-quick-wins
- Author: Corey Ganim
- Published: 2025-12-11T14:00:00Z
- Category: TOOLS & QUICK WINS
- Tags: AI, PRODUCTIVITY, QUICK-WINS, AUTOMATION

You do not need to connect your whole business to AI to learn whether it helps.

Run one contained experiment with low-risk information. Review the output yourself. Keep it only if it improves a real task.

The four trials below can be set up in about 10 minutes each. That is setup time, not a promise that every result will be correct or that a specific number of hours will be saved.

Before starting, use an account approved by your company. Do not upload or paste confidential, client, employee, payment, health, or regulated information unless your policy and account controls explicitly allow it. AI output is a draft until a person reviews it.

## 1. Capture a voice note with Wispr Flow

**Use it for:** a rough internal update, outline, or idea you would otherwise lose.

1. Install Wispr Flow on a supported device.
2. Choose one low-risk note.
3. Dictate for two minutes.
4. Edit the transcription before saving or sharing it.

Wispr offers free access, but platform-specific limits apply. Check the [official Wispr pricing page](https://wisprflow.ai/pricing) rather than assuming one quota applies everywhere. Its [Smart Formatting guide](https://docs.wisprflow.ai/articles/5373093536-how-do-i-use-smart-formatting-and-backtrack) explains current dictation behavior.

Review [Wispr’s data controls](https://wisprflow.ai/data-controls) before using workplace information. Context Awareness and Privacy Mode affect what surrounding text may be processed and how data may be used.

[Watch the voice-to-text demonstration →](https://www.loom.com/share/9f8bea03b1994bb8bdca225b8743c69f)

**Keep it if:** the edited note reaches the right place faster and still sounds like you.

## 2. Draft a checklist from a short process recording

**Use it for:** capturing company knowledge that currently lives in one person’s head.

1. Record a three-to-five-minute Loom of a low-risk repetitive task.
2. Upload the supported video to Gemini.
3. Ask: “Draft a numbered checklist from this recording. Mark decisions, missing information, and steps that need human approval.”
4. Replay the video and correct the checklist.
5. Ask someone else to follow it before calling it an SOP.

Gemini accepts video files subject to plan-dependent file-count, size, and duration limits in the [Gemini Apps upload guide](https://support.google.com/gemini/answer/14903178). The model may miss a step or infer one that was never shown. Human review is required before the checklist becomes operating guidance.

[Watch the checklist demonstration →](https://www.loom.com/share/d27e53b863c04a0e8f5a27ceac19adc4)

**Keep it if:** another person can follow the reviewed checklist and the process owner accepts responsibility for updating it.

For a fuller approach to turning process capture into owned company knowledge, use the [workflow-first AI guide](/blog/10-hour-week-ai-tools).

## 3. Draft one reusable email template

**Use it for:** a repeated, low-risk email such as scheduling or requesting missing information.

1. Choose one email type.
2. Remove names, contact details, confidential terms, and client-specific facts from three examples.
3. In an approved AI account, ask for one concise template with brackets around every field a person must customize.
4. Verify the tone, facts, promises, and placeholders.
5. Save it as a draft—not an automatic send.

Prompt:

> Compare these examples. Draft one reusable template in the same tone. Put brackets around every fact a person must review or customize. Do not add claims, dates, prices, or promises that are not in the examples.

[Watch the email-template demonstration →](https://www.loom.com/share/25518d756c9243d4b022a9bdb4e0e4c5)

**Keep it if:** the reviewer catches every variable before sending and the template improves response consistency.

For a service business, that consistency matters beyond convenience. A clear, reviewed reply can protect the handoff from inquiry to the person who owns the next step. Customer-facing messages still need a responsible human.

## 4. Prepare for one meeting from supplied context

**Use it for:** organizing public or approved information before a call.

Do not assume ChatGPT can read a LinkedIn URL. Login walls, robots controls, account state, and page visibility can block retrieval.

1. Paste only the relevant public profile text you can lawfully use.
2. Add the approved purpose of the meeting and a sanitized summary of prior context.
3. Ask for three possible questions and a short list of facts to verify.
4. Check every name, role, company, and factual statement against the source.

Prompt:

> Based only on the supplied text, draft three useful questions for this meeting. Separate source facts from suggestions. If information is missing, say so instead of guessing.

[Watch the meeting-prep demonstration →](https://www.loom.com/share/ca80eef8b9854749952726f0b854de06)

**Keep it if:** the preparation is accurate, relevant, and easy to verify.

The [prompt library](/prompt-library) offers more reusable starting points, but context, policy, and review matter more than clever wording.

## Decide what happens after the trial

After one week, ask:

- Did the task occur often enough to matter?
- Was the output accurate after review?
- Did it improve revenue support, company knowledge, reliability, or capacity?
- Who owns the next version?
- What information should never enter the workflow?

If the trial is useful and low-risk, keep it simple. If it needs multiple systems, automatic actions, sensitive data, or exception handling, score it with the [AI implementation-readiness guide](/blog/stop-wasting-time-wrong-problems) before expanding.

You can browse more real build demonstrations on the [Build With AI podcast](/podcast). One careful experiment is enough for today.

### The Beginner AI Tool Stack: A Minimal Decision Guide

- URL: https://returnmytime.com/blog/beginner-ai-tool-stack
- Author: Corey Ganim
- Published: 2025-12-11T10:00:00Z
- Category: TOOLS & QUICK WINS
- Tags: AI TOOLS, STRATEGY, BEGINNER, PRODUCTIVITY

A useful beginner AI stack may contain one tool, not five.

Start with the job that creates the most repeated friction. Choose one tool for that job, run a contained trial, and keep it only if the team can review its output and own the workflow.

This guide covers five common jobs: capturing ideas, digesting long material, thinking through documents, documenting meetings, and editing images. It is a decision guide—not a recommendation to buy every product.

If you have not chosen a workflow yet, begin with the [AI project readiness scorecard](/blog/stop-wasting-time-wrong-problems). If you already know the workflow, use the [workflow-first implementation sequence](/blog/10-hour-week-ai-tools) to introduce tools in stages.

## 1. Voice input: Wispr Flow

**Choose it when:** typing slows down drafts, notes, or internal process capture.

**Skip it when:** the work is mostly structured data entry, or company policy does not approve voice processing for the information involved.

Wispr Flow supports Mac, Windows, iPhone, and Android and works in most apps with a text field. Features and limits vary by platform. The [official features page](https://wisprflow.ai/features) and [supported-device guide](https://docs.wisprflow.ai/articles/1036674442-supported-devices-and-system-requirements) are better buying references than a fixed comparison table.

Free access has platform-specific conditions. Wispr currently lists different desktop and mobile limits, so check the [current pricing page](https://wisprflow.ai/pricing) before relying on a quota. Do not treat a soft cap or promotion as a bypass.

Before using company information, review [Wispr’s data controls](https://wisprflow.ai/data-controls). Context Awareness may process relevant active-window text, and Privacy Mode changes how dictation data is treated.

**First trial:** dictate one low-risk internal update, then compare the edited result with your normal process.

## 2. Long-form summaries: NoteGPT

**Choose it when:** your work requires extracting information from videos, audio, documents, presentations, or links.

**Skip it when:** you need a source of truth rather than a summary, or nobody will verify the output against the original.

NoteGPT supports several content types across summarization, transcription, translation, and chat, with feature-specific limits described in its [terms](https://notegpt.io/terms) and [help center](https://help.notegpt.io/).

Its free plan uses **quotas**, not interchangeable queries. Different actions can consume different amounts. NoteGPT also uses “Unlimited” for basic quotas while premium credits, advanced models, file sizes, and processing limits still apply. Check the [official pricing page](https://notegpt.io/pricing) instead of assuming the plan name covers every feature.

**First trial:** summarize one non-confidential training item, verify the important points against the original, and decide whether the summary helps someone act.

## 3. Document thinking: Claude

**Choose it when:** you need help drafting, comparing, organizing, or questioning documents.

**Skip it when:** the task requires an authoritative answer without source review, or the account is not approved for the data.

Claude supports common document, spreadsheet, text, and image formats, subject to file and context limits. Anthropic lists the current formats in its [file-upload guide](https://support.claude.com/en/articles/8241126-upload-files-to-claude).

Claude Pro is currently listed at $20 per month in the United States. Anthropic says Pro provides at least five times free usage per session, but session, weekly, model, feature, and capacity limits still apply. See [Claude plan options](https://support.claude.com/en/articles/11049762-choose-a-claude-plan) and the [Pro plan limits](https://support.claude.com/en/articles/8325606-what-is-the-pro-plan). Start free and upgrade only after your actual work reaches a useful limit.

**First trial:** give it one approved document, ask for a structured draft, and verify every consequential claim.

## 4. Meeting documentation: Fathom

**Choose it when:** meetings contain decisions and action items that are routinely lost.

**Skip it when:** recording is not permitted, attendees have not received the required notice, or no one will own the follow-up.

Fathom supports Zoom, Google Meet, and Microsoft Teams and offers recording, transcription, summaries, sharing, and transcript questions. Its individual Free plan includes unlimited recordings, storage, and transcription. After the first five calls, free users retain basic summaries while advanced summary use is limited. Paid individual plans add features such as advanced summaries, action items, follow-up emails, and conversational tools. Plan prices and features can change, so compare the [Free-versus-Premium guide](https://help.fathom.video/en/articles/5290881) with [official pricing](https://fathom.video/pricing) before choosing a plan.

Before recording or sharing, follow applicable consent law, meeting-platform notices, company policy, and attendee expectations.

**First trial:** use it on one approved internal meeting, compare the summary with the team’s decisions, and assign every action item.

## 5. Plain-language image editing: Nano Banana

**Choose it when:** you need a fast draft or a basic image edit that a person can review.

**Skip it when:** the image requires exact brand, legal, accessibility, or factual control that the workflow cannot verify.

Google’s Gemini image tools can create and edit images from natural-language instructions, including targeted edits and character consistency. Google currently brands the consumer capability as Nano Banana 2, with Nano Banana Pro available to eligible paid subscribers. The [Gemini image guide](https://support.google.com/gemini/answer/14286560) describes current access and controls.

Free access is limited, not unlimited. Daily quotas and account, age, language, country, and model conditions can apply; paid plans provide higher limits. Check [Gemini limits and upgrades](https://support.google.com/gemini/answer/16275805).

**First trial:** edit a non-sensitive internal draft, then review accuracy, rights, brand fit, and accessibility before publication.

## How to choose without building another tool pile

Score each candidate from 1 to 3:

| Question | 1 | 3 |
|---|---|---|
| Does it solve a frequent job? | Rarely | Weekly or daily |
| Can a person review the output? | Unclear | Named reviewer |
| Is the data approved? | Unknown | Policy and account approved |
| Can the team own it? | No owner | Owner and fallback named |
| Does it improve a business result? | Novelty | Revenue, knowledge, reliability, or capacity |

Choose the highest-scoring **job**, then trial one tool. The [four 10-minute AI quick wins](/blog/four-10-minute-ai-quick-wins) provide low-risk experiments for several of these categories.

For a service operator, the stack should strengthen a business system. Meeting notes should become owned decisions. Process drafts should become reviewed company knowledge. Lead-response support should become a reliable handoff—not another account that nobody monitors.

You can see practical implementations on the [Build With AI podcast](/podcast), or explore [managed implementation](/services) when the workflow crosses systems, handles sensitive information, or needs ongoing ownership.

The right stack is the smallest one your business can use, review, and maintain.

### Stop Wasting Time on the Wrong AI Problems

- URL: https://returnmytime.com/blog/stop-wasting-time-wrong-problems
- Author: Corey Ganim
- Published: 2025-12-04T10:00:00Z
- Category: MANAGED AI
- Tags: STRATEGY, PRIORITIZATION, AI

A long list of AI ideas is not a roadmap. It is an inventory.

The Effort versus Impact Matrix is a useful first filter, but it cannot tell you whether a project has a responsible owner, usable data, acceptable risk, or a definition of done. Those questions decide whether an attractive idea is ready to implement.

For a service business with steady inbound, the difference matters. “Automate lead follow-up” may look high impact. But if nobody owns the response rules, lead sources are inconsistent, or the team cannot handle exceptions, a fast automation can create a faster reliability problem. The revenue opportunity is real; so is the need for operational ownership.

## Start with effort and impact

![Effort vs Impact Matrix diagram showing four quadrants](/images/blog/effort-impact-matrix.png)

Place each candidate on two axes:

- **Effort:** implementation time, integration complexity, policy work, training, and ongoing maintenance.
- **Impact:** revenue protected or created, company knowledge retained, service quality improved, or meaningful capacity released.

The four quadrants remain useful:

- **High impact, low effort:** investigate first.
- **High impact, high effort:** plan and resource deliberately.
- **Low impact, low effort:** delegate, batch, or leave manual.
- **Low impact, high effort:** decline.

Do not turn the top-left quadrant into an automatic “build now” list. Run each candidate through the readiness checks below.

## The seven-factor readiness scorecard

### 1. Business impact

Name the business result in plain language. Good answers include:

- respond to qualified inquiries consistently;
- keep delivery knowledge when a team member is unavailable;
- reduce missed handoffs between sales and operations;
- make a recurring report more reliable.

“Use AI” is not an outcome.

### 2. Implementation effort

Count the full effort, not only configuration. Include source cleanup, permissions, testing, training, monitoring, and exception handling. A ten-minute demo may still lead to an owned production system.

For low-risk learning, start with one of these [four contained AI experiments](/blog/four-10-minute-ai-quick-wins). They help you learn without confusing a demonstration with a finished business process.

### 3. Named owner

One person must be accountable for:

- approving the workflow;
- reviewing exceptions;
- updating rules and source material;
- deciding when the system should pause.

“The team” is not an owner. A vendor can manage the system, but someone inside the company still owns the business decision.

### 4. Data readiness

Ask:

- Where does the source information live?
- Is it current and complete?
- Which system is authoritative?
- Does it contain client, employee, payment, or regulated information?
- Which accounts and policies approve its use?

Scattered knowledge is often the real constraint. Automation built on conflicting documents simply produces conflict faster.

### 5. Risk and reversibility

Separate low-risk drafting from high-risk action.

An internal draft that a person reviews is easier to reverse than an automatic client promise, payment action, record update, or deletion. The higher the consequence, the stronger the approval, logging, fallback, and access controls should be.

### 6. Volume and frequency

Measure how often the work occurs. A small annoyance repeated 100 times a month may deserve attention before an impressive task performed twice a year.

For lead operations, count qualified inbound by source, current response patterns, and the points where ownership breaks. Do not invent revenue recovered. Use your own lead volume, close rate, and average value to decide whether the opportunity is material.

### 7. Acceptance criteria

Write the test before choosing the tool. Examples:

- Every test lead creates one complete record and alerts the assigned owner.
- The response uses only approved service-area and scheduling information.
- A failed handoff is visible within a defined time.
- A draft procedure can be followed successfully by someone other than its author.

If you cannot describe an acceptable output, the project is not decision-ready.

## A simple decision table

| Question | Ready | Not ready |
|---|---|---|
| Is the business result specific? | Measurable outcome | General interest in AI |
| Is the source information usable? | Current source of truth | Scattered or conflicting inputs |
| Is the owner named? | One accountable person | “The team” |
| Is the risk bounded? | Review, fallback, permissions | Irreversible action without controls |
| Is volume known? | Observed frequency | Assumption |
| Is success testable? | Written acceptance criteria | “It feels better” |

When several answers land in “not ready,” the next move is discovery and process cleanup—not buying another tool.

## Choose the right implementation path

Use a simple do-it-yourself tool when the workflow is deterministic, low-risk, and easy for your team to own. The [Zapier automation guide](/blog/zapier-automations-small-business) shows that boundary in practice.

Use a staged workflow plan when the work is understood but the team needs a sequence. Start with the [workflow-first AI implementation guide](/blog/10-hour-week-ai-tools).

Consider managed implementation when the workflow is revenue-critical, crosses multiple systems, contains sensitive information, or lacks an internal technical owner. The value is not merely configuration. It is design, testing, monitoring, exception handling, and clear accountability. [Return My Time’s services](/services) are designed for operators who prefer that work handled for them.

## Your next 20 minutes

1. List five recurring operational problems.
2. Plot effort and impact.
3. Score the strongest two across owner, data, risk, frequency, and acceptance criteria.
4. Choose one next decision: decline, gather evidence, run a contained trial, or prepare for implementation.

If you want help finding the best starting point, [request a free AI Opportunity Assessment](/services/free-ai-opportunity-assessment). The goal is not to automate the most exciting problem. It is to choose a valuable problem that your business is ready to own.

### The 10-Hour Week: A Workflow-First AI Plan for Business Owners

- URL: https://returnmytime.com/blog/10-hour-week-ai-tools
- Author: Corey Ganim
- Published: 2025-12-03T10:00:00Z
- Category: TOOLS & QUICK WINS
- Tags: PRODUCTIVITY, NO-CODE, AUTOMATION

A shorter workweek does not start with downloading four tools. It starts with finding one recurring workflow that deserves attention.

The “10-hour week” is a direction, not a promised result. Your outcome depends on the work, its frequency, the quality of your source information, and whether someone owns the system after launch.

For a lead-rich service business, the first useful workflow may be inbound follow-up. For another team, it may be turning delivery knowledge into a repeatable handoff. Both connect to business value: revenue is protected when leads receive a reliable response, and company knowledge becomes more durable when it does not live only in one employee’s head.

Before choosing a tool, use the [AI project prioritization guide](/blog/stop-wasting-time-wrong-problems) to identify a workflow with enough volume and impact to matter.

## Stage 1: Define the workflow and finish line

Write down five things:

1. What starts the workflow?
2. What information does it need?
3. What result should it produce?
4. Who reviews exceptions?
5. What would make you trust it?

“Help with email” is too vague. “Draft a response to new estimate requests using our approved service-area and scheduling rules, then route it to the sales coordinator for review” is testable.

Capture a simple baseline for one week: how often the work occurs, how long it takes, where errors happen, and what a delay costs. That gives you something more useful than an assumed savings number.

## Stage 2: Reduce input friction

If typing slows down drafts, notes, or process capture, voice input can be a useful first layer. Wispr Flow supports voice typing across desktop and mobile apps, generally wherever there is a text field. Its current platform features and limits are documented on the [official Wispr Flow features page](https://wisprflow.ai/features).

Use it for low-risk material first: a meeting recap, an internal outline, or the rough steps of a process. Review the transcription before it becomes a client message or operating instruction.

Also check the [Wispr data controls](https://wisprflow.ai/data-controls) before using company information. Context Awareness and Privacy Mode affect what surrounding text may be processed and how dictation data may be used. Sensitive, client, employee, or regulated data should only enter an account and workflow approved by your company.

## Stage 3: Add an AI assistant to one bounded task

ChatGPT calls connected services **Apps**, not connectors. Gmail and Google Calendar use cases can include finding messages, summarizing email, and reading a schedule, but availability varies by plan, region, workspace policy, enabled actions, and approved scopes. OpenAI documents those conditions in [Apps in ChatGPT](https://help.openai.com/en/articles/11487775-connectors-in) and its [App use-case guide](https://help.openai.com/en/articles/12084614).

Start with read-only work:

- Summarize the unread messages that match a defined client or project.
- Find the thread connected to a proposal.
- List tomorrow’s appointments and flag missing preparation notes.

Do not assume every account can create or change a calendar event. Write actions depend on the action available to your account and the scopes you granted, and they require confirmation. Review the proposed change before approving it.

Connecting company email or calendars is a data decision, not just a convenience setting. OpenAI explains indexing, retention after disconnecting, memory, training treatment, and workspace controls in its [Google App data-controls FAQ](https://help.openai.com/en/articles/10408842). Use approved accounts and grant only the access the workflow needs.

## Stage 4: Turn a real process into reviewed company knowledge

Use an account approved by your company and choose a low-risk process recording that excludes sensitive client, employee, payment, or regulated information. Record a short Loom while the process owner completes the task. Upload the supported video to Gemini and ask for:

- a numbered draft procedure;
- required inputs and expected outputs;
- decisions that need a human;
- gaps or ambiguous steps.

Gemini supports video uploads, subject to file-count, size, duration, and plan limits described in the [Gemini Apps upload guide](https://support.google.com/gemini/answer/14903178). It is one option, not the only model that can accept video.

Treat the generated procedure as a draft. Replay the recording, verify every step, test the procedure with someone other than the author, and assign an owner and review date. The output becomes useful company knowledge only after a person accepts responsibility for its accuracy. The [AI SOP generator guide](/blog/ai-sop-generator) explains how to capture, review, and maintain that procedure.

The [four quick AI experiments](/blog/four-10-minute-ai-quick-wins) offer a smaller version of this exercise if you want to test the approach before changing a live process.

## Stage 5: Automate only the stable handoff

Once the inputs, decisions, and expected result are clear, automate the deterministic steps: move an approved form submission to the right system, create a task, or notify an owner.

The [small-business Zapier guide](/blog/zapier-automations-small-business) explains where a simple trigger-and-action workflow is enough—and when monitoring, exception handling, and ongoing ownership are necessary.

Keep a human review step for judgment, promises, pricing, client communication, and changes to company records. A workflow is not reliable because it ran once. It is reliable when someone can see failures, correct exceptions, and confirm the output still meets its acceptance criteria.

## A practical four-week sequence

**Week 1:** Measure one recurring workflow and choose a clear success condition.

**Week 2:** Reduce input friction and test one bounded AI task with non-sensitive information.

**Week 3:** Document the process, review the draft, and test it with another person.

**Week 4:** Automate one stable handoff, assign an owner, and review the results.

If the workflow touches lead response, company knowledge, or multiple systems, the hard part is usually not the tool. It is design, policy, testing, and ownership. [Return My Time’s managed AI services](/services) are built for operators who want that implementation handled rather than another platform to maintain.

The useful goal is not “install AI.” It is one calmer, measurable workflow that the team can trust. Repeat from there.

## Prompt Library

The published library contains 84 complete prompts across the available categories. Every prompt below includes its full instructions and supporting content.

### Create Business Model Canvas

- URL: https://returnmytime.com/prompt-library/create-business-model-canvas
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: business-creation, strategy, ai
- Published: 2026-07-24
- Provenance: current
- Source: https://returnmytime.com/prompt-library/create-business-model-canvas
- Source captured: 2026-07-24T00:00:00Z


Turn your business details into a complete nine-part Business Model Canvas covering customers, value, channels, revenue, resources, and costs.

#### What it does

- Organizes supplied business details into the nine sections of a Business Model Canvas.

#### When to use

- When pressure-testing how a business creates, delivers, and captures value.


#### Required inputs

- DESCRIBE YOUR BUSINESS

#### Prompt

#CONTEXT:
You are adopting the role of a business model canvas architect. Your task is to help the user develop a detailed business model canvas for their given business, mapping out the key components and ensuring they are coherent, aligned with business objectives, and grounded in market realities.

#ROLE:
You are a business model canvas architect, leveraging deep expertise in business strategy, market analysis, and value proposition design to craft comprehensive business model canvases.

#RESPONSE GUIDELINES:
The business model canvas should be organized into the following sections:

1. Customer Segments 
● List 3 key customer segments
2. Value Propositions
● Identify 3 main value propositions 
3. Channels
● Outline 3 channels for reaching customers
4. Customer Relationships 
● Describe 2 types of customer relationships
5. Revenue Streams
● List 2 primary revenue streams 
6. Key Resources
● Identify 3 critical resources 
7. Key Activities  
● Describe 3 essential activities
8. Key Partnerships
● List 2 important partnerships
9. Cost Structure 
● Break down 3 main costs

Focus on ensuring the canvas elements are coherent, aligned with stated business objectives, and grounded in market realities. Avoid making the canvas overly complex or theoretical.

#TASK CRITERIA:
1. The business model canvas should comprehensively cover the 9 key components 
2. Each component should include the requested number of items
3. The items should be specific and relevant to the given business
4. Prioritize elements that are most critical to the business' success
5. Ensure overall alignment and coherence across the components
6. Ground the canvas in market realities and business objectives
7. Avoid overly theoretical or generic elements 

#INFORMATION ABOUT ME:
● My business: [DESCRIBE YOUR BUSINESS]

#RESPONSE FORMAT:
Customer Segments:
1. [Customer segment 1]
2. [Customer segment 2] 
3. [Customer segment 3]

Value Propositions:
1. [Value proposition 1]
2. [Value proposition 2]
3. [Value proposition 3]

Channels:
1. [Channel 1]  
2. [Channel 2]
3. [Channel 3]

Customer Relationships:
1. [Relationship type 1]
2. [Relationship type 2] 

Revenue Streams:
1. [Revenue stream 1]
2. [Revenue stream 2]

Key Resources:  
1. [Resource 1]
2. [Resource 2]
3. [Resource 3]

Key Activities:
1. [Activity 1] 
2. [Activity 2]
3. [Activity 3]

Key Partnerships:
1. [Partnership 1]
2. [Partnership 2]

Cost Structure:
1. [Cost 1]
2. [Cost 2] 
3. [Cost 3]

#### Expected output

- A nine-section Business Model Canvas with the requested number of specific entries.

#### Customization tips

- Understand Each Segment: Thoroughly explore each of the nine segments (Key Partners, Key Activities, Key Resources, Value Propositions, Customer Relationships, Channels, Customer Segments, Cost Structure, and Revenue Streams) to ensure a comprehensive understanding of how your business operates and creates value.
- Iterate and Validate: Regularly update and refine your business model canvas based on feedback from stakeholders and changes in the market. This iterative process helps in adapting to new insights and maintaining a competitive edge.
- Utilize Visual Tools: Employ visual aids like sticky notes or digital tools for mapping out the canvas. This enhances collaboration among team members and stakeholders, making it easier to see connections and dependencies across different segments of the business.
- Identify Key Assumptions: Clearly define the assumptions underlying each segment of your business model canvas. This will help you identify potential risks and uncertainties that need to be addressed.
- Seek External Input: Don't rely solely on your own perspective. Seek input from external experts, mentors, or advisors who can provide valuable insights and challenge your assumptions.
- Prioritize Value Propositions: Focus on understanding and refining your value propositions to ensure they align with customer needs and differentiate your business from competitors.
- Test and Validate: Conduct market research, surveys, or interviews to gather feedback and validate your assumptions and value propositions. This will help you make informed decisions and refine your business model.
- Continuously Monitor and Adapt: Regularly review and update your business model canvas to reflect changes in the market, customer preferences, and industry trends. Stay agile and be willing to adapt your business model as needed.

#### Guardrails

- Treat the canvas as a working hypothesis and validate important assumptions with customers and operating evidence.

#### Example input

#INFORMATION ABOUT ME:
● My business: God of Prompt, the biggest collection of easy-to-follow AI resources, AI Prompts & How-to Guides for busy Small Business Owners.

#### Related prompts

- https://returnmytime.com/prompt-library/conduct-market-research-validation
- https://returnmytime.com/prompt-library/create-financial-projections


### Conduct Market Research Validation

- URL: https://returnmytime.com/prompt-library/conduct-market-research-validation
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: business-creation, research, ai
- Published: 2026-07-24
- Provenance: current
- Source: https://returnmytime.com/prompt-library/conduct-market-research-validation
- Source captured: 2026-07-24T00:00:00Z


Converts user input into a structured task for generating instructions to validate a business idea.

#### What it does

- Structures a sourced market, competitor, customer-needs, and opportunity assessment for a business idea.

#### When to use

- Before investing heavily in a new offer, market, or business idea.


#### Required inputs

- INSERT BUSINESS IDEA

#### Prompt

#CONTEXT:
You are tasked with helping a user validate a business idea through comprehensive market research and analysis. The user will provide their business idea, and you will need to thoroughly investigate the target market, competition, customer needs, and market opportunity to determine the idea's viability.

#ROLE:
Adopt the role of an expert business strategist and market researcher skilled at validating business ideas through comprehensive market analysis.

#RESPONSE GUIDELINES:
Organize the market research findings into the following sections:

1. Business Idea Overview: Briefly summarize the user's business idea.

2. Market Overview: 
   ● Market Size: Provide data on the target market size, citing reputable sources.
   ● Market Trends: Identify key market trends, citing reputable sources.
   ● Target Customer Segments: Describe the target customer segments, citing reputable sources.

3. Competitive Landscape:
   ● Key Competitors: List the main competitors in the market, citing reputable sources.
   ● Competitive Advantages: Analyze the business idea's competitive advantages, citing reputable sources.
   ● Barriers to Entry: Discuss potential barriers to entering the market, citing reputable sources.

4. Customer Needs Analysis:
   ● Customer Pain Points: Identify the main pain points and challenges faced by target customers, citing reputable sources.
   ● Unmet Needs: Highlight any unmet needs in the market that the business idea could address, citing reputable sources.
   ● Idea Value Proposition: Explain how the business idea provides value to customers and meets their needs, citing reputable sources.

5. Market Opportunity:
   ● Market Growth Potential: Assess the market's growth potential, citing reputable sources.
   ● Estimated Market Share: Provide an estimated market share the business idea could capture, citing reputable sources.
   ● Revenue Potential: Estimate the revenue potential for the business idea, citing reputable sources.

6. Validation Summary: Summarize the key findings and insights from the market research, validating or invalidating the business idea's viability.

7. Recommendations: Offer data-driven recommendations based on the market research findings to help the user refine their business idea and increase its chances of success.

#TASK CRITERIA:
● Focus on providing comprehensive, data-driven insights backed by reputable sources.
● Analyze the market from multiple angles, including size, trends, competition, customer needs, and growth potential.
● Identify key opportunities and challenges for the business idea within the target market.
● Provide a clear validation summary and actionable recommendations based on the market research findings.
● Avoid making assumptions or presenting information without proper citations.

#INFORMATION ABOUT ME:
● Business idea: [INSERT BUSINESS IDEA]

#RESPONSE FORMAT:
Business Idea Overview:
[Brief summary of the business idea]

Market Overview:
● Market Size: [Market size data] (Source: [Citation])
● Market Trends: [Key market trends] (Source: [Citation])
● Target Customer Segments: [Description of target customer segments] (Source: [Citation])

Competitive Landscape:
● Key Competitors: [List of main competitors] (Source: [Citation])
● Competitive Advantages: [Analysis of competitive advantages] (Source: [Citation])
● Barriers to Entry: [Discussion of barriers to entry] (Source: [Citation])

Customer Needs Analysis:
● Customer Pain Points: [Identification of customer pain points] (Source: [Citation])
● Unmet Needs: [Highlight of unmet needs in the market] (Source: [Citation])
● Idea Value Proposition: [Explanation of the business idea's value proposition] (Source: [Citation])

Market Opportunity:
● Market Growth Potential: [Assessment of market growth potential] (Source: [Citation])
● Estimated Market Share: [Estimated market share the business idea could capture] (Source: [Citation])
● Revenue Potential: [Estimation of revenue potential] (Source: [Citation])

Validation Summary:
[Summary of key findings and insights validating or invalidating the business idea's viability]

Recommendations:
[Data-driven recommendations to refine the business idea and increase its chances of success]

#### Expected output

- A cited validation report with market findings, risks, recommendations, and a clear viability summary.

#### Customization tips

- Add a geography, customer segment, and price range to narrow the research.
- Ask for publication dates beside sources when market conditions change quickly.

#### Guardrails

- Verify sources, dates, and quoted figures independently; label estimates and unavailable evidence instead of inventing citations.

#### Example input

[INSERT BUSINESS IDEA]: A mobile bicycle-repair service for commuters in Columbus, Ohio.

#### Related prompts

- https://returnmytime.com/prompt-library/analyze-business-ideas
- https://returnmytime.com/prompt-library/create-business-model-canvas


### Generate Catchy Business Names

- URL: https://returnmytime.com/prompt-library/generate-catchy-business-names
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 1-2 hours
- Tags: business-creation, branding, ai
- Published: 2026-07-24
- Provenance: current
- Source: https://returnmytime.com/prompt-library/generate-catchy-business-names
- Source captured: 2026-07-24T00:00:00Z


Generates creative and catchy business names tailored to the specified niche.

#### What it does

- Generates ten brand-name options and explains how each fits the supplied audience and positioning.

#### When to use

- When exploring names for a new business, product, or offer.


#### Required inputs

- DESCRIBE YOUR NICHE OR INDUSTRY
- DESCRIBE YOUR TARGET AUDIENCE
- DESCRIBE YOUR BRAND PERSONALITY
- LIST YOUR KEY DIFFERENTIATORS

#### Prompt

#CONTEXT:
Adopt the role of an expert business name generator with a knack for creating catchy, relevant, and memorable names. Your task is to help the user generate a list of creative business name ideas tailored to the given niche or industry.

#ROLE:
You are an expert business name generator with a knack for creating catchy, relevant, and memorable names.

#RESPONSE GUIDELINES:
1. Provide a diverse range of options that are easy to pronounce, memorable, and evocative of the brand's essence.
2. Consider the target audience, brand personality, and key differentiators when brainstorming names.
3. Provide rationales for each name idea to explain how it aligns with the brand's characteristics and objectives.
4. Offer additional tips for choosing and refining the perfect business name.

#BUSINESS NAME CRITERIA:
1. Names should be catchy, relevant, and memorable.
2. Names should be easy to pronounce and evocative of the brand's essence.
3. Names should consider the target audience, brand personality, and key differentiators.
4. Avoid names that are too long, difficult to spell, or similar to existing brands.

#INFORMATION ABOUT ME:
● My niche or industry: [DESCRIBE YOUR NICHE OR INDUSTRY]
● My target audience: [DESCRIBE YOUR TARGET AUDIENCE]
● My brand personality: [DESCRIBE YOUR BRAND PERSONALITY]
● My key differentiators: [LIST YOUR KEY DIFFERENTIATORS]

#RESPONSE FORMAT:
Niche Description:
[NICHE DESCRIPTION]

Target Audience:
[TARGET AUDIENCE]

Brand Personality:
[BRAND PERSONALITY]

Key Differentiators:
[KEY DIFFERENTIATORS]

Name Ideas:
1. [NAME IDEA 1]
2. [NAME IDEA 2]
3. [NAME IDEA 3]
4. [NAME IDEA 4]
5. [NAME IDEA 5]
6. [NAME IDEA 6]
7. [NAME IDEA 7] 
8. [NAME IDEA 8]
9. [NAME IDEA 9]
10. [NAME IDEA 10]

Name Rationales:
1. [NAME RATIONALE 1]
2. [NAME RATIONALE 2]
3. [NAME RATIONALE 3]
4. [NAME RATIONALE 4]
5. [NAME RATIONALE 5]
6. [NAME RATIONALE 6]
7. [NAME RATIONALE 7]
8. [NAME RATIONALE 8] 
9. [NAME RATIONALE 9]
10. [NAME RATIONALE 10]

Additional Tips:
[ADDITIONAL TIPS]

#### Expected output

- Ten memorable business-name ideas with matching rationales and selection guidance.

#### Customization tips

- Specify words, tones, or associations to include or avoid.
- Shortlist names only after checking domains, business registries, and relevant trademark databases.

#### Guardrails

- Generated names are creative options, not clearance opinions; complete professional trademark review before adopting one.

#### Example input

[DESCRIBE YOUR NICHE OR INDUSTRY]: Mobile bicycle repair
[DESCRIBE YOUR TARGET AUDIENCE]: Time-poor urban commuters
[DESCRIBE YOUR BRAND PERSONALITY]: Dependable, energetic, practical
[LIST YOUR KEY DIFFERENTIATORS]: Same-day curbside service; transparent pricing

#### Related prompts

- https://returnmytime.com/prompt-library/create-compelling-mission-statement
- https://returnmytime.com/prompt-library/trademark-registration-guide


### Create Compelling Mission Statement

- URL: https://returnmytime.com/prompt-library/create-compelling-mission-statement
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 1-2 hours
- Tags: business-creation, strategy, ai
- Published: 2026-07-24
- Provenance: current
- Source: https://returnmytime.com/prompt-library/create-compelling-mission-statement
- Source captured: 2026-07-24T00:00:00Z


Assists in creating a mission statement that reflects the core values and objectives of a business.

#### What it does

- Turns a business purpose, offering, values, and goals into a concise mission-statement draft.

#### When to use

- When aligning a team or clarifying a new brand's purpose and commitments.


#### Required inputs

- BUSINESS NAME
- BUSINESS INDUSTRY
- TARGET CUSTOMERS
- PRODUCT OR SERVICE OFFERING
- UNIQUE SELLING PROPOSITION
- VALUE 1
- VALUE 2
- VALUE 3
- GOAL 1
- GOAL 2
- GOAL 3

#### Prompt

#CONTEXT:
You are an expert mission statement architect, crafting compelling mission statements that align with business values and goals. Your task is to help the user create a powerful, inspiring mission statement based on the provided business details. The mission statement should succinctly capture the essence of the business, its purpose, and its commitments to its stakeholders. Ensure the mission statement is memorable, differentiating, and aligns with the specified business values and goals.

#ROLE:
As an expert mission statement architect, your role is to analyze the provided business details and craft a compelling mission statement that captures the essence of the business, its purpose, and its commitments to its stakeholders. You should ensure that the mission statement aligns with the specified business values and goals while being memorable and differentiating.

#RESPONSE GUIDELINES:
1. Analyze the provided business details, including the business name, industry, target customers, product or service offering, unique selling proposition, values, and goals.
2. Identify the key elements that should be included in the mission statement based on the business details.
3. Craft a concise, powerful, and inspiring mission statement that captures the essence of the business, its purpose, and its commitments to its stakeholders.
4. Ensure that the mission statement aligns with the specified business values and goals.
5. Review and refine the mission statement to make it memorable and differentiating.

#TASK CRITERIA:
1. The mission statement should be concise, typically no more than one or two sentences.
2. It should capture the essence of the business, its purpose, and its commitments to its stakeholders.
3. The mission statement should align with the specified business values and goals.
4. It should be memorable and differentiating, setting the business apart from its competitors.
5. Avoid using jargon or overly complex language; the mission statement should be easily understood by all stakeholders.

#INFORMATION ABOUT ME:
● My business name: [BUSINESS NAME]
● My business industry: [BUSINESS INDUSTRY]
● My target customers: [TARGET CUSTOMERS]
● My product or service offering: [PRODUCT OR SERVICE OFFERING]
● My unique selling proposition: [UNIQUE SELLING PROPOSITION]
● My business values: 
  1. [VALUE 1]
  2. [VALUE 2] 
  3. [VALUE 3]
● My business goals:
  1. [GOAL 1]
  2. [GOAL 2]
  3. [GOAL 3]

#RESPONSE FORMAT:
[BUSINESS NAME]

Mission Statement:
[MISSION STATEMENT]

#### Expected output

- A concise mission statement aligned with the supplied values, goals, customers, and differentiation.

#### Customization tips

- Replace abstract values with language customers and employees can recognize in daily decisions.
- Read the result aloud and remove jargon or claims the business cannot consistently support.

#### Guardrails

- Keep commitments truthful, specific, and supportable by current operations.

#### Example input

[BUSINESS NAME]: PedalReady
[BUSINESS INDUSTRY]: Mobile bicycle repair
[TARGET CUSTOMERS]: Urban commuters
[PRODUCT OR SERVICE OFFERING]: Same-day workplace and curbside repairs
[UNIQUE SELLING PROPOSITION]: Transparent quotes before every repair
[VALUE 1]: Reliability
[VALUE 2]: Respect for customers' time
[VALUE 3]: Safe workmanship
[GOAL 1]: Keep commuters riding
[GOAL 2]: Reduce repair uncertainty
[GOAL 3]: Build long-term customer trust

#### Related prompts

- https://returnmytime.com/prompt-library/generate-catchy-business-names
- https://returnmytime.com/prompt-library/create-business-model-canvas


### Create Financial Projections

- URL: https://returnmytime.com/prompt-library/create-financial-projections
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: business-creation, finance, ai
- Published: 2026-07-24
- Provenance: current
- Source: https://returnmytime.com/prompt-library/create-financial-projections
- Source captured: 2026-07-24T00:00:00Z


Converts user input into detailed financial projections for the first year of a business.

#### What it does

- Builds an assumption-led first-year revenue, expense, cash-flow, break-even, and financing model.

#### When to use

- When preparing an initial operating budget or testing the economics of a business plan.


#### Required inputs

- INSERT KEY DETAILS ABOUT THE BUSINESS

#### Prompt

#CONTEXT:
You are an expert financial analyst and business planner tasked with creating comprehensive and detailed financial projections and budgets for the first year of a given business. Your goal is to provide projections in a clear, well-structured format that can be easily understood and used for decision making and planning.

#ROLE:
Adopt the role of an expert financial analyst and business planner with deep knowledge of financial modeling, budgeting, and forecasting for startups and small businesses.

#RESPONSE GUIDELINES:
● Begin with a list of key assumptions used in creating the financial projections
● Provide a table of monthly revenue projections for the first year, including a total annual revenue figure
● Break down expense projections into categories, showing monthly averages and annual totals for each category, as well as overall total expenses
● Present monthly and cumulative cash flow projections for the first year in a table format
● Perform a break-even analysis, indicating the projected break-even point in terms of month and revenue
● Highlight key financial metrics, including Year 1 revenue, expenses, profit/loss, and profit margin
● Estimate startup costs, recommended initial working capital, and total financing needed

#TASK CRITERIA:
● Focus on creating comprehensive and detailed financial projections that account for all relevant factors and assumptions
● Present information in a clear, well-structured format using tables where appropriate
● Ensure projections are realistic and based on sound financial modeling principles
● Avoid making overly optimistic or pessimistic assumptions without justification
● Provide sufficient detail to support decision making and planning, but avoid unnecessary complexity

#INFORMATION ABOUT ME:
● My business details: [INSERT KEY DETAILS ABOUT THE BUSINESS]

#RESPONSE FORMAT:
Assumptions:
● Assumption 1
● Assumption 2
● Assumption 3

Revenue Projections:
Month | Revenue ($)
1 | Revenue Month 1
2 | Revenue Month 2
... | ...
12 | Revenue Month 12
Total | Total Annual Revenue

Expense Projections:
Expense Category | Monthly Average ($) | Annual Total ($)
Expense Category 1 | Monthly Average 1 | Annual Total 1
Expense Category 2 | Monthly Average 2 | Annual Total 2
... | ... | ...
Total | Total Monthly Average | Total Annual Expenses

Cash Flow Projections:
Month | Net Cash Flow ($) | Cumulative Cash Flow ($)
1 | Net Cash Flow 1 | Cumulative Cash Flow 1
2 | Net Cash Flow 2 | Cumulative Cash Flow 2
... | ... | ...
12 | Net Cash Flow 12 | Cumulative Cash Flow 12

Break-Even Analysis:
● Projected break-even point: Month X
● Projected break-even revenue: $X

Key Financial Metrics:
● Year 1 revenue: $X
● Year 1 expenses: $X
● Year 1 profit (loss): $X
● Year 1 profit margin: X%

Financing Needs:
● Estimated startup costs: $X
● Recommended initial working capital: $X
● Total financing needed: $X

#### Expected output

- A 12-month projection with explicit assumptions, tables, break-even analysis, key metrics, and estimated financing needs.

#### Customization tips

- Provide price, volume, seasonality, staffing, tax, and startup-cost assumptions.
- Request conservative, base, and optimistic scenarios when decisions depend on uncertain demand.

#### Guardrails

- Treat projections as illustrative planning estimates, verify formulas and assumptions, and obtain review from a qualified accountant or financial professional before relying on them.

#### Example input

[INSERT KEY DETAILS ABOUT THE BUSINESS]: Mobile bicycle repair in Columbus; $95 average order; 45 jobs per month at launch; one owner-technician; $18,000 vehicle and equipment costs; $2,400 monthly fixed expenses.

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas
- https://returnmytime.com/prompt-library/create-professional-pitch-decks


### Create Professional Pitch Decks

- URL: https://returnmytime.com/prompt-library/create-professional-pitch-decks
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: business-creation, strategy, ai
- Published: 2026-07-24
- Provenance: current
- Source: https://returnmytime.com/prompt-library/create-professional-pitch-decks
- Source captured: 2026-07-24T00:00:00Z


Converts user input into a structured pitch deck for investors or partners.

#### What it does

- Creates a structured investor or partner pitch-deck narrative from supplied product, audience, and business-goal details.

#### When to use

- When drafting the first content outline for a funding or partnership presentation.


#### Required inputs

- INSERT INFORMATION ABOUT THE PRODUCT/SERVICE
- DESCRIBE YOUR TARGET AUDIENCE
- OUTLINE YOUR BUSINESS GOALS

#### Prompt

#CONTEXT:
You are tasked with creating a comprehensive, visually engaging pitch deck for a given product or service. The pitch deck should tell a compelling story, highlight the unique value proposition, and persuade investors or partners to come on board. Use the provided structure and guidelines to craft an effective pitch deck that will secure funding and partnerships.

#ROLE:
Adopt the role of a master pitch deck creator, specializing in crafting compelling presentations to secure funding and partnerships. Your task is to help the user create a persuasive pitch deck for their product or service.

#RESPONSE GUIDELINES:
The pitch deck should follow this structure:

1. Company Overview
   ● Mission
   ● Team
   ● Key Milestones

2. Problem
   ● Problem Statement
   ● Problem Impact
   ● Current Solutions

3. Solution
   ● Product Description
   ● Key Features
   ● User Journey
   ● Value Proposition

4. Market Opportunity
   ● Target Market
   ● Market Size
   ● Market Trends
   ● Competitive Landscape

5. Business Model
   ● Revenue Model
   ● Pricing Strategy
   ● Sales Channels
   ● Partnerships

6. Traction and Validation
   ● Key Metrics
   ● Customer Testimonials
   ● Awards and Recognition

7. Financials
   ● Revenue Forecast
   ● Expense Breakdown
   ● Funding Requirements
   ● Use of Funds

8. Roadmap
   ● Short-term Goals
   ● Long-term Vision
   ● Key Milestones

9. Investment Opportunity
   ● Investment Terms
   ● Equity Offering
   ● Exit Strategy

10. Appendix
    ● Detailed Financials
    ● Team Bios
    ● References

#PITCH DECK CRITERIA:
1. The pitch deck should be visually engaging, with clear and impactful visuals that support the message.
2. The story should be compelling, highlighting the unique value proposition and persuading investors or partners to come on board.
3. The structure should be clear and easy to follow, guiding the audience through the key aspects of the company and its offering.
4. Focus on the most critical information and avoid overwhelming the audience with too many details.
5. Use data and metrics to support your claims and demonstrate traction and validation.
6. Tailor the pitch deck to the specific audience, whether it's investors, partners, or other stakeholders.

#INFORMATION ABOUT ME:
● My product/service: [INSERT INFORMATION ABOUT THE PRODUCT/SERVICE]
● My target audience: [DESCRIBE YOUR TARGET AUDIENCE]
● My business goals: [OUTLINE YOUR BUSINESS GOALS]

#RESPONSE FORMAT:
The pitch deck should be formatted as a visually appealing presentation, using slides with a consistent design and layout. Use bullet points, images, charts, and graphs to convey information effectively. Ensure that the presentation flows logically and tells a compelling story.

#### Expected output

- A slide-by-slide pitch-deck outline covering the problem, solution, market, business model, traction, team, financials, and ask.

#### Customization tips

- Supply verified traction, unit economics, team credentials, and a precise funding or partnership ask.
- Replace generated visual suggestions with evidence-backed charts and brand-approved assets.

#### Guardrails

- Verify every market, traction, financial, and team claim before presenting it to investors or partners.

#### Example input

[INSERT INFORMATION ABOUT THE PRODUCT/SERVICE]: PedalReady provides bookable same-day mobile bicycle repair at homes and workplaces.
[DESCRIBE YOUR TARGET AUDIENCE]: Urban commuters and employers supporting bike-to-work programs.
[OUTLINE YOUR BUSINESS GOALS]: Launch in Columbus, reach 180 monthly repairs, then expand through trained local operators.

#### Related prompts

- https://returnmytime.com/prompt-library/create-financial-projections
- https://returnmytime.com/prompt-library/launch-startup-success-strategy


### Generate Legal Business Requirements

- URL: https://returnmytime.com/prompt-library/generate-legal-business-requirements
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: business-creation, legal, ai
- Published: 2026-07-24
- Provenance: current
- Source: https://returnmytime.com/prompt-library/generate-legal-business-requirements
- Source captured: 2026-07-24T00:00:00Z


Researches and lists all legal requirements necessary for starting a business in the specified location.

#### What it does

- Produces a location-specific research checklist for registrations, licenses, permits, taxes, employment, and compliance.

#### When to use

- At the beginning of legal-requirements research for a new business or operating location.


#### Required inputs

- ENTER BUSINESS LOCATION HERE

#### Prompt

#CONTEXT:
Adopt the role of an expert legal researcher and business advisor. Your task is to help the user identify the legal requirements for starting a business in a specified location.

#ROLE:
You are an expert legal researcher and business advisor, specializing in identifying legal requirements for starting businesses in various jurisdictions.

#RESPONSE GUIDELINES:
Thoroughly research and comprehensively list out all the legal requirements and steps necessary for starting a business in the specified location. Cite reputable sources for each requirement listed. Organize the information in a clear, easy-to-follow format.

For each requirement, provide:
● Description
● Process 
● Timeline
● Cost
● Source URL

Include a list of additional resources at the end.

Add a disclaimer stating that the information is for general guidance only and not legal advice, and to consult with a qualified attorney and relevant government agencies for specific advice.

#BUSINESS LOCATION CRITERIA:
● Focus on providing comprehensive and accurate information specific to the given location
● Cite reputable sources for each requirement
● Organize the information in a clear, easy-to-follow format
● Include a disclaimer that the information is for general guidance only and not legal advice

#INFORMATION ABOUT ME:
● My business location: [ENTER BUSINESS LOCATION HERE]

#RESPONSE FORMAT:
Legal Requirements for Starting a Business in [location]:

1. Requirement 1
   Description: [description]
   Process: [process]
   Timeline: [timeline] 
   Cost: [cost]
   Source: [source_url]

2. Requirement 2
   Description: [description]
   Process: [process] 
   Timeline: [timeline]
   Cost: [cost]
   Source: [source_url]

3. Requirement 3
   Description: [description]
   Process: [process]
   Timeline: [timeline]
   Cost: [cost] 
   Source: [source_url]

...

Additional Resources:
1. [resource1_url]
2. [resource2_url] 
3. [resource3_url]

Disclaimer:
This information is provided for general guidance only and should not be considered as legal advice. Consult with a qualified attorney and relevant government agencies for specific advice pertaining to your unique situation.

#### Expected output

- A categorized legal-requirements checklist with responsible authorities, cited sources, and next steps.

#### Customization tips

- Include city, state or province, and country because requirements vary by jurisdiction.
- Add the business structure, industry, premises, and hiring plans to surface specialized obligations.

#### Guardrails

- Use current official government sources and have a qualified local attorney or compliance professional review the results; this prompt does not provide legal advice.

#### Example input

[ENTER BUSINESS LOCATION HERE]: Columbus, Ohio, United States

#### Related prompts

- https://returnmytime.com/prompt-library/trademark-registration-guide
- https://returnmytime.com/prompt-library/launch-startup-success-strategy


### Trademark Registration Guide

- URL: https://returnmytime.com/prompt-library/trademark-registration-guide
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: business-creation, legal, ai
- Published: 2026-07-24
- Provenance: current
- Source: https://returnmytime.com/prompt-library/trademark-registration-guide
- Source captured: 2026-07-24T00:00:00Z


Converts user input into a detailed guide on trademark registration for a business name and logo in a specified location.

#### What it does

- Outlines a location-specific trademark search and registration process for a proposed business name and logo.

#### When to use

- When preparing questions and documents for trademark clearance and professional review.


#### Required inputs

- BUSINESS NAME
- LOGO DESCRIPTION
- BUSINESS LOCATION

#### Prompt

#CONTEXT:
You are an expert intellectual property lawyer specializing in trademark registration. Your task is to help a business owner trying to register trademarks for their business name and logo in a specific location.

#ROLE:
Adopt the role of an expert intellectual property lawyer specializing in trademark registration. Provide a comprehensive guide on the trademark registration process, tailored to the specific location and needs of the business owner.

#RESPONSE GUIDELINES:
1. Begin with an overview of the applicable laws and government agencies involved in the trademark registration process in the specified location.
2. Provide a list of the required documents for the trademark application.
3. Outline the step-by-step application process, including any necessary forms, submissions, and examinations.
4. Give an estimated timeline for the registration process, breaking it down into key stages and their respective durations.
5. Detail the estimated costs involved in registering a trademark, including filing fees, attorney fees, and any other associated expenses.
6. Offer tips to strengthen the trademark application, such as conducting thorough searches and providing detailed descriptions of the mark.
7. Highlight potential challenges that may arise during the registration process, such as objections or oppositions from third parties.
8. Include a list of relevant sources and references for further information on trademark registration in the specified location.

#TASK CRITERIA:
1. Tailor the guide specifically to the location provided by the business owner.
2. Focus on providing actionable advice and step-by-step instructions for the trademark registration process.
3. Ensure that all information is accurate, up-to-date, and compliant with the relevant laws and regulations in the specified location.
4. Avoid using overly technical or legal jargon, and explain any complex concepts in plain language.
5. Emphasize the importance of conducting thorough searches and providing detailed descriptions of the mark to strengthen the application.

#INFORMATION ABOUT ME:
● My business name: [BUSINESS NAME]
● My business logo description: [LOGO DESCRIPTION] 
● My business location: [BUSINESS LOCATION]

#RESPONSE FORMAT:
Applicable Laws:
● [Law 1]
● [Law 2] 
● [Law 3]

Government Agencies Involved:
● [Agency 1]
● [Agency 2]

Required Documents:
● [Document 1]
● [Document 2]
● [Document 3]
● [Document 4]

Application Process:
Step 1: [Step 1 Description]
Step 2: [Step 2 Description]
Step 3: [Step 3 Description] 
Step 4: [Step 4 Description]
Step 5: [Step 5 Description]

Timeline:
The typical trademark registration process in [BUSINESS LOCATION] takes [DURATION], broken down as follows:
● [Stage 1]: [Stage 1 Duration]
● [Stage 2]: [Stage 2 Duration]
● [Stage 3]: [Stage 3 Duration]

Costs:
The estimated costs for registering a trademark in [BUSINESS LOCATION] are:
● [Cost 1 Description]: [Cost 1 Amount]
● [Cost 2 Description]: [Cost 2 Amount] 
● [Cost 3 Description]: [Cost 3 Amount]
Total estimated cost: [Total Cost]

Tips to Strengthen Application:
1. [Tip 1]
2. [Tip 2] 
3. [Tip 3]

Potential Challenges:
● [Challenge 1]
● [Challenge 2]
● [Challenge 3]

Sources:
1. [Source 1]
2. [Source 2] 
3. [Source 3]

#### Expected output

- A step-by-step trademark research and filing guide covering searches, classes, documents, fees, timelines, and post-filing responsibilities.

#### Customization tips

- Describe the goods and services the mark will cover, not only the logo.
- Search confusingly similar names and designs before spending on branding or filing.

#### Guardrails

- Trademark law and registries change; verify current official guidance and obtain advice from a qualified trademark attorney before filing or relying on clearance results.

#### Example input

[BUSINESS NAME]: PedalReady
[LOGO DESCRIPTION]: A simple green bicycle wheel with a small wrench forming one spoke
[BUSINESS LOCATION]: Columbus, Ohio, United States

#### Related prompts

- https://returnmytime.com/prompt-library/generate-legal-business-requirements
- https://returnmytime.com/prompt-library/generate-catchy-business-names


### Generate AI Business Ideas

- URL: https://returnmytime.com/prompt-library/generate-ai-business-ideas
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 1-2 hours
- Tags: business-creation, strategy, ai
- Published: 2026-07-24
- Modified: 2026-07-31
- Provenance: current
- Source: https://returnmytime.com/prompt-library/generate-ai-business-ideas
- Source captured: 2026-07-24T00:00:00Z


Generate five practical AI business ideas for a solopreneur, including costs, required skills, time to market, and credible supporting sources.

#### What it does

- Generates five AI-enabled business concepts constrained by the operator's expertise, resources, and available time.

#### When to use

- When exploring feasible AI business directions before selecting one for validation.


#### Required inputs

- INSERT YOUR RELEVANT EXPERTISE
- DESCRIBE YOUR AVAILABLE RESOURCES
- DESCRIBE YOUR TIME CONSTRAINTS

#### Prompt

#CONTEXT:
Adopt the role of an AI business ideas generator with deep knowledge of cutting-edge technologies, business models, and market trends. Your task is to help a solopreneur with limited resources brainstorm innovative, low-cost business ideas.

#ROLE:
You are an AI business ideas generator specializing in ideating innovative, low-cost business ideas tailored for solopreneurs.

#RESPONSE GUIDELINES:
Return a table with 5 rows and 5 columns:
● Business Idea
● Description
● Cost-Efficiency (1-5, where 5 is the most cost-efficient)
● Key Skills Required
● Time to Market

Provide a list of credible sources to support your ideas and analysis.

#TASK CRITERIA:
1. Focus on AI-powered business ideas suitable for a solopreneur with limited resources.
2. Evaluate each idea's cost-efficiency, required skills, and estimated time to market.
3. Cite credible sources to support your ideas and analysis.

#INFORMATION ABOUT ME:
● My expertise: [INSERT YOUR RELEVANT EXPERTISE]
● My available resources: [DESCRIBE YOUR AVAILABLE RESOURCES]
● My time constraints: [DESCRIBE YOUR TIME CONSTRAINTS]

#RESPONSE FORMAT:
Business Idea | Description | Cost-Efficiency (1-5) | Key Skills Required | Time to Market
--● | --● | --● | --● | ---
$business_idea1 | $description1 | $cost_efficiency1 | $skills1 | $time_to_market1
$business_idea2 | $description2 | $cost_efficiency2 | $skills2 | $time_to_market2
$business_idea3 | $description3 | $cost_efficiency3 | $skills3 | $time_to_market3
$business_idea4 | $description4 | $cost_efficiency4 | $skills4 | $time_to_market4
$business_idea5 | $description5 | $cost_efficiency5 | $skills5 | $time_to_market5

Sources:
1. $source1
2. $source2
3. $source3

#### Expected output

- Five practical AI business ideas with customer, value, cost, skill, time-to-market, revenue, risk, and source considerations.

#### Customization tips

- Name the workflows and customer problems you understand first-hand.
- Treat generated cost and market estimates as hypotheses to validate with buyers and current sources.

#### Guardrails

- Verify cited sources and avoid representing generated market, cost, or revenue estimates as established facts.

#### Example input

[INSERT YOUR RELEVANT EXPERTISE]: Ten years managing independent dental practices
[DESCRIBE YOUR AVAILABLE RESOURCES]: $8,000 budget, laptop, industry contacts, no employees
[DESCRIBE YOUR TIME CONSTRAINTS]: 15 hours per week for six months

#### Related prompts

- https://returnmytime.com/prompt-library/analyze-business-ideas
- https://returnmytime.com/prompt-library/launch-startup-success-strategy

#### Related articles

- https://returnmytime.com/blog/ai-assessment-before-build

### Launch Startup Success Strategy

- URL: https://returnmytime.com/prompt-library/launch-startup-success-strategy
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: business-creation, strategy, ai
- Published: 2026-07-24
- Provenance: current
- Source: https://returnmytime.com/prompt-library/launch-startup-success-strategy
- Source captured: 2026-07-24T00:00:00Z


Evaluate a startup idea's market, risks, MVP, customer acquisition, monetization, and path from concept to first revenue.

#### What it does

- Maps a product or service idea through market validation, risks, MVP design, acquisition, monetization, and first revenue.

#### When to use

- When converting a selected startup idea into a practical early-stage action plan.


#### Required inputs

- NEW PRODUCT/SERVICE IDEA

#### Prompt

#CONTEXT:
Adopt the role of a seasoned entrepreneur who has built multiple successful companies from the ground up. Your task is to help the user analyze a novel product or service idea and provide guidance on how to turn it into a successful startup.

#ROLE:
You are an entrepreneur with a keen eye for identifying untapped market opportunities and turning innovative ideas into thriving businesses. 

#RESPONSE GUIDELINES:
Provide a comprehensive analysis of the product/service idea, covering the following key areas:

1. Market Validation: Analyze the market opportunity and target customer
2. Value Proposition: Highlight the unique value proposition of the product/service
3. Risks and Challenges: Identify key risks and challenges in bringing this idea to market
4. MVP Advice: Provide advice for developing a minimum viable product (MVP)
5. Customer Acquisition: Offer strategies for acquiring initial customers and driving growth
6. Monetization: Discuss potential monetization models and revenue streams
7. Idea to Revenue: Give step-by-step guidance to go from initial idea to generating revenue

#TASK CRITERIA:
1. Validate the market opportunity thoroughly
2. Identify and address potential risks and challenges
3. Provide actionable advice for each stage of the startup journey
4. Focus on the most critical aspects of building a successful business
5. Avoid generic or overly simplistic recommendations

#INFORMATION ABOUT ME:
● Product/Service Idea: [NEW PRODUCT/SERVICE IDEA]

#RESPONSE FORMAT:
Organize the response using clear headings for each section outlined in the #RESPONSE GUIDELINES. Use bullet points and concise paragraphs to present the information in an easily digestible format. Avoid using XML tags in the response.

#### Expected output

- A structured startup strategy with validation, risk, MVP, acquisition, monetization, and staged execution recommendations.

#### Customization tips

- Add launch geography, target buyer, budget, deadline, and existing evidence.
- Convert recommendations into small tests with owners, costs, dates, and pass/fail criteria.

#### Guardrails

- Separate evidence from assumptions and validate demand before committing material capital or making performance claims.

#### Example input

[NEW PRODUCT/SERVICE IDEA]: A subscription mobile bicycle-repair service sold to midsize employers as a commuter benefit.

#### Related prompts

- https://returnmytime.com/prompt-library/analyze-business-ideas
- https://returnmytime.com/prompt-library/create-professional-pitch-decks


### Analyze Business Ideas

- URL: https://returnmytime.com/prompt-library/analyze-business-ideas
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 1-2 hours
- Tags: business-creation, strategy, ai
- Published: 2026-07-24
- Provenance: current
- Source: https://returnmytime.com/prompt-library/analyze-business-ideas
- Source captured: 2026-07-24T00:00:00Z


Analyze a business idea with a structured framework covering market opportunity, positioning, risks, execution priorities, and next steps.

#### What it does

- Applies a pragmatic framework to a business idea's market, customer personas, risks, alternative models, and overall viability.

#### When to use

- When seeking a blunt first-pass critique before deeper market research.


#### Required inputs

- Business idea to analyze (provide it alongside the prompt)

#### Prompt

#CONTEXT:
You are adopting the role of a pragmatic business strategist with expertise in dissecting business ideas for real-world applicability. Your task is to analyze a given business idea objectively, considering its genuine merits and potential pitfalls. Assume the roles of theoretical personas, offering realistic feedback on the idea's utility or lack thereof. Provide a blunt, balanced validation and recommendation.

#ROLE:
Pragmatic business strategist with expertise in dissecting business ideas for real-world applicability.

#RESPONSE GUIDELINES: 
1. Provide an overview of the business idea.
2. Identify potential markets for the idea.
3. Create three theoretical personas, each with:
   ● Age
   ● Occupation
   ● Pain points
   ● Validation statement including the problem, solution effectiveness, and opinion
4. Discuss market risks.
5. Suggest alternative business models.
6. Give a final validation and recommendation.

#BUSINESS IDEA CRITERIA:
1. Analyze the idea objectively, considering both merits and pitfalls.
2. Provide realistic feedback from the perspective of theoretical personas.
3. Offer a blunt, balanced validation and recommendation.
4. Focus on the idea's real-world applicabilit

#### Expected output

- A balanced idea analysis with market possibilities, three theoretical personas, risks, alternatives, and a recommendation.

#### Customization tips

- State the buyer, user, geography, price, delivery model, and evidence already gathered.
- Use the result to define validation questions rather than treating fictional personas as customer research.

#### Guardrails

- Theoretical personas and recommendations are hypotheses; validate them with real customers and current market evidence.

#### Example input

BUSINESS IDEA: A membership that gives urban commuters priority access to mobile bicycle repairs at their workplace.

#### Related prompts

- https://returnmytime.com/prompt-library/conduct-market-research-validation
- https://returnmytime.com/prompt-library/generate-ai-business-ideas


### Generate Customer Refund Emails

- URL: https://returnmytime.com/prompt-library/generate-customer-refund-emails
- Category: Customer Support
- Workflow: customer-support-service-recovery
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Quality
- Estimated time saved: 2+ hours
- Tags: customer-support, refunds, email
- Published: 2026-07-30
- Modified: 2026-07-31
- Provenance: archive
- Source: https://web.archive.org/web/20260307083326id_/https://returnmytime.com/prompt-library/generate-customer-refund-emails
- Source captured: 2026-03-07T08:33:26Z
- Source hash: bedb3e871ea6fd90c81b86613698cd7f15e720f70aacbc6bb0e37a66d667a65d


Draft a customer refund email from verified policy, approved eligibility, amount, timing, and next steps, then send it only after human review.

#### What it does

- Turns an approved refund decision and verified case details into a clear customer email draft.
- Organizes the acknowledgement, refund details, next steps, support path, and professional close.

#### When to use

- After an authorized person has reviewed the request and approved the refund decision and customer-facing details.

#### When not to use

- Do not use it to determine eligibility, choose a refund amount, override policy, change an account, or send a message automatically.

#### Required inputs

- CUSTOMER REQUEST
- REFUND POLICY
- APPROVED ELIGIBILITY DECISION
- APPROVED REFUND AMOUNT
- APPROVED ITEMS
- PROCESSING TIME
- PAYMENT METHOD
- APPROVED NEXT STEPS
- SUPPORT CONTACT
- COMPANY TONE
- SIGNATURE

#### Prompt

#CONTEXT:
You are drafting a response to a customer refund request. Use only the verified case facts and approvals supplied below. The result is a draft for human review, not authorization to issue a refund or contact the customer.

#ROLE:
Act as an empathetic customer-service email writer. Be concise, specific, and professional.

#RESPONSE GUIDELINES:
1. Acknowledge the customer's actual request without adding facts.
2. State only the supplied eligibility decision, approved refund amount, approved items, processing time, and payment method.
3. Present the approved next steps clearly.
4. Include the supplied support contact and signature.
5. If an essential fact is missing or conflicting, flag it for the reviewer instead of guessing.
6. Do not send the email. Label the response "DRAFT — HUMAN APPROVAL REQUIRED."

#INFORMATION ABOUT ME:
- Customer request: [CUSTOMER REQUEST]
- Applicable refund policy: [REFUND POLICY]
- Approved eligibility decision: [APPROVED ELIGIBILITY DECISION]
- Approved refund amount: [APPROVED REFUND AMOUNT]
- Approved items: [APPROVED ITEMS]
- Verified processing time: [PROCESSING TIME]
- Payment method description: [PAYMENT METHOD]
- Approved next steps: [APPROVED NEXT STEPS]
- Support contact: [SUPPORT CONTACT]
- Company tone: [COMPANY TONE]
- Signature: [SIGNATURE]

#RESPONSE FORMAT:
DRAFT — HUMAN APPROVAL REQUIRED
Subject: [SUBJECT]

Email:
[CUSTOMER-READY DRAFT]

Reviewer checks:
- Eligibility, amount, items, timing, and policy match the approved inputs
- No unsupported promise, admission, exception, or account action
- Approved next steps and contact details are accurate

#### Expected output

- A clearly labeled refund email draft plus a reviewer checklist for the approved case facts.

#### Customization tips

- Add the customer name and order number when they are verified.
- Replace broad policy text with the exact rule that an authorized reviewer applied.

#### Guardrails

- Human approval is required for refund amounts, policy exceptions, admissions, promises, account changes, retention offers, and external sending.
- Do not expose full payment details or other unnecessary personal information.

#### Example input

Customer request: Order 1048 arrived damaged and the customer asked for a refund. Policy: damaged items qualify after photo review. Approved decision: full refund approved after photo verification. Approved amount: $48. Approved item: ceramic desk lamp. Processing time: 5–7 business days. Payment method: original card. Next steps: no return required; retain the confirmation email. Support: support@example.com. Tone: calm and direct. Signature: Maya, Customer Care.

#### Related prompts

- https://returnmytime.com/prompt-library/reply-to-customer-complaints

#### Related articles

- https://returnmytime.com/blog/four-10-minute-ai-quick-wins

### Reply to Customer Complaints

- URL: https://returnmytime.com/prompt-library/reply-to-customer-complaints
- Category: Customer Support
- Workflow: customer-support-service-recovery
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Quality
- Estimated time saved: 2+ hours
- Tags: customer-support, complaints, service-recovery
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260308003059id_/https://returnmytime.com/prompt-library/reply-to-customer-complaints
- Source captured: 2026-03-08T00:30:59Z
- Source hash: 9753b29d1586e5a26e164b041ffc88f51d8523626da768eed7decca182dd7db3


Draft a customer complaint reply from verified facts, policy, and approved resolution options. Flag gaps for a reviewer before anyone sends it.

#### What it does

- Structures a complaint response that acknowledges the issue and provides only approved next steps.
- Separates missing facts and escalation needs from customer-ready draft language.

#### When to use

- When a complaint needs a consistent first draft after the known facts, policy, and permitted resolution options have been gathered.

#### When not to use

- Do not use it to admit liability, invent a resolution, make a retention offer, change an account, or bypass escalation rules.

#### Required inputs

- CUSTOMER COMPLAINT
- VERIFIED CASE FACTS
- COMPANY POLICY
- APPROVED RESOLUTION OPTIONS
- ESCALATION OWNER
- COMPANY TONE
- SUPPORT CONTACT
- SIGNATURE

#### Prompt

#CONTEXT:
You are drafting a response to a customer complaint. Use only the supplied complaint, verified facts, policy, and approved resolution options. The response is a draft for human review.

#ROLE:
Act as an empathetic customer-service representative who writes in the company's stated tone without mirroring hostility or making unsupported commitments.

#RESPONSE GUIDELINES:
1. Acknowledge the specific complaint and its reported impact.
2. Express empathy without making a legal admission or claiming unverified fault.
3. Separate confirmed facts from questions that still need investigation.
4. Offer only an approved resolution option or route the case to the named escalation owner.
5. Include the supplied support contact and signature.
6. Do not send the response. Label it "DRAFT — HUMAN APPROVAL REQUIRED."

#INFORMATION ABOUT ME:
- Customer complaint: [CUSTOMER COMPLAINT]
- Verified case facts: [VERIFIED CASE FACTS]
- Applicable company policy: [COMPANY POLICY]
- Approved resolution options: [APPROVED RESOLUTION OPTIONS]
- Escalation owner: [ESCALATION OWNER]
- Company tone: [COMPANY TONE]
- Support contact: [SUPPORT CONTACT]
- Signature: [SIGNATURE]

#RESPONSE FORMAT:
DRAFT — HUMAN APPROVAL REQUIRED
Subject: [SUBJECT]

Customer response:
[CUSTOMER-READY DRAFT]

Open questions:
- [MISSING OR CONFLICTING FACTS]

Reviewer checks:
- The response matches verified facts and policy
- Any resolution or escalation is explicitly approved
- No unsupported admission, promise, exception, account action, or offer

#### Expected output

- A labeled complaint-response draft, a list of unresolved questions, and a human-review checklist.

#### Customization tips

- Include the customer's preferred name and case number when verified.
- State a response deadline only when an owner has approved it.

#### Guardrails

- Human approval is required for refund amounts, policy exceptions, admissions, promises, account changes, retention offers, and external sending.
- Do not infer intent, fault, legal responsibility, or compensation from the complaint alone.

#### Example input

Complaint: A customer says two scheduled coaching calls started late. Verified facts: one call began 12 minutes late; the second began on time. Policy: service concerns are reviewed by the program manager. Approved options: apologize for the verified delay and offer a manager follow-up; no credit has been approved. Escalation owner: Program Manager. Tone: direct, calm, friendly. Support: support@example.com. Signature: Maya, Customer Care.

#### Related prompts

- https://returnmytime.com/prompt-library/generate-customer-refund-emails


### Create Targeted Instagram Ads

- URL: https://returnmytime.com/prompt-library/create-targeted-instagram-ads
- Category: Marketing Funnel
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 2+ hours
- Tags: marketing-funnel, instagram, advertising
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307091006id_/https://returnmytime.com/prompt-library/create-targeted-instagram-ads
- Source captured: 2026-03-07T09:10:06Z
- Source hash: c2a5aefb0d51cf77ef56c8b7c8186ce659a1258d49f07e7500e80f420d1e97f4


Draft Instagram ad concepts around a defined audience, verified benefits, approved proof, offer, and next step.

#### What it does

- Creates review-ready Instagram ad-copy variations from supplied offer and audience evidence.
- Separates the draft copy from targeting assumptions and pre-publication checks.

#### When to use

- When you have a defined offer, target audience, verified benefits, approved proof, and a clear next step to test.

#### When not to use

- Do not use it as a substitute for platform-policy review, audience research, legal review, or campaign testing.

#### Required inputs

- PRODUCT OR SERVICE
- TARGET AUDIENCE
- VERIFIED BENEFITS
- APPROVED PROOF
- APPROVED OFFER
- CALL TO ACTION
- BRAND TONE
- CAMPAIGN CONSTRAINTS
- SUCCESS METRIC
- REVIEW DATE

#### Prompt

#CONTEXT:
You are creating Instagram ad-copy concepts for a specific product or service and target audience. Use only the supplied benefits, proof, offer, and constraints. The concepts are drafts to test, not performance predictions.

#ROLE:
Act as a direct-response copywriter who prioritizes clear, concise claims that can be verified before publication.

#RESPONSE GUIDELINES:
1. Create three distinct ad concepts, each built around one supplied audience need.
2. Include a headline, primary text, call to action, creative direction, and the verified benefit used.
3. Use approved proof only; omit social proof when none is supplied.
4. Do not invent scarcity, testimonials, results, audience attributes, or targeting capabilities.
5. Flag platform specifications, policy, disclosure, and character limits for current verification before launch.

#INFORMATION ABOUT ME:
- Product or service: [PRODUCT OR SERVICE]
- Target audience: [TARGET AUDIENCE]
- Verified benefits: [VERIFIED BENEFITS]
- Approved proof: [APPROVED PROOF]
- Approved offer: [APPROVED OFFER]
- Call to action: [CALL TO ACTION]
- Brand tone: [BRAND TONE]
- Campaign constraints: [CAMPAIGN CONSTRAINTS]
- User-provided success metric: [SUCCESS METRIC]
- User-provided review date: [REVIEW DATE]

#RESPONSE FORMAT:
Concept [NUMBER]
- Audience need: [NEED]
- Headline: [HEADLINE]
- Primary text: [COPY]
- Call to action: [CTA]
- Creative direction: [VISUAL IDEA]
- Verified claim used: [SUPPORTED CLAIM]
- Pre-launch checks: [POLICY, SPECIFICATION, DISCLOSURE, AND TRACKING CHECKS]

Test plan:
- [VARIABLE TO TEST]
- [SUCCESS METRIC]
- [REVIEW DATE]

#### Expected output

- Three evidence-bounded Instagram ad concepts with creative directions and a current-policy verification checklist.

#### Customization tips

- Test one audience problem or benefit at a time so results are interpretable.
- Provide approved proof verbatim rather than asking the model to create it.

#### Guardrails

- Verify current Instagram specifications, advertising policies, disclosures, and targeting options before publication.
- Do not invent testimonials, urgency, outcomes, or sensitive audience attributes.

#### Example input

Product: a 12-week career-transition coaching program. Audience: U.S. corporate managers exploring a first consulting business. Verified benefits: weekly coaching, planning templates, and peer sessions. Approved proof: none. Offer: free fit call; no discount or deadline. CTA: Book a fit call. Tone: direct and encouraging. Constraints: avoid income claims; confirm current Instagram ad specifications and policies before launch. Success metric: qualified fit-call booking rate. Review date: August 30, 2026.

#### Related prompts

- https://returnmytime.com/prompt-library/brainstorm-lead-magnets
- https://returnmytime.com/prompt-library/create-engaging-social-media-bios


### Create Engaging Social Media Bios

- URL: https://returnmytime.com/prompt-library/create-engaging-social-media-bios
- Category: Brand & Audience Building
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 2+ hours
- Tags: brand-audience-building, social-media, bios
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307094526id_/https://returnmytime.com/prompt-library/create-engaging-social-media-bios
- Source captured: 2026-03-07T09:45:26Z
- Source hash: 9e22542adb99c8e378c445656c4783471e72da75350805dda4eb1f8f17c973f4


Draft platform-specific social bios from a verified role, audience, value proposition, credentials, tone, and next step.

#### What it does

- Creates several concise bio options for each named social platform.
- Maps every credibility statement to user-provided facts and flags current platform constraints for review.

#### When to use

- When a person or brand has a clear audience, verified credentials, value proposition, tone, and desired next step.

#### When not to use

- Do not use it to invent credentials, awards, audience results, keywords, or current platform limits.

#### Required inputs

- NAME OR BRAND
- ROLE
- TARGET AUDIENCE
- VERIFIED CREDENTIALS
- VALUE PROPOSITION
- BIO GOAL
- TARGET PLATFORMS
- DESIRED TONE
- CALL TO ACTION

#### Prompt

#CONTEXT:
You are drafting concise social-media bios for a person or brand. Each bio must use only the supplied facts and fit the stated audience, goal, and tone.

#ROLE:
Act as a social profile copywriter. Prioritize clarity and specificity over slogans.

#RESPONSE GUIDELINES:
1. Create three bio variations for each target platform.
2. Include the role, audience relevance, value proposition, and call to action when space permits.
3. Use verified credentials only. Omit any fact that is not supplied.
4. Explain the tradeoff behind each variation.
5. Treat character limits, platform features, and profile conventions as items to verify at publication time.

#INFORMATION ABOUT ME:
- Name or brand: [NAME OR BRAND]
- Role: [ROLE]
- Target audience: [TARGET AUDIENCE]
- Verified credentials: [VERIFIED CREDENTIALS]
- Value proposition: [VALUE PROPOSITION]
- Bio goal: [BIO GOAL]
- Target platforms: [TARGET PLATFORMS]
- Desired tone: [DESIRED TONE]
- Call to action: [CALL TO ACTION]

#RESPONSE FORMAT:
Platform: [PLATFORM]

Option [NUMBER]
- Bio: [BIO]
- Approximate character count: [COUNT]
- Positioning choice: [RATIONALE]
- Facts used: [SUPPORTED FACTS]

Publication review:
- Current platform limit and feature check: [CHECK]
- Credential and claim verification: [CHECK]
- Link and call-to-action verification: [CHECK]

#### Expected output

- Three fact-bounded bio options per platform with rationale and a publication-review checklist.

#### Customization tips

- Supply a single, audience-specific value proposition instead of a broad list of services.
- Verify character counts in the live platform editor before publishing.

#### Guardrails

- Review credentials, claims, platform limits, link behavior, and profile features before publication.
- Do not invent achievements, results, keywords, hashtags, or endorsements.

#### Example input

Name or brand: Northstar Operations. Role: operations consultant. Audience: owners of 10–50 person professional-services firms. Credentials: 12 years in service operations; no awards or client-result claims approved. Value proposition: helps teams document handoffs and reduce avoidable rework. Goal: make the profile relevant to qualified owners. Platforms: LinkedIn and Instagram. Tone: direct and warm. CTA: Read the operations checklist.

#### Related prompts

- https://returnmytime.com/prompt-library/create-targeted-instagram-ads
- https://returnmytime.com/prompt-library/brainstorm-lead-magnets


### Brainstorm Lead Magnets

- URL: https://returnmytime.com/prompt-library/brainstorm-lead-magnets
- Category: Marketing Funnel
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 1-2 hours
- Tags: marketing-funnel, lead-magnets, audience
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307092842id_/https://returnmytime.com/prompt-library/brainstorm-lead-magnets
- Source captured: 2026-03-07T09:28:42Z
- Source hash: 5543bec1aefa1d47fb331c9ef6b5e2ca5f267692310b78dff3553a80297b8af2


Generate lead-magnet concepts tied to a defined audience need, an approved paid offer, verified expertise, and realistic delivery constraints.

#### What it does

- Generates several lead-magnet concepts with titles, content outlines, delivery formats, and offer alignment.
- Makes assumptions and validation questions visible before content production begins.

#### When to use

- When an audience profile and paid offer are clear enough to compare useful opt-in ideas.

#### When not to use

- Do not use it to claim audience demand, conversion performance, or expertise that has not been researched or verified.

#### Required inputs

- NICHE
- IDEAL SUBSCRIBER
- PAID OFFERS
- VERIFIED EXPERTISE
- DELIVERY CONSTRAINTS

#### Prompt

#CONTEXT:
You are helping an online business owner brainstorm lead magnets that address a defined subscriber need and connect naturally to an approved paid offer.

#ROLE:
Act as a lead-magnet strategist. Generate useful, bounded concepts that the business can validate before producing.

#RESPONSE GUIDELINES:
1. Brainstorm three to five distinct concepts.
2. For each, provide a benefit-led working title, the problem addressed, a short content outline, a realistic delivery format, and the connection to a paid offer.
3. Use only the supplied expertise and audience information.
4. State assumptions and propose a lightweight way to test demand.
5. Avoid unsupported performance claims, false urgency, or promises that the asset will convert.

#INFORMATION ABOUT ME:
- Niche: [NICHE]
- Ideal subscriber, including pain points and desired outcomes: [IDEAL SUBSCRIBER]
- Paid offers: [PAID OFFERS]
- Verified expertise and source material: [VERIFIED EXPERTISE]
- Delivery constraints: [DELIVERY CONSTRAINTS]

#RESPONSE FORMAT:
Concept [NUMBER]
- Working title: [TITLE]
- Subscriber problem: [PROBLEM]
- Useful outcome: [OUTCOME]
- Content outline: [OUTLINE]
- Delivery format: [FORMAT]
- Connection to paid offer: [CONNECTION]
- Assumptions to verify: [ASSUMPTIONS]
- Demand test: [SMALL TEST]

Comparison:
- Best concept to test first: [CONCEPT]
- Evidence needed before production: [EVIDENCE]

#### Expected output

- Three to five lead-magnet concepts with outlines, offer connections, assumptions, and small demand tests.

#### Customization tips

- Provide real customer questions or interview notes to improve specificity.
- Choose a delivery format based on production capacity and observed audience preference.

#### Guardrails

- Validate audience demand and offer alignment with real feedback; do not present model-generated ideas as market evidence.
- Review all claims and source material before publishing the lead magnet.

#### Example input

Niche: career-transition coaching. Ideal subscriber: corporate professionals who struggle to choose a first consulting offer and want a low-risk validation path. Paid offer: a 12-week career-pivot coaching program. Verified expertise: an approved workshop transcript on offer selection and five anonymized client questions. Delivery constraints: a PDF or three-email mini-course that can be produced in one week.

#### Related prompts

- https://returnmytime.com/prompt-library/create-targeted-instagram-ads
- https://returnmytime.com/prompt-library/generate-product-unique-selling-proposition


### Generate a Product Unique Selling Proposition

- URL: https://returnmytime.com/prompt-library/generate-product-unique-selling-proposition
- Category: Product Development
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: product-development, positioning, usp
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260308010809id_/https://returnmytime.com/prompt-library/generate-product-unique-selling-proposition
- Source captured: 2026-03-08T01:08:09Z
- Source hash: 3c424ccb1898f82c4048aff793e4c687bc1d798c8f80da06783cb5b2a2808c82


Draft and compare product USP options from verified features, benefits, customer evidence, and competitor source material.

#### What it does

- Turns product facts and customer evidence into several concise USP options.
- Separates supported differentiators from assumptions that require testing.

#### When to use

- When product features, audience needs, customer evidence, and relevant competitor material are available for comparison.

#### When not to use

- Do not use it to declare an unverified market advantage or replace customer and competitor research.

#### Required inputs

- PRODUCT DESCRIPTION
- TARGET AUDIENCE
- VERIFIED FEATURES AND BENEFITS
- CUSTOMER EVIDENCE
- COMPETITOR SOURCE MATERIAL

#### Prompt

#CONTEXT:
You are helping define a product's unique selling proposition using supplied product facts, customer evidence, and competitor source material.

#ROLE:
Act as a product-positioning strategist. Distinguish supported facts from assumptions and write options that can be tested.

#RESPONSE GUIDELINES:
1. Summarize the product, audience, and customer problem from the supplied inputs.
2. Map each verified feature to a practical or emotional benefit supported by customer evidence.
3. Compare only against the provided competitor source material and cite the relevant source beside each comparison.
4. Draft three concise USP options with different positioning angles.
5. Explain the evidence supporting each option and identify assumptions to test.
6. Do not claim "best," "only," category leadership, or assured outcomes without substantiation.

#INFORMATION ABOUT ME:
- Product description: [PRODUCT DESCRIPTION]
- Target audience: [TARGET AUDIENCE]
- Verified features and benefits: [VERIFIED FEATURES AND BENEFITS]
- Customer evidence: [CUSTOMER EVIDENCE]
- Competitor source material: [COMPETITOR SOURCE MATERIAL]

#RESPONSE FORMAT:
Evidence summary:
- Product and audience: [SUMMARY]
- Supported benefits: [BENEFITS WITH EVIDENCE]
- Competitor comparison: [COMPARISON WITH PROVIDED SOURCES]
- Unresolved assumptions: [ASSUMPTIONS]

USP option [NUMBER]:
- Statement: [USP]
- Positioning angle: [ANGLE]
- Supporting evidence: [EVIDENCE]
- Risk or assumption: [RISK]
- Test: [VALIDATION TEST]

Recommendation:
- Best option to test: [OPTION]
- Why: [RATIONALE]
- Evidence still needed: [EVIDENCE GAP]

#### Expected output

- Three evidence-linked USP options with assumptions, risks, and proposed validation tests.

#### Customization tips

- Provide dated competitor pages and direct customer language rather than summaries alone.
- Test the leading USP in interviews or a small message experiment before standardizing it.

#### Guardrails

- Review every differentiator and comparative claim against current source material before publication.
- Do not treat model-generated positioning as proof of customer demand or product superiority.

#### Example input

Product: a workshop and template kit for documenting service handoffs. Audience: owners of professional-services firms with 10–50 employees. Verified features: facilitated mapping workshop, editable SOP templates, and a 30-day review checklist. Customer evidence: five discovery interviews describe repeated rework during handoffs. Competitor sources: dated links and notes for two publicly available workflow-template products.

#### Related prompts

- https://returnmytime.com/prompt-library/brainstorm-lead-magnets


### Generate Actionable Business Ideas

- URL: https://returnmytime.com/prompt-library/generate-actionable-business-ideas
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: business-creation, generate, actionable, business
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307095209id_/https://returnmytime.com/prompt-library/generate-actionable-business-ideas
- Source captured: 2026-03-07T09:52:09Z
- Source hash: 4c73c1576d511e5d1d0be7e9af8afa026c783cbab73bde57749b62ec786681d2


Generate a comprehensive business plan with the prompt template, tailored to your skills and resources. It provides 20 actionable business ideas, complete with creation steps and alignment to your capabilities, all organized in an easy-to-read table format.

#### What it does

- Generates a list of 20 actionable business ideas tailored to the user's specific skills and resources.
- Explains the steps to create each business idea and why it aligns with the user's skillset.
- Presents the information in a table format for clear and organized viewing.

#### When to use

- Use when you have a defined skillset and resource inventory and need 20 executable business concepts to compare.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting actionable business ideas.

#### Required inputs

- MY SKILLSET
- MY AVAILABLE RESOURCES

#### Prompt

#CONTEXT:
Adopt the role of an expert consultant specializing in career development, skills assessment, and generating personalized business ideas tailored to an individual's unique profile. Your task is to help the user generate highly relevant, actionable business ideas based on their skillset and available resources.

#ROLE:
You are an expert consultant specializing in career development, skills assessment, and generating personalized business ideas tailored to an individual's unique profile.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

- Thoroughly analyze the provided skillset and available resources
- Generate 20 highly relevant, actionable business ideas
- For each idea, explain the specific steps to create the business
- Provide a clear rationale for why each idea is well-suited to the individual's skillset
- Present the information in a well-organized table format with the following columns:
  - Business Idea Name
  - Creation Steps (5 steps)
  - Skillset Match Rationale

#TASK CRITERIA:
- Focus on generating ideas that leverage the individual's unique skillset and available resources
- Provide detailed, step-by-step guidance for creating each business
- Explain why each idea fits the individual's supplied skills and resources
- Avoid generic or one-size-fits-all ideas
- Ensure each idea is actionable and feasible given the provided information

#INFORMATION ABOUT ME:
- My skillset: [MY SKILLSET]
- My available resources: [MY AVAILABLE RESOURCES]

#RESPONSE FORMAT:
Expected output: A 20-row business-idea table with five creation steps and a skillset-match rationale for each idea.

| Business Idea Name | Creation Steps | Skillset Match Rationale |
|--------------------|----------------|--------------------------|
| Idea 1             | 1. Step 1      | Rationale for Idea 1     |
|                    | 2. Step 2      |                          |
|                    | 3. Step 3      |                          |
|                    | 4. Step 4      |                          |
|                    | 5. Step 5      |                          |
| Idea 2             | 1. Step 1      | Rationale for Idea 2     |
|                    | 2. Step 2      |                          |
|                    | 3. Step 3      |                          |
|                    | 4. Step 4      |                          |
|                    | 5. Step 5      |                          |
| ...                | ...            | ...                      |
| Idea 20            | 1. Step 1      | Rationale for Idea 20    |
|                    | 2. Step 2      |                          |
|                    | 3. Step 3      |                          |
|                    | 4. Step 4      |                          |
|                    | 5. Step 5      |                          |

#### Expected output

- A 20-row business-idea table with five creation steps and a skillset-match rationale for each idea.

#### Customization tips

- Fill in the “my skillset” and “my available resources” placeholders with specific details about your skills and the resources you have access to.
- Example: If your skillset includes digital marketing and graphic design, and your available resources include a high-end computer and design software, mention these details to tailor the business ideas to your capabilities.
- Use the prompt to generate a list of business ideas that are directly aligned with what you can realistically start and manage based on your personal qualifications and assets.
- Request the creation of a table format to organize the business ideas, steps to create them, and the reasons they match your skillset, ensuring clarity and ease of reference.
- Identify Your Skills: Before using this prompt, list your skills and resources clearly to ensure the business ideas provided are tailored accurately to your capabilities.
- Research Market Demand: After receiving business ideas, research the current market demand for these services or products to ensure viability and potential profitability.
- Develop a Business Plan: For each business idea, create a detailed business plan that includes steps for implementation, marketing strategies, and financial projections. Identify Your Target Audience: Understand the specific demographic or market segment that your business idea will cater to, and tailor your products or services accordingly.
- Consider Scalability: Evaluate the potential for growth and scalability of each business idea, ensuring that it can be expanded or adapted as your business grows.
- Seek Feedback: Share your business ideas with trusted mentors, advisors, or potential customers to gather valuable feedback and insights for improvement.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My skillset: Facilitation, curriculum design, career coaching, and B2B workshop delivery
- My available resources: $8,000 launch budget, 15 hours per week, an email list of 1,200, and two contractor partners

#### Related prompts

- https://returnmytime.com/prompt-library/analyze-business-ideas
- https://returnmytime.com/prompt-library/conduct-market-research-validation


### Generate Honest Idea Feedback

- URL: https://returnmytime.com/prompt-library/generate-honest-idea-feedback
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 1-2 hours
- Tags: business-creation, generate, honest, idea
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307082440id_/https://returnmytime.com/prompt-library/generate-honest-idea-feedback
- Source captured: 2026-03-07T08:24:40Z
- Source hash: d82db4dbfa57142d9920da618bc25139244c445c8c7a7634a8c8419edaffcd6f


Use the prompt template to obtain expert-level, unbiased feedback on your business ideas. This structured feedback process involves in-depth questioning and detailed evaluation, helping solopreneurs make informed decisions about their ventures.

#### What it does

- Evaluates the potential of a solopreneur's new idea through a structured feedback process.
- Provides objective and unbiased feedback, including a numerical score and detailed advantages and disadvantages.
- Guides the solopreneur on next steps with creative and unconventional advice tailored to the specific idea.

#### When to use

- Use when you want an idea challenged through diagnostic questions before investing time or money.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting honest idea feedback.

#### Required inputs

- MY IDEA

#### Prompt

#CONTEXT:
You are Honest Feedback GPT, a seasoned Solopreneur who helps Solopreneurs get honest feedback on their ideas. You are a world-class expert in identifying the advantages and disadvantages of any idea.

#ROLE:
Adopt the role of an objective, unbiased expert who provides specific, actionable, and honest feedback on Solopreneurs' ideas to help them make informed decisions.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Ask 5 in-depth questions about the idea to understand how promising it is
2. Provide honest feedback, including:
   - Idea score from 0 to 10
   - Advantages
   - Disadvantages
   - Recommended next steps
3. Be concise, straightforward, and avoid platitudes or meaningless phrases
4. Recommend creative and unconventional next steps, not trivial advice

#HONEST FEEDBACK CRITERIA:
- Be as objective and unbiased as possible
- Ask questions that help understand the idea's potential
- Provide specific and actionable feedback, even if negative
- Avoid flattery and meaningless phrases
- Recommend creative and unconventional next steps

#INFORMATION ABOUT ME:
- My idea: [MY IDEA]

#RESPONSE FORMAT:
Expected output: Five diagnostic questions followed by a 0–10 idea score, advantages, disadvantages, and recommended next steps.

1. Greet the Solopreneur and confirm readiness to start
2. Ask 5 questions about the idea and wait for answers
3. Provide honest feedback in the following format:
   - Idea score: <0-10>
   - Advantages:
     - <Advantage 1>
     - <Advantage 2>
     ...
   - Disadvantages:
     - <Disadvantage 1>
     - <Disadvantage 2>
     ...
   - Recommended next steps:
     - <Step 1>
     - <Step 2>
     ...

#### Expected output

- Five diagnostic questions followed by a 0–10 idea score, advantages, disadvantages, and recommended next steps.

#### Customization tips

- Fill in the “idea” placeholder with a concise description of your new idea. Example: "My new idea is a subscription-based platform that connects freelance graphic designers with small businesses needing design services."
- Use the structured feedback process outlined in the prompt to guide your interaction. Begin by sharing your idea, then follow the steps of asking questions, receiving answers, and providing a detailed feedback score, advantages, disadvantages, and recommended next steps.
- Maintain the criteria for honest feedback throughout the interaction by focusing on objectivity, asking in-depth questions, and providing specific, actionable, and creative advice without flattery or trivial suggestions.
- Prepare insightful questions: Develop questions that delve into the scalability, market demand, and unique value proposition of the idea to thoroughly understand its potential.
- Focus on critical analysis: When giving feedback, emphasize both the strengths and weaknesses clearly and objectively, ensuring the feedback is balanced and actionable.
- Suggest innovative next steps: Propose next steps that challenge conventional approaches, such as testing the idea in an unconventional market or using a novel marketing strategy. Be an active listener: Pay close attention to the information shared by the entrepreneur and ask follow-up questions to gain a deeper understanding of the idea.
- Consider the market potential: Evaluate the idea's fit in the current market landscape, considering factors such as competition, target audience, and potential demand.
- Provide specific and actionable feedback: Avoid generic or vague feedback and instead offer concrete suggestions for improvement or areas to capitalize on.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My idea: A subscription service that turns recorded client-delivery calls into reviewed SOP drafts for boutique consulting firms

#### Related prompts

- https://returnmytime.com/prompt-library/analyze-business-ideas
- https://returnmytime.com/prompt-library/conduct-market-research-validation


### Generate Industry Gap Analysis

- URL: https://returnmytime.com/prompt-library/generate-industry-gap-analysis
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: business-creation, generate, industry, analysis
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260308004311id_/https://returnmytime.com/prompt-library/generate-industry-gap-analysis
- Source captured: 2026-03-08T00:43:11Z
- Source hash: d37d02ffaeb9c1be70706b5f3c55b3171041d6d651f21d5cbfea232b17890583


Generate a comprehensive analysis using the archived prompt template. to identify gaps in your industry, complete with interactive charts and graphs. This tool provides deep research insights and strategic recommendations tailored for solopreneurs, helping you effectively penetrate and fill market gaps.

#### What it does

- Identifies gaps in the specified industry through detailed research.
- Generates clear chart specifications and graphs to visually represent the research findings.
- Offers strategic recommendations for solopreneurs to penetrate and fill identified industry gaps, including source citations.

#### When to use

- Use when selecting an opportunity within a named industry and current market evidence can be supplied or verified.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting industry gap analysis.

#### Required inputs

- MY TARGET INDUSTRY

#### Prompt

#CONTEXT:
Adopt the role of an expert business strategist and researcher with deep knowledge across industries. Your task is to help a solopreneur trying to identify potential gaps and opportunities in a given industry through comprehensive research and analysis, and provide strategic recommendations to effectively enter the market and capitalize on the identified gaps.

#ROLE:
You are an expert business strategist and researcher with deep knowledge across industries.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

Organize the information in the following sections:

1. Industry Overview
Goal: Provide a high-level overview of the target industry.

2. Market Size and Growth
Goal: Present the current market size and projected growth using interactive charts.

3. Key Players and Market Share
Goal: Identify the major players in the industry and their market share using interactive graphs.

4. Identified Gaps and Opportunities
Goal: Describe 3 potential gaps or opportunities in the market, including:
- Description of the gap
- Estimated market size of the gap
- Potential of the gap

5. Recommendations for Solopreneur
Goal: Provide strategic recommendations for the solopreneur to capitalize on the most promising gap, including:
- Description of the target gap
- Proposed product or service offering
- Unique value proposition
- Go-to-market strategy
- Key partnerships and resources needed

6. Sources
Goal: Cite all sources used in the research.

#TASK CRITERIA:
1. Conduct comprehensive research and analysis of the given industry.
2. Identify potential gaps and opportunities based on the research.
3. Provide data-driven insights and recommendations.
4. Present findings using interactive data visualizations where relevant.
5. Focus on actionable strategies for a solopreneur to enter the market.
6. Cite all sources to ensure credibility of the information provided.

#INFORMATION ABOUT ME:
- My target industry: [MY TARGET INDUSTRY]

#RESPONSE FORMAT:
Expected output: An industry overview with market size and growth, key players, three evidence-backed gaps, opportunity estimates, and source notes.

Industry Overview
<Insert industry overview>

Market Size and Growth
<Insert market size and growth information>
<Insert interactive chart showing market size and projected growth>

Key Players and Market Share
<Insert key players and market share information>
<Insert interactive graph displaying major players and their market share>

Identified Gaps and Opportunities
Gap 1:
Description: <Insert gap description>
Market Size: <Insert estimated market size of the gap>
Potential: <Insert potential of the gap>

Gap 2:
Description: <Insert gap description>
Market Size: <Insert estimated market size of the gap>
Potential: <Insert potential of the gap>

Gap 3:
Description: <Insert gap description>
Market Size: <Insert estimated market size of the gap>
Potential: <Insert potential of the gap>

Recommendations for Solopreneur
Target Gap: <Insert most promising gap>

Product/Service Offering:
<Insert proposed product or service description>

Unique Value Proposition:
<Insert unique value proposition>

Go-to-Market Strategy:
<Insert go-to-market strategy>

Key Partnerships and Resources:
<Insert key partnerships needed>
<Insert key resources needed>

Sources:
<Insert source 1>
<Insert source 2>
<Insert source 3>

#### Expected output

- An industry overview with market size and growth, key players, three evidence-backed gaps, opportunity estimates, and source notes.

#### Customization tips

- Fill in the “industry” placeholder with the specific industry you are involved in.
- Example: If you are a solopreneur in the renewable energy sector, replace “industry” with "renewable energy."
- Use the prompt to conduct a thorough analysis of your industry, focusing on identifying unmet needs or less competitive areas where a solopreneur can effectively enter and grow.
- Use compact comparison tables when they make sourced trends and gaps easier to evaluate; do not invent chart data.
- Identify Key Industry Players: Research and analyze the strategies and weaknesses of the top businesses in your industry to identify common gaps that you can exploit.
- Utilize Advanced Data Visualization Tools: Employ tools like Tableau or Microsoft Power BI to create interactive charts and graphs that clearly illustrate the gaps and opportunities in your industry.
- Develop a Strategic Action Plan: Based on the research findings, outline a detailed strategy that includes specific steps, timelines, and resources needed to effectively penetrate and fill the identified gaps as a solopreneur. Conduct In-Depth Market Research: Dive deep into market trends, consumer behavior, and competitor analysis to gain a comprehensive understanding of the gaps in your industry.
- Leverage Online Surveys and Interviews: Gather insights directly from your target audience through online surveys and interviews to identify specific pain points and unmet needs.
- Stay Updated with Industry News and Reports: Regularly follow industry news, reports, and publications to stay informed about the latest trends and developments in your industry, which can help you identify emerging gaps.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My target industry: Business Coaching

#### Related prompts

- https://returnmytime.com/prompt-library/analyze-business-ideas
- https://returnmytime.com/prompt-library/conduct-market-research-validation


### Generate Mobile App Ideas

- URL: https://returnmytime.com/prompt-library/generate-mobile-app-ideas
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: business-creation, generate, mobile, ideas
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307092231id_/https://returnmytime.com/prompt-library/generate-mobile-app-ideas
- Source captured: 2026-03-07T09:22:31Z
- Source hash: 9fcb55c06c2822681d7bfd3cfba3096eedb1e6444b569487363203c242bfadd8


Use this prompt template to generate innovative mobile app business ideas, helping entrepreneurs identify lucrative opportunities and develop unique concepts for the mobile marketplace.

#### What it does

- Generates innovative business ideas centered around mobile applications.
- Explores various market needs and technological possibilities for new apps.
- Focuses on usability, profitability, and niche targeting for each business idea.

#### When to use

- Use when your expertise, preferred market, and development resources are known and you need app concepts with business-model detail.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting mobile app ideas.

#### Required inputs

- MY AREAS OF EXPERTISE
- MY TARGET MARKET PREFERENCES
- MY DEVELOPMENT RESOURCES

#### Prompt

#CONTEXT:
You are an entrepreneur looking to generate innovative mobile app business ideas that solve real-world problems and capitalize on emerging opportunities. Your task is to ideate a diverse set of app ideas across various domains, providing key details for each one.

#ROLE:
You are an innovative entrepreneur with a knack for generating unique mobile app business ideas that have global scalability potential. Adopt the perspective of a creative problem-solver who can identify niche and mass-market opportunities.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

For each mobile app business idea, provide the following information:

1. Description: A concise overview of the app idea
2. Key Features: The main functionalities and unique selling points of the app
3. Target Market: The intended user base and their characteristics
4. Revenue Model: How the app will generate income (e.g., subscriptions, ads, in-app purchases)
5. Potential Challenges: Any obstacles or difficulties in developing or scaling the app

Aim to generate 5 distinct app ideas, covering a range of domains and target markets.

#TASK CRITERIA:
1. Focus on app ideas that address real-world problems or capitalize on emerging trends
2. Prioritize ideas with global scalability potential
3. Provide a balanced mix of niche and mass-market app ideas
4. Avoid ideas that are overly complex or require extensive resources to develop
5. Ensure each idea has a clear target market and viable revenue model

#INFORMATION ABOUT ME:
- My areas of expertise: [MY AREAS OF EXPERTISE]
- My target market preferences: [MY TARGET MARKET PREFERENCES]
- My development resources: [MY DEVELOPMENT RESOURCES]

#REVIEW REQUIREMENTS:
1. Treat every cost, price, revenue, funding, profitability, and market-potential figure as a scenario until the primary-source review below is complete.

#RESPONSE FORMAT:
Expected output: Five mobile-app business ideas with descriptions, key features, target markets, revenue models, and potential challenges.

For each mobile app business idea, present the information in the following format:

Business Idea <NUMBER>:

Description: <CONCISE OVERVIEW OF THE APP IDEA>

Key Features:
- <FEATURE 1>
- <FEATURE 2>
- <FEATURE 3>

Target Market: <DESCRIPTION OF THE INTENDED USER BASE>

Revenue Model: <EXPLANATION OF HOW THE APP WILL GENERATE INCOME>

Potential Challenges:
- <CHALLENGE 1>
- <CHALLENGE 2>
- <CHALLENGE 3>

#### Expected output

- Five mobile-app business ideas with descriptions, key features, target markets, revenue models, and potential challenges.

#### Customization tips

- Fill in the “ideate businesses that can be built around mobile applications” placeholder with specific types of businesses or industries you are interested in exploring for mobile app development.
- Example: If you are interested in health and fitness, you might specify "Generate mobile app business ideas for online fitness coaching and diet tracking."
- Use this prompt to brainstorm innovative mobile app ideas that align with current market trends and consumer needs within the specified industry or interest area.
- Keep each app concept concise while preserving the requested audience, problem, key features, revenue model, and evidence gaps.
- Identify Market Gaps: Research existing mobile apps to find gaps or unmet needs in various sectors that your business could fill.
- Focus on User Experience: Design your app ideas with a strong emphasis on intuitive user interfaces and seamless user experiences to stand out in the competitive market.
- Leverage Emerging Technologies: Incorporate cutting-edge technologies like AR, AI, or blockchain to enhance functionality and offer unique features that can attract tech-savvy users. Conduct Market Research: Gather data and insights on target audience preferences, market trends, and competitor analysis to validate the demand for your mobile app business idea.
- Identify Monetization Strategies: Explore different revenue streams such as in-app purchases, subscriptions, or advertising to generate income from your mobile app.
- Test and Iterate: Conduct user testing and gather feedback to continuously improve your mobile app's features, design, and functionality.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.
- Current primary sources required: Use current app-store policies, provider pricing pages, written development estimates, and dated customer research as primary sources for revenue-model and cost assumptions.

#### Example input

## Information about me

- My areas of expertise: Career coaching, curriculum design, customer interviews, and workshop facilitation
- My target market preferences: U.S. corporate professionals ages 35–50 preparing for entrepreneurship
- My development resources: A $25,000 prototype budget, one product designer, one contract developer, and 12 weeks

#### Related prompts

- https://returnmytime.com/prompt-library/analyze-business-ideas
- https://returnmytime.com/prompt-library/conduct-market-research-validation


### Generate Niche Market Opportunities

- URL: https://returnmytime.com/prompt-library/generate-niche-market-opportunities
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: business-creation, generate, niche, market
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307084725id_/https://returnmytime.com/prompt-library/generate-niche-market-opportunities
- Source captured: 2026-03-07T08:47:25Z
- Source hash: 5723936e141519ae58d3ed6eb4555324607a7610617a39efb5f534e8b9f32609


Use this prompt template to identify niche markets with low investment opportunities, create a comprehensive table, and conduct detailed research with cited sources. Ideal for entrepreneurs looking to strategically enter new industries.

#### What it does

- Identifies potential niche markets suitable for low investment entry.
- Creates a comprehensive table summarizing the research findings and opportunities.
- Conducts detailed research on identified niches with proper citation of sources.

#### When to use

- Use when matching your skills, resources, and budget to five low-investment niche opportunities.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting niche market opportunities.

#### Required inputs

- MY SKILLS
- MY AVAILABLE RESOURCES
- MY BUDGET

#### Prompt

#CONTEXT:
You are an expert business strategist and researcher tasked with identifying niche market opportunities that require minimal upfront investment. Your goal is to provide a user with 5 promising niche markets that present viable business opportunities.

#ROLE:
As an expert business strategist and researcher specializing in identifying niche market opportunities with low investment costs, your role is to thoroughly research and provide insights into promising niche markets that offer potential for success with minimal startup costs, based on my avaialable resources and skills.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

Compile your findings into a comprehensive table with the following columns:
- Niche Market
- Key Characteristics
- Target Audience
- Potential Products/Services
- Estimated Startup Costs

Provide 5 rows, each representing a different promising niche market. Cite reputable sources to support each identified opportunity.

#TASK CRITERIA:
1. Focus on niche markets that require minimal upfront investment.
2. Provide a clear and concise overview of each niche market, its key characteristics, target audience, and potential products/services.
3. Estimate the startup costs associated with each niche market.
4. Cite reputable sources to support your findings.
5. Avoid discussing niche markets that require significant upfront investment or have limited potential for success.

#INFORMATION ABOUT ME:
My skills: [MY SKILLS]
My available resources: [MY AVAILABLE RESOURCES]
My budget: [MY BUDGET]

#REVIEW REQUIREMENTS:
1. Treat every cost, price, revenue, funding, profitability, and market-potential figure as a scenario until the primary-source review below is complete.

#RESPONSE FORMAT:
Expected output: A five-row niche-opportunity table covering characteristics, target audiences, products or services, startup-cost estimates, and sources.

| Niche Market | Key Characteristics | Target Audience | Potential Products/Services | Estimated Startup Costs |
|--------------|---------------------|-----------------|-----------------------------|-----------------------|
| $niche1      | $characteristics1   | $audience1      | $products_services1         | $startup_costs1       |
| $niche2      | $characteristics2   | $audience2      | $products_services2         | $startup_costs2       |
| $niche3      | $characteristics3   | $audience3      | $products_services3         | $startup_costs3       |
| $niche4      | $characteristics4   | $audience4      | $products_services4         | $startup_costs4       |
| $niche5      | $characteristics5   | $audience5      | $products_services5         | $startup_costs5       |

Sources:
1. $source1
2. $source2
3. $source3
4. $source4
5. $source5
...

#### Expected output

- A five-row niche-opportunity table covering characteristics, target audiences, products or services, startup-cost estimates, and sources.

#### Customization tips

- Fill in the “your skills in detail” placeholder with the skillset you have.
- Fill in “your resources” with your readily available resources that you could potentially utilize to start a business in a certain niche.
- Fill in “your budget” with the actual budget you’re willing to invest into the new business venture.
- Identify Emerging Trends: Use tools like Google Trends or industry reports to spot rising interests and underserved niches that align with your skills and resources.
- Utilize Competitive Analysis: Examine competitors in your target market to identify gaps in their offerings that you can exploit with innovative solutions or better value propositions.
- Focus on Scalability: Select niche markets that not only require low investment to enter but also offer potential for growth and scalability without exponential increases in costs. Conduct Thorough Market Research: Gather data on market size, customer demographics, and competition to identify niche markets with high potential for success.
- Identify Unique Selling Points: Determine what sets your product or service apart from competitors and highlight these unique features to attract customers in niche markets.
- Build Relationships with Industry Experts: Network with professionals in your industry to gain insights and advice on identifying niche markets and penetrating them effectively.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.
- Current primary sources required: Use written vendor quotes, current provider pricing pages, and the SBA market-research guide at https://www.sba.gov/business-guide/plan-your-business/market-research-competitive-analysis as primary sources for startup-cost estimates.

#### Example input

## Information about me

- My skills: Career coaching, facilitation, curriculum design, and B2B workshop delivery
- My available resources: 15 hours per week, an email list of 1,200, two contractor partners, and existing workshop materials
- My budget: $8,000

#### Related prompts

- https://returnmytime.com/prompt-library/analyze-business-ideas
- https://returnmytime.com/prompt-library/conduct-market-research-validation


### Generate Tailored Bootstrapping Advice

- URL: https://returnmytime.com/prompt-library/generate-tailored-bootstrapping-advice
- Category: Business Creation
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: business-creation, generate, tailored, bootstrapping
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307085321id_/https://returnmytime.com/prompt-library/generate-tailored-bootstrapping-advice
- Source captured: 2026-03-07T08:53:21Z
- Source hash: 1102262d6c98f9a0f502b377ebc0cb8cb4252a324013b525a42a75f141c425f5


Use this prompt template to explore bootstrapping methods tailored to your new business venture. It intelligently asks relevant questions, ensuring the advice you receive is customized and actionable for your specific needs.

#### What it does

- Converts the user's request into a structured set of questions tailored to exploring bootstrapping methods for a new business.
- Provides expert advice based on the user's responses to the tailored questions.
- Keeps each question and recommendation focused on the venture's documented constraints.

#### When to use

- Use when a founder has a defined model, audience, funding need, and skills and wants non-equity funding tactics to assess.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting tailored bootstrapping advice.

#### Required inputs

- MY BUSINESS MODEL
- MY TARGET MARKET
- MY CURRENT FUNDING NEEDS
- MY KEY FOUNDER SKILLS

#### Prompt

#CONTEXT:
Adopt the role of an expert business advisor with deep knowledge in bootstrapping methods and strategies for new ventures. Your task is to help the user explore bootstrapping approaches tailored to their specific business context.

#ROLE:
You are an expert business advisor with deep knowledge in bootstrapping methods and strategies for new ventures.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

- Engage in a Q&A session to explore bootstrapping approaches
- Ask probing questions to understand the business model, target market, funding needs, and founder skills
- Provide practical, actionable advice on lean startup techniques, creative funding sources, and cost-effective growth hacks
- Include 3 specific bootstrapping tactics with the following details for each:
  - Tactic name
  - Tactic description
  - Tactic benefits
  - Tactic considerations
- Summarize key takeaways with 3 main points
- Outline recommended next steps for the user

#BOOTSTRAPPING CRITERIA:
1. Advice should be tailored to the specific business context provided by the user
2. Focus on lean startup techniques, creative funding sources, and cost-effective growth hacks
3. Provide practical and actionable advice that the user can readily implement
4. Avoid generic or overly theoretical recommendations

#INFORMATION ABOUT ME:
- My business model: [MY BUSINESS MODEL]
- My target market: [MY TARGET MARKET]
- My current funding needs: [MY CURRENT FUNDING NEEDS]
- My key founder skills: [MY KEY FOUNDER SKILLS]

#REVIEW REQUIREMENTS:
1. Treat every cost, price, revenue, funding, profitability, and market-potential figure as a scenario until the primary-source review below is complete.

#RESPONSE FORMAT:
Expected output: Three diagnostic questions and three tailored bootstrapping tactics with benefits and considerations.

Question 1: <QUESTION>
Answer 1: <ANSWER>

Bootstrapping Tactic 1:
- Tactic Name: <NAME>
- Tactic Description: <DESCRIPTION>
- Tactic Benefits: <BENEFITS>
- Tactic Considerations: <CONSIDERATIONS>

Question 2: <QUESTION>
Answer 2: <ANSWER>

Bootstrapping Tactic 2:
- Tactic Name: <NAME>
- Tactic Description: <DESCRIPTION>
- Tactic Benefits: <BENEFITS>
- Tactic Considerations: <CONSIDERATIONS>

Question 3: <QUESTION>
Answer 3: <ANSWER>

Bootstrapping Tactic 3:
- Tactic Name: <NAME>
- Tactic Description: <DESCRIPTION>
- Tactic Benefits: <BENEFITS>
- Tactic Considerations: <CONSIDERATIONS>

Key Takeaways:
1. <TAKEAWAY 1>
2. <TAKEAWAY 2>
3. <TAKEAWAY 3>

Next Steps:
<OUTLINE NEXT STEPS>

#### Expected output

- Three diagnostic questions and three tailored bootstrapping tactics with benefits and considerations.

#### Customization tips

- Fill in the “documented inputs” by specifying your business type and key challenges. Example: "My new business venture is an online eco-friendly apparel store, and I'm struggling with limited startup capital and sourcing sustainable materials."
- Use the prompt to engage in a detailed Q&A session where the AI asks targeted questions based on the information provided about your business venture, ensuring the advice is customized and relevant.
- Answer each question with the concrete resources, constraints, and evidence needed for focused bootstrapping advice.
- Prepare a set of questions that identify the current resources available to the user, such as skills, network, and initial capital.
- Develop questions that assess the user's industry knowledge and market understanding to provide more customized and effective bootstrapping strategies.
- Inquire about the user's long-term business goals and immediate needs to align your advice with their specific entrepreneurial vision and timeline. Identify the user's target audience and their pain points to tailor bootstrapping strategies that address specific customer needs.
- Ask about the user's willingness to take risks and explore unconventional methods to maximize limited resources and achieve business growth.
- Inquire about the user's previous experience with bootstrapping or starting businesses to provide advice based on their level of familiarity and expertise.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.
- Current primary sources required: Use current cash-flow records and the SBA loan-program guidance at https://www.sba.gov/funding-programs/loans as primary sources for funding scenarios.
- A named qualified financial professional serving as the human reviewer must review and approve every pricing, funding, revenue, cost, and projection conclusion before use.

#### Example input

## Information about me

- My business model: Cohort coaching plus private advisory and digital templates
- My target market: U.S. corporate professionals ages 35–50 preparing to start a business
- My current funding needs: $25,000 for contractor capacity, customer-acquisition tests, and a six-month operating reserve
- My key founder skills: Facilitation, curriculum design, career coaching, and B2B workshop delivery

#### Related prompts

- https://returnmytime.com/prompt-library/analyze-business-ideas
- https://returnmytime.com/prompt-library/conduct-market-research-validation


### Automate Business Processes

- URL: https://returnmytime.com/prompt-library/automate-business-processes
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 1-2 hours
- Tags: business-operations-growth, automate, business, processes
- Published: 2026-07-30
- Modified: 2026-07-31
- Provenance: archive
- Source: https://web.archive.org/web/20260307090051id_/https://returnmytime.com/prompt-library/automate-business-processes
- Source captured: 2026-03-07T09:00:51Z
- Source hash: 6825bf9b53e7b4f63b9f00b079c9ec1aa716a6132669a53cf1544e21ad84db00


Use this prompt template to examine a one-person business process, identify responsible automation opportunities, and separate verified setup guidance from steps that require current provider research.

#### What it does

- Identifies specific inefficiencies in the described business process and provides targeted solutions to address them.
- Outlines detailed, step-by-step instructions for setting up automation tools and AI systems to streamline and systematize the business process.
- Explains the time-saving benefits and efficiency gains of each suggested automation or systematization, ensuring the advice is actionable and identifies assumptions and research still required.

#### When to use

- Use when a named process, its current steps, and its bottlenecks are documented well enough to design automations.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting business processes.

#### Required inputs

- PROCESS NAME
- CURRENT PROCESS
- CURRENT INEFFICIENCIES

#### Prompt

#CONTEXT:
You will act as "Arnold Automator", an AI that specializes in helping one-person businesses systematize, streamline, and automate their tasks using AI and automation tools. Your utmost aim is to help one-person businesses save time and spend less effort on their business.

#ROLE:
Adopt the role of an expert in business process automation and AI tools. Your task is to provide uncommon, expert advice to help the user streamline and automate a specific process in their one person business.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

- Prioritize uncommon, expert advice
- Be specific, but separate verified setup steps from steps that still require provider-documentation research
- Describe exact setup steps only when they are supported by current official provider documentation
- Explain IN DETAIL why your solutions save the one person business time
- Suggest removing individual steps in the process to streamline it, if applicable

#PROCESS CRITERIA:
1. A great process automation solution is specific, actionable, and saves the user significant time and effort.
2. Focus on providing detailed, step-by-step instructions for setting up automations and AI tools.
3. If current official provider documentation has not been supplied or verified, do not provide exact setup instructions; instead list the documentation questions and verification steps required.
4. Always explain in detail how your solutions will save the one person business time and effort.

#INFORMATION ABOUT ME:
- Process Name: [PROCESS NAME]
- Current Process: [CURRENT PROCESS]
- Current Inefficiencies: [CURRENT INEFFICIENCIES]

#RESPONSE FORMAT:
Expected output: An automation plan with solution-by-solution setup instructions, estimated time savings, and process-streamlining recommendations.

Provide your response in the following format:

Suggested Automation Solutions:
1. <Solution 1>
   - Step-by-step setup instructions
   - Detailed explanation of time and effort savings

2. <Solution 2>
   - Step-by-step setup instructions
   - Detailed explanation of time and effort savings

<...>

Suggested Process Streamlining:
- <Suggestion 1>
- <Suggestion 2>
- <...>

#### Expected output

- An automation plan with solution-by-solution setup instructions, estimated time savings, and process-streamlining recommendations.

#### Customization tips

- Fill in the “your process name”, “step by step how you do it today”, and “inefficiencies” placeholders with specific details about your business process, current execution steps, and identified inefficiencies.
- Example: If your process is "Email Marketing Campaigns", describe it step by step like "I manually collect email addresses, create email content, and send emails through a standard email client. I spend too much time on manual entry and lack automation in scheduling and tracking responses."
- Use the detailed descriptions you provide to tailor Arnold Automator's advice specifically to the nuances of your process. This ensures that the automation and AI tool recommendations are directly applicable and can effectively address the inefficiencies you're experiencing.
- After filling in the details, ask Arnold Automator for step-by-step automation setups. This includes choosing the right tools, integrating them into your existing workflow, and configuring them to automate repetitive tasks, thus saving time and reducing effort in your one-person business.
- Identify Repetitive Tasks: Examine your current process to pinpoint repetitive tasks that do not require decision-making or creativity, such as data entry, scheduling, or sending follow-up emails. These are prime candidates for automation.
- Implement Automation Tools: For tasks like scheduling, use Calendly to automate appointment bookings directly linked to your calendar. For email, set up automated responses or follow-up emails using tools like Mailchimp or ActiveCampaign, which can trigger based on specific actions taken by your customers.
- Streamline Communication: Utilize a CRM tool like HubSpot to integrate all customer interactions into one platform. This will save time by avoiding the need to switch between different tools and platforms, and by automating tasks like lead tracking and customer follow-ups, you can ensure nothing falls through the cracks. Create Standard Operating Procedures (SOPs): Document step-by-step instructions for your process, including any specific tools or software used. This will ensure consistency and make it easier to delegate or automate tasks in the future.
- Explore AI-Powered Task Management: Consider using AI-powered task management tools like Trello or Asana to streamline your process. These tools can automatically assign tasks, set deadlines, and provide reminders, saving you time and effort in managing and tracking your tasks.
- Automate Data Entry: Use tools like Zapier or Integromat to automate data entry tasks by connecting different apps and platforms. For example, you can automatically transfer data from a form submission to a spreadsheet or CRM, eliminating the need for manual data entry.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.
- Verify every named tool's setup, permissions, limits, and integration steps against current official provider documentation before implementation.
- Do not provide exact setup instructions without current official provider documentation; mark the setup as research required instead.

#### Example input

## Information about me

- Process Name: Lead intake and discovery-call booking
- Current Process: A website form emails the owner; the owner copies the lead into the CRM, sends a qualification email, shares a booking link, and manually updates the pipeline stage.
- Current Inefficiencies: Duplicate data entry, inconsistent follow-up, and a two-day response delay

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas

#### Related articles

- https://returnmytime.com/blog/10-hour-week-ai-tools
- https://returnmytime.com/blog/zapier-automations-small-business

### Conduct Industry Trends Research

- URL: https://returnmytime.com/prompt-library/conduct-industry-trends-research
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 1-2 hours
- Tags: business-operations-growth, conduct, industry, trends
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260308011626id_/https://returnmytime.com/prompt-library/conduct-industry-trends-research
- Source captured: 2026-03-08T01:16:26Z
- Source hash: a6469ac2b66e75fd1007b25366b47985ddc444fc1ad16b7f65a2f15fe8a82982


Generate a detailed analysis of industry trends relevant to your business venture using the archived prompt template. This tool helps identify opportunities, ensures comprehensive research, and organizes findings in a table with cited sources for enhanced clarity and reference.

#### What it does

- Conducts comprehensive research on industry trends relevant to the user's business interest.
- Identifies and highlights opportunities within these trends for responsible business exploration.
- Organizes the research findings and cited sources into a table for clear presentation.

#### When to use

- Use when evaluating how current industry trends could create opportunities or threats for a defined venture.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting industry trends research.

#### Required inputs

- MY BUSINESS VENTURE

#### Prompt

#CONTEXT:
You are an industry analyst tasked with helping a business identify market trends and opportunities relevant to their venture. Your analysis should include a comprehensive overview of the industry, key trends to be aware of, and specific opportunities to capitalize on those trends.

#ROLE:
Adopt the role of a highly knowledgeable and thorough industry analyst with expertise in identifying market trends and opportunities.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

Conduct comprehensive research on industry trends relevant to the given business venture. Identify specific opportunities to capitalize on these trends.

Present your findings in the following format:

Industry Overview: Provide a high-level summary of the current state of the industry.

Key Trends:
1. Identify 3 key trends shaping the industry
2. Provide a brief description of each trend

Opportunities:
Create a table with the following columns:
- Opportunity: Identify a specific opportunity
- Description: Provide details on the opportunity
- Potential Impact: Assess the potential impact of capitalizing on this opportunity

Cite all sources used in your research and analysis.

#VENTURE CRITERIA:
- Focus on trends and opportunities directly relevant to the provided business venture
- Prioritize opportunities with the highest potential impact
- Avoid speculation or opinions not backed by credible sources
- Ensure all sources are properly cited

#INFORMATION ABOUT ME:
- My business venture: [MY BUSINESS VENTURE]

#RESPONSE FORMAT:
Expected output: An industry overview, three key trends, a three-row opportunity table, and a source list.

Industry Overview:
<Provide industry overview here>

Key Trends:
1. <Trend 1>
2. <Trend 2>
3. <Trend 3>

Opportunities:
| Opportunity | Description | Potential Impact |
|-------------|-------------|------------------|
| <Opportunity 1> | <Description 1> | <Impact 1> |
| <Opportunity 2> | <Description 2> | <Impact 2> |
| <Opportunity 3> | <Description 3> | <Impact 3> |

Sources:
1. <Source 1>
2. <Source 2>
3. <Source 3>

#### Expected output

- An industry overview, three key trends, a three-row opportunity table, and a source list.

#### Customization tips

- Fill in the “industry trends” and “business venture” placeholders with specific industry trends and details about your business venture.
- Example: If you are interested in the renewable energy sector and your business venture involves developing solar panel solutions, specify these details.
- Ensure that the request for a table and citation of sources is clear and specific to the information you need. For instance, ask for a table summarizing key trends and opportunities in the renewable energy sector, followed by a list of cited sources.
- Prioritize the trends and opportunities with the clearest sourced implications for the specific venture.
- Identify Key Trends: Focus on identifying emerging and enduring trends within your industry that align with your business goals and values.
- Utilize Diverse Sources: Gather information from a variety of credible sources such as industry reports, academic journals, expert interviews, and market analysis to ensure a comprehensive understanding of the trends.
- Visualize Data Effectively: Use tools like Microsoft Excel, Google Sheets, or Tableau for creating clear and informative tables that summarize your research findings, making it easier to analyze and present. Create a Research Plan: Outline a clear plan for your research, including specific objectives, methodologies, and timelines to ensure a systematic and organized approach.
- Engage with Industry Experts: Reach out to professionals and thought leaders in your industry for insights and perspectives on the trends you are exploring. Their expertise can provide valuable information and guidance.
- Stay Updated: Continuously monitor and track industry news, publications, and online forums to stay up-to-date with the latest trends and developments. Set up Google Alerts or subscribe to relevant newsletters to receive regular updates.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My business venture: A coaching practice helping corporate professionals transition to entrepreneurship through 1:1 coaching and group programs

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Conduct SWOT Analysis

- URL: https://returnmytime.com/prompt-library/conduct-swot-analysis
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 2+ hours
- Tags: business-operations-growth, conduct, swot, analysis
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307085907id_/https://returnmytime.com/prompt-library/conduct-swot-analysis
- Source captured: 2026-03-07T08:59:07Z
- Source hash: f5b029dc395a4a77bbe6cd424dbc1e838cbcb6a98f008380da6de3ca7c69a6b7


Use this prompt template to conduct a comprehensive SWOT analysis for your new business venture, identifying strengths, weaknesses, opportunities, and threats to enhance strategic planning and decision-making.

#### What it does

- Identifies and lists the strengths of the new business venture to leverage them for strategic advantage.
- Analyzes and outlines the weaknesses of the business to address and mitigate potential risks.
- Explores opportunities and threats in the market to formulate a comprehensive strategic plan.

#### When to use

- Use when a venture needs an evidence-based review of business capabilities and market conditions before choosing priorities.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting SWOT analysis.

#### Required inputs

- MY BUSINESS VENTURE

#### Prompt

#CONTEXT:
Adopt the role of an expert business strategist with deep knowledge in conducting comprehensive SWOT analyses for businesses across various industries. Your task is to help the user conduct a thorough SWOT analysis for their given business venture, critically examining its Strengths, Weaknesses, Opportunities, and Threats. Provide insightful and actionable recommendations based on the analysis to help strategize effectively for success.

#ROLE:
You are an expert business strategist with deep knowledge in conducting comprehensive SWOT analyses for businesses across various industries.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

Organize the SWOT analysis into the following sections:

Business Overview: Provide a brief overview of the business venture.

Strengths: List 3 key strengths of the business.
- Strength 1
- Strength 2
- Strength 3

Weaknesses: List 3 key weaknesses of the business.
- Weakness 1
- Weakness 2
- Weakness 3

Opportunities: List 3 key opportunities for the business.
- Opportunity 1
- Opportunity 2
- Opportunity 3

Threats: List 3 key threats to the business.
- Threat 1
- Threat 2
- Threat 3

Key Success Factors: List 3 key factors critical for the business's success.
1. Key Success Factor 1
2. Key Success Factor 2
3. Key Success Factor 3

Key Risk Factors: List 3 key risk factors that could hinder the business's success.
1. Key Risk Factor 1
2. Key Risk Factor 2
3. Key Risk Factor 3

Recommendations: Provide 3 actionable recommendations based on the SWOT analysis.
1. Recommendation 1
2. Recommendation 2
3. Recommendation 3

Next Steps: Outline 3 next steps the business should take based on the recommendations.
1. Next Step 1
2. Next Step 2
3. Next Step 3

#TASK CRITERIA:
1. Conduct a comprehensive SWOT analysis, critically examining the business's Strengths, Weaknesses, Opportunities, and Threats.
2. Provide insightful and actionable recommendations based on the analysis to help strategize effectively for success.
3. Focus on the most important factors and prioritize the recommendations and next steps based on their potential impact and feasibility.

#INFORMATION ABOUT ME:
- My business venture: [MY BUSINESS VENTURE]

#RESPONSE FORMAT:
Expected output: A structured SWOT analysis with success factors, risk factors, prioritized recommendations, and next steps.

Organize the response into clearly defined sections with headings for each part of the SWOT analysis, as outlined in the RESPONSE GUIDELINES. Use bullet points for lists and number the key success factors, key risk factors, recommendations, and next steps. Avoid using XML tags in the response.

#### Expected output

- A structured SWOT analysis with success factors, risk factors, prioritized recommendations, and next steps.

#### Customization tips

- Fill in the details of your new business venture, including its unique strengths, potential weaknesses, market opportunities, and external threats.
- Example: If your business is a tech startup focusing on sustainable energy solutions, identify strengths like innovative technology, weaknesses such as limited funding, opportunities in growing environmental awareness, and threats from established competitors.
- Use the SWOT analysis to guide strategic planning, ensuring that you leverage strengths, address weaknesses, capitalize on opportunities, and mitigate threats.
- Keep each SWOT item concise, specific, and linked to supplied evidence or an explicitly labeled assumption.
- Identify Core Strengths: Focus on unique resources and capabilities that give your business a competitive advantage.
- Analyze Internal Weaknesses: Objectively assess areas where your business may lack resources or capabilities, impacting performance.
- Explore External Opportunities and Threats: Examine market trends, economic shifts, and competitor strategies to align your business planning and risk management. Identify Key Opportunities: Look for potential areas of growth or new markets to tap into.
- Mitigate Threats: Take proactive measures to minimize risks and protect your business from potential threats.
- Prioritize Actionable Strategies: Focus on developing strategies that can be implemented effectively and have a significant impact on your business.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My business venture: A coaching practice helping corporate professionals transition to entrepreneurship through 1:1 coaching and group programs

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Create Actionable Business Roadmap

- URL: https://returnmytime.com/prompt-library/create-actionable-business-roadmap
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 1-2 hours
- Tags: business-operations-growth, create, actionable, business
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307085709id_/https://returnmytime.com/prompt-library/create-actionable-business-roadmap
- Source captured: 2026-03-07T08:57:09Z
- Source hash: 55cf74d5ca787250d76250b1328e42cde371907481367552edacd1a7171bb6d5


Create a tailored, actionable business roadmap using the archived prompt template, designed to meet specific business goals within a 12-week timeframe. This tool provides a structured approach, breaking down goals into manageable, quantifiable weekly KPIs, ensuring clarity and progress tracking.

#### What it does

- Converts user input into a detailed 3-month business roadmap.
- Ensures each week's goals and actions are quantifiable and aligned with the monthly and overall 3-month objectives.
- Provides two separate, comprehensive roadmaps based on the user's specific business goals and context.

#### When to use

- Use when a three-month goal and current business context are clear enough to sequence weekly actions and KPIs.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting actionable business roadmap.

#### Required inputs

- MY 3-MONTH GOAL
- MY BUSINESS CONTEXT

#### Prompt

#CONTEXT:
You are an expert business strategist tasked with helping the user create a 12-week actionable business roadmap to achieve their 3-month goal. The roadmap should be based on the specifics of the user's business goal and context.

#ROLE:
As an expert business strategist, your role is to reverse-engineer the user's goal into a 3-month actionable roadmap that they can follow to reach their desired outcome. You should provide guidance and support to ensure the user has a clear and quantifiable plan to achieve their goal.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

- Create two roadmaps for the user to follow
- Each week's KPIs should be quantifiable
- Format each of the 3 months like this:
  # Month X: <BIG GOAL OF MONTH X>
  - Summarize the month's goal and KPIs
  ## Week X: <BIG GOAL OF WEEK X>
  - Summarize and quantify weekly actions the user needs to take to reach the week's goal (in a way that they can put it into their calendar)
- If the user's goal is not action-oriented and quantifiable, tell them: "Retry with a quantifiable and action-oriented goal. You can do this!"

#TASK CRITERIA:
- Focus on creating an actionable and quantifiable roadmap
- Break down the user's 3-month goal into smaller, achievable weekly goals
- Provide specific, measurable actions for each week that the user can put into their calendar
- Avoid vague or general advice; be as specific and detailed as possible
- If the user's goal is not action-oriented or quantifiable, prompt them to retry with a more suitable goal

#INFORMATION ABOUT ME:
- My 3-month goal: [MY 3-MONTH GOAL]
- My business context: [MY BUSINESS CONTEXT]

#RESPONSE FORMAT:
Expected output: A 12-week roadmap organized by month and week with actions, measurable KPIs, dependencies, and progress checkpoints.

- Use markdown formatting for headers and bullet points
- Organize the roadmap by months and weeks
- Provide clear, concise, and actionable steps for each week
- Avoid using XML tags in the response

#### Expected output

- A 12-week roadmap organized by month and week with actions, measurable KPIs, dependencies, and progress checkpoints.

#### Customization tips

- Fill in the “goal” and “context” placeholders with your specific business goal and the context surrounding it.
- Example: If your goal is "Increase online sales by 30%" and your context is "Current online sales are steady but have plateaued in the last quarter despite increased marketing efforts."
- Use the structured format provided to break down your goal into monthly and weekly segments, ensuring each segment has clear, quantifiable KPIs that align with the overall 3-month goal.
- Example: For Month 1, set a big goal like "Improve website conversion rate by 10%," and for Week 1, a goal could be "Implement A/B testing on landing pages to increase conversions."
- If the goal provided is not quantifiable or action-oriented, utilize the built-in instruction to prompt a revision of the goal, ensuring it meets the criteria for effective planning and measurable outcomes.
- Example: If the input goal is "Improve business visibility," advise to "Retry with a quantifiable and action-oriented goal. You can do this!" to ensure the roadmap can be actionable and results can be measured effectively.
- Clarify and Quantify the 3-Month Goal: Ensure the main goal is specific, measurable, achievable, relevant, and time-bound (SMART) to facilitate effective roadmap creation.
- Break Down the Goal into Monthly Objectives: Identify key milestones for each month that directly contribute to achieving the overall 3-month goal, ensuring each has clear, quantifiable targets.
- Develop Weekly Action Items: For each week, define specific tasks that progress towards the monthly objectives, including detailed actions and how they can be integrated into daily schedules. Identify Key Resources: Determine the necessary resources, such as budget, personnel, and technology, needed to execute the roadmap effectively.
- Set Realistic Timelines: Ensure that the timelines for each week's goals and actions are feasible and take into account any potential constraints or limitations.
- Regularly Review and Adjust: Continuously monitor progress and make necessary adjustments to the roadmap based on feedback, changes in the business environment, or new insights.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.
- Treat named tools and platforms as archived examples; verify current capabilities, terms, pricing, and policies in each provider's official documentation before use.

#### Example input

## Information about me

- My 3-month goal: Scale from 1:1 coaching to a hybrid model with group programs and digital courses, while maintaining personal touch
- My business context: Clarity Coaching Co. — A coaching practice helping corporate professionals transition to entrepreneurship through 1:1 coaching and group programs

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Create Business Automation Workflows

- URL: https://returnmytime.com/prompt-library/create-business-automation-workflows
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 2+ hours
- Tags: business-operations-growth, create, business, automation
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307092955id_/https://returnmytime.com/prompt-library/create-business-automation-workflows
- Source captured: 2026-03-07T09:29:55Z
- Source hash: 6dc78b222dc8b40ccc8ca2cb9b3b0abfd69cad832a05c53096d6177a3bc16cb0


Use this prompt template to design and implement business automation workflows that streamline repetitive tasks using tools like Zapier, Make.com, or custom scripts, enhancing efficiency and productivity.

#### What it does

- Identifies repetitive tasks within a business that can be automated.
- Suggests specific automation workflows using tools like Zapier, Make.com, or custom scripts.
- Provides a detailed implementation strategy for these automation workflows.

#### When to use

- Use when current operations are documented and repetitive tasks need tool-specific workflow designs.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting business automation workflows.

#### Required inputs

- MY CURRENT BUSINESS OPERATIONS

#### Prompt

#CONTEXT:
Adopt the role of an expert in business automation, optimization, and process efficiency. Your task is to help the user analyze their current business operations, identify repetitive tasks that can be automated, and develop comprehensive automation workflows to improve efficiency.

#ROLE:
You are an expert in business automation, optimization, and process efficiency.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Begin with an overview of the current business operations.
2. Identify repetitive, time-consuming tasks that are prime candidates for automation.
3. For each identified task, develop a comprehensive, step-by-step automation workflow:
   - Specify the task being automated
   - Identify the automation tool to be used (e.g., Zapier, Make.com, or custom scripts)
   - Outline the steps involved in the automation process
   - Estimate the expected efficiency gains from implementing the automation
4. Provide a detailed implementation strategy for the suggested automations, broken down into phases.
5. Discuss potential challenges that may arise during the implementation process.
6. Recommend solutions to address the identified challenges.

#TASK CRITERIA:
1. Focus on identifying the most repetitive and time-consuming tasks that are suitable for automation.
2. Develop automation workflows that are comprehensive, detailed, and easy to understand.
3. Prioritize automation tools that are user-friendly and cost-effective.
4. Ensure that the implementation strategy is realistic and achievable.
5. Identify potential challenges and provide practical solutions to overcome them.

#INFORMATION ABOUT ME:
- My current business operations: [MY CURRENT BUSINESS OPERATIONS]

#RESPONSE FORMAT:
Expected output: A current-operations summary, three repetitive tasks, and three tool-specific automation workflows with setup steps and expected efficiency gains.

Current Business Operations Overview:
<Insert overview of current business operations>

Identified Repetitive Tasks:
1. <Repetitive Task 1>
2. <Repetitive Task 2>
3. <Repetitive Task 3>

Automation Workflow 1:
Task: <Repetitive Task 1>
Tool: <Automation Tool>
Steps:
1. <Step 1>
2. <Step 2>
3. <Step 3>
Expected Efficiency Gains: <Estimated efficiency gains>

Automation Workflow 2:
Task: <Repetitive Task 2>
Tool: <Automation Tool>
Steps:
1. <Step 1>
2. <Step 2>
3. <Step 3>
Expected Efficiency Gains: <Estimated efficiency gains>

Automation Workflow 3:
Task: <Repetitive Task 3>
Tool: <Automation Tool>
Steps:
1. <Step 1>
2. <Step 2>
3. <Step 3>
Expected Efficiency Gains: <Estimated efficiency gains>

Implementation Strategy:
Phase 1: <Description of Phase 1>
Phase 2: <Description of Phase 2>
Phase 3: <Description of Phase 3>

Potential Challenges:
1. <Potential Challenge 1>
2. <Potential Challenge 2>
3. <Potential Challenge 3>

Recommended Solutions:
1. <Recommended Solution 1>
2. <Recommended Solution 2>
3. <Recommended Solution 3>

#### Expected output

- A current-operations summary, three repetitive tasks, and three tool-specific automation workflows with setup steps and expected efficiency gains.

#### Customization tips

- Fill in the “identify repetitive tasks” placeholder with specific tasks that are frequently performed in your business that could be automated. Example: "Processing customer inquiries, scheduling appointments, and updating inventory are repetitive tasks in my business."
- Use the prompt to explore different automation tools like Zapier, Make.com, or even custom scripts, depending on the complexity and customization needs of the tasks you've identified.
- Keep the workflow concise enough to implement while retaining triggers, actions, exceptions, owners, and verification steps.
- Map Out Workflow Details: Before automating, visually map out the current workflow of repetitive tasks to identify bottlenecks and inefficiencies that automation can resolve.
- Select the Right Tools: Choose automation tools that integrate seamlessly with your existing software stack and are scalable to your business needs, whether it's Zapier for simple integrations or custom scripts for complex workflows.
- Monitor and Optimize: After implementing automation workflows, continuously monitor their performance and impact on business efficiency. Use data-driven insights to tweak and optimize the workflows for better results. Document Current Processes: Take the time to thoroughly document all of your current repetitive tasks and workflows before attempting to automate them. This will help you identify any gaps or areas for improvement.
- Start Small: Begin by automating one or two simple tasks or workflows before tackling larger, more complex processes. This will allow you to gain experience and confidence in using automation tools.
- Involve Your Team: Get input from your team members who are directly involved in the repetitive tasks. They may have valuable insights and suggestions for improving the automation workflows.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.
- Treat named tools and platforms as archived examples; verify current capabilities, terms, pricing, and policies in each provider's official documentation before use.

#### Example input

## Information about me

- My current business operations: A coaching practice helping corporate professionals transition to entrepreneurship through 1:1 coaching and group programs

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Create Customer Loyalty Program

- URL: https://returnmytime.com/prompt-library/create-customer-loyalty-program
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 2+ hours
- Tags: business-operations-growth, create, customer, loyalty
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307092539id_/https://returnmytime.com/prompt-library/create-customer-loyalty-program
- Source captured: 2026-03-07T09:25:39Z
- Source hash: d84232398eb5de5820fe8f0649b3aec0c62c8c105bd0793dff5f53f5e82c67d0


Create a tailored loyalty reward program for your niche business to enhance customer retention and satisfaction. this prompt template helps you design a comprehensive scheme that rewards returning customers and boosts your brand loyalty.

#### What it does

- Converts user input into a structured plan for creating a loyalty reward program.
- Focuses on tailoring the program specifically to the business's niche.
- Outlines key components such as reward tiers, customer eligibility, and redemption processes.

#### When to use

- Use when a business knows its niche, customer demographics, and offers and wants to design a measurable loyalty program.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting customer loyalty program.

#### Required inputs

- MY BUSINESS NICHE
- MY TARGET CUSTOMER DEMOGRAPHICS
- MY PRODUCTS/SERVICES

#### Prompt

#CONTEXT:
You are a seasoned loyalty rewards program expert with deep expertise in designing effective, engaging customer loyalty programs across various industries. Your task is to help the user design a comprehensive, multi-tiered loyalty rewards program tailored to their specific business niche, considering target customer demographics, purchasing behaviors, and preferences.

#ROLE:
Adopt the role of a seasoned loyalty rewards program expert with deep expertise in designing effective, engaging customer loyalty programs across various industries.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Begin with a high-level program overview.
2. Provide a detailed target customer profile, including demographics, purchasing behaviors, and preferences.
3. Outline the points-based system, covering how points are earned and the redemption tiers and rewards.
4. Describe gamification elements to drive engagement.
5. Discuss program promotion strategies.
6. Explain customer retention strategies.
7. Provide insights on program evolution and optimization.

#TASK CRITERIA:
1. The loyalty program should be tailored to the specific business niche and target customer demographics.
2. The points system should be easy to understand and appealing to customers.
3. Gamification elements should be engaging and drive customer participation.
4. Promotion strategies should effectively reach and entice the target audience.
5. Retention strategies should focus on long-term customer engagement and loyalty.
6. Program evolution and optimization should be based on data-driven insights and customer feedback.

#INFORMATION ABOUT ME:
- My business niche: [MY BUSINESS NICHE]
- My target customer demographics: [MY TARGET CUSTOMER DEMOGRAPHICS]
- My products/services: [MY PRODUCTS/SERVICES]

#RESPONSE FORMAT:
Expected output: A loyalty-program design with customer profile, points rules, reward tiers, personalized offers, launch steps, and measurement metrics.

Program Overview:

Target Customer Profile:
Demographics: <demographics>
Purchasing Behaviors: <purchasing_behaviors>
Preferences: <preferences>

Points System:
How to Earn Points: <earn_points>

Redemption Tiers and Rewards:
Tier 1:
Name: <name>
Points Range: <points_range>
Rewards: <rewards>

Tier 2:
Name: <name>
Points Range: <points_range>
Rewards: <rewards>

Tier 3:
Name: <name>
Points Range: <points_range>
Rewards: <rewards>

Gamification Elements:
<gamification_elements>

Program Promotion Strategies:
<program_promotion_strategies>

Customer Retention Strategies:
<customer_retention_strategies>

Program Evolution and Optimization:
<program_evolution_and_optimization>

#### Expected output

- A loyalty-program design with customer profile, points rules, reward tiers, personalized offers, launch steps, and measurement metrics.

#### Customization tips

- Fill in the “your business” and “niche” placeholders with specific details about your business and its niche.
- Example: If your business is a coffee shop and your niche is organic and locally sourced products, you would specify this in the placeholders.
- Consider the unique aspects of your business and niche that can be leveraged in a loyalty program, such as exclusive discounts on new products or special events for repeat customers.
- Describe each loyalty-program feature concisely, including its eligibility rule, customer value, business cost, and measurement plan.
- Focus on tiered rewards: Create multiple levels of loyalty rewards that encourage continued engagement and spending, such as bronze, silver, and gold tiers.
- Personalize rewards: Use customer data to offer personalized rewards that cater specifically to the interests and purchasing habits of your customers.
- Promote exclusivity: Design special rewards or exclusive events for top-tier loyalty members to foster a sense of exclusivity and value, enhancing customer retention. Create a referral program: Encourage existing customers to refer new customers by offering rewards or discounts for successful referrals.
- Offer surprise rewards: Occasionally surprise loyal customers with unexpected rewards or discounts to show appreciation and keep them engaged.
- Track and analyze customer behavior: Use analytics tools to track customer behavior and preferences, allowing you to tailor your loyalty program and rewards to better meet their needs.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My business niche: Career-transition coaching for corporate professionals
- My target customer demographics: U.S. corporate professionals ages 35–50, earning at least $100,000, who are preparing to start a business
- My products/services: A 12-week group coaching program, private coaching, and implementation workshops

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Generate Daily Productivity Blueprint

- URL: https://returnmytime.com/prompt-library/generate-daily-productivity-blueprint
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 2+ hours
- Tags: business-operations-growth, generate, daily, productivity
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307090305id_/https://returnmytime.com/prompt-library/generate-daily-productivity-blueprint
- Source captured: 2026-03-07T09:03:05Z
- Source hash: ab5afa0b9587c76ae4a6cfae77cff30aa4c4a1205fdd5a451fc9da9e5dcb3f0d


Use this prompt template to create a customized daily productivity blueprint tailored to your needs. It includes morning routines, focused work sessions, task batching, supportive environments, and quality downtime to enhance your productivity and well-being.

#### What it does

- Creates a daily productivity blueprint tailored to individual needs and preferences.
- Incorporates techniques like the Pomodoro Technique, task batching, and themed days to enhance focus and efficiency.
- Emphasizes the balance between productive work sessions and quality downtime to maintain overall well-being and sustained productivity.

#### When to use

- Use when current productivity problems and the desired workday are specific enough to design a daily routine.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting daily productivity blueprint.

#### Required inputs

- MY CURRENT PRODUCTIVITY CHALLENGES
- MY IDEAL PRODUCTIVE DAY

#### Prompt

#CONTEXT:
You are seeking to enhance your productivity and achieve maximum output in less time. To accomplish this, you need a comprehensive daily productivity blueprint that incorporates proven techniques and personal well-being practices.

#ROLE:
As your productivity enhancer, my role is to provide you with a tailored, effective productivity blueprint that combines best practices and personalized recommendations to help you optimize your daily routine and achieve your goals.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

The productivity blueprint should include the following components:

1. Introduction: A brief overview of the purpose and benefits of the productivity blueprint.

2. Morning Routine: A step-by-step guide to an energizing and mentally preparing morning routine.

3. Focused Work Sessions: Recommendations for implementing focused work sessions, such as the Pomodoro Technique, along with suggestions for restorative breaks to maintain high energy levels.

4. Task Batching and Theme Days: Strategies for reducing context switching and enhancing deep work through task batching and theme days.

5. Supportive Environment: Tips for creating a distraction-minimizing environment that encourages concentration.

6. Quality Downtime: Emphasis on the importance of quality downtime for recharging and returning to work with renewed vigor and clarity.

7. Personalized Advice: Two questions to assess your current situation and provide tailored advice based on your responses.

#PRODUCTIVITY BLUEPRINT CRITERIA:
1. The blueprint should be comprehensive, covering all aspects of a productive day, from morning routine to quality downtime.
2. Each component of the blueprint should be actionable, providing clear steps and strategies for implementation.
3. The blueprint should emphasize the importance of both focused work and restorative practices.
4. Personalized advice should be based on your responses to the two questions and tailored to your unique situation and needs.

#INFORMATION ABOUT ME:
- My current productivity challenges: [MY CURRENT PRODUCTIVITY CHALLENGES]
- My ideal productive day: [MY IDEAL PRODUCTIVE DAY]

#RESPONSE FORMAT:
Expected output: A daily productivity blueprint with priorities, time blocks, routines, distraction controls, and personalized implementation advice.

The productivity blueprint should be presented in a clear, easy-to-follow format, utilizing a combination of paragraphs, bullet points, and numbered lists to organize the information. The personalized advice section should be distinctly separated from the main blueprint components.

#### Expected output

- A daily productivity blueprint with priorities, time blocks, routines, distraction controls, and personalized implementation advice.

#### Customization tips

- Fill in the “ask me 2 questions to figure out my current situation” placeholder with specific questions that will help tailor the productivity blueprint to your unique needs.
- Example: "What are your top three productivity challenges?" and "How many hours do you currently dedicate to focused work each day?"
- Use the blueprint to structure your day around proven productivity techniques and personal well-being practices, ensuring each recommended action is actionable and specific to your lifestyle and work demands.
- Regularly update and adjust the blueprint as you answer the tailored questions and implement the strategies, tracking which techniques improve your productivity and well-being the most.
- Create a morning routine that includes meditation or light exercise to boost mental clarity and physical readiness for the day's tasks.
- Implement the Pomodoro Technique, working in 25-minute blocks followed by 5-minute breaks to sustain concentration and prevent burnout.
- Organize tasks into batches or assign theme days for similar tasks to minimize context switching and promote deeper focus on each task type.
- Design a workspace that limits interruptions, perhaps using noise-cancelling headphones and having a clear, organized desk, to foster a productive environment.
- Incorporate regular, scheduled downtime that involves activities completely unrelated to work, such as hobbies or spending time with loved ones, to enhance overall well-being and productivity.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My current productivity challenges: Reactive inbox work, fragmented client follow-up, and unprotected project time
- My ideal productive day: Client calls from 10 a.m. to 1 p.m., a two-hour creation block, one 30-minute admin block, and no meetings after 3 p.m.

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Generate Growth Hacking Strategies

- URL: https://returnmytime.com/prompt-library/generate-growth-hacking-strategies
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 2+ hours
- Tags: business-operations-growth, generate, growth, hacking
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260308003652id_/https://returnmytime.com/prompt-library/generate-growth-hacking-strategies
- Source captured: 2026-03-08T00:36:52Z
- Source hash: 2be2764925eac8fb3739e4e588a5b96e6f88f0a30b478ec518139de60298e155


Use this prompt template to uncover potent and budget-friendly growth hacking strategies tailored for niche markets. It guides thorough research, organizes findings in a table, and ensures proper citation of all sources.

#### What it does

- Identifies growth hacking opportunities for specific business, product, or service.
- Conducts thorough research to support the identification of these opportunities.
- Presents findings in a table format with well-cited sources.

#### When to use

- Use when a defined offer, niche, and audience need five testable acquisition experiments with costs and metrics.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting growth hacking strategies.

#### Required inputs

- MY BUSINESS/PRODUCT/SERVICE
- MY NICHE
- MY TARGET AUDIENCE

#### Prompt

#CONTEXT:
You are a growth hacking expert with deep knowledge in business strategy, digital marketing, and data analysis. Your task is to help the user identify the most effective and affordable growth hacking strategies for their specific business, product, or service in its niche.

#ROLE:
Adopt the role of a growth hacking expert with extensive experience in business strategy, digital marketing, and data analysis. Provide insights and recommendations from this perspective.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Begin with a niche overview, including a description of the niche, target audience demographics, and key competitors.
2. Provide a table of growth hacking opportunities with the following columns:
   - Strategy
   - Description
   - Cost
   - Potential Impact
   - Implementation Steps
3. Include at least 5 growth hacking strategies in the table.
4. End with a list of 5 reliable sources used in your research and analysis.

#GROWTH HACKING CRITERIA:
1. Strategies should be effective and affordable for the specific business, product, or service in its niche.
2. Conduct extensive research, analyzing competitors, target audience, and industry trends.
3. Provide a comprehensive overview of each growth hacking opportunity, backed by reliable sources.
4. Focus on strategies that have the highest potential impact and are feasible to implement.

#INFORMATION ABOUT ME:
- My business/product/service: [MY BUSINESS/PRODUCT/SERVICE]
- My niche: [MY NICHE]
- My target audience: [MY TARGET AUDIENCE]

#REVIEW REQUIREMENTS:
1. Treat every cost, price, revenue, funding, profitability, and market-potential figure as a scenario until the primary-source review below is complete.

#RESPONSE FORMAT:
Expected output: A niche overview and five-row growth-strategy table with cost, potential impact, implementation steps, sources, and success metrics.

Niche Overview:
- Niche Description
- Target Audience Demographics
- Key Competitors

Growth Hacking Opportunities:

| Strategy | Description | Cost | Potential Impact | Implementation Steps |
|----------|-------------|------|------------------|----------------------|
| Strategy 1 | Description 1 | Cost 1 | Impact 1 | Steps 1 |
| Strategy 2 | Description 2 | Cost 2 | Impact 2 | Steps 2 |
| Strategy 3 | Description 3 | Cost 3 | Impact 3 | Steps 3 |
| Strategy 4 | Description 4 | Cost 4 | Impact 4 | Steps 4 |
| Strategy 5 | Description 5 | Cost 5 | Impact 5 | Steps 5 |

Sources:
1. Source 1
2. Source 2
3. Source 3
4. Source 4
5. Source 5

#### Expected output

- A niche overview and five-row growth-strategy table with cost, potential impact, implementation steps, sources, and success metrics.

#### Customization tips

- Fill in the “my business/product/service” and “niche” placeholders with specific details about your business and its market niche.
- Example: If your business is an online educational platform, and your niche is competitive exam preparation, you would specify "online educational platform for competitive exam preparation" as your business/product/service and "competitive exam preparation" as your niche.
- Use the prompt to explore growth hacking strategies tailored to your specific business type and niche, focusing on both innovative and cost-effective methods.
- Ensure the response includes a well-organized table summarizing the growth hacking opportunities and cites sources accurately to validate the research.
- Identify complementary businesses or influencers in your niche and propose mutually beneficial partnerships or collaborations.
- Utilize A/B testing on different growth strategies to identify the most effective approach for your specific audience within the niche.
- Develop a feedback loop where insights and data gathered from the research are continuously used to refine and optimize the growth hacking strategies. Create a comprehensive list of potential growth hacking opportunities specific to your niche.
- Utilize social media platforms to engage with your target audience and build brand awareness.
- Implement referral programs to incentivize existing customers to refer new customers.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.
- Current primary sources required: Use current channel rate cards, vendor pricing pages, and dated experiment data as primary sources for strategy-cost and impact estimates.

#### Example input

## Information about me

- My business/product/service: The Career Pivot Program - a 12-week coaching program helping corporate professionals build their first business
- My niche: Career transition coaching for corporate professionals
- My target audience: Corporate professionals (ages 35-50) feeling stuck in their careers, earning $100K+ but unfulfilled, ready to build something of their own

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Generate Outsourcing Strategy Plan

- URL: https://returnmytime.com/prompt-library/generate-outsourcing-strategy-plan
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 2+ hours
- Tags: business-operations-growth, generate, outsourcing, strategy
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307092423id_/https://returnmytime.com/prompt-library/generate-outsourcing-strategy-plan
- Source captured: 2026-03-07T09:24:23Z
- Source hash: f48dfd0d7611c8715357d1793b5397e4ab17f8b670a1c7be9b269de1785ff175


Use this prompt template to streamline your outsourcing strategy by identifying tasks that can be outsourced and organizing them into a manageable plan. This tool helps you categorize and table your tasks efficiently, enhancing productivity and oversight.

#### What it does

- Identifies tasks suitable for outsourcing and categorizes them.
- Creates a structured plan for managing outsourced tasks effectively.
- Organizes tasks into a table format for clear visualization and tracking.

#### When to use

- Use when a task inventory, business goals, and current team structure are available for an outsourcing decision.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting outsourcing strategy plan.

#### Required inputs

- MY TASK LIST
- MY BUSINESS GOALS
- MY CURRENT TEAM STRUCTURE

#### Prompt

#CONTEXT:
Adopt the role of an expert business consultant specializing in outsourcing strategies and optimization. Your task is to help the user analyze a given task list and create a comprehensive outsourcing plan to optimize their business operations.

#ROLE:
You are an expert business consultant specializing in outsourcing strategies and optimization. Your goal is to provide specific, actionable recommendations for effectively managing outsourced work based on a thorough analysis of the user's task list.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Begin with a thorough analysis of the provided task list, identifying opportunities for outsourcing.
2. Create an outsourcing suitability table with the following columns:
   - Task
   - Outsourcing Suitability (score from 1-5, with 5 being the most suitable)
   - Rationale
3. Provide specific outsourcing recommendations based on the analysis.
4. Outline vendor selection criteria to ensure the best fit for the outsourced tasks.
5. Develop a communication and monitoring plan to manage the outsourced work effectively.
6. Suggest quality control measures to maintain high standards for the outsourced tasks.
7. Identify potential risks and provide mitigation strategies.
8. Create a timeline for implementing the outsourcing plan.
9. Estimate the budget required for the outsourcing initiative.
10. Calculate the expected return on investment (ROI) from outsourcing.

#TASK CRITERIA:
1. Focus on identifying tasks that can be effectively outsourced to improve business efficiency and cost-effectiveness.
2. Provide clear rationales for outsourcing suitability scores.
3. Ensure that the outsourcing recommendations are specific, actionable, and aligned with the user's business goals.
4. Avoid generic advice and tailor the recommendations to the user's unique situation and task list.

#INFORMATION ABOUT ME:
- My task list: [MY TASK LIST]
- My business goals: [MY BUSINESS GOALS]
- My current team structure: [MY CURRENT TEAM STRUCTURE]

#REVIEW REQUIREMENTS:
1. Treat every cost, price, revenue, funding, profitability, and market-potential figure as a scenario until the primary-source review below is complete.
2. Worker classification and employment treatment vary by facts and jurisdiction. The current-source and named-counsel review requirements below govern this draft.

#RESPONSE FORMAT:
Expected output: A task-suitability table with 1–5 scores, outsourcing recommendations, vendor criteria, communication plan, risks, and implementation steps.

Task List Analysis:
<Thorough analysis of the provided task list, identifying outsourcing opportunities>

Outsourcing Suitability Table:
| Task | Outsourcing Suitability (1-5) | Rationale |
|------|-------------------------------|-----------|
| <Task 1> | <Score> | <Rationale> |
| <Task 2> | <Score> | <Rationale> |
| ... | ... | ... |

Outsourcing Recommendations:
<Specific, actionable recommendations based on the analysis>

Vendor Selection Criteria:
<Criteria for selecting the best-fit vendors for outsourced tasks>

Communication and Monitoring Plan:
<Plan for effectively managing and monitoring outsourced work>

Quality Control Measures:
<Measures to ensure high-quality output from outsourced tasks>

Risk Mitigation Strategies:
<Strategies to mitigate potential risks associated with outsourcing>

Outsourcing Timeline:
<Timeline for implementing the outsourcing plan>

Outsourcing Budget:
<Estimated budget required for the outsourcing initiative>

Expected ROI:
<Calculation of the expected return on investment from outsourcing>

#### Expected output

- A task-suitability table with 1–5 scores, outsourcing recommendations, vendor criteria, communication plan, risks, and implementation steps.

#### Customization tips

- Fill in the “get my tasks list” placeholder with a detailed list of your current tasks. Example: If you are running a digital marketing agency, your tasks list might include "social media management, content creation, email marketing, and client reporting."
- Organize the tasks into categories that can be easily outsourced, such as repetitive tasks or tasks requiring specialized skills not available in-house. Use a table format to clearly distinguish between different types of tasks and potential outsourcing opportunities.
- Develop a management plan for overseeing outsourced work, ensuring it includes communication strategies, quality control measures, and timelines to efficiently integrate outsourced tasks with your in-house operations.
- Identify Core vs. Non-Core Activities: Determine which tasks are essential to your core business operations and which are supportive. Outsource non-core tasks to focus more on strategic growth areas.
- Choose the Right Partners: Select outsourcing partners who have proven expertise in the specific tasks you need help with. Assess their credentials, past work, and client testimonials to ensure quality and reliability.
- Establish Clear Communication Channels: Set up regular meetings and reporting systems to stay updated on the progress of outsourced tasks. Use tools like Slack, Trello, or Asana to facilitate smooth communication and project management. Create a Detailed Task List: Break down your tasks into specific and manageable components to ensure clarity and accuracy when outsourcing.
- Define Clear Expectations: Clearly communicate your expectations, deadlines, and desired outcomes to your outsourcing partners to avoid misunderstandings and ensure alignment.
- Monitor and Evaluate Performance: Regularly review the performance of your outsourcing partners to ensure they are meeting your standards and delivering quality work. Provide feedback and make necessary adjustments as needed.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.
- Current primary sources required: Use current vendor proposals, published provider rate cards, and internal task-time records as primary sources for outsourcing cost comparisons.
- Verify worker-classification and employment claims against current primary official sources for the relevant jurisdiction, including https://www.dol.gov/agencies/whd/fact-sheets/13-flsa-employment-relationship and https://www.irs.gov/businesses/small-businesses-self-employed/independent-contractor-self-employed-or-employee for U.S. use, and record the review date.
- A named employment counsel serving as the human reviewer must review and approve every worker-classification, pay, contract, and employment conclusion before use.

#### Example input

## Information about me

- My task list: Inbox triage, call-note cleanup, CRM updates, client scheduling, and weekly reporting
- My business goals: Increase qualified discovery calls while preserving 15 weekly hours for client delivery
- My current team structure: One founder, one part-time virtual assistant, and two specialist contractors

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Identify New Business Opportunities

- URL: https://returnmytime.com/prompt-library/identify-new-business-opportunities
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 1-2 hours
- Tags: business-operations-growth, identify, business, opportunities
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307093340id_/https://returnmytime.com/prompt-library/identify-new-business-opportunities
- Source captured: 2026-03-07T09:33:40Z
- Source hash: 230d82e6f370712941a20716f0d0411123b6c12d26adc95c85862107ddf32519


Use the prompt template to harness Google's 8 principles of innovation, identifying new business opportunities and addressing blind spots. This tool guides innovation consultants in brainstorming actionable ideas, evaluating their impact and feasibility, and developing structured action plans to drive growth and innovation.

#### What it does

- Generates a detailed innovation strategy using Google's 8 principles.
- Prioritizes ideas based on impact and feasibility scores.
- Creates an actionable plan for each top opportunity.

#### When to use

- Use when a business can describe its current constraints, offer, and customers and needs adjacent opportunities ranked.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting new business opportunities.

#### Required inputs

- MY CURRENT SITUATION AND CHALLENGES
- MY BUSINESS
- MY TARGET CUSTOMERS

#### Prompt

#CONTEXT:
I'm looking to find new opportunities and blind spots in my business using Google's 8 principles of innovation. I want to reframe my mission, pursue moonshot ideas, embrace experimentation, seek diverse inspiration, share knowledge, balance imagination with data, open up my capabilities to others, and take smart risks.

#ROLE:
Innovation consultant applying Google's 8 principles of innovation to identify growth opportunities and push creative boundaries.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

For each of Google's 8 principles:
- Brainstorm 2-3 specific ideas to implement or explore
- Provide examples to bring ideas to life

Prioritize the top 3 opportunities based on:
- Impact score from 0 to 10 (10 — highest)
- Feasibility score from 0 to 10 (10 — highest)

Provide an action plan for each top opportunity.

The ideas should:
- Be creative and push outside comfort zone
- Inspire action
- Avoid playing it safe

#TASK CRITERIA:
- Focus on identifying customer pain points, bold moonshot ideas, areas to experiment more, neglected sources of inspiration, opportunities to share knowledge, imaginative ideas to pressure-test, assets others could build on, and smart risks worth taking
- Avoid analysis paralysis, information hoarding, dismissing bold ideas without data, or failing to reframe failure as learning
- Prioritize ideas based on potential impact and feasibility

#INFORMATION ABOUT ME:
- My current situation and challenges: [MY CURRENT SITUATION AND CHALLENGES]
- My business: [MY BUSINESS]
- My target customers: [MY TARGET CUSTOMERS]

#RESPONSE FORMAT:
Expected output: Eight principle-based opportunity sections plus a scored table of the top three opportunities by impact and feasibility.

- Organize response into sections for each of the 8 principles
- Use bullet points for brainstormed ideas and action plans
- Create a table with columns for top 3 opportunities, impact scores, and feasibility scores
- Use emojis to highlight key points:
💡 Creative ideas
🎯 High-impact opportunities
✅ Feasible to implement

#### Expected output

- Eight principle-based opportunity sections plus a scored table of the top three opportunities by impact and feasibility.

#### Customization tips

- Fill in the “context”, “your business”, and “your target customers” placeholders with specific details about your current business challenges, the nature of your business, and the characteristics of your target customers.
- Example: For “context”, you might write "Facing increased competition and market saturation"; for “your business”, "We are a tech startup focusing on AI-driven solutions"; for “your target customers”, "Tech-savvy millennials looking for automation tools."
- Use the structured response guidelines to methodically apply each of Google's 8 principles of innovation. This involves brainstorming specific ideas, providing examples, and developing an action plan for each principle, ensuring that each idea is evaluated for impact and feasibility.
- Emphasize creativity and practicality in your brainstorming sessions. Push the boundaries of conventional thinking as suggested, but also keep the ideas actionable and grounded in the realities of your business environment. Use the emojis effectively to make the key points stand out and to visually categorize the ideas and opportunities.
- Map Out the Innovation Landscape: Create a visual map or chart that categorizes your current projects, initiatives, and ideas according to Google's 8 principles. This will help you see where you're already active and where there are gaps to fill.
- Set Up Innovation Labs: Dedicate spaces or times where your team can freely explore and experiment with new ideas without the pressure of immediate results. This encourages embracing experimentation and taking smart risks.
- Develop a Feedback Loop: Implement a system where insights and learnings from applying these principles are regularly collected and reviewed. This helps in refining approaches and sharing knowledge across the organization. Identify Customer Pain Points: Conduct surveys, interviews, and feedback sessions to understand your customers' pain points and areas for improvement. This will help you reframe your mission and prioritize ideas that address these pain points.
- Collaborate with External Partners: Seek partnerships with other businesses, organizations, or individuals who can bring diverse inspiration and fresh perspectives to your innovation process. This can lead to new opportunities and unique ideas.
- Encourage Cross-Functional Collaboration: Create opportunities for employees from different departments or teams to collaborate and share knowledge. This can spark creativity, foster innovation, and lead to breakthrough ideas.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My current situation and challenges: Private coaching is at capacity, lead flow is inconsistent, and delivery knowledge lives in the founder's head
- My business: A boutique career-transition coaching practice offering private and cohort programs
- My target customers: U.S. corporate professionals ages 35–50 who earn at least $100,000 and are validating a first business

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Master Your Niche

- URL: https://returnmytime.com/prompt-library/master-your-niche
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 1-2 hours
- Tags: business-operations-growth, master, your, niche
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307092514id_/https://returnmytime.com/prompt-library/master-your-niche
- Source captured: 2026-03-07T09:25:14Z
- Source hash: 1252c48560af9a1ecd6e86da38d37ce23b7601147d9385234fd0681ce12f8162


Use this prompt template to develop a comprehensive niche domination strategy that enhances your unique strengths and positions you as a leader in your market. This strategy includes detailed market analysis, a branding overhaul, content marketing plans, and strategic collaborations, all tailored to ensure your business's growth and market leadership.

#### What it does

- Conducts a detailed market analysis to identify untapped sub-niches and emerging trends for strategic advantage.
- Recommends a comprehensive branding overhaul to enhance market presence and attract a dedicated customer base.
- Develops a content marketing and thought leadership plan to position the business as a leading authority in the niche.

#### When to use

- Use when niche, differentiated strengths, and current market position are known and a focused authority strategy is needed.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting niche.

#### Required inputs

- MY CURRENT NICHE
- MY UNIQUE STRENGTHS
- MY CURRENT MARKET POSITION

#### Prompt

#CONTEXT:
You are an expert niche mastery advisor. Your task is to help the user develop a niche domination strategy that leverages their unique strengths and sets them apart in a crowded market, ensuring a steady rise in income.

#ROLE:
Expert niche mastery advisor

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Conduct an in-depth market analysis to pinpoint untapped sub-niches or emerging trends the user can capitalize on.
2. Recommend a branding overhaul that authentically communicates the user's specialized expertise and attracts a loyal customer base.
3. Devise content marketing and thought leadership initiatives that establish the user's business as the go-to authority in their niche.
4. Outline collaboration and joint venture opportunities that can provide quick wins and market penetration.
5. Present this strategy as a fearless pursuit of excellence, empowering the user to strive for a top position in their market with conviction and finesse.
6. Ask the user 2 questions to figure out their current situation, then provide tailored advice.

#NICHE DOMINATION STRATEGY CRITERIA:
1. The strategy should be tailored to the user's unique strengths and niche.
2. It should focus on setting the user apart in a crowded market.
3. The strategy should aim to ensure a steady rise in income for the user.
4. It should be comprehensive, covering market analysis, branding, content marketing, thought leadership, collaborations, and joint ventures.
5. The strategy should empower the user to fearlessly pursue excellence and strive for a top position in their market.

#INFORMATION ABOUT ME:
- My current niche: [MY CURRENT NICHE]
- My unique strengths: [MY UNIQUE STRENGTHS]
- My current market position: [MY CURRENT MARKET POSITION]

#RESPONSE FORMAT:
Expected output: A niche-positioning strategy with audience insight, differentiation, authority-building actions, channel priorities, and execution steps.

Provide the niche domination strategy in a clear, structured format using headings, subheadings, and bullet points as needed. Ensure that each section of the strategy is easily identifiable and digestible for the user.

#### Expected output

- A niche-positioning strategy with audience insight, differentiation, authority-building actions, channel priorities, and execution steps.

#### Customization tips

- Fill in the “unique strengths”, “current market position”, and “niche” placeholders with your specific strengths, the market you operate in, and the niche you want to dominate.
- Example: If your unique strengths are in innovative tech solutions, your current market is the tech industry, and your niche is cybersecurity solutions for small businesses.
- Use the prompt to guide the development of a comprehensive strategy that not only highlights your unique strengths but also addresses specific market needs and opportunities. Ensure each section of the strategy is interconnected and builds upon your core competencies.
- Tailor the final presentation of the strategy to reflect a bold and confident approach, emphasizing your commitment to excellence and leadership in the niche. This will help in positioning yourself as a thought leader and a reliable authority in the market.
- Conduct a thorough market analysis to identify gaps and emerging trends, using tools like Google Trends, industry reports, and competitor analysis to find opportunities where your unique strengths can shine.
- Revamp your branding to reflect your deep expertise and unique value proposition, focusing on creating a consistent visual and verbal identity that resonates with your target audience and differentiates you from competitors.
- Implement a robust content marketing strategy that positions you as a thought leader, using blogs, whitepapers, webinars, and social media to share valuable insights that address your audience's pain points and needs. Identify your target audience: Clearly define who your ideal customers are and what their specific needs and pain points are. This will help you tailor your niche domination strategy to effectively meet their demands.
- Stay updated on industry trends: Continuously monitor your niche market to identify emerging trends, new technologies, and changing customer preferences. This will allow you to adapt your strategy and stay ahead of the competition.
- Build strategic partnerships: Look for opportunities to collaborate with other businesses or influencers in your niche. This can help you expand your reach, gain credibility, and tap into new customer segments.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My current niche: Career-transition coaching for experienced corporate professionals
- My unique strengths: A repeatable validation framework, executive marketing experience, 50 client transitions, and strong workshop facilitation
- My current market position: A boutique U.S. practice known through referrals but with low unaided awareness beyond its email audience

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Optimize AI Business Integration

- URL: https://returnmytime.com/prompt-library/optimize-ai-business-integration
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 2+ hours
- Tags: business-operations-growth, optimize, business, integration
- Published: 2026-07-30
- Modified: 2026-07-31
- Provenance: archive
- Source: https://web.archive.org/web/20260307090602id_/https://returnmytime.com/prompt-library/optimize-ai-business-integration
- Source captured: 2026-03-07T09:06:02Z
- Source hash: cb1bb60c8e9f823ac61ddde5e8b0121408793f914e0b5daa5f378f8aa1e2aa7c


Use this prompt template to optimize AI integration in your business, focusing on low-cost solutions to enhance productivity and growth. Tailored to your business description, tech stack, and goals, it delivers actionable AI tool recommendations that align with your strategic objectives.

#### What it does

- Analyzes the specific needs of the business based on the provided description, tech stack, and goals.
- Generates tailored recommendations for AI tools and strategies to enhance productivity and growth.
- Summarizes the top AI recommendations and outlines the next steps for implementation and integration.

#### When to use

- Use when a business, its current technology stack, and measurable goals are documented before evaluating AI tools.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting AI business integration.

#### Required inputs

- MY BUSINESS
- MY CURRENT TECH STACK
- MY BUSINESS GOALS

#### Prompt

#CONTEXT:
You are an expert AI consultant hired to help the user best leverage AI for their business. Your task is to research and recommend low-cost and low-barrier-to-entry AI tools to boost the user's productivity and business growth.

#ROLE:
Expert AI consultant

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Analyze the business description to identify key processes and areas that could benefit from AI and automation.
2. Consider the current tech stack and identify any gaps or opportunities to integrate AI tools.
3. Prioritize recommendations based on alignment with stated business goals and potential for efficiency gains and ROI.
4. Provide top 3 recommendations, each with a justification.
5. Summarize key next steps, including prioritizing initiatives, defining use cases and requirements, and identifying necessary AI tools and technology partners.

#BUSINESS CRITERIA:
1. Focus on low-cost and low-barrier-to-entry AI tools.
2. Prioritize recommendations that align with the user's stated business goals.
3. Aim to enhance efficiency, productivity, and business growth.
4. Avoid recommending complex or expensive AI solutions that may not fit the user's needs or budget.

#INFORMATION ABOUT ME:
- My business: [MY BUSINESS]
- My current tech stack: [MY CURRENT TECH STACK]
- My business goals: [MY BUSINESS GOALS]

#REVIEW REQUIREMENTS:
1. Treat every cost, price, revenue, funding, profitability, and market-potential figure as a scenario until the primary-source review below is complete.

#RESPONSE FORMAT:
Expected output: Three ranked AI-integration recommendations with justifications, use cases, required tools, implementation considerations, and next steps.

Recommendation 1:
- Recommendation: <RECOMMENDATION>
- Justification: <JUSTIFICATION>

Recommendation 2:
- Recommendation: <RECOMMENDATION>
- Justification: <JUSTIFICATION>

Recommendation 3:
- Recommendation: <RECOMMENDATION>
- Justification: <JUSTIFICATION>

Summary:
In summary, my top recommendations to leverage AI in your business are:
1. <RECOMMENDATION_1>
2. <RECOMMENDATION_2>
3. <RECOMMENDATION_3>

The key next steps are to prioritize these initiatives, define specific use cases and requirements, and identify the necessary AI tools and technology partners. With the right implementation roadmap and change management, AI can help enhance <BUSINESS_GOALS> for your business.

#### Expected output

- Three ranked AI-integration recommendations with justifications, use cases, required tools, implementation considerations, and next steps.

#### Customization tips

- Fill in the placeholders “business”, “current tech stack”, and “business goals” with specific details about your business. For example, if your business is an online retail store specializing in eco-friendly products, your tech stack includes Shopify and Google Analytics, and your goals are to increase online sales and improve customer retention.
- Use the <scratchpad> section to critically analyze how AI can enhance specific areas of your business based on the description and tech stack you provided. This analysis should help in identifying the most impactful AI tools.
- In the <recommendations> and <summary> sections, clearly outline AI tools and strategies that align with your business goals. Ensure these recommendations are actionable and include a justification for why each is suitable, focusing on potential ROI and efficiency improvements.
- Identify Key Processes for AI Integration: Analyze your business operations to pinpoint areas where AI can streamline workflows, reduce errors, and save time.
- Select AI Tools That Integrate With Existing Tech: Choose AI solutions that seamlessly integrate with your current tech stack to avoid disruption and leverage existing data.
- Focus on Scalable AI Solutions: Prioritize AI tools that can accommodate business growth and evolving needs without assuming an unverified return on investment.
- Review Data and Permissions: Document the data each proposed tool receives, the permissions it needs, and the accountable owner before implementation.
- Pilot Before Scaling: Define a bounded pilot with a baseline, success measure, review date, and rollback plan before expanding an AI integration.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.
- Current primary sources required: Use current official vendor documentation, pricing pages, security terms, and measured process baselines as primary sources for AI cost and ROI assumptions.

#### Example input

## Information about me

- My business: A boutique career-transition coaching practice with one founder and a part-time assistant
- My current tech stack: HubSpot Starter, Google Workspace, Calendly, Zoom, Stripe, Notion, and Zapier
- My business goals: Respond to qualified leads within 15 minutes, reduce CRM administration by five hours per week, and standardize client handoffs

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas

#### Related articles

- https://returnmytime.com/blog/ai-assessment-before-build

### Optimize Business Process Efficiency

- URL: https://returnmytime.com/prompt-library/optimize-business-process-efficiency
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 2+ hours
- Tags: business-operations-growth, optimize, business, process
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260308000549id_/https://returnmytime.com/prompt-library/optimize-business-process-efficiency
- Source captured: 2026-03-08T00:05:49Z
- Source hash: 8f5f3cf9832a6c0971c37905ce5861c92e323a752ef96b460c3c42bf4799a209


Use this prompt template to design a comprehensive system overhaul proposal that enhances business efficiency. It helps identify bottlenecks in workflows, integrates lean and Six Sigma methodologies, and establishes a sustainable continuous improvement culture, ensuring smoother operations and significant resource savings.

#### What it does

- Maps out existing workflows, identifying inefficiencies and areas prone to errors.
- Suggests implementation of lean practices or Six Sigma methods to streamline processes.
- Provides a framework for implementing new procedures, including pilot tests and feedback mechanisms.

#### When to use

- Use when current processes and improvement goals are documented and a future-state process proposal is needed.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting business process efficiency.

#### Required inputs

- MY BUSINESS
- MY CURRENT PROCESSES
- MY IMPROVEMENT GOALS

#### Prompt

#CONTEXT:
Adopt the role of a process improvement specialist with expertise in lean practices, Six Sigma methods, and change management. Your task is to help the user design a comprehensive system overhaul proposal to revamp current business processes for maximum efficiency and minimal waste.

#ROLE:
You are a process improvement specialist with expertise in lean practices, Six Sigma methods, and change management.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Analyze the current state
- Map out existing workflows
- Identify inefficiencies including delay points, redundancies, and error-prone steps
2. Propose targeted improvements
- Suggest streamlining techniques using lean practices and Six Sigma methods
- Provide a revised process map
- Estimate the expected impact in terms of time savings, resource optimization, and error reduction
3. Outline an implementation plan
- Recommend pilot tests and feedback loops
- Anticipate resistance to change and recommend mitigation strategies
- Provide a rollout timeline
4. Foster a culture of continuous improvement
- Suggest sustainability measures, ongoing optimization, and ways to embed an improvement culture

#TASK CRITERIA:
- Focus on maximizing efficiency and minimizing waste
- Provide a comprehensive, end-to-end plan from analysis through implementation and ongoing improvement
- Give specific, actionable recommendations grounded in lean and Six Sigma methodologies
- Anticipate and address change management challenges
- Avoid vague generalizations; be detailed and precise in your analysis and recommendations

#INFORMATION ABOUT ME:
- My business: [MY BUSINESS]
- My current processes: [MY CURRENT PROCESSES]
- My improvement goals: [MY IMPROVEMENT GOALS]

#RESPONSE FORMAT:
Expected output: A proposal with current-state process map, inefficiencies, revised process, expected impact, implementation plan, metrics, risks, and mitigations.

Organize the proposal using the following structure:
## Current State Analysis
### Process Map
### Inefficiencies
- Delay Points
- Redundancies
- Error-Prone Steps

## Proposed Improvements
### Streamlining Techniques
- Lean Practices
- Six Sigma Methods
### Revised Process Map
### Expected Impact
- Time Savings
- Resource Optimization
- Error Reduction

## Implementation Plan
### Pilot Tests
### Feedback Loops
### Change Management
- Anticipated Resistance
- Mitigation Strategies
### Rollout Timeline

## Continuous Improvement
### Sustainability Measures
### Ongoing Optimization
### Improvement Culture

#### Expected output

- A proposal with current-state process map, inefficiencies, revised process, expected impact, implementation plan, metrics, risks, and mitigations.

#### Customization tips

- Fill in the “your company name” and “specific processes to be improved” placeholders with the name of your company and the specific business processes you want to overhaul.
- Example: If your company is "XYZ Corp" and you want to improve inventory management and customer service processes, mention these explicitly in the proposal.
- Use the bullet points provided in the prompt to structure your proposal, ensuring each section is detailed and tailored to your company’s needs. Focus on providing clear, actionable steps in each section.
- Keep the proposal concise and free of unnecessary jargon while preserving evidence, owners, implementation steps, and success measures.
- Map out existing workflows: Begin by conducting a thorough analysis of current business processes to pinpoint inefficiencies, unnecessary redundancies, or error-prone stages. Use flowcharts or process mapping tools to visualize these workflows clearly.
- Implement Lean or Six Sigma: Integrate Lean management and Six Sigma techniques to eliminate waste and reduce variability in business processes. Focus on value stream mapping to identify and remove non-value-added activities, improving overall process flow and quality.
- Develop a change management strategy: Address potential resistance by preparing a detailed change management plan. This should include communication strategies, training sessions for employees on new processes, and a clear timeline for implementation. Ensure there is a support system in place to help employees adapt to changes. Identify Key Performance Indicators (KPIs): Define measurable metrics that will help track the success of the system overhaul and identify areas for further improvement.
- Involve Stakeholders: Engage employees and key stakeholders throughout the process to gain their buy-in and valuable insights. This will increase the chances of successful implementation and adoption of the new processes.
- Monitor and Evaluate: Continuously monitor and evaluate the effectiveness of the new procedures. Regularly review KPIs and gather feedback from employees to identify any issues or areas that need adjustment.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My business: A coaching practice helping corporate professionals transition to entrepreneurship through 1:1 coaching and group programs
- My current processes: Leads submit a form, receive a manual email, and wait for a calendar link
- My improvement goals: Scale from 1:1 coaching to a hybrid model with group programs and digital courses, while maintaining personal touch

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Optimize Problem-Solving Strategies

- URL: https://returnmytime.com/prompt-library/optimize-problem-solving-strategies
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 1-2 hours
- Tags: business-operations-growth, optimize, problem, solving
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307083046id_/https://returnmytime.com/prompt-library/optimize-problem-solving-strategies
- Source captured: 2026-03-07T08:30:46Z
- Source hash: 36b0a0de476186588bc686246c6aea3cff6ca639d3a9f906a7bed29f00094af6


Use first-principles thinking to dissect business challenges, challenge assumptions, identify core components, and draft direct, ambitious, evidence-bounded solutions.

#### What it does

- Breaks down complex business problems into fundamental principles to identify core issues.
- Generates innovative, evidence-based solutions by reassembling identified components.
- Provides direct, first-principles strategic advice focused on efficiency, evidence, and ambitious but bounded options.

#### When to use

- Use when a specific business problem and its context need first-principles analysis rather than analogy or habit.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting problem-solving strategies.

#### Required inputs

- THE CONTEXT OF MY SITUATION
- THE SITUATION OR PROBLEM I'M FACING

#### Prompt

#CONTEXT:
You are an expert in first principles thinking, specializing in breaking down problems and finding solutions. The user will provide context about a business situation or problem they are facing, and your task is to help them solve it using first principles thinking.

#ROLE:
Act as a direct, first-principles business strategist who questions assumptions and proposes ambitious but evidence-bounded options.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Question the underlying assumptions the user has about their situation or problem, which may be causing them to perceive it as problematic.
2. Break down the situation or problem using first principles thinking.
3. Identify the most basic, evidence-based components that a potential solution could have.
4. Assemble these components in novel ways to help the user solve their business problem.
5. Provide business strategy advice in a direct, concise, first-principles tone.

#FIRST PRINCIPLES THINKING CRITERIA:
1. Avoid relying on analogic thinking or assumptions.
2. Focus on breaking down the problem into its most fundamental components.
3. Use evidence and facts to guide your thinking and problem-solving approach.
4. Question the user's underlying assumptions about their situation or problem.
5. Aim to provide novel and innovative solutions by assembling the basic components in new ways.

#INFORMATION ABOUT ME:
- The context of my situation: [THE CONTEXT OF MY SITUATION]
- The situation or problem I'm facing: [THE SITUATION OR PROBLEM I'M FACING]

#RESPONSE FORMAT:
Expected output: A concise first-principles analysis of assumptions, fundamental components, solution options, and recommended next steps.

Respond in clear, concise paragraphs with line breaks for readability. Use the tone of voice and a direct, concise, first-principles communication style throughout your response.

#### Expected output

- A concise first-principles analysis of assumptions, fundamental components, solution options, and recommended next steps.

#### Customization tips

- Fill in the “context” and “explain your business situation/problem” placeholders with specific details about the environment and the issue you're facing.
- Example: For “context”, you might write "In the rapidly evolving tech industry" and for “explain your business situation/problem”, you could specify "My company is struggling to scale production efficiently."
- Use the prompt to engage in a structured problem-solving session where you systematically question and break down the assumptions underlying your business issue, leading to clearer insights and potential solutions.
- Keep the analysis concise and focused on the assumptions, evidence, root causes, and next tests that materially affect the problem.
- Question Assumptions Rigorously: Always start by challenging the initial assumptions of the business problem. For instance, if the problem is low sales, question whether it's truly due to product quality, market reach, or perhaps pricing strategies.
- Identify Underlying Factors: Break down the problem into its fundamental elements. If the issue is customer retention, dissect factors like customer service, product usability, and customer engagement strategies.
- Innovate Through Synthesis: After identifying the basic components, creatively recombine them to form innovative solutions. For instance, if improving user experience and enhancing technical support are identified components, consider developing a comprehensive customer education program that integrates both elements. Focus on the Root Cause: Dig deep to identify the underlying cause of the problem, rather than just addressing the symptoms. This will help you develop more effective and long-lasting solutions.
- Think Outside the Box: Challenge conventional thinking and explore unconventional approaches to solving the problem. Don't limit yourself to what has been done before, but instead, push the boundaries of innovation.
- Leverage Cross-Disciplinary Knowledge: Draw upon knowledge and insights from different fields or industries to gain new perspectives and ideas. Sometimes the best solutions come from combining ideas from seemingly unrelated areas.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- The context of my situation: A coaching practice receives 40 qualified leads per month through referrals and content
- The situation or problem I'm facing: Discovery-call bookings fell from 30% to 18% after adding two qualification steps, but lead quality did not improve

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Optimize Task Management

- URL: https://returnmytime.com/prompt-library/optimize-task-management
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 1-2 hours
- Tags: business-operations-growth, optimize, task, management
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307084427id_/https://returnmytime.com/prompt-library/optimize-task-management
- Source captured: 2026-03-07T08:44:27Z
- Source hash: 239139e8d4fdaf73c5cf557e2870c0e966f3fe3a1347b4819ae86a32d94e6262


Use the prompt template to streamline your workflow by identifying and eliminating draining tasks, delegating them efficiently, and replacing them with work that ignites your passion. This tailored approach enhances productivity, job satisfaction, and overall business performance.

#### What it does

- Converts user input into actionable strategies for eliminating, delegating, or automating draining tasks.
- Generates creative replacements for these tasks with activities the user is passionate about.
- Explains the expected operational and personal impact of each proposed task change.

#### When to use

- Use when draining tasks and high-energy tasks are identified and work should be eliminated, delegated, automated, or redesigned.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting task management.

#### Required inputs

- MY BUSINESS
- MY MOST HATED, DRAINING TASKS
- MY ABSOLUTE FAVORITE, PASSION TASKS

#### Prompt

#CONTEXT:
Adopt the role of a business optimization expert. Your task is to help the user identify and eliminate their most draining tasks, then replace them with work they are truly passionate about.

#ROLE:
You are a business optimization expert who specializes in helping entrepreneurs and business owners streamline their workflows and focus on the tasks that bring them the most joy and fulfillment.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

For each of the user's most hated, draining tasks, provide:
1. A creative strategy to eliminate, delegate, or automate it (so it's off their plate for good)
2. An inspiring way to replace it with one of their passion tasks (or something similar)
3. The specific potential impact this swap could have on their business, time, and satisfaction

#TASK CRITERIA:
1. Focus on providing specific, actionable strategies for eliminating or delegating draining tasks.
2. Emphasize the positive impact of replacing draining tasks with passion tasks.
3. Avoid generic advice or cliches. Tailor your suggestions to the user's unique situation and goals.

#INFORMATION ABOUT ME:
- My business: [MY BUSINESS]
- My most hated, draining tasks: [MY MOST HATED, DRAINING TASKS]
  1. <DRAINING TASK 1>
  2. <DRAINING TASK 2>
  3. <DRAINING TASK 3>
- My absolute favorite, passion tasks: [MY ABSOLUTE FAVORITE, PASSION TASKS]
  1. <PASSION TASK 1>
  2. <PASSION TASK 2>
  3. <PASSION TASK 3>

#RESPONSE FORMAT:
Expected output: For each draining task, elimination, delegation, or automation options, a replacement passion task, and the expected impact.

For each draining task, provide:
- A bullet point list of creative strategies to eliminate, delegate, or automate it
- A paragraph describing an inspiring way to replace it with a passion task (or something similar)
- A paragraph describing the specific potential impact this swap could have on the user's business, time, and satisfaction

#### Expected output

- For each draining task, elimination, delegation, or automation options, a replacement passion task, and the expected impact.

#### Customization tips

- Complete “My business,” “My most hated, draining tasks,” and “My absolute favorite, passion tasks” with specific operating context and concrete activities.
- Example: Describe a small marketing agency, list scheduling social posts as a draining task, and list creating campaign strategy as a passion task.
- Use the provided structure to address each disliked task systematically: suggest how to eliminate, delegate, or automate it, then propose a more fulfilling replacement. This approach helps you redirect energy toward work that matters most.
- For each recommendation, state the owner, dependencies, risks, and evidence needed before changing the workflow.
- Identify and Prioritize: Clearly list all tasks you perform, rating them from most to least enjoyable. This helps in understanding where your energy is being drained and where it's being boosted.
- Delegate or Outsource: For tasks that are necessary but draining, look into delegating them to other team members or outsourcing to freelancers or agencies who specialize in these areas.
- Integrate Passion Projects: Gradually replace time spent on draining tasks with higher-energy work, scheduling it during peak productivity hours.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My business: A coaching practice helping corporate professionals transition to entrepreneurship through 1:1 coaching and group programs
- My most hated, draining tasks:
  1. Copy/pasting lead details into the CRM
  2. Writing repetitive follow-up emails
  3. Scheduling calls with back-and-forth messages
- My absolute favorite, passion tasks:
  1. Coaching sessions and client strategy
  2. Creating frameworks and templates
  3. Publishing educational content

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Prioritize Your Tasks

- URL: https://returnmytime.com/prompt-library/prioritize-your-tasks
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 2+ hours
- Tags: business-operations-growth, prioritize, your, tasks
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307095205id_/https://returnmytime.com/prompt-library/prioritize-your-tasks
- Source captured: 2026-03-07T09:52:05Z
- Source hash: ea53444ab1040fd1707b96c3ecf41b99d1295d7d2dda589ecddd4082c88677a4


Rank current work against a stated business goal, then identify tasks to focus, eliminate, automate, or delegate with clear rationale.

#### What it does

- Ranks user's tasks based on their importance to the specified business goal, assigning a numerical importance score to each.
- Identifies tasks that can be eliminated due to their low impact on achieving the business goal.
- Provides strategies for automating or delegating remaining tasks to optimize productivity and focus.

#### When to use

- Use when a current goal and complete activity list are available and work must be ranked, eliminated, delegated, or automated.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting tasks.

#### Required inputs

- MY CURRENT BUSINESS GOAL
- MY CURRENT ACTIVITIES/TASKS

#### Prompt

#CONTEXT:
You are acting as a neutral prioritization expert. Your task is to help the user rank tasks by their contribution to the current business goal, then identify responsible opportunities to focus, eliminate, automate, or delegate work.

#ROLE:
Act as a neutral prioritization expert. Explain the evidence and assumptions behind each score and recommendation.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Rank the user's activities in order from most important to least important. Assign each activity a score from 0 (not important) to 100 (extremely important).
2. Suggest activities the user could eliminate because they have low or no impact on their current progress towards their business goal.
3. For any remaining low-leveraged activities, come up with a brief plan on how to automate them (ideally) or delegate the task to someone else.
4. If any of the remaining high-impact tasks can be automated or delegated, inform the user of this as well.
5. Write a short, decisive, motivational explanation about what's important right now and what the user should focus on.
6. If missing sufficient information to prioritize, ask the user for that information and then update the assessment.

#TASK CRITERIA:
1. Focus on identifying the most impactful tasks that directly contribute to the user's current business goal.
2. Provide clear and concise recommendations for eliminating, automating, or delegating low-impact tasks.
3. Give decisive, motivational guidance that keeps the user focused on high-priority tasks.
4. Avoid overwhelming the user with too many tasks or details; keep the advice streamlined and actionable.

#INFORMATION ABOUT ME:
- My current business goal: [MY CURRENT BUSINESS GOAL]
- My current activities/tasks: [MY CURRENT ACTIVITIES/TASKS]

#RESPONSE FORMAT:
Expected output: A 0–100 activity ranking, elimination list, delegation and automation plans, high-impact tasks, and an implementation sequence.

1. Ranking of activities (most important to least important)
   - Activity 1: Score (0-100)
   - Activity 2: Score (0-100)
   - ...

2. Activities to eliminate
   - Activity X
   - Activity Y
   - ...

3. Low-leveraged activities to automate or delegate
   - Activity A: Automation/delegation plan
   - Activity B: Automation/delegation plan
   - ...

4. High-impact tasks that can be automated or delegated
   - Task M: Automation/delegation suggestion
   - Task N: Automation/delegation suggestion
   - ...

5. Motivational explanation and focus points
   - Key point 1
   - Key point 2
   - ...

#### Expected output

- A 0–100 activity ranking, elimination list, delegation and automation plans, high-impact tasks, and an implementation sequence.

#### Customization tips

- Fill in the “business goal” and “tasks” placeholders with your specific business goal and the list of tasks you are currently engaged in.
- Example: If your business goal is "Increase annual revenue by 20%" and your tasks include "Email marketing, Weekly team meetings, Client follow-ups, Social media management," input these details into the placeholders.
- Use the completed prompt to guide the AI in accurately assessing the importance of each task relative to your stated business goal, enabling it to provide a prioritized list and actionable advice on elimination and delegation.
- After receiving the AI's recommendations, review the prioritization and suggested strategies for automation or delegation to ensure they align with your understanding of your business operations and goals. Adjust your approach based on this analysis to optimize productivity and focus on high-impact activities.
- Identify Key Performance Indicators: Before ranking tasks, define clear Key Performance Indicators (KPIs) related to your business goal to measure the impact of each task accurately.
- Utilize Decision-Making Frameworks: Apply frameworks like Eisenhower Box or Pareto Principle (80/20 rule) to categorize and prioritize tasks based on urgency and importance.
- Leverage Technology and Outsourcing: Evaluate appropriate automation and delegation options for lower-value work. Break complex tasks into smaller, reviewable steps before assigning them.
- Evaluate Skillsets: Assess the skills and strengths of your team members or potential outsourced help to determine who is best suited for each task.
- Set Clear Expectations: Clearly communicate your expectations and desired outcomes for each task to ensure everyone is on the same page.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.
- Treat named tools and platforms as archived examples; verify current capabilities, terms, pricing, and policies in each provider's official documentation before use.

#### Example input

## Information about me

- My current business goal: Enroll 20 qualified customers in the next cohort by September 30
- My current activities/tasks: Interview five prospects, revise the sales page, follow up with 18 warm leads, reconcile bookkeeping, redesign the logo, document onboarding, and schedule social posts

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Scale Your Product/Service

- URL: https://returnmytime.com/prompt-library/scale-your-productservice
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 1-2 hours
- Tags: business-operations-growth, scale, your, productservice
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307091300id_/https://returnmytime.com/prompt-library/scale-your-productservice
- Source captured: 2026-03-07T09:13:00Z
- Source hash: b05aa4bcaf175a7664e896a0985300a1de8a0841f89ec7794ec47becb39258d7


Generate five tailored, high-impact strategies to scale your product using the archived prompt template. This tool focuses on proven revenue-boosting tactics including user acquisition, conversion optimization, and market expansion, specifically designed for solopreneurs ready to move beyond early adopters.

#### What it does

- Generates five detailed, actionable ideas for scaling a product, focusing on user acquisition, conversion rate optimization, and market expansion.
- Ensures each idea is backed by source-backed, testable tactics to increase revenue, with specific steps and holistic thinking.
- Formats the response in Markdown, adhering to the character limit and structuring the content for clarity and effectiveness.

#### When to use

- Use when an offer, audience, current revenue, and target revenue are known and a scaling path must be evaluated.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting product/service.

#### Required inputs

- MY PRODUCT
- MY TARGET AUDIENCE
- MY CURRENT REVENUE
- MY TARGET REVENUE

#### Prompt

#CONTEXT:
You are Scale GPT, a professional digital marketer who helps solopreneurs scale their products after getting initial traction. Your task is to help the user generate actionable ideas to grow their product and go beyond early adopters to win a bigger part of the market.

#ROLE:
As a world-class expert in generating actionable ideas to grow internet products, you will provide specific and in-depth advice on proven tactics to increase revenue. Adopt the perspective of a top 1% CMO hired to grow sales.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

Generate 5 actionable ideas for scaling the user's product, focusing on:
1. User acquisition
2. Conversion rate optimization
3. Product development
4. Operations
5. Market expansion

For each idea:
- Provide specific, self-explanatory descriptions of what to do and how to approach the tactic
- Make assumptions about the product when necessary to provide comprehensive advice

#SCALE IDEAS CRITERIA:
- Focus on proven tactics that increase revenue for internet products that have already demonstrated traction
- Think holistically, covering user acquisition, conversion rate optimization, product development, operations, and market expansion
- Be specific and provide exact steps and approaches for each tactic
- Make necessary assumptions about the product to provide advice as a top 1% CMO focused on growing sales

#INFORMATION ABOUT ME:
- My product: [MY PRODUCT]
- My target audience: [MY TARGET AUDIENCE]
- My current revenue: [MY CURRENT REVENUE]
- My target revenue: [MY TARGET REVENUE]

#REVIEW REQUIREMENTS:
1. Treat every cost, price, revenue, funding, profitability, and market-potential figure as a scenario until the primary-source review below is complete.

#RESPONSE FORMAT:
Expected output: A scaling strategy covering current state and revenue gap plus offer, channel, operations, risk, milestone, and metric recommendations.

Use Markdown to format the response into a clear, readable structure. Avoid using XML tags.

#### Expected output

- A scaling strategy covering current state and revenue gap plus offer, channel, operations, risk, milestone, and metric recommendations.

#### Customization tips

- Fill in the “description of your product/service”, “description of your target audience”, “your current revenue”, and “your target revenue” placeholders with specific details about your product, its users, and financial figures.
- Example: "My product is an AI-powered content creation tool designed for small business owners. My target audience is small business owners in the digital marketing field. My current revenue is $50,000 per month, and my target revenue is $100,000 per month."
- Use the provided criteria to generate ideas that are holistic and cover various aspects of scaling a business, such as user acquisition, conversion rate optimization, and market expansion. Ensure that each idea is actionable and includes specific steps or strategies.
- Format your response using Markdown to enhance readability and organization. This includes using bullet points, headers, and bold text to emphasize key points and organize the content logically.
- Leverage Retargeting Campaigns: Implement retargeting strategies to re-engage visitors who didn't purchase initially. Use dynamic ads tailored to the specific products they viewed to increase conversion rates.
- Expand Affiliate Networks: Develop a robust affiliate program to tap into new customer bases through influencers and bloggers who align with your target audience. Offer competitive commissions to incentivize high-quality referrals.
- Optimize Product Bundling: Create product bundles that offer value to your customers while increasing average order value. Analyze purchasing data to identify complementary products and promote these bundles with special discounts. Create a Referral Program: Encourage your existing customers to refer their friends and family to your product by offering incentives such as discounts, exclusive access, or rewards. Make it easy for them to share their referral links and track their progress.
- Implement A/B Testing: Continuously test different variations of your website, landing pages, and marketing campaigns to identify what resonates best with your target audience. Use data-driven insights to optimize your messaging, design, and user experience.
- Leverage Social Media Advertising: Utilize targeted social media ads to reach a wider audience and drive traffic to your website. Experiment with different platforms and ad formats to find the most effective channels for your product.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.
- Current primary sources required: Use current financial statements, capacity records, acquisition data, and written vendor estimates as primary sources for the revenue-gap and scaling scenarios.
- A named qualified financial professional serving as the human reviewer must review and approve every pricing, funding, revenue, cost, and projection conclusion before use.

#### Example input

## Information about me

- My product: The Career Pivot Program - a 12-week coaching program helping corporate professionals build their first business
- My target audience: Corporate professionals (ages 35-50) feeling stuck in their careers, earning $100K+ but unfulfilled, ready to build something of their own
- My current revenue: $42,000 monthly recurring and project revenue
- My target revenue: $70,000 per month within 12 months

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Solve Complex Issues

- URL: https://returnmytime.com/prompt-library/solve-complex-issues
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 2+ hours
- Tags: business-operations-growth, solve, complex, issues
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307084331id_/https://returnmytime.com/prompt-library/solve-complex-issues
- Source captured: 2026-03-07T08:43:31Z
- Source hash: 766e31830e42832d9f4785bd27efdeca134f1151e12caf46e69340a197ffd902


Use the prompt template to integrate multiple strategic frameworks into a comprehensive problem-solving tool. This prompt combines techniques like the Five Whys, TRIZ, Job-To-Be-Done, Ishikawa Diagram, Zwicky Box, Affinity Diagram, Ansoff Matrix, and Impact/Effort Matrix to offer a multidimensional approach to analyzing and solving complex issues across various topics.

#### What it does

- Combines multiple problem-solving and analytical techniques into a unified approach for addressing complex issues.
- Facilitates a comprehensive analysis of a topic by integrating methods like the Five Whys, TRIZ, and Ishikawa Diagram among others.
- Aims to provide deep insights, innovative solutions, and strategic plans tailored to specific business or organizational needs.

#### When to use

- Use when a clearly bounded topic, goal, and constraints need analysis through multiple problem-solving frameworks.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting complex issues.

#### Required inputs

- MY TOPIC
- MY GOAL
- MY CONSTRAINTS

#### Prompt

#CONTEXT:
Adopt the role of an expert in various problem-solving techniques and frameworks. Your task is to help the user solve problems related to a specific topic by applying these techniques systematically.

#ROLE:
Act as a specialist in the following areas:
- Root Cause Analysis
- Inventive Problem Solving (TRIZ)
- Customer Insight (Jobs-to-be-Done Framework)
- Quality Improvement (Ishikawa/Fishbone Diagram)
- Complex Systems Analysis (Zwicky Box/Morphological Analysis)
- Organizational Psychology (Affinity Diagram)
- Growth Strategy (Ansoff Matrix)
- Project Management (Impact/Effort Matrix)

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Identify the problem or topic to be analyzed
2. Apply the relevant problem-solving technique or framework
3. Break down the problem into its components or causes
4. Explore root causes, relationships, and potential solutions
5. Provide insights, strategies, and implementation plans
6. Organize ideas and concepts based on natural relationships or categories
7. Utilize tables, impact/effort scores, or other visual aids for readability and clarity

#TASK CRITERIA:
1. Focus on a systematic, step-by-step approach to problem-solving
2. Dive deep into the root causes and underlying factors contributing to the problem
3. Explore all possible solutions and strategies, considering their impact and feasibility
4. Provide clear explanations and rationales for each step of the process
5. Avoid surface-level analysis or generic advice; aim for specific and actionable insights

#INFORMATION ABOUT ME:
- My topic: [MY TOPIC]
- My goal: [MY GOAL]
- My constraints: [MY CONSTRAINTS]

#RESPONSE FORMAT:
Expected output: A framework-led, step-by-step problem analysis with insights, solution strategies, implementation plan, and key recommendations.

1. Introduce the problem or topic to be analyzed
2. Identify the specific problem-solving technique or framework being applied
3. Present a step-by-step analysis, using bullet points, numbered lists, or tables as appropriate
4. Provide clear insights, strategies, and implementation plans
5. Summarize key findings and recommendations
6. Use emojis, formatting, and visual aids to enhance readability and clarity

#### Expected output

- A framework-led, step-by-step problem analysis with insights, solution strategies, implementation plan, and key recommendations.

#### Customization tips

- Fill in the “topic” placeholder with the specific subject you want to analyze using the combined methodologies.
- Example: If your topic is "customer retention in e-commerce," use this as the topic across all methodologies to create a comprehensive analysis.
- Use this prompt template to generate a detailed, multi-faceted analysis of the chosen topic by applying different problem-solving frameworks. This approach allows for a holistic view, identifying various aspects of the problem and potential solutions.
- Structure your response by addressing each methodology separately within your analysis. Begin with the Five Whys Technique, followed by TRIZ, and continue in the order listed to maintain clarity and organization in your findings.
- Integrate Multiple Analysis Techniques: Use the Five Whys, TRIZ, and Ishikawa Diagram in sequence to deeply analyze and solve complex problems, ensuring a thorough exploration of causes, contradictions, and innovative solutions.
- Enhance Decision-Making Tools: Combine the Ansoff Matrix, Impact/Effort Matrix, and Zwicky Box to evaluate and prioritize growth strategies, efforts, and potential solutions, optimizing resource allocation and strategic direction.
- Focus on Customer-Centric Approaches: Employ the Job-To-Be-Done Framework and Affinity Diagram together to gain a nuanced understanding of customer needs and expectations, aligning product development and marketing strategies to better meet these demands. Use the Affinity Diagram and Ishikawa Diagram together to identify and categorize the root causes of a problem related to “topic”, providing a comprehensive analysis of the underlying issues.
- Combine the Impact/Effort Matrix and Ansoff Matrix to prioritize growth strategies for “topic”, considering both their potential impact and the effort required for implementation, ensuring efficient resource allocation.
- Integrate the Zwicky Box and TRIZ principles to systematically explore and generate innovative solutions for “topic”, leveraging the dimensions, attributes, and interactions within the system.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My topic: Reducing the delay between a qualified lead inquiry and the first human response
- My goal: Bring median response time below 15 minutes without adding a full-time employee
- My constraints: A $1,500 monthly budget, HubSpot Starter, one part-time assistant, and no automated outbound messages without owner approval

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Summarize Weekly Notes

- URL: https://returnmytime.com/prompt-library/summarize-weekly-notes
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 2+ hours
- Tags: business-operations-growth, summarize, weekly, notes
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307084113id_/https://returnmytime.com/prompt-library/summarize-weekly-notes
- Source captured: 2026-03-07T08:41:13Z
- Source hash: 9ed66886593c40c0efb4b5d1665812061d30db99c3049db615fddff38522572e


Use the prompt template to transform your weekly business notes into a structured summary, highlighting key accomplishments and identifying areas for improvement. This tool aids in optimizing your business strategy by providing clear, actionable steps for the upcoming week, ensuring continuous growth and success.

#### What it does

- Analyzes and summarizes a business owner's weekly accomplishments and areas for improvement based on their daily notes.
- Structures the summary into specific sections including Week Overview, Key Accomplishments, Challenges Faced, Insights Gained, Areas for Improvement, and a Motivational Quote.
- Focuses on clarity and specificity, providing actionable steps for improvement and avoiding irrelevant information.

#### When to use

- Use when a complete week of daily notes needs to become a structured review and next-week action plan.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting weekly notes.

#### Required inputs

- MY DAILY NOTES FOR THE WEEK

#### Prompt

#CONTEXT:
You are an expert personal assistant specializing in analyzing and summarizing the accomplishments and areas for improvement for business owners based on their daily notes. Your task is to help the user carefully review their provided daily notes for the week, identifying key accomplishments, challenges, and insights, and synthesize this information into a concise, structured summary that highlights the most significant achievements and suggests actionable ways to optimize performance in the upcoming week.

#ROLE:
Adopt the role of an expert personal assistant, specializing in analyzing and summarizing the accomplishments and areas for improvement for business owners based on their daily notes.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

Organize the information into the following sections:

1. Week Overview
   - Provide a brief recap of the week

2. Key Accomplishments
   - List the top three accomplishments

3. Challenges Faced
   - List the top three challenges

4. Insights Gained
   - List the top three insights

5. Areas for Improvement
   - List three areas for improvement
   - Provide specific action steps for each area

6. Motivational Quote
   - Include a relevant motivational quote and its author

#TASK CRITERIA:
1. Carefully review the provided daily notes for the week.
2. Identify key accomplishments, challenges, and insights.
3. Synthesize the information into a concise, structured summary.
4. Highlight the most significant achievements.
5. Suggest actionable ways to optimize performance in the upcoming week.
6. Avoid including irrelevant or unnecessary information.
7. Focus on providing clear and specific action steps for improvement.

#INFORMATION ABOUT ME:
- My daily notes for the week: [MY DAILY NOTES FOR THE WEEK]

#RESPONSE FORMAT:
Expected output: A weekly review with overview, accomplishments, challenges, insights, improvement areas, priorities, and next-week actions.

Week Overview:
<Brief recap of the week>

Key Accomplishments:
- <Accomplishment 1>
- <Accomplishment 2>
- <Accomplishment 3>

Challenges Faced:
- <Challenge 1>
- <Challenge 2>
- <Challenge 3>

Insights Gained:
- <Insight 1>
- <Insight 2>
- <Insight 3>

Areas for Improvement:
1. <Improvement Area 1>
   <Specific action steps>
2. <Improvement Area 2>
   <Specific action steps>
3. <Improvement Area 3>
   <Specific action steps>

Motivational Quote:
"<Relevant motivational quote>" - <Quote Author>

#### Expected output

- A weekly review with overview, accomplishments, challenges, insights, improvement areas, priorities, and next-week actions.

#### Customization tips

- Fill in the “your daily notes for the week here” placeholder with your specific details. Example: "Met with 5 potential clients, finalized the quarterly budget, attended a leadership seminar."
- Use the structured sections provided in the prompt to organize your summary effectively. Ensure each section is filled with concise, relevant information that reflects the week's activities, challenges, and learnings.
- Focus on actionable insights and clear steps in the "Areas for Improvement" section to foster real progress in the upcoming week. Tailor the motivational quote to inspire and align with the week's experiences or upcoming challenges.
- Utilize a structured template: Adopt a consistent format for each week's summary to streamline the review process and ensure all critical aspects are covered.
- Implement a tracking system: Use tools like Trello or Asana to track weekly accomplishments and challenges, making it easier to reflect and prepare the summary.
- Encourage reflection: Prompt the business owner to regularly update their daily notes with reflections on outcomes and feelings to deepen the insights for the weekly summary. Prioritize the most significant achievements: Highlight the accomplishments that had the greatest impact on the business or moved it closer to its goals.
- Be specific in identifying challenges: Clearly articulate the specific challenges faced during the week, including any obstacles or roadblocks encountered.
- Provide actionable steps for improvement: Offer concrete and practical action steps for each area of improvement to guide the business owner in optimizing performance.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My daily notes for the week:
  - Monday: Reviewed new leads and followed up on 6 inquiries.
  - Tuesday: Updated onboarding emails and cleaned up CRM tags.
  - Wednesday: Recorded a short training video for clients.
  - Thursday: Drafted next week’s content and scheduled posts.
  - Friday: Held 3 coaching calls and wrote action plans.

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Understand Complex Code

- URL: https://returnmytime.com/prompt-library/understand-complex-code
- Category: Business Operations & Growth
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Efficiency
- Estimated time saved: 2+ hours
- Tags: business-operations-growth, understand, complex, code
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260308011239id_/https://returnmytime.com/prompt-library/understand-complex-code
- Source captured: 2026-03-08T01:12:39Z
- Source hash: e42e57be438a3693ab9be9aa12b813fea26b6c8efbeb2eb6ba398469d00823d9


Use this prompt template to transform complex software engineering concepts into simple, relatable explanations. Ideal for educators and developers, it breaks down code snippets, identifies key elements, and explains functionality using everyday language and analogies, making technical knowledge accessible to all.

#### What it does

- Converts complex technical code analysis into a structured, easy-to-understand format for non-technical audiences.
- Provides a detailed breakdown of the code's functionality, usage of programming concepts, and real-world applications.
- Offers practical insights into the code's benefits, limitations, and suggestions for best practices or improvements.

#### When to use

- Use when a concrete code snippet needs a plain-language explanation, risk review, and improvement suggestions.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting complex code.

#### Required inputs

- CODE SNIPPET

#### Prompt

#CONTEXT:
You are an expert software engineer and educator tasked with analyzing a provided code snippet, identifying its key components, logic, algorithms, and design patterns, and explaining its purpose and functionality in plain, non-technical language that a layperson can understand. Provide clear analogies and relatable examples, then explain potential use cases, benefits, limitations, and applicable best practices or improvements.

#ROLE:
Adopt the role of an expert software engineer and educator with deep knowledge of programming languages, software design patterns, algorithms, and the ability to break down and explain complex technical concepts in easy-to-understand, simple terms with metaphors.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Begin with a code snippet section where the user can insert their code.
2. Provide a language and concepts section that identifies the programming language, key concepts, and algorithms/design patterns used in the code snippet.
3. Give a detailed explanation with a high-level overview, a detailed breakdown, and analogies or examples to aid understanding.
4. Include an insights section that lists potential use cases, benefits, limitations, and best practices or improvements.
5. End with another code snippet section for the user to insert their code.

#TASK CRITERIA:
1. Focus on breaking down complex technical concepts into easy-to-understand language suitable for a layperson.
2. Use clear analogies and relatable examples to aid comprehension.
3. Provide insights into potential use cases, benefits, limitations, and best practices or improvements.
4. Avoid using overly technical jargon or assuming prior knowledge of programming concepts.

#INFORMATION ABOUT ME:
- Code Snippet: [CODE SNIPPET]

#RESPONSE FORMAT:
Expected output: A code explanation identifying the language, concepts, algorithms or patterns, logic breakdown, use cases, risks, and improvements.

```
<INSERT CODE SNIPPET HERE>
```

Language and Concepts:
- Programming Language: <language>
- Key Concepts: <key_concepts>
- Algorithms/Design Patterns: <algorithms_patterns>

Explanation:
<high_level_overview>

<detailed_breakdown>

<analogies_examples>

Insights:
Potential Use Cases:
- <use_case_1>
- <use_case_2>

Benefits:
- <benefit_1>
- <benefit_2>

Limitations:
- <limitation_1>
- <limitation_2>

Best Practices/Improvements:
- <best_practice_1>
- <improvement_1>

```
<INSERT YOUR CODE SNIPPET HERE>
```

#### Expected output

- A code explanation identifying the language, concepts, algorithms or patterns, logic breakdown, use cases, risks, and improvements.

#### Customization tips

- Fill in the placeholders “code snippet here”, $language, $key_concepts, $algorithms_patterns, $high_level_overview, $detailed_breakdown, $analogies_examples, $use_case_1, $use_case_2, $benefit_1, $benefit_2, $limitation_1, $limitation_2, $best_practice_1, and $improvement_1 with specific details about the code and its context.
- Example: For $language, you might input "Python"; for $key_concepts, input "Object-Oriented Programming"; for $algorithms_patterns, input "Singleton Design Pattern"; and similarly, fill other placeholders with relevant details and examples.
- Use the structured response format to guide your analysis and explanation of the code snippet, ensuring each section is filled out comprehensively to provide a full understanding of the code's functionality and implications.
- Keep your explanations in the <explanation> and <insights> sections clear and jargon-free, using metaphors and simple language to make the technical content accessible to non-experts, as specified in the role and task descriptions.
- Paste in code snippets and get a detailed breakdown of what it means
- Ask your AI assistant to provide a simpler breakdown and explanation if it’s not understandable
- Ask your AI assistant to offer insights on how the code might be optimized or adapted, giving you a deeper understanding of best practices in software development. Break Down the Code: Start by identifying the main components, logic, algorithms, and design patterns used in the code snippet.
- Use Analogies and Examples: Explain the purpose and functionality of the code using relatable examples and metaphors to make it easier for non-technical individuals to understand.
- Highlight Real-World Use Cases: Discuss how the code can be applied in practical scenarios, emphasizing its relevance and potential benefits.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- Code Snippet:
```javascript
function calculateRenewalDate(startDate, termMonths) {
  const renewal = new Date(startDate);
  renewal.setMonth(renewal.getMonth() + termMonths);
  return renewal.toISOString().slice(0, 10);
}
```

#### Related prompts

- https://returnmytime.com/prompt-library/create-business-model-canvas


### Conduct Reddit Market Research

- URL: https://returnmytime.com/prompt-library/conduct-reddit-market-research
- Category: Product Development
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: product-development, conduct, reddit, market
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260308005457id_/https://returnmytime.com/prompt-library/conduct-reddit-market-research
- Source captured: 2026-03-08T00:54:57Z
- Source hash: 9e7d9e74e8d774b32c01fb92d7c1f29bada86d207171e78fe7172b195fcbd38d


Use this prompt template to conduct comprehensive Reddit market research tailored to your product launch ideas, ensuring you understand audience sentiments and trends effectively.

#### What it does

- Converts product launch ideas into specific research questions for Reddit.
- Gathers and analyzes data from Reddit to understand market trends and audience preferences.
- Summarizes research findings to aid in refining and validating product launch strategies.

#### When to use

- Use when a product-launch idea needs source-linked audience and competitor signals from relevant Reddit communities.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting Reddit market research.

#### Required inputs

- MY PRODUCT LAUNCH IDEA

#### Prompt

#CONTEXT:
You are adopting the role of an expert market researcher and strategist. Your task is to help the user conduct comprehensive market research on Reddit to validate a given product launch idea and inform the product strategy.

#ROLE:
As an expert market researcher and strategist, your role is to leverage Reddit to gain valuable insights into potential markets, target audiences, and consumer sentiment for new product ideas. You will analyze user discussions, compile key findings, and provide recommendations to guide the product strategy.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

Organize the research findings and recommendations into the following sections:

1. Product Idea: Restate the given product launch idea.

2. Relevant Subreddits: Identify 3 relevant subreddits for the market research.

3. Target Audience Insights: Provide 3 key insights about the target audience based on the Reddit analysis.

4. Competitor Analysis: Analyze 3 competitors and list their strengths and weaknesses.

5. Consumer Sentiment: Summarize 3 main consumer sentiments regarding the product idea.

6. Potential Barriers: Identify 3 potential barriers to the product's success.

7. Opportunities: Highlight 3 opportunities for the product based on the research.

8. Key Findings: Summarize the 3 most important findings from the market research.

9. Recommendations: Provide 3 actionable recommendations based on the research insights.

10. Source Ledger: For every factual finding, provide the Reddit discussion URL, community, source date, verification date, and the claim it supports.

#TASK CRITERIA:
1. Focus on identifying the most relevant subreddits for the given product idea.
2. Analyze user discussions to gather insights into the target audience, competitors, consumer sentiment, potential barriers, and opportunities.
3. Prioritize the most critical findings and actionable recommendations that will inform the product strategy.
4. Avoid making assumptions or including information not directly supported by the Reddit research.
5. Add a date caveat to every conclusion: Reddit discussions are time-bound signals and may have changed since the cited source date.

#INFORMATION ABOUT ME:
- My product launch idea: [MY PRODUCT LAUNCH IDEA]

#RESPONSE FORMAT:
Expected output: A Reddit research brief with relevant communities, audience insights, competitor observations, sentiment, pain points, opportunities, source links, source dates, and freshness caveats.

Product Idea:
<Product launch idea>

Relevant Subreddits:
1. <Subreddit 1>
2. <Subreddit 2>
3. <Subreddit 3>

Target Audience Insights:
- <Insight 1>
- <Insight 2>
- <Insight 3>

Competitor Analysis:
- <Competitor 1>: <Strengths>, <Weaknesses>
- <Competitor 2>: <Strengths>, <Weaknesses>
- <Competitor 3>: <Strengths>, <Weaknesses>

Consumer Sentiment:
- <Sentiment 1>
- <Sentiment 2>
- <Sentiment 3>

Potential Barriers:
- <Barrier 1>
- <Barrier 2>
- <Barrier 3>

Opportunities:
- <Opportunity 1>
- <Opportunity 2>
- <Opportunity 3>

Key Findings:
1. <Finding 1>
2. <Finding 2>
3. <Finding 3>

Recommendations:
1. <Recommendation 1>
2. <Recommendation 2>
3. <Recommendation 3>

Source Ledger:
| Finding | Source URL | Community | Source date | Verification date | Freshness caveat |
|---|---|---|---|---|---|
| <Supported finding> | <VALID REDDIT URL> | <SUBREDDIT> | <YYYY-MM-DD> | <YYYY-MM-DD> | This signal may have changed since the source date. |

#### Expected output

- A Reddit research brief with relevant communities, audience insights, competitor observations, sentiment, pain points, opportunities, source links, source dates, and freshness caveats.

#### Customization tips

- Fill in the “product launch idea” placeholder with specific product ideas you are considering for market research.
- Example: If you are exploring the launch of eco-friendly packaging solutions, replace “product launch idea” with "eco-friendly packaging solutions."
- Use this prompt to gather insights from Reddit about public opinion, potential demand, and competitor analysis related to your specified product launch ideas.
- Keep the brief focused on the strongest source-linked findings, while retaining dates and caveats needed to judge freshness.
- Identify Relevant Subreddits: Focus on finding subreddits that align with your product's niche or target audience to gather specific insights and feedback.
- Review Public Discussion Context: Read the surrounding public discussion before coding a statement as evidence so that quotations and sentiment are not stripped of context.
- Analyze Feedback Trends: Use tools like sentiment analysis to quantify the positive and negative feedback from Reddit discussions, helping refine your product launch strategy based on community input. Research Popular Threads: Look for threads that have a high number of upvotes and comments, as these indicate a strong interest in the topic and can provide valuable insights for your product launch.
- Monitor Competitor Mentions: Keep an eye out for mentions of your competitors in Reddit discussions, as this can give you a better understanding of their strategies and help you differentiate your own product.
- Check Sampling Limits: Record community selection, date range, thread selection criteria, and missing segments so the brief does not overstate representativeness.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.
- Verify current platform rules against the primary official source https://redditinc.com/policies/content-policy and record the review date.
- Treat named tools and platforms as archived examples; verify current capabilities, terms, pricing, and policies in each provider's official documentation before use.

#### Example input

## Information about me

- My product launch idea: The Career Pivot Program - a 12-week coaching program helping corporate professionals build their first business

#### Related prompts

- https://returnmytime.com/prompt-library/conduct-market-research-validation
- https://returnmytime.com/prompt-library/generate-product-unique-selling-proposition


### Create Low-Cost MVPs

- URL: https://returnmytime.com/prompt-library/create-low-cost-mvps
- Category: Product Development
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 1-2 hours
- Tags: product-development, create, cost, mvps
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307233404id_/https://returnmytime.com/prompt-library/create-low-cost-mvps
- Source captured: 2026-03-07T23:34:04Z
- Source hash: 74c6bb9b5ef56d5cccdec77fcee605d9eedebf97b88c44bf39b2ab6d0b99ed96


Use the prompt template to strategize and develop a cost-effective, no-code Minimum Viable Product (MVP) tailored specifically for startups. This tool offers actionable advice on building, measuring success, and enhancing MVPs based on specific audience insights, all within tight constraints.

#### What it does

- Generates a list of 10 feasible ideas for constructing a low-cost MVP under specified constraints.
- Provides methodologies to measure the success of the MVP focusing on key metrics and user engagement.
- Outlines strategies for gathering customer feedback effectively and enhancing the core concept based on audience insights.

#### When to use

- Use when a lean product idea, audience, and operating constraints need ten rapid no-code validation concepts.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting low-cost mvps.

#### Required inputs

- MY LEAN PRODUCT IDEA
- ADDITIONAL CONTEXT
- MY TARGET AUDIENCE

#### Prompt

#CONTEXT:
I am working on a startup idea and need help with constructing a low-cost MVP, measuring its success, gathering customer feedback, and generating ideas to add more value to my core concept based on my target audience.

#ROLE:
Adopt the role of a startup advisor with expertise in lean product development and MVP testing.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Provide a list of 10 ideas for constructing a low-cost MVP
   - Each idea should cost $0 to build
   - Each idea should be small enough for a rapid validation sprint
   - Each idea should be built without code

2. Recommend ways to measure the success of the MVP
   - Identify the One Metric That Matters (OMTM) for the MVP
   - Provide ideas on measuring user engagement
   - Define outcomes that mark a successful vs. unsuccessful test

3. Provide ideas on gathering customer feedback for the MVP

4. Generate a list of 10 ideas to test with the target audience on adding more value to the core concept

#MVP CRITERIA:
1. Focus on ideas that are quick, easy, and inexpensive to implement
2. Prioritize ideas that can effectively test the core value proposition
3. Avoid ideas that require significant time, resources, or technical expertise

#INFORMATION ABOUT ME:
- My lean product idea: [MY LEAN PRODUCT IDEA]
- Additional context: [ADDITIONAL CONTEXT]
- My target audience: [MY TARGET AUDIENCE]

#RESPONSE FORMAT:
Expected output: Ten rapid, no-code MVP ideas constrained to zero build cost, plus success metrics and customer-feedback methods.

Answer me the following things:

1) Come up with 10 ideas on how can I construct a low-cost MVP for this

Constraints:

- Costs $0 to build
- Small enough for a rapid validation sprint
- Can be built without code

2) Ideas on how I can Measure success of the MVP

Guiding questions:

- What's the OMTM (one metric that matters) of this MVP?
- How can I measure user engagement
- What outcomes could mark a successful vs unsuccessful test? (Draw a line in the sand)

3) Ideas on how I can get customer feedback for this MVP

4) Come up with 10 ideas I can test with my audience on how I can add more value to the core concept based on my audience

#### Expected output

- Ten rapid, no-code MVP ideas constrained to zero build cost, plus success metrics and customer-feedback methods.

#### Customization tips

- Fill in the “your lean product idea”, “context”, and “audience context” placeholders with specific details about your startup idea, its background, and your target audience. Example: If your lean product idea is "a virtual fitness coaching app," your context might be "aimed at busy professionals looking to stay fit," and your audience context could be "health-conscious individuals aged 20-40 who enjoy technology."
- Use the detailed information you provide in the placeholders to tailor the AI's responses to your specific needs, ensuring that the ideas for MVP construction, success measurement, and customer feedback are relevant and actionable for your particular audience and product concept.
- Focus on the constraints and guiding questions provided in the prompt to ensure that the AI-generated ideas are feasible (e.g., cost-effective, quick to build, and measurable) and align with your strategic goals for the MVP, enhancing the likelihood of gaining valuable insights from the MVP testing phase.
- Create a Landing Page: Use platforms like Carrd or Wix to set up a simple, informative landing page that captures email sign-ups and gauges interest.
- Establish Success Criteria: Define clear, quantifiable goals (e.g., 100 sign-ups in the first week) to help determine the success or need for adjustments in your MVP.
- Conduct User Interviews: Reach out to users who interact with your MVP for more in-depth feedback via virtual meetings or phone calls. Create a User Feedback Forum: Set up a dedicated online forum or community where users can provide feedback, ask questions, and share ideas.
- Offer Incentives for Feedback: Encourage users to provide feedback by offering incentives such as discounts, freebies, or exclusive access to new features.
- Utilize User Testing Platforms: Use platforms like UserTesting or TryMyUI to gather feedback from real users who can test your MVP and provide valuable insights.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My lean product idea: A paid 90-minute career-pivot planning workshop with a take-home validation worksheet
- Additional context: One founder, a $1,500 test budget, four weeks, and a goal of enrolling ten participants before building a full course
- My target audience: U.S. corporate professionals ages 35–50 who are considering entrepreneurship

#### Related prompts

- https://returnmytime.com/prompt-library/conduct-market-research-validation
- https://returnmytime.com/prompt-library/generate-product-unique-selling-proposition


### Generate Branded Merchandise Ideas

- URL: https://returnmytime.com/prompt-library/generate-branded-merchandise-ideas
- Category: Product Development
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: product-development, generate, branded, merchandise
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260308004408id_/https://returnmytime.com/prompt-library/generate-branded-merchandise-ideas
- Source captured: 2026-03-08T00:44:08Z
- Source hash: 209ec8791451607bf23a9f7c87f6bb769393812dab99c7418687783b8f582ac4


Generate unique and appealing branded merchandise ideas tailored specifically for your business niche and target audience using the archived prompt template. This tool streamlines the creative process, ensuring your promotional items resonate well and enhance brand visibility.

#### What it does

- Generates creative ideas for branded merchandise tailored to a specific business niche and target audience.
- Develops these ideas into actionable design concepts and product suggestions.
- Focuses on aligning the merchandise with brand identity and appeal to the target audience.

#### When to use

- Use when a business niche and audience are defined and merchandise concepts must fit both brand and buyer.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting branded merchandise ideas.

#### Required inputs

- MY BUSINESS NICHE
- MY TARGET AUDIENCE

#### Prompt

#CONTEXT:
Adopt the role of an expert brand merchandise designer. Your task is to help the user brainstorm creative, practical, and appealing branded merchandise ideas tailored to their specific business niche and target audience.

#ROLE:
You are an expert brand merchandise designer with a knack for creating unique, impactful branded products that resonate with target audiences.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

For each merchandise idea, provide the following information:

1. Product Name
2. Product Description
3. Materials
4. Customization Options
5. Brand Alignment: Explain how the merchandise item aligns with the brand identity.
6. Target Audience Appeal: Explain how the merchandise item appeals to the target audience's preferences and needs.

Generate a total of 5 merchandise ideas.

#TASK CRITERIA:
1. Brainstorm a wide range of creative, practical, and appealing branded merchandise ideas.
2. Tailor each idea specifically to the given business niche and target audience.
3. Provide detailed product descriptions, materials, and customization options.
4. Explain how each merchandise item aligns with the brand identity and appeals to the target audience's preferences and needs.
5. Avoid generic or unrelated merchandise ideas that do not align with the business niche or target audience.

#INFORMATION ABOUT ME:
- My business niche: [MY BUSINESS NICHE]
- My target audience: [MY TARGET AUDIENCE]

#RESPONSE FORMAT:
Expected output: Five merchandise concepts with product details, materials, customization, brand alignment, and audience appeal.

Merchandise Idea 1:
Product Name:
Product Description:
Materials:
Customization Options:
Brand Alignment:
Target Audience Appeal:

Merchandise Idea 2:
Product Name:
Product Description:
Materials:
Customization Options:
Brand Alignment:
Target Audience Appeal:

Merchandise Idea 3:
Product Name:
Product Description:
Materials:
Customization Options:
Brand Alignment:
Target Audience Appeal:

Merchandise Idea 4:
Product Name:
Product Description:
Materials:
Customization Options:
Brand Alignment:
Target Audience Appeal:

Merchandise Idea 5:
Product Name:
Product Description:
Materials:
Customization Options:
Brand Alignment:
Target Audience Appeal:

#### Expected output

- Five merchandise concepts with product details, materials, customization, brand alignment, and audience appeal.

#### Customization tips

- Fill in the “my business”, “niche”, and “target audience” placeholders with specific details about your business, its niche, and the intended audience.
- Example: "My business is a boutique coffee shop, niche in artisanal brews, target audience is coffee enthusiasts aged 20-35 who appreciate specialty blends."
- Use this prompt to generate creative and unique merchandise ideas that align closely with your brand identity and appeal directly to your target audience.
- Keep each merchandise concept concise while retaining its audience fit, brand connection, production considerations, and validation question.
- Identify Niche Appeal: Focus on designing merchandise that resonates specifically with the interests and preferences of your target audience within your niche.
- Leverage Brand Identity: Ensure all merchandise designs align closely with your brand's visual identity and values to strengthen brand recognition and loyalty.
- Test Market Response: Start with a limited release of your new merchandise items to gauge customer interest and feedback before committing to larger production runs. Consider Practicality: Design merchandise that is not only visually appealing but also functional and useful for your target audience.
- Stay on Trend: Keep up with the latest trends and popular designs in your niche to create merchandise that is relevant and appealing to your target audience.
- Collaborate with Influencers: Partner with influencers or industry experts in your niche to co-create merchandise that will resonate with their followers and expand your reach.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My business niche: Career transition coaching for corporate professionals
- My target audience: Corporate professionals (ages 35-50) feeling stuck in their careers, earning $100K+ but unfulfilled, ready to build something of their own

#### Related prompts

- https://returnmytime.com/prompt-library/conduct-market-research-validation
- https://returnmytime.com/prompt-library/generate-product-unique-selling-proposition


### Generate Customer Feedback System

- URL: https://returnmytime.com/prompt-library/generate-customer-feedback-system
- Category: Product Development
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: product-development, generate, customer, feedback
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307083016id_/https://returnmytime.com/prompt-library/generate-customer-feedback-system
- Source captured: 2026-03-07T08:30:16Z
- Source hash: 254e6baa0f190011ed62e106f96ff1ce85afd57338bc06493dfb4763c762a215


Use this prompt template to establish a comprehensive customer feedback system that collects and analyzes responses efficiently, helping you improve your product or service based on user insights.

#### What it does

- Converts any user input into a structured bullet point list.
- Focuses on setting up a system to gather and analyze customer feedback.
- Ensures the feedback system is tailored to the specific needs of the product or service.

#### When to use

- Use when a product or service needs a repeatable multi-channel collection, analysis, reporting, and action loop.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting customer feedback system.

#### Required inputs

- PRODUCT/SERVICE DETAILS

#### Prompt

#CONTEXT:
You are an expert in designing comprehensive systems for gathering, analyzing, and acting on customer insights. Your task is to help the user design a robust, multi-channel customer feedback system that captures, processes, and responsibly applies customer sentiment and input for a given product or service.

#ROLE:
Adopt the role of a customer feedback expert with deep knowledge in designing comprehensive systems for gathering, analyzing and actioning customer insights.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

The response should be organized into the following sections:

1. Target Customer Segments: List the key customer segments to focus feedback efforts on.

2. Key Feedback Channels: Identify the most effective channels for gathering customer feedback.

3. Feedback Collection Methods: Outline specific methods and tactics for collecting feedback through each channel.

4. Feedback Analysis Approach: Describe how the collected feedback will be analyzed and processed to extract actionable insights.

5. Key Metrics to Track: List the most important metrics to monitor for gauging customer sentiment and feedback effectiveness.

6. Insight Reporting Cadence: Specify the frequency and format of reporting customer feedback insights to stakeholders.

7. Feedback Action Plan: Provide a step-by-step plan for translating customer feedback insights into tangible improvements and actions.

#TASK CRITERIA:
- Focus on designing a comprehensive, multi-channel system that covers the entire feedback lifecycle from collection to action.
- Tailor the feedback system to the specific product/service and target customer segments.
- Emphasize actionability and closing the loop with customers.
- Avoid vague or overly generic recommendations.

#INFORMATION ABOUT ME:
- Product/Service Details: [PRODUCT/SERVICE DETAILS]

#RESPONSE FORMAT:
Expected output: A feedback system with customer segments, five channels, collection methods, analysis approach, metrics, reporting cadence, and action plan.

Target Customer Segments:
- <segment1>
- <segment2>
- <segment3>

Key Feedback Channels:
- <channel1>
- <channel2>
- <channel3>
- <channel4>
- <channel5>

Feedback Collection Methods:
- <method1>
- <method2>
- <method3>
- <method4>

Feedback Analysis Approach:
- <analysis1>
- <analysis2>
- <analysis3>

Key Metrics to Track:
- <metric1>
- <metric2>
- <metric3>
- <metric4>

Insight Reporting Cadence:
- <report1>
- <report2>

Feedback Action Plan:
1. <action1>
2. <action2>
3. <action3>
4. <action4>
5. <action5>

#### Expected output

- A feedback system with customer segments, five channels, collection methods, analysis approach, metrics, reporting cadence, and action plan.

#### Customization tips

- Fill in the “my product/service” placeholder with the specific product or service you are seeking feedback on.
- Example: If your product is a mobile app for budget tracking, replace “my product/service” with "mobile app for budget tracking."
- Use this prompt to design a detailed plan or a set of instructions aimed at collecting and analyzing feedback effectively, ensuring it covers various feedback channels like surveys, online reviews, and direct customer interviews.
- Consider specifying the type of analysis you will perform on the feedback, such as sentiment analysis, trend identification, or feature request aggregation, to gain actionable insights from the data collected.
- Implement a Multi-Channel Collection Strategy: Use various platforms like email, social media, and your website to collect feedback, ensuring you reach customers where they are most active.
- Develop a Structured Analysis Plan: Organize feedback into categories such as product features, customer service, and user experience to easily identify areas for improvement.
- Act on Feedback and Communicate Changes: Regularly update customers on how their feedback has been implemented to build trust and encourage continuous engagement. Create a Feedback Loop: Establish a system to continuously gather and incorporate customer feedback into your product/service development process.
- Encourage Honest and Constructive Feedback: Make it clear that you value all types of feedback and provide a safe space for customers to share their thoughts and suggestions.
- Utilize Feedback Tools and Software: Explore the use of feedback management tools and software to streamline the collection, analysis, and reporting of customer feedback.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- Product/Service Details: The Career Pivot Program - a 12-week coaching program helping corporate professionals build their first business

#### Related prompts

- https://returnmytime.com/prompt-library/conduct-market-research-validation
- https://returnmytime.com/prompt-library/generate-product-unique-selling-proposition


### Generate Digital Product Ideas

- URL: https://returnmytime.com/prompt-library/generate-digital-product-ideas
- Category: Product Development
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: product-development, generate, digital, product
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307095247id_/https://returnmytime.com/prompt-library/generate-digital-product-ideas
- Source captured: 2026-03-07T09:52:47Z
- Source hash: 40e92036afef75042206e14d2d30eac9d7c4a31a575d831364448305b543219c


Use this prompt template to generate innovative digital product ideas within a $100,000 budget, complete with a detailed matrix table assessing effort, scores, time, and costs, simplifying project planning and budget management.

#### What it does

- Transforms the user input into actionable ideas for a digital product within a $100,000 budget.
- Creates a matrix table categorizing each idea by effort, scores, time, and costs.
- Separates scenario estimates and assumptions from sourced facts.

#### When to use

- Use when budget and product goal are fixed and three concepts need effort, cost, timeline, and market-potential comparison.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting digital product ideas.

#### Required inputs

- MY BUDGET
- MY GOAL

#### Prompt

#CONTEXT:
You are an expert product strategist tasked with generating innovative ideas for a high-value digital product within the supplied budget. Your goal is to compare each idea's market-potential scenario, development-effort assumption, timeline scenario, and cost scenario in a clear decision matrix.

#ROLE:
As an expert product strategist with deep knowledge in digital product ideation, market analysis, and creating product roadmaps, your role is to provide strategic insights and recommendations for developing a successful digital product within the given budget constraints.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

- Generate 3 innovative digital product ideas
- For each idea, provide:
  - Product name
  - Product description
  - Target market
  - Unique value proposition
  - Development effort score (1-5, where 5 is the highest effort)
  - Market potential score (1-5, where 5 is the highest potential)
  - Estimated development time
  - Estimated cost breakdown
- Create a comparison matrix table with the following columns:
  - Product Idea
  - Development Effort (1-5)
  - Market Potential (1-5)
  - Estimated Timeline
  - Estimated Cost
- Provide a final recommendation based on the analysis

#TASK CRITERIA:
- Focus on generating high-value digital product ideas that can be developed within the $100,000 budget
- Ensure each idea has a clear target market and unique value proposition
- Present development effort, market potential, timeline, and cost estimates as scenarios with explicit assumptions, confidence, and evidence gaps
- Present the findings in a clear and concise manner to facilitate decision-making
- Avoid ideas that are overly complex or require significantly more resources than the given budget allows

#INFORMATION ABOUT ME:
- My budget: [MY BUDGET]
- My goal: [MY GOAL]

#REVIEW REQUIREMENTS:
1. Treat every cost, price, revenue, funding, profitability, and market-potential figure as a scenario until the primary-source review below is complete.

#RESPONSE FORMAT:
Expected output: Three digital-product concepts with target markets, value propositions, effort and potential scores, timelines, cost breakdowns, comparison matrix, and recommendation.

## Digital Product Ideas

### Idea 1
- Product Name: <Product Name>
- Product Description: <Product Description>
- Target Market: <Target Market>
- Unique Value Proposition: <Unique Value Proposition>
- Development Effort Score: <1-5>
- Market Potential Score: <1-5>
- Estimated Development Time: <Estimated Development Time>
- Estimated Cost Breakdown: <Estimated Cost Breakdown>

### Idea 2
- Product Name: <Product Name>
- Product Description: <Product Description>
- Target Market: <Target Market>
- Unique Value Proposition: <Unique Value Proposition>
- Development Effort Score: <1-5>
- Market Potential Score: <1-5>
- Estimated Development Time: <Estimated Development Time>
- Estimated Cost Breakdown: <Estimated Cost Breakdown>

### Idea 3
- Product Name: <Product Name>
- Product Description: <Product Description>
- Target Market: <Target Market>
- Unique Value Proposition: <Unique Value Proposition>
- Development Effort Score: <1-5>
- Market Potential Score: <1-5>
- Estimated Development Time: <Estimated Development Time>
- Estimated Cost Breakdown: <Estimated Cost Breakdown>

## Comparison Matrix
| Product Idea | Development Effort (1-5) | Market Potential (1-5) | Estimated Timeline | Estimated Cost |
|--------------|--------------------------|------------------------|--------------------|-----------------
| <Idea 1 Name>| <Idea 1 Dev Effort>      | <Idea 1 Market Score>  | <Idea 1 Timeline>  | <Idea 1 Cost>   |
| <Idea 2 Name>| <Idea 2 Dev Effort>      | <Idea 2 Market Score>  | <Idea 2 Timeline>  | <Idea 2 Cost>   |
| <Idea 3 Name>| <Idea 3 Dev Effort>      | <Idea 3 Market Score>  | <Idea 3 Timeline>  | <Idea 3 Cost>   |

## Recommendation
<Final recommendation based on the analysis>

#### Expected output

- Three digital-product concepts with target markets, value propositions, effort and potential scores, timelines, cost breakdowns, comparison matrix, and recommendation.

#### Customization tips

- Fill in the “documented inputs” such as the specific type of digital product you are considering, and any constraints or goals you have in mind.
- Example: If you are considering developing an educational app, specify that the digital product is an "Educational App" and include details like target age group or subject matter.
- Use the matrix table to systematically evaluate each product idea by assigning values to effort, scores, time, and costs. This helps in comparing different ideas on a quantifiable basis.
- Keep the comparison concise while retaining the assumptions and evidence needed to interpret each score, timeline, and cost scenario.
- Identify Market Gaps: Before suggesting ideas, conduct thorough market research to identify gaps or needs that are not currently being met by existing digital products.
- Utilize Expertise: Involve stakeholders and experts in relevant fields to brainstorm and validate ideas, ensuring the suggestions are feasible and valuable.
- Focus on Scalability: Consider the potential for scaling each product idea when evaluating effort, scores, time, and costs to ensure long-term viability and profitability. Consider User Experience: Prioritize creating a digital product that provides a seamless and intuitive user experience, as this will greatly impact its success and adoption.
- Research Competitive Pricing: Analyze the pricing strategies of similar digital products in the market to determine the optimal price point for your $100,000 product, ensuring it is competitive yet profitable.
- Test and Iterate: Conduct user testing and gather feedback throughout the development process to continuously improve and refine your digital product, ensuring it meets the needs and expectations of your target audience.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.
- Current primary sources required: Use current provider pricing pages, written vendor estimates, and dated customer research as primary sources for development-cost and market-potential assumptions.

#### Example input

## Information about me

- My budget: $100,000
- My goal: To develop a high-value digital product

#### Related prompts

- https://returnmytime.com/prompt-library/conduct-market-research-validation
- https://returnmytime.com/prompt-library/generate-product-unique-selling-proposition


### Generate Product Ideas

- URL: https://returnmytime.com/prompt-library/generate-product-ideas
- Category: Product Development
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: product-development, generate, product, ideas
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260308000523id_/https://returnmytime.com/prompt-library/generate-product-ideas
- Source captured: 2026-03-08T00:05:23Z
- Source hash: c603bd32b8604ff91c5baf6c902ad5e34bfa766764cb679a1024e2a708c9e6d2


Generate innovative product ideas tailored to your niche and target audience with the prompt template. This tool helps identify market needs and provides well-cited sources, ensuring your brainstorming is informed and effective.

#### What it does

- Converts user input into a structured list of potential product ideas tailored to specific market needs and target audiences.
- Cites sources for each listed product idea to ensure credibility and traceability.
- Separates source-supported market needs from hypotheses that still require validation.

#### When to use

- Use when a niche and audience are known and product concepts must trace back to sourced market needs.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting product ideas.

#### Required inputs

- MY NICHE
- MY TARGET AUDIENCE
- MY DATED MARKET SOURCES

#### Prompt

#CONTEXT:
Adopt the role of a creative ideation expert specializing in product development. Your task is to help the user generate innovative product ideas that address key market needs and pain points for a given niche and target audience.

#ROLE:
You are a creative ideation expert with deep knowledge of market trends, consumer psychology, and innovation strategies.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Identify the niche and target audience.
2. Use the supplied dated market sources to list key market needs and pain points.
3. Generate three innovative product ideas that address the identified market needs and pain points.
4. For each product idea, provide:
   - Product name
   - Description
   - Key features
   - Unique value proposition
   - Target market need addressed
5. For each supported claim, provide the source title, publisher, source date, source URL, and supported claim.
6. If a claim lacks a supplied source, label it "Hypothesis — validation required"; never invent a citation.

#PRODUCT IDEATION CRITERIA:
1. Consider emerging trends, disruptive technologies, and unique value propositions when generating product ideas.
2. Ensure each product idea addresses a specific market need or pain point identified through research.
3. Cite only sources supplied by the user. Treat any need without a supplied source as a hypothesis requiring validation.
4. Focus on creating innovative and unique product ideas that differentiate from existing offerings in the market.

#INFORMATION ABOUT ME:
- My niche: [MY NICHE]
- My target audience: [MY TARGET AUDIENCE]
- My dated market sources: [MY DATED MARKET SOURCES]

#RESPONSE FORMAT:
Expected output: A source-supported or hypothesis-labeled list of market needs and three product ideas with names, descriptions, three features, value propositions, and need alignment.

Niche: <NICHE>
Target Audience: <TARGET AUDIENCE>

Market Needs:
1. <MARKET NEED 1>
   - Evidence status: <Source-supported | Hypothesis — validation required>
   - Source title: <SOURCE TITLE OR "Not supplied">
   - Publisher: <PUBLISHER OR "Not supplied">
   - Source date: <YYYY-MM-DD OR "Not supplied">
   - Source URL: <VALID URL OR "Not supplied">
   - Supported claim: <EXACT CLAIM OR HYPOTHESIS>
2. <MARKET NEED 2 WITH THE SAME EVIDENCE FIELDS>
3. <MARKET NEED 3 WITH THE SAME EVIDENCE FIELDS>

Product Ideas:
1. <PRODUCT 1>
   Name: <PRODUCT NAME 1>
   Description: <PRODUCT DESCRIPTION 1>
   Key Features:
     - <FEATURE 1>
     - <FEATURE 2>
     - <FEATURE 3>
   Value Proposition: <VALUE PROPOSITION 1>
   Target Market Need: <TARGET MARKET NEED 1>

2. <PRODUCT 2>
   Name: <PRODUCT NAME 2>
   Description: <PRODUCT DESCRIPTION 2>
   Key Features:
     - <FEATURE 1>
     - <FEATURE 2>
     - <FEATURE 3>
   Value Proposition: <VALUE PROPOSITION 2>
   Target Market Need: <TARGET MARKET NEED 2>

3. <PRODUCT 3>
   Name: <PRODUCT NAME 3>
   Description: <PRODUCT DESCRIPTION 3>
   Key Features:
     - <FEATURE 1>
     - <FEATURE 2>
     - <FEATURE 3>
   Value Proposition: <VALUE PROPOSITION 3>
   Target Market Need: <TARGET MARKET NEED 3>

Sources:
1. <SOURCE TITLE — PUBLISHER — YYYY-MM-DD — VALID URL — SUPPORTED CLAIM>
2. <SOURCE TITLE — PUBLISHER — YYYY-MM-DD — VALID URL — SUPPORTED CLAIM>
3. <SOURCE TITLE — PUBLISHER — YYYY-MM-DD — VALID URL — SUPPORTED CLAIM>

#### Expected output

- A source-supported or hypothesis-labeled list of market needs and three product ideas with names, descriptions, three features, value propositions, and need alignment.

#### Customization tips

- Complete “My niche,” “My target audience,” and “My dated market sources” with specific context and source title, publisher, date, URL, and supported claim.
- Example: "Market needs might include sustainable packaging solutions, niche could be eco-friendly beauty products, and target audience might be environmentally conscious consumers aged 25-40."
- Use the prompt to generate a list of innovative product ideas that directly address the identified market needs, ensuring they are tailored to appeal to your specified niche and target audience.
- Check every citation against the supplied source. If no supplied source supports a market need, keep it labeled as a hypothesis instead of creating a citation.
- Focus on Market Gaps: Identify unmet needs within your niche by analyzing customer feedback on forums, reviews, and social media to generate targeted product ideas.
- Utilize Competitive Analysis: Examine competitors’ offerings and identify areas they are overlooking. Use tools like SWOT analysis to find opportunities for innovation and differentiation.
- Leverage Data Analytics: Employ data analytics tools to analyze market trends and consumer behavior. Use insights from Google Trends, industry reports, and case studies to validate and refine your product ideas. Collaborate with Your Target Audience: Engage with your potential customers through surveys, focus groups, or online communities to gather their input and ideas for product development.
- Stay Updated on Industry News: Regularly read industry publications, blogs, and newsletters to stay informed about the latest trends, technologies, and market needs that can inspire new product ideas.
- Experiment with Prototypes: Create prototypes or minimum viable products (MVPs) to test and validate your product ideas before investing significant resources. Gather feedback from early adopters to refine and improve your offerings.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My niche: Career transition coaching for corporate professionals
- My target audience: Corporate professionals (ages 35-50) feeling stuck in their careers, earning $100K+ but unfulfilled, ready to build something of their own
- My dated market sources: Customer interview synthesis — Clarity Coaching Co. — 2026-07-15 — https://example.com/research/customer-interviews — recurring concerns about time, risk, and validation

#### Related prompts

- https://returnmytime.com/prompt-library/conduct-market-research-validation
- https://returnmytime.com/prompt-library/generate-product-unique-selling-proposition


### Generate SaaS Product Ideas

- URL: https://returnmytime.com/prompt-library/generate-saas-product-ideas
- Category: Product Development
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: product-development, generate, saas, product
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307082248id_/https://returnmytime.com/prompt-library/generate-saas-product-ideas
- Source captured: 2026-03-07T08:22:48Z
- Source hash: cb4ec31b9bc11b4f373fd4d913ab9f8354633c189cc22a512f31ca04cd663fc4


Generate a high-potential SaaS product idea using the archived prompt template, tailored for aspiring entrepreneurs. This tool guides you through a systematic process, leveraging emerging technologies, market trends, and your unique skills to develop a compelling product vision with significant market potential.

#### What it does

- Guides the entrepreneur through a systematic process to generate and refine a SaaS product idea.
- Focuses on aligning the product concept with emerging technologies, market needs, and the founder's strengths.
- Develops the idea into a compelling vision and value proposition with a structured approach.

#### When to use

- Use when founder skills, entrepreneurial goals, and desired impact are clear enough to assess SaaS opportunities.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting SaaS product ideas.

#### Required inputs

- MY BACKGROUND AND SKILLS
- MY ENTREPRENEURIAL ASPIRATIONS
- THE TYPE OF IMPACT I WANT TO HAVE

#### Prompt

#CONTEXT:
Adopt the role of an expert product innovation consultant. Your task is to help an aspiring entrepreneur generate and refine a truly innovative, high-potential SaaS product idea. Guide them through a systematic ideation process that considers emerging technologies, evolving customer needs, industry trends, and the founder's unique skills and passions.

#ROLE:
Act as an experienced product innovation consultant with deep expertise in SaaS business models, emerging technologies, and customer-centric design thinking methodologies. Your role is to provide strategic guidance, creative prompts, and analytical frameworks to help the aspiring entrepreneur navigate the ideation process and develop a compelling product vision.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

The response should be organized into the following sections:

1. Founder Profile
- Brief background on the founder's skills, domain expertise, and entrepreneurial aspirations
- Identification of the founder's unique strengths, values, and motivations
- Reflection on the scale and type of impact the founder wants to have

2. Market Trends
- Overview of major technology and market trends shaping the future of software
- Identification of key shifts in customer needs, behaviors, and expectations
- Discussion of emerging challenges or opportunities these trends present

3. Problem Spaces
- Synthesis of key problem spaces or unmet needs that arise from the identified trends
- Prioritization of problems based on scale, urgency, and alignment with founder's interests
- "How Might We" questions to frame the problems as opportunities for innovation

4. Solution Brainstorm
- Guided brainstorming exercises to generate a wide range of potential solutions
- Encouragement to combine and remix ideas across domains for novel solutions
- Provocations to inspire radically different approaches and challenge assumptions

5. Idea Shortlist
- Criteria for filtering and prioritizing the most promising product ideas
- Evaluation of each idea's feasibility, differentiation, and market potential
- Selection of the top 3-5 ideas for further development and validation

6. Concept Development
- Framework for refining the selected ideas into more detailed product concepts
- Articulation of each concept's unique value proposition and key features
- Guidance on defining the target user, use cases, and benefits
- Recommendations for rapid prototyping and user testing to validate the concepts

7. Vision Narrative
- Crafting a compelling narrative around the most promising product concept
- Painting an inspiring long-term vision of the product's potential impact and future
- Aligning the product vision with the founder's personal mission and values
- Tips for communicating the vision to inspire and attract others

8. Next Steps
- Identifying key assumptions to test and validate for the selected concept
- Recommendations for conducting market research and competitive analysis
- Defining success metrics and creating a learning plan to iterate and improve
- Guidance on building a MVP, forming a team, and acquiring early customers

#IDEATION CRITERIA:
1. The ideation process should be comprehensive, covering the founder's background, market trends, problem spaces, potential solutions, idea prioritization, concept refinement, vision crafting, and next steps.

2. The focus should be on identifying a technically feasible product concept tied to a clearly stated, evidence-backed market need.

3. The founder's unique skills, passions, and desired impact should be incorporated into the ideation process.

4. Emerging technologies and evolving customer needs should be carefully considered when generating and evaluating ideas.

5. The output should enable the founder to develop the most promising idea into a compelling vision and value proposition.

#INFORMATION ABOUT ME:
- My background and skills: [MY BACKGROUND AND SKILLS]
- My entrepreneurial aspirations: [MY ENTREPRENEURIAL ASPIRATIONS]
- The type of impact I want to have: [THE TYPE OF IMPACT I WANT TO HAVE]

#RESPONSE FORMAT:
Expected output: A structured SaaS opportunity analysis with founder fit, problem, audience, product concept, differentiation, validation path, and key takeaways.

The response should be formatted as follows:

# <SECTION TITLE>
<Section content in paragraph form>

- <Bullet points for lists or key takeaways>

## <Subsection Title>
<Subsection content>

\```
<Code snippets or examples>
\```

**<Bold text for emphasis>**

*<Italic text for emphasis>*

> <Blockquotes for highlighting key insights or quotes>

#### Expected output

- A structured SaaS opportunity analysis with founder fit, problem, audience, product concept, differentiation, validation path, and key takeaways.

#### Customization tips

- Fill in the “brief background on founder” placeholder with specific details about the founder's skills, domain expertise, and entrepreneurial aspirations.
- Example: "I have a background in software engineering with expertise in cloud computing and a strong passion for environmental sustainability. My aspiration is to create a SaaS product that helps businesses reduce their carbon footprint."
- Use the structured sections provided in the prompt (founder profile, market trends, problem spaces, etc.) to systematically develop the SaaS product idea. Each section guides you through different stages of the ideation and development process, ensuring a comprehensive approach.
- Leverage the final sections (concept development, vision narrative, next steps) to refine the selected idea, articulate its value proposition, and plan for prototyping, market research, and initial customer acquisition. This structured approach helps in transforming an idea into a viable product and business strategy.
- Explore Emerging Technologies: Investigate cutting-edge technologies like AI, blockchain, or IoT to find opportunities where the entrepreneur's skills can be leveraged to create innovative solutions.
- Analyze Market Gaps: Conduct thorough market research to identify unmet needs or pain points in industries the entrepreneur is passionate about, which could lead to high-impact SaaS solutions.
- Develop Unique Value Propositions: Focus on creating a SaaS product that not only fills a market need but also stands out through unique features or by serving a niche market effectively. Leverage Founder's Expertise: Identify the founder's specific skills, domain expertise, and entrepreneurial aspirations to guide the ideation process towards areas where they can add unique value.
- Stay Updated on Industry Trends: Continuously monitor and analyze emerging technologies, customer needs, and market trends to stay ahead of the curve and identify new opportunities for innovation.
- Collaborate and Remix Ideas: Encourage collaboration with others, both within and outside the industry, to combine and remix ideas from different domains, fostering the creation of novel and disruptive solutions.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My background and skills: Former marketing director with experience in career coaching, curriculum design, customer research, and workflow automation
- My entrepreneurial aspirations: Build a focused B2B SaaS company that can reach $1 million in annual recurring revenue with a small team
- The type of impact I want to have: Help small professional-services firms document delivery knowledge and reduce rework during employee transitions

#### Related prompts

- https://returnmytime.com/prompt-library/conduct-market-research-validation
- https://returnmytime.com/prompt-library/generate-product-unique-selling-proposition


### Generate Target Audience Fit Analysis

- URL: https://returnmytime.com/prompt-library/generate-target-audience-fit-analysis
- Category: Product Development
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 1-2 hours
- Tags: product-development, generate, target, audience
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307083959id_/https://returnmytime.com/prompt-library/generate-target-audience-fit-analysis
- Source captured: 2026-03-07T08:39:59Z
- Source hash: 283525aba45cce2a4982d4c27735bf2004d19797ec8ef6ee931b2e45e79f1e53


Use the prompt template to expertly assess the viability of a proposed product or service for a specific target audience. This tool guides you through a structured analysis, including audience characteristics, product alignment, optimization strategies, and a final recommendation, ensuring a tailored and strategic approach to product development.

#### What it does

- Analyzes the key characteristics, needs, and pain points of a specified target audience to assess product or service suitability.
- Evaluates how well the proposed product or service aligns with the target audience's needs, identifying areas of strong match and potential gaps.
- Suggests improvements or modifications to the product or service to better meet the needs and address the pain points of the target audience.

#### When to use

- Use when a proposed product and specific audience need a structured fit, gap, and validation review.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting target audience fit analysis.

#### Required inputs

- MY TARGET AUDIENCE
- MY PROPOSED PRODUCT/SERVICE

#### Prompt

#CONTEXT:
You will be acting as a product/service developer tasked with determining if a proposed product or service is a good fit for a specific target audience.

#ROLE:
Product/service developer analyzing the fit between a proposed offering and a target market.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Target Audience Analysis section:
- Brainstorm a bulleted list of key characteristics, needs, and pain points relevant for product/service development.

2. Product/Service Alignment section:
- Analyze how well the proposed product or service aligns with and serves the needs of the target audience.
- Note areas where it matches their needs well and any areas where it may miss the mark.

3. Product/Service Optimization section:
- Brainstorm ways the proposed product or service could be improved, modified or pivoted to better meet the target audience's needs and address their pain points.

4. Recommendation section:
- Provide an overall recommendation on whether or not to pursue developing this product or service for the target audience.
- Justify the recommendation based on the analysis above.

#TASK CRITERIA:
- Focus on aspects of the target audience most relevant for developing products and services for them.
- Provide a comprehensive analysis covering alignment, optimization opportunities, and an overall recommendation.
- Justify the recommendation based on the insights from the analysis.

#INFORMATION ABOUT ME:
- My target audience: [MY TARGET AUDIENCE]
- My proposed product/service: [MY PROPOSED PRODUCT/SERVICE]

#RESPONSE FORMAT:
Expected output: A target-audience fit analysis with needs, product alignment, gaps, evidence, risks, and validation recommendations.

Use markdown formatting to structure the response with clear headings for each section. Use bullet points where appropriate to improve readability. Avoid using any XML tags in the response.

#### Expected output

- A target-audience fit analysis with needs, product alignment, gaps, evidence, risks, and validation recommendations.

#### Customization tips

- Fill in the “target audience” and “product service description” placeholders with specific details about your target audience and the proposed product or service.
- Example: If your target audience is "young professionals interested in fitness," and your proposed product is "a subscription-based meal planning app tailored for busy lifestyles," use these details to fill in the placeholders.
- Use the <target_audience_analysis> section to deeply explore and list the characteristics, needs, and pain points of your target audience. This should guide the development and refinement of the product or service to ensure it is highly tailored and relevant.
- In the <product_service_optimization> and <recommendation> sections, critically assess the alignment between the product/service and the audience's needs. Suggest realistic improvements and conclude with a well-justified recommendation on the feasibility of the product/service based on your analysis.
- Conduct Thorough Audience Research: Gather detailed demographic, psychographic, and behavioral data about the target audience to understand their preferences, lifestyle, and consumption habits.
- Validate Product-Market Fit: Use surveys, focus groups, and beta testing to collect feedback directly from the target audience, ensuring the product or service meets their expectations and solves their specific problems.
- Develop Tailored Marketing Strategies: Create marketing messages and campaigns that resonate deeply with the target audience's values and needs, using the channels where they are most active. Identify Unique Selling Points: Determine what sets your product or service apart from competitors and highlight those features that specifically address the target audience's needs and pain points.
- Consider Pricing Strategy: Analyze the target audience's purchasing power and willingness to pay, and adjust the pricing of your product or service accordingly to ensure it is accessible and attractive to them.
- Seek Continuous Feedback: Establish channels for ongoing communication with the target audience to gather feedback and insights that can inform future improvements and iterations of the product or service.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.

#### Example input

## Information about me

- My target audience: Corporate professionals (ages 35-50) feeling stuck in their careers, earning $100K+ but unfulfilled, ready to build something of their own
- My proposed product/service: The Career Pivot Program - a 12-week coaching program helping corporate professionals build their first business

#### Related prompts

- https://returnmytime.com/prompt-library/conduct-market-research-validation
- https://returnmytime.com/prompt-library/generate-product-unique-selling-proposition


### Generate User Testing Strategy

- URL: https://returnmytime.com/prompt-library/generate-user-testing-strategy
- Category: Product Development
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Type: Effectiveness
- Estimated time saved: 2+ hours
- Tags: product-development, generate, user, testing
- Published: 2026-07-30
- Provenance: archive
- Source: https://web.archive.org/web/20260307081329id_/https://returnmytime.com/prompt-library/generate-user-testing-strategy
- Source captured: 2026-03-07T08:13:29Z
- Source hash: 9406e96b8f966bd71fe8e8a450f130b91f7994b8497f6c8cc076169878987d97


Use this prompt template to meticulously outline a comprehensive user testing plan for your MVP. It guides you in developing a detailed strategy to gather accurate and actionable feedback, ensuring product validation and improvement.

#### What it does

- Creates a detailed plan for user testing an MVP (Minimum Viable Product).
- Describes methods to collect feedback effectively and comprehensively from test users.
- Outlines strategies to analyze and implement the feedback for product improvement.

#### When to use

- Use when an MVP is concrete enough to define participants, methods, metrics, timeline, and analysis.

#### When not to use

- Do not use this when the required context is unavailable or when an accountable owner cannot review the assumptions and resulting user testing strategy.

#### Required inputs

- MY MVP DESCRIPTION

#### Prompt

#CONTEXT:
Adopt the role of an expert user experience researcher and product testing strategist. Your task is to help the user create a comprehensive user testing plan for a given MVP, outlining a detailed strategy to gather accurate and thorough user feedback.

#ROLE:
You are an expert user experience researcher and product testing strategist, providing guidance on creating effective user testing plans for MVPs.

#RESPONSE GUIDELINES:
Review status: Draft for human review.

1. Begin with an overview of the MVP to provide context.
2. Clearly state the testing objectives.
3. Outline three user testing methods, including:
   - Name of the method
   - Description of the method
   - Participant criteria
   - Testing environment
   - Data collection techniques
4. Describe the participant recruitment process, including:
   - Screening criteria
   - Incentives
   - Recruitment channels
5. Provide a testing timeline with three phases, each including:
   - Name of the phase
   - Duration
   - Key activities
6. Outline the data analysis plan, covering:
   - Quantitative analysis
   - Qualitative analysis
   - Insights synthesis
7. List the deliverables of the user testing plan.

#TASK CRITERIA:
1. Consider various user testing methods to gather comprehensive feedback.
2. Ensure participant recruitment criteria align with the target audience of the MVP.
3. Choose testing environments that closely resemble real-world usage scenarios.
4. Select data collection techniques that capture both quantitative and qualitative insights.
5. Develop a structured testing plan optimized for the specific MVP.
6. Focus on gathering accurate and thorough user feedback to inform product development.

#INFORMATION ABOUT ME:
- My MVP description: [MY MVP DESCRIPTION]

#RESPONSE FORMAT:
Expected output: An MVP testing strategy with objectives, three testing methods, participant criteria, environments, data collection, timeline, metrics, and analysis plan.

MVP Overview:
<MVP overview>

Testing Objectives:
<Testing objectives>

User Testing Methods:
Method 1:
Name: <Name>
Description: <Description>
Participant Criteria: <Participant criteria>
Testing Environment: <Testing environment>
Data Collection: <Data collection>

Method 2:
Name: <Name>
Description: <Description>
Participant Criteria: <Participant criteria>
Testing Environment: <Testing environment>
Data Collection: <Data collection>

Method 3:
Name: <Name>
Description: <Description>
Participant Criteria: <Participant criteria>
Testing Environment: <Testing environment>
Data Collection: <Data collection>

Participant Recruitment:
Screening Criteria: <Screening criteria>
Incentives: <Incentives>
Recruitment Channels: <Recruitment channels>

Testing Timeline:
Phase 1:
Name: <Name>
Duration: <Duration>
Key Activities: <Key activities>

Phase 2:
Name: <Name>
Duration: <Duration>
Key Activities: <Key activities>

Phase 3:
Name: <Name>
Duration: <Duration>
Key Activities: <Key activities>

Data Analysis Plan:
Quantitative Analysis: <Quantitative analysis>
Qualitative Analysis: <Qualitative analysis>
Insights Synthesis: <Insights synthesis>

Deliverables: <Deliverables>

#### Expected output

- An MVP testing strategy with objectives, three testing methods, participant criteria, environments, data collection, timeline, metrics, and analysis plan.

#### Customization tips

- Fill in the “mvp” placeholder with the specific product or service you are developing.
- Example: If your MVP is a new mobile app for budget tracking, specify this in your user testing plan.
- Develop a detailed strategy that includes methods such as surveys, interviews, and usability tests to gather comprehensive feedback from users.
- Ensure the plan includes a timeline and criteria for selecting participants to ensure that the feedback is representative of your target audience.
- Define Clear Objectives: Establish specific goals for what you want to learn from the user testing, such as usability issues, user satisfaction, or specific feature feedback.
- Create a Diverse User Group: Select a varied group of participants that represent your target demographic to ensure a wide range of insights and feedback.
- Implement Structured and Unstructured Testing: Use both scripted tasks to observe specific interactions and allow free use to uncover unexpected insights or issues. Prioritize User Experience: Design your user testing plan with a focus on understanding how users interact with your MVP and identify any pain points or areas for improvement.
- Utilize a Mix of Testing Methods: Combine different testing methods such as remote testing, in-person interviews, and surveys to gather a comprehensive range of feedback from your users.
- Establish a Feedback Loop: Create a system to collect and organize user feedback, ensuring that it is easily accessible and actionable for your team.

#### Guardrails

- Use only supplied evidence or clearly labeled assumptions; do not invent customer facts, research findings, citations, or implementation results.
- The accountable business owner must review the draft, validate assumptions, and decide what to use.
- A named human research owner must review and approve every participant incentive before any action.

#### Example input

## Information about me

- My MVP description: A clickable prototype that turns a recorded client handoff into an editable SOP draft, with upload, review, and export screens

#### Related prompts

- https://returnmytime.com/prompt-library/conduct-market-research-validation
- https://returnmytime.com/prompt-library/generate-product-unique-selling-proposition


### Write an Instant Lead Response

- URL: https://returnmytime.com/prompt-library/write-instant-lead-response
- Category: Revenue Operations
- Workflow: lead-response-revenue-recovery
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Efficiency
- Estimated time saved: 20-30 minutes per response
- Tags: lead-response, speed-to-lead, revenue-operations
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Draft a fast, useful first response to a new service inquiry using only verified offer, fit, location, and scheduling facts.

#### What it does

- Turns supplied inquiry context into a concise first-response draft with a clear next step.
- Separates usable facts, missing information, and escalation needs before anyone contacts the lead.

#### When to use

- When a new inbound inquiry needs a prompt, personalized acknowledgment and the team has verified the offer and next-step path.

#### When not to use

- Do not use it to confirm service availability, quote a price, decide eligibility, or send a message without the approved business rules and a named human reviewer.

#### Required inputs

- OFFER
- LEAD SOURCE
- RESPONSE WINDOW
- SERVICE AREA
- QUALIFICATION FACTS
- TONE
- SCHEDULING PATH
- ESCALATION OWNER

#### Prompt

#CONTEXT:
Draft the first response to a new inbound lead for a service business. Use only the supplied facts. This is a draft for review, not a message to send.

#INPUTS:
- Service offer: [OFFER]
- Lead source and inquiry context: [LEAD SOURCE]
- Target response window: [RESPONSE WINDOW]
- Verified service area or coverage rules: [SERVICE AREA]
- Verified qualification facts and explicit unknowns: [QUALIFICATION FACTS]
- Company tone: [TONE]
- Approved scheduling or next-step path: [SCHEDULING PATH]
- Named escalation owner: [ESCALATION OWNER]

#INSTRUCTIONS:
1. Extract the lead's stated need and every supplied fact that may shape the reply.
2. List material missing or conflicting facts. Do not guess them.
3. Draft a concise acknowledgment that sounds human, names the inquiry, and explains only a verified next step.
4. Ask no more than two necessary clarification questions.
5. If fit, coverage, scheduling, or another rule is unknown, recommend the named escalation owner as the destination instead of resolving it.
6. Do not promise or claim availability when availability rules or current capacity were not supplied.
7. Do not promise, claim, or infer eligibility from incomplete qualification facts.
8. Do not invent prices, discounts, response statistics, credentials, policies, customer results, calendar openings, or commitments.
9. Do not send the reply. A named human owner must review and approve the exact draft before external sending.

#RESPONSE FORMAT:
## Known facts
- Fact — source input

## Missing or conflicting facts
- Unknown — why it matters — owner to confirm

## Immediate reply draft
DRAFT — HUMAN APPROVAL REQUIRED
[Concise lead-facing response with one approved next step]

## Reviewer checklist
- Claims match supplied facts
- Availability and eligibility remain unconfirmed unless explicitly supported
- Scheduling path is approved
- Named human approval recorded before sending

#### Expected output

- An immediate reply draft labeled for human approval, plus known facts, missing facts, and reviewer checks.

#### Customization tips

- Paste the lead's own wording into the source context when company policy permits.
- Replace a booking link with an approved callback path when real-time availability is not verified.

#### Guardrails

- A named human owner must approve the exact draft before external sending; the assistant must not send it.
- A named human owner must approve every price, promise, policy exception, permission, legal commitment, and external message before use.
- Do not promise availability or claim eligibility unless the supplied rules and current facts explicitly support that statement.
- Keep every missing or unknown fact explicit instead of filling gaps with assumptions.

#### Example input

Fictional example — OFFER: Residential heat-pump consultation. LEAD SOURCE: Website form from Jordan Lee asking about replacing a 15-year-old system. RESPONSE WINDOW: Draft within 10 minutes. SERVICE AREA: Cedar Ridge and Pine County; exact address not supplied. QUALIFICATION FACTS: Homeowner status and property address unknown; no eligibility decision made. TONE: Calm, practical, neighborly. SCHEDULING PATH: Invite the lead to choose a callback window using https://example.com/cedar-callback. ESCALATION OWNER: Priya, Service Coordinator.

#### Related prompts

- https://returnmytime.com/prompt-library/create-after-hours-lead-response
- https://returnmytime.com/prompt-library/create-missed-lead-recovery-sequence
- https://returnmytime.com/prompt-library/audit-lead-response-time

#### Related articles

- https://returnmytime.com/blog/ai-lead-qualification-routing

### Create a Missed-Lead Recovery Sequence

- URL: https://returnmytime.com/prompt-library/create-missed-lead-recovery-sequence
- Category: Revenue Operations
- Workflow: lead-response-revenue-recovery
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Effectiveness
- Estimated time saved: 1-2 hours
- Tags: lead-recovery, follow-up, revenue-operations
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Create a respectful, timed follow-up sequence for inbound leads who did not receive a timely first response.

#### What it does

- Builds a reviewable recovery cadence around the lead's actual inquiry and prior contact history.
- Defines stop conditions, escalation points, and missing facts before follow-up begins.

#### When to use

- When an inbound lead went unanswered or stalled and the team wants a bounded recovery plan based on verified contact history.

#### When not to use

- Do not use it when consent, channel permissions, suppression status, or prior contact history is unknown.

#### Required inputs

- OFFER
- LEAD SOURCE
- MISSED LEAD AGE
- PRIOR CONTACT HISTORY
- SERVICE AREA
- QUALIFICATION FACTS
- TONE
- SCHEDULING PATH
- RECOVERY CADENCE
- ESCALATION OWNER

#### Prompt

#CONTEXT:
Create a draft recovery sequence for an inbound service lead whose first response was missed or whose conversation stalled. Use only supplied facts and permissions. Do not contact the lead.

#INPUTS:
- Service offer: [OFFER]
- Lead source and inquiry context: [LEAD SOURCE]
- Age of missed inquiry: [MISSED LEAD AGE]
- Verified prior contact history, consent, and suppression status: [PRIOR CONTACT HISTORY]
- Verified service area or coverage rules: [SERVICE AREA]
- Verified qualification facts and explicit unknowns: [QUALIFICATION FACTS]
- Company tone: [TONE]
- Approved scheduling or next-step path: [SCHEDULING PATH]
- Approved timing, channel limits, and maximum touches: [RECOVERY CADENCE]
- Named escalation owner: [ESCALATION OWNER]

#INSTRUCTIONS:
1. Summarize the evidence showing what happened, what was already sent, and what remains unknown.
2. Stop and flag the plan as blocked when consent, suppression status, or an approved channel is missing.
3. Create a limited sequence within the supplied cadence. For each touch, state timing, allowed channel, purpose, draft copy, and stop condition.
4. Acknowledge the delayed response plainly without manufacturing urgency, blaming the team, or inventing a reason.
5. End the sequence when the lead replies, opts out, books, becomes ineligible under supplied rules, or reaches the approved touch limit.
6. For unclear fit, coverage, policy, or sensitive requests, recommend escalation to the named owner.
7. Do not promise or claim availability when current capacity or calendar facts were not supplied.
8. Do not promise, claim, or infer eligibility from incomplete qualification facts.
9. Do not invent results, prices, discounts, deadlines, consent, or prior conversations.
10. Do not send or schedule any touch. A named human owner must review and approve each exact message, channel, recipient, and time before external sending.

#RESPONSE FORMAT:
## Recovery evidence
- Known fact — source input
- Missing or conflicting fact — required owner

## Timed recovery sequence
| Touch | Timing | Allowed channel | Purpose | Draft | Stop condition |
| --- | --- | --- | --- | --- | --- |

## Escalations and blocks
- Decision or blocker — named owner — decision required

## Human approval checklist
- Consent and suppression status verified
- Every channel, recipient, time, and draft approved before sending

#### Expected output

- A timed recovery sequence with evidence, draft messages, stop conditions, escalation blocks, and human-approval checks.

#### Customization tips

- Use the shortest approved cadence that gives the lead a clear next step.
- Include suppression and consent fields from the system of record rather than assumptions.

#### Guardrails

- A named human owner must approve every exact message, recipient, channel, and time before external sending; the assistant must not send or schedule it.
- A named human owner must approve every price, promise, policy exception, permission, legal commitment, and external message before use.
- Do not promise availability or claim eligibility unless supplied rules and facts explicitly support the statement.
- List missing or unknown consent, contact, fit, and coverage facts as blockers.

#### Example input

Fictional business example — OFFER: Commercial landscaping consultation. LEAD SOURCE: Referral form from Morgan Chen requesting seasonal grounds maintenance. MISSED LEAD AGE: 3 business days. PRIOR CONTACT HISTORY: One form submission; no reply sent; email consent captured; SMS consent not captured; not suppressed. SERVICE AREA: Harbor City commercial properties only; site address unknown. QUALIFICATION FACTS: Commercial property confirmed; acreage and start date unknown. TONE: Direct and courteous. SCHEDULING PATH: Approved request-a-call form at https://example.com/harbor-call. RECOVERY CADENCE: Email on day 0 and day 3, maximum two touches, weekdays only. ESCALATION OWNER: Luis, Sales Manager.

#### Related prompts

- https://returnmytime.com/prompt-library/write-instant-lead-response
- https://returnmytime.com/prompt-library/audit-lead-response-time

#### Related articles

- https://returnmytime.com/blog/ai-lead-qualification-routing

### Audit Your Lead Response Time

- URL: https://returnmytime.com/prompt-library/audit-lead-response-time
- Category: Revenue Operations
- Workflow: lead-response-revenue-recovery
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Effectiveness
- Estimated time saved: 2-3 hours
- Tags: lead-response-time, revenue-audit, sales-operations
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Audit timestamped inbound-lead records to expose response delays, missing data, source-level patterns, and accountable next actions.

#### What it does

- Calculates response-time performance only when supplied timestamps support it.
- Separates observed delays, data-quality gaps, and process hypotheses so an owner can prioritize investigation.

#### When to use

- When a service business has timestamped lead and response records and needs a defensible baseline before changing its process.

#### When not to use

- Do not use it to estimate performance from anecdotes, infer missing timestamps, or claim revenue impact without supplied conversion evidence.

#### Required inputs

- RESPONSE LOG
- TARGET RESPONSE WINDOW
- LEAD SOURCE DEFINITIONS
- BUSINESS HOURS
- ESCALATION OWNER

#### Prompt

#CONTEXT:
Audit lead-response time for a service business using only the supplied timestamped records and definitions. Do not contact leads or change a workflow.

#INPUTS:
- Response log with lead ID, received timestamp, first human or approved automated response timestamp, source, and status: [RESPONSE LOG]
- Approved target response window and measurement unit: [TARGET RESPONSE WINDOW]
- Lead-source definitions and exclusions: [LEAD SOURCE DEFINITIONS]
- Business hours, time zone, and after-hours treatment: [BUSINESS HOURS]
- Named escalation owner: [ESCALATION OWNER]

#INSTRUCTIONS:
1. Validate the fields, timestamp formats, time zone, duplicate handling, exclusions, and definition of first response.
2. List missing, invalid, conflicting, or unknown values. Never impute a timestamp or silently discard a record.
3. For calculable records, show the formula and calculate elapsed response time consistently.
4. Report total records, calculable records, excluded records by reason, median, range, and target-window pass rate. Do not claim a percentile unless the sample supports the requested calculation.
5. Break out results by supplied lead source and business-hours status. Suppress or label any segment too incomplete to interpret.
6. Identify observed bottlenecks separately from hypotheses that require investigation.
7. Recommend bounded next checks, each with evidence needed, a named owner, and a verification measure.
8. Do not promise or claim availability when capacity or calendar data was not supplied.
9. Do not promise, claim, or infer lead eligibility; this audit measures response time, not fit.
10. Do not invent conversion lift, revenue recovered, benchmarks, or causal claims.
11. Do not send messages or execute workflow changes. A named human owner must approve any external sending or operational change.

#RESPONSE FORMAT:
## Data-quality report
| Issue | Affected records | Impact | Owner question |

## Evidence-based response-time audit
- Scope and definitions
- Calculable versus excluded record counts
- Median, range, and target-window pass rate with formulas
- Source and business-hours breakdowns

## Observations versus hypotheses
| Type | Finding | Evidence | Confidence | Missing evidence |

## Prioritized next checks
| Priority | Check | Evidence needed | Named owner | Verification measure |

## Approval boundary
- No lead contact or workflow change is authorized by this audit

#### Expected output

- An evidence-based response-time audit with data-quality findings, transparent calculations, source breakdowns, hypotheses, and prioritized next checks.

#### Customization tips

- Export stable lead IDs so duplicate and reopened conversations can be audited.
- Define whether an automated acknowledgment counts as a first response before calculating the baseline.

#### Guardrails

- A named human owner must approve any external sending or operational change; the assistant must not send messages or modify workflows.
- A named human owner must approve every price, promise, policy exception, permission, legal commitment, and external message before use.
- Do not promise availability or claim eligibility; neither conclusion is supported by response-time records alone.
- Keep missing or unknown timestamps and definitions visible, and never fabricate a complete sample.

#### Example input

Fictional example — RESPONSE LOG: L-101, received 2026-07-07 09:02 ET, first approved response 09:14 ET, web form, open; L-102, received 2026-07-07 18:40 ET, first response 2026-07-08 08:12 ET, referral, booked; L-103, received 2026-07-08 11:05 ET, first response missing, web form, open. TARGET RESPONSE WINDOW: 15 elapsed minutes during business hours. LEAD SOURCE DEFINITIONS: Web form and referral are included; test records excluded. BUSINESS HOURS: Monday-Friday, 08:00-18:00 ET; after-hours timing begins at next opening. ESCALATION OWNER: Dana, Revenue Operations Lead.

#### Related prompts

- https://returnmytime.com/prompt-library/write-instant-lead-response
- https://returnmytime.com/prompt-library/create-missed-lead-recovery-sequence
- https://returnmytime.com/prompt-library/create-after-hours-lead-response

#### Related articles

- https://returnmytime.com/blog/ai-lead-qualification-routing

### Create an After-Hours Lead Response

- URL: https://returnmytime.com/prompt-library/create-after-hours-lead-response
- Category: Revenue Operations
- Workflow: lead-response-revenue-recovery
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Efficiency
- Estimated time saved: 45-60 minutes
- Tags: after-hours, lead-routing, revenue-operations
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Draft an after-hours acknowledgment and routing plan that sets truthful expectations without inventing capacity or fit decisions.

#### What it does

- Creates a transparent acknowledgment for inquiries received outside approved operating hours.
- Maps routine, unclear, and urgent-but-supported cases to a named next step without pretending a person or appointment is available.

#### When to use

- When a service business wants a consistent draft and internal handoff plan for inquiries received outside staffed hours.

#### When not to use

- Do not use it as an emergency service, to classify safety-critical situations, or to confirm an appointment without current approved availability data.

#### Required inputs

- OFFER
- LEAD SOURCE
- AFTER-HOURS WINDOW
- SERVICE AREA
- QUALIFICATION FACTS
- TONE
- SCHEDULING PATH
- ESCALATION OWNER
- BUSINESS HOURS

#### Prompt

#CONTEXT:
Create a draft after-hours acknowledgment and internal routing plan for a service-business inquiry. Use only supplied policies and facts. Do not send, schedule, or route anything.

#INPUTS:
- Service offer: [OFFER]
- Lead source and inquiry context: [LEAD SOURCE]
- Definition of the after-hours window: [AFTER-HOURS WINDOW]
- Verified service area or coverage rules: [SERVICE AREA]
- Verified qualification facts and explicit unknowns: [QUALIFICATION FACTS]
- Company tone: [TONE]
- Approved scheduling or next-step path: [SCHEDULING PATH]
- Named escalation owner and approved escalation conditions: [ESCALATION OWNER]
- Staffed business hours, time zone, and truthful response expectation: [BUSINESS HOURS]

#INSTRUCTIONS:
1. Extract the received time, source, stated need, supplied location, and known qualification facts.
2. List every material missing or conflicting fact before drafting.
3. Draft a short acknowledgment that confirms receipt and gives only the supplied staffed-hours expectation.
4. Do not say that a person is reviewing the inquiry now unless that fact was supplied.
5. Do not present a scheduling link as confirmed availability. Describe it only as a request path unless current slots are supplied and approved.
6. Do not promise or claim availability when current staffing, capacity, or calendar data was not supplied.
7. Do not promise, claim, or infer eligibility when service-area or qualification facts are incomplete.
8. Create an internal routing table for routine, missing-information, outside-area, and supplied escalation conditions. Keep unknown cases with the named escalation owner.
9. Do not invent emergency capabilities, prices, discounts, response times, policies, or outcomes.
10. Do not send, schedule, or route the draft. A named human owner must review and approve the exact acknowledgment, recipient, channel, timing, and routing rules before use.

#RESPONSE FORMAT:
## Known facts and unknowns
- Known fact — source input
- Missing or conflicting fact — why it matters

## After-hours acknowledgment
DRAFT — HUMAN APPROVAL REQUIRED
[Lead-facing receipt confirmation and truthful next-step expectation]

## Routing plan
| Condition | Evidence required | Destination | Response expectation | Escalation owner |

## Human approval checklist
- Business hours and expectation verified
- No unsupported availability or eligibility statement
- Exact message and routing rules approved before use

#### Expected output

- An after-hours acknowledgment and routing plan with known facts, explicit unknowns, conditional paths, and human-approval checks.

#### Customization tips

- Use separate acknowledgment language for staffed, unstaffed, and truly on-call periods.
- Add an emergency disclaimer only when the business has approved wording and an authoritative destination.

#### Guardrails

- A named human owner must approve the exact message, recipient, channel, timing, and routing rules; the assistant must not send, schedule, or route anything.
- A named human owner must approve every price, promise, policy exception, permission, legal commitment, and external message before use.
- Do not promise availability or claim eligibility unless supplied current facts and approved rules explicitly support the statement.
- State missing or unknown coverage, contract, urgency, and scheduling facts instead of guessing.

#### Example input

Fictional business example — OFFER: In-office network troubleshooting for small businesses. LEAD SOURCE: Website chat received Tuesday at 8:20 p.m. from Avery Singh reporting intermittent office Wi-Fi. AFTER-HOURS WINDOW: Weekdays after 6:00 p.m. CT. SERVICE AREA: Lakeview business district; customer address unknown. QUALIFICATION FACTS: Business customer stated; contract status, device count, and safety impact unknown. TONE: Reassuring and concise. SCHEDULING PATH: Request a callback at https://example.com/lakeview-support; this does not confirm a slot. ESCALATION OWNER: On-call Operations Manager, but only verified active-contract outages qualify for after-hours escalation. BUSINESS HOURS: Monday-Friday, 8:00 a.m.-6:00 p.m. CT; team reviews new non-emergency requests after opening.

#### Related prompts

- https://returnmytime.com/prompt-library/write-instant-lead-response
- https://returnmytime.com/prompt-library/audit-lead-response-time

#### Related articles

- https://returnmytime.com/blog/ai-lead-qualification-routing

### Create a Lead Qualification Framework

- URL: https://returnmytime.com/prompt-library/create-lead-qualification-framework
- Category: Revenue Operations
- Workflow: lead-qualification-routing
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Effectiveness
- Estimated time saved: 60-90 minutes per framework draft
- Tags: lead-qualification, ideal-customer-profile, revenue-operations
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Turn verified ideal-customer criteria and disqualifiers into a reviewable lead-qualification framework with explicit evidence and unknown states.

#### What it does

- Translates supplied fit criteria into observable qualification questions and evidence requirements.
- Keeps qualified, disqualified, and insufficient-data outcomes separate for human review.

#### When to use

- When a service business needs a consistent, explainable first-pass qualification framework before changing its intake or CRM process.

#### When not to use

- Do not use it to make a live eligibility decision, reject a lead, change a policy, update an account, or contact anyone without a named human reviewer.

#### Required inputs

- IDEAL CUSTOMER CRITERIA
- DISQUALIFIERS
- OBSERVED LEAD FIELDS
- QUALIFICATION OWNER

#### Prompt

#CONTEXT:
Draft a lead-qualification framework for a service business. Use only the supplied business rules and observed lead fields. This is a recommendation for human review, not an eligibility decision or live workflow.

#INPUTS:
- Approved ideal-customer criteria and definitions: [IDEAL CUSTOMER CRITERIA]
- Approved disqualifiers and their business rationale: [DISQUALIFIERS]
- Fields actually observed in the lead record, including blanks: [OBSERVED LEAD FIELDS]
- Named human qualification owner: [QUALIFICATION OWNER]

#INSTRUCTIONS:
1. Separate each criterion into its observable evidence, qualifying condition, disqualifying condition, and unknown condition.
2. Identify criteria that cannot be evaluated from the observed fields. Mark the outcome "Insufficient data"; do not guess, invent, infer, or fabricate a fact.
3. Distinguish a verified disqualifier from a missing answer. Missing evidence is not evidence of disqualification.
4. Do not qualify, disqualify, score, or route a person based on a protected trait or a proxy for a protected trait. Flag any criterion that could function as a proxy for removal and human review.
5. Explain the service-business reason for every retained criterion without creating new policy.
6. Recommend no adverse decision when the evidence is unknown, conflicting, or insufficient.
7. Do not access or update a CRM, contact a lead, send a message, change a rule, or execute routing.
8. A named human owner must review and approve every qualification, disqualification, scoring, or routing rule change before operational use.

#RESPONSE FORMAT:
## Assumptions and source limits
- Supplied fact — source input
- Unknown — why it matters

## Qualification framework
| Criterion | Observable evidence | Qualifying condition | Disqualifying condition | Unknown / insufficient-data state | Rationale |
|---|---|---|---|---|---|

## Criteria requiring removal or revision
- Criterion — protected-trait, proxy, unsupported, or ambiguous concern — human review needed

## Human decision checklist
- Qualification owner named
- Every rule traceable to supplied policy
- Insufficient-data path preserved
- Rule changes approved before operational use

#### Expected output

- A qualification framework with evidence-backed conditions, explicit unknown and insufficient-data states, flagged criteria, and a human decision checklist.

#### Customization tips

- Replace broad fit language with observable evidence the intake team can consistently collect.
- Review disqualifiers quarterly so an old capacity or territory rule does not become hidden policy.

#### Guardrails

- Do not qualify, disqualify, score, route, or recommend an adverse decision based on a protected trait or any proxy for a protected trait.
- Do not invent missing lead facts; preserve Unknown and Insufficient data as valid outcomes.
- A named human owner must review and approve every qualification, disqualification, scoring, routing rule change, policy exception, account update, external message, price, promise, permission, and legal commitment before any operational use.
- The assistant must not access or change a live CRM, route a lead, contact anyone, or send an external message.

#### Example input

Fictional example — IDEAL CUSTOMER CRITERIA: Northstar Field Services serves commercial facilities within 40 miles of Cedar Junction; requests should concern planned HVAC maintenance for one or more sites. DISQUALIFIERS: Residential-only request or verified location outside the approved coverage map. OBSERVED LEAD FIELDS: Organization name: Harbor Workshop; request: quarterly HVAC maintenance; site type: commercial; address: blank; number of sites: unknown. QUALIFICATION OWNER: Elena, Revenue Operations Manager.

#### Related prompts

- https://returnmytime.com/prompt-library/build-lead-scoring-rubric
- https://returnmytime.com/prompt-library/identify-missing-lead-information

#### Related articles

- https://returnmytime.com/blog/ai-lead-qualification-routing

### Build a Lead-Scoring Rubric

- URL: https://returnmytime.com/prompt-library/build-lead-scoring-rubric
- Category: Revenue Operations
- Workflow: lead-qualification-routing
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Effectiveness
- Estimated time saved: 60-90 minutes per rubric draft
- Tags: lead-scoring, qualification-rubric, revenue-operations
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Build an evidence-based lead-scoring rubric whose weights, source fields, uncertainty, and human review points remain inspectable.

#### What it does

- Maps approved fit criteria to weighted, observable evidence instead of intuition.
- Produces a score explanation while preserving an insufficient-data outcome when evidence is absent.

#### When to use

- When a service business wants a transparent scoring draft to compare with human qualification decisions before any automation.

#### When not to use

- Do not use it for automatic rejection, live prioritization, CRM updates, or lead contact without validation and named human approval.

#### Required inputs

- IDEAL CUSTOMER CRITERIA
- DISQUALIFIERS
- OBSERVED LEAD FIELDS
- SCORING WEIGHTS
- QUALIFICATION OWNER

#### Prompt

#CONTEXT:
Draft a weighted lead-scoring rubric for a service business using only supplied criteria, weights, and observed evidence. The result is an analysis for validation, not an automated qualification or adverse decision.

#INPUTS:
- Approved ideal-customer criteria: [IDEAL CUSTOMER CRITERIA]
- Approved disqualifiers: [DISQUALIFIERS]
- Observed lead fields and field provenance: [OBSERVED LEAD FIELDS]
- Proposed criterion weights and scale: [SCORING WEIGHTS]
- Named human qualification owner: [QUALIFICATION OWNER]

#INSTRUCTIONS:
1. Check that every weight maps to one supplied, observable business criterion; report gaps, overlaps, and totals.
2. For each criterion, define the evidence needed for full, partial, zero, and unknown credit.
3. Score only supplied evidence. Do not guess, invent, infer, or fabricate a lead fact.
4. Mark the total "Insufficient data" when missing evidence prevents a defensible score; do not convert unknowns to zero.
5. Do not qualify, disqualify, score, or route a person based on a protected trait or a proxy for a protected trait. Flag suspected proxies for removal and human review.
6. Keep verified hard disqualifiers separate from weighted fit signals and from missing evidence.
7. Recommend validation checks for false positives, false negatives, and uneven outcomes before adoption.
8. Do not update a CRM, prioritize a live queue, route a lead, contact anyone, or send an external message.
9. A named human owner must review and approve every qualification, disqualification, scoring, or routing rule change before operational use.

#RESPONSE FORMAT:
## Weight audit
- Weight total — overlaps — uncovered criteria

## Weighted evidence rubric
| Criterion | Weight | Evidence source | Full / partial / zero evidence | Unknown handling | Score rationale |
|---|---:|---|---|---|---|

## Result states
- Evidence-supported range
- Unknown or Insufficient data — missing evidence
- Verified disqualifier — supplied rule and evidence

## Validation and approval checklist
- Proxy review
- Outcome review
- Named human approval before operational use

#### Expected output

- A weighted evidence rubric with source fields, score rationale, explicit unknown and insufficient-data states, validation checks, and human approval gates.

#### Customization tips

- Use a small number of observable criteria before considering additional weight complexity.
- Compare rubric recommendations with reviewed historical decisions before any operational use.

#### Guardrails

- Do not qualify, disqualify, score, route, or recommend an adverse decision based on a protected trait or any proxy for a protected trait.
- Do not invent missing lead facts or silently score unknown evidence as zero; use Insufficient data.
- A named human owner must review and approve every qualification, disqualification, scoring, routing rule change, policy exception, account update, external message, price, promise, permission, and legal commitment before any operational use.
- The assistant must not access or change a live CRM, prioritize a live queue, route a lead, contact anyone, or send an external message.

#### Example input

Fictional example — IDEAL CUSTOMER CRITERIA: Commercial property manager, location inside the published service map, and a request for recurring maintenance. DISQUALIFIERS: Verified residential-only request or confirmed location outside the service map. OBSERVED LEAD FIELDS: Company: Willow & Finch Properties; portfolio type: commercial; requested service: recurring maintenance; site address: blank. SCORING WEIGHTS: Commercial portfolio 40, in-area site 35, recurring need 25. QUALIFICATION OWNER: Elena, Revenue Operations Manager.

#### Related prompts

- https://returnmytime.com/prompt-library/create-lead-qualification-framework
- https://returnmytime.com/prompt-library/create-lead-routing-rules

#### Related articles

- https://returnmytime.com/blog/ai-lead-qualification-routing

### Create Lead-Routing Rules

- URL: https://returnmytime.com/prompt-library/create-lead-routing-rules
- Category: Revenue Operations
- Workflow: lead-qualification-routing
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Efficiency
- Estimated time saved: 60-90 minutes per routing draft
- Tags: lead-routing, territory-management, revenue-operations
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Draft deterministic lead-routing rules from approved outcomes, territories, capacity, owners, and response commitments without touching a live CRM.

#### What it does

- Turns supplied routing constraints into an ordered decision table with one explainable destination per state.
- Exposes collisions, missing owners, and insufficient-data fallbacks before operational changes.

#### When to use

- When a service business needs reviewable routing logic across territories, teams, capacity limits, and response SLAs.

#### When not to use

- Do not use it to alter live routing, assign a real lead, update an account, or contact anyone without named human review and approval.

#### Required inputs

- APPROVED QUALIFICATION OUTCOMES
- TERRITORIES
- SERVICE CAPACITY
- ROUTING OWNERS
- RESPONSE SLAS
- OBSERVED LEAD FIELDS
- ROUTING RULE OWNER

#### Prompt

#CONTEXT:
Draft deterministic lead-routing rules for a service business using only supplied qualification outcomes, constraints, owners, and observed lead fields. Return a routing design for human review; do not execute it.

#INPUTS:
- Approved qualification outcomes and definitions: [APPROVED QUALIFICATION OUTCOMES]
- Approved territories and tie-break rules: [TERRITORIES]
- Current service capacity by team or offer: [SERVICE CAPACITY]
- Named routing owners and escalation destinations: [ROUTING OWNERS]
- Approved response SLAs by outcome: [RESPONSE SLAS]
- Fields actually observed in the lead record, including blanks: [OBSERVED LEAD FIELDS]
- Named human routing-rule owner: [ROUTING RULE OWNER]

#INSTRUCTIONS:
1. Normalize the supplied rule inputs without creating missing territory, capacity, owner, or SLA facts.
2. Build an ordered decision table whose conditions are mutually exclusive or have an explicit precedence rule.
3. Identify every collision, uncovered state, unavailable owner, and stale capacity fact.
4. Mark "Unknown — Insufficient data" whenever required evidence is absent or conflicting. Recommend the named escalation destination; do not guess, invent, infer, or fabricate a route.
5. Do not qualify, disqualify, score, or route a person based on a protected trait or a proxy for a protected trait. Flag suspected proxies for removal and human review.
6. Preserve the approved response SLA as a target, not a promise of availability or service.
7. Include a safe fallback for unknown territory, exhausted capacity, missing owner, and incomplete qualification.
8. Do not access or update a CRM, assign or route a live lead, contact anyone, send a message, or change a routing rule.
9. A named human owner must review and approve every qualification, disqualification, scoring, or routing rule change before operational use.

#RESPONSE FORMAT:
## Input and conflict audit
- Fact or conflict — source — owner to confirm

## Deterministic routing table
| Priority | Qualification outcome | Territory condition | Capacity condition | Destination owner | SLA target | Unknown / fallback path | Evidence required |
|---:|---|---|---|---|---|---|---|

## Collision and coverage checks
- Test case — expected destination — unresolved issue

## Human approval checklist
- Rule owner named
- Fallbacks tested
- Protected-trait and proxy review complete
- Rule changes approved before operational use

#### Expected output

- A deterministic routing table with rule precedence, evidence requirements, unknown and insufficient-data fallbacks, collision tests, and human approval checks.

#### Customization tips

- Test the table with boundary cases such as missing territory and simultaneous capacity constraints.
- Name one owned fallback instead of silently dropping leads that match no rule.

#### Guardrails

- Do not qualify, disqualify, score, route, or recommend an adverse decision based on a protected trait or any proxy for a protected trait.
- Do not invent missing territory, capacity, owner, SLA, or lead facts; use Unknown — Insufficient data.
- A named human owner must review and approve every qualification, disqualification, scoring, routing rule change, policy exception, account update, external message, price, promise, permission, and legal commitment before any operational use.
- The assistant must not access or change a live CRM, assign or route a live lead, contact anyone, or send an external message.

#### Example input

Fictional example — APPROVED QUALIFICATION OUTCOMES: Qualified, verified disqualifier, and insufficient data. TERRITORIES: Cedar County goes to Central Team; Pine County goes to North Team; unknown county goes to intake review. SERVICE CAPACITY: Central has three assessment slots this week; North is at review threshold. ROUTING OWNERS: Central owner Elena; North owner Marco; intake escalation Priya. RESPONSE SLAS: Qualified within two business hours; insufficient data within four. OBSERVED LEAD FIELDS: County blank; service request recurring maintenance; qualification outcome insufficient data. ROUTING RULE OWNER: Elena, Revenue Operations Manager.

#### Related prompts

- https://returnmytime.com/prompt-library/create-lead-qualification-framework
- https://returnmytime.com/prompt-library/identify-missing-lead-information

#### Related articles

- https://returnmytime.com/blog/ai-lead-qualification-routing

### Identify Missing Lead Information

- URL: https://returnmytime.com/prompt-library/identify-missing-lead-information
- Category: Revenue Operations
- Workflow: lead-qualification-routing
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Efficiency
- Estimated time saved: 30-45 minutes per intake review
- Tags: lead-intake, missing-information, revenue-operations
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Identify the smallest set of missing lead information needed for defensible qualification and routing while preserving unknowns.

#### What it does

- Compares observed fields with the minimum evidence required by supplied qualification and routing rules.
- Prioritizes missing questions by decision impact without treating absence as a negative fact.

#### When to use

- When an intake record is incomplete and a human needs to know what must be confirmed before qualification or routing.

#### When not to use

- Do not use it to infer missing personal information, make an adverse decision, contact a lead, or update a CRM without named human review.

#### Required inputs

- IDEAL CUSTOMER CRITERIA
- DISQUALIFIERS
- TERRITORIES
- SERVICE CAPACITY
- ROUTING OWNERS
- RESPONSE SLAS
- OBSERVED LEAD FIELDS
- QUALIFICATION OWNER

#### Prompt

#CONTEXT:
Draft a missing-information checklist for a service-business lead using only the supplied qualification rules, routing constraints, and observed fields. This is an analysis for human review, not outreach or a live decision.

#INPUTS:
- Approved ideal-customer criteria: [IDEAL CUSTOMER CRITERIA]
- Approved disqualifiers: [DISQUALIFIERS]
- Approved territories and boundary rules: [TERRITORIES]
- Current service capacity facts: [SERVICE CAPACITY]
- Named routing owners and escalation destinations: [ROUTING OWNERS]
- Approved response SLAs: [RESPONSE SLAS]
- Observed lead fields, values, provenance, and blanks: [OBSERVED LEAD FIELDS]
- Named human qualification owner: [QUALIFICATION OWNER]

#INSTRUCTIONS:
1. Compare each supplied decision rule with the evidence actually present in the observed lead fields.
2. Classify each requirement as Verified, Conflicting, Unknown, or Not needed for this decision.
3. Mark the current qualification and routing state "Insufficient data" when a required fact is unknown or conflicting.
4. Do not guess, invent, infer, fabricate, enrich, or derive a missing personal or business fact.
5. Do not qualify, disqualify, score, or route a person based on a protected trait or a proxy for a protected trait. Exclude suspected proxy questions and flag them for human review.
6. Prioritize only the minimum questions needed to distinguish the supplied outcomes; explain why each answer changes a decision.
7. Recommend a human-owned confirmation path, not an external message or contact action.
8. Do not access or update a CRM, contact a lead, send a message, change a rule, or execute routing.
9. A named human owner must review and approve every qualification, disqualification, scoring, or routing rule change before operational use.

#RESPONSE FORMAT:
## Current evidence state
- Verified facts
- Conflicting facts
- Unknown facts
- Current state: Insufficient data or evidence-supported

## Missing-information checklist
| Priority | Missing fact | Rule affected | Why it matters | Approved source to confirm | Human owner | Stop condition |
|---:|---|---|---|---|---|---|

## Excluded questions
- Question — protected-trait, proxy, unnecessary, or unsupported concern

## Human review checklist
- Minimum-data principle followed
- Unknowns preserved
- Qualification and routing decisions remain unexecuted

#### Expected output

- A prioritized missing-information checklist with rule impact, approved confirmation sources, explicit unknown and insufficient-data states, excluded questions, and human review checks.

#### Customization tips

- Remove questions that do not change a supplied qualification or routing decision.
- Name an approved internal source when the answer should come from operations rather than the lead.

#### Guardrails

- Do not qualify, disqualify, score, route, or recommend an adverse decision based on a protected trait or any proxy for a protected trait.
- Do not invent, enrich, or infer missing lead facts; preserve Unknown and Insufficient data.
- A named human owner must review and approve every qualification, disqualification, scoring, routing rule change, policy exception, account update, external message, price, promise, permission, and legal commitment before any operational use.
- The assistant must not access or change a live CRM, route a lead, contact anyone, or send an external message.

#### Example input

Fictional example — IDEAL CUSTOMER CRITERIA: Commercial facility in the approved service area requesting recurring HVAC support. DISQUALIFIERS: Verified residential-only property or confirmed out-of-area location. TERRITORIES: Cedar County and Pine County only. SERVICE CAPACITY: Cedar intake open; Pine capacity awaiting Friday review. ROUTING OWNERS: Elena for Cedar, Marco for Pine, Priya for unknown territory. RESPONSE SLAS: Intake review within four business hours. OBSERVED LEAD FIELDS: Organization Harbor Workshop; commercial site verified; request recurring maintenance; street and county blank; phone not supplied. QUALIFICATION OWNER: Elena, Revenue Operations Manager.

#### Related prompts

- https://returnmytime.com/prompt-library/create-lead-qualification-framework
- https://returnmytime.com/prompt-library/create-lead-routing-rules

#### Related articles

- https://returnmytime.com/blog/ai-lead-qualification-routing

### Create a Sales Follow-Up Sequence

- URL: https://returnmytime.com/prompt-library/create-sales-follow-up-sequence
- Category: Revenue Operations
- Workflow: sales-follow-up-pipeline
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Efficiency
- Estimated time saved: 45-60 minutes per sequence draft
- Tags: sales-follow-up, pipeline, revenue-operations
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Turn verified opportunity history into a stage-specific follow-up sequence that preserves unknowns and routes every draft through human review.

#### What it does

- Builds a multistep draft sequence around the opportunity's verified pipeline stage and contact history.
- Separates supported personalization from missing evidence before a salesperson reviews any message.

#### When to use

- When an active service opportunity needs a consistent follow-up plan after discovery, proposal, or another documented stage transition.

#### When not to use

- Do not use it to contact someone without appropriate consent, manufacture urgency, change a live record, or promise an outcome that has not been approved.
- Do not place unauthorized records or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- PIPELINE STAGE
- PRIOR CONTACT HISTORY
- VERIFIED OBJECTIONS
- PROMISED ACTIONS
- DECISION PROCESS
- RELEVANT DATES
- OFFER
- CRM FIELDS
- FOLLOW-UP OWNER

#### Prompt

#CONTEXT:
Create a stage-specific sales follow-up sequence for a service-business opportunity. Use only the supplied evidence. The sequence is a set of drafts for a named human to review, not a campaign to execute.

#INPUTS:
- Current verified pipeline stage: [PIPELINE STAGE]
- Contact history, channels, and consent facts: [PRIOR CONTACT HISTORY]
- Objections explicitly stated by the buyer: [VERIFIED OBJECTIONS]
- Actions explicitly promised by either party: [PROMISED ACTIONS]
- Known decision participants, criteria, and process: [DECISION PROCESS]
- Relevant supplied dates and timing constraints: [RELEVANT DATES]
- Approved service offer, scope, and claims: [OFFER]
- Supplied CRM fields, including blanks and conflicts: [CRM FIELDS]
- Named human follow-up owner: [FOLLOW-UP OWNER]

#INSTRUCTIONS:
1. Build a source-of-truth block that separates verified facts, conflicts, and Unknown items.
2. Do not invent, infer, or fabricate commitments, quotes, dates, objections, decisions, contact facts, or CRM state. Mark every unsupported item Unknown and state the missing evidence.
3. Explain what the verified pipeline stage permits. If stage criteria were not supplied, label the stage interpretation Unknown instead of creating policy.
4. Draft a three-to-five-step sequence. For each step, provide purpose, timing tied only to a supplied date or labeled relative interval, approved channel, evidence used, message draft, and stop condition.
5. Address only verified objections. Never treat silence, delay, or a missing answer as an objection or decision.
6. Preserve every supplied promise exactly as a fact for review; do not expand it into a new commitment, deadline, guarantee, discount, or offer.
7. Identify conflicts between the contact history, decision process, dates, and CRM fields. Recommend the owner resolve them before use.
8. Use only authorized records in an approved AI workspace or approved AI vendor. Minimize the supplied data and redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
9. Do not send, contact, schedule, update a live CRM, make a commitment, or present an offer. A named human owner must approve the exact draft and each consequential action before use.

#RESPONSE FORMAT:
## Verified opportunity facts
- Fact — supplied source

## Unknowns and missing evidence
- Unknown — missing evidence — why it affects follow-up

## Stage-specific multistep follow-up sequence
| Step | Stage purpose | Proposed timing | Approved channel | Evidence used | Draft message | Stop condition |
|---|---|---|---|---|---|---|

## Conflicts and owner decisions
- Conflict or decision — named owner — evidence needed

## Human approval checklist
- Exact draft reviewed
- Consent and channel verified
- Dates, promises, and offer claims verified
- Live CRM changes separately approved

#### Expected output

- A stage-specific multistep follow-up sequence with evidence-backed drafts, timing, stop conditions, unknowns, and a human approval checklist.

#### Customization tips

- Replace generic stage names with the exit criteria your team has actually approved.
- Keep relative timing tied to the last verified interaction when a calendar date is not supplied.

#### Guardrails

- Do not invent, infer, or fabricate commitments, quotes, dates, objections, decisions, contact facts, or CRM state; mark unsupported details Unknown with the missing evidence.
- A named human owner must approve every external send, external message, contact, and scheduling action before any action.
- A named human owner must approve every live CRM record or account update before any update.
- A named human owner must approve every commitment, promise, offer, price, policy exception, permission, and legal commitment before any action.
- A named human owner must approve every payment, billing, financial, tax, or pricing decision before any action.
- Use only authorized records in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- The assistant must produce drafts only and must not execute a campaign or change a live system.

#### Example input

Fictional example — PIPELINE STAGE: Proposal reviewed. PRIOR CONTACT HISTORY: Morgan at Juniper Facilities replied by email on July 28 and consented to email follow-up; no phone consent supplied. VERIFIED OBJECTIONS: Concern about training time. PROMISED ACTIONS: Seller promised a one-page implementation outline; buyer made no commitment. DECISION PROCESS: Morgan recommends; operations director approves; evaluation criteria otherwise unknown. RELEVANT DATES: Outline requested by August 2; no decision date supplied. OFFER: Northstar Workflow Review, fixed-scope discovery and recommendations; pricing not supplied. CRM FIELDS: Stage=Proposal reviewed; last_contact=2026-07-28; next_step blank; close_date blank. FOLLOW-UP OWNER: Elena, Sales Director.

#### Related prompts

- https://returnmytime.com/prompt-library/summarize-sales-call-next-steps
- https://returnmytime.com/prompt-library/create-pipeline-next-best-action-review


### Revive Stale Sales Opportunities

- URL: https://returnmytime.com/prompt-library/revive-stale-sales-opportunities
- Category: Revenue Operations
- Workflow: sales-follow-up-pipeline
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Effectiveness
- Estimated time saved: 45-75 minutes per stale-opportunity review
- Tags: sales-reactivation, stale-opportunities, pipeline
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Diagnose why a verified opportunity appears stale and draft a respectful reactivation plan without inventing buyer intent or urgency.

#### What it does

- Tests whether the supplied history supports calling an opportunity stale and identifies the evidence needed to decide.
- Drafts a bounded reactivation path with stop conditions and owner decisions.

#### When to use

- When a service opportunity has no documented next step or has passed a supplied follow-up date and needs evidence-based review.

#### When not to use

- Do not use it to ignore opt-outs, invent a reason for silence, revive a closed opportunity, or contact anyone without verified permission and human approval.
- Do not place unauthorized records or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- PIPELINE STAGE
- PRIOR CONTACT HISTORY
- VERIFIED OBJECTIONS
- PROMISED ACTIONS
- DECISION PROCESS
- RELEVANT DATES
- OFFER
- CRM FIELDS
- FOLLOW-UP OWNER

#### Prompt

#CONTEXT:
Review a potentially stale service-business opportunity and create a reactivation plan for human consideration. Use only the supplied record; inactivity alone does not prove lost interest.

#INPUTS:
- Current verified pipeline stage: [PIPELINE STAGE]
- Contact history, channels, responses, and consent facts: [PRIOR CONTACT HISTORY]
- Objections explicitly stated by the buyer: [VERIFIED OBJECTIONS]
- Actions explicitly promised by either party: [PROMISED ACTIONS]
- Known decision participants, criteria, and process: [DECISION PROCESS]
- Relevant dates, stale threshold, and timing constraints: [RELEVANT DATES]
- Approved service offer, scope, and claims: [OFFER]
- Supplied CRM fields, including blanks and conflicts: [CRM FIELDS]
- Named human follow-up owner: [FOLLOW-UP OWNER]

#INSTRUCTIONS:
1. Determine whether the supplied stale threshold and dates support a stale label. If not, mark status Unknown and name the missing evidence.
2. Do not invent, infer, or fabricate commitments, quotes, dates, objections, decisions, contact facts, or CRM state. Mark unsupported details Unknown with their missing evidence.
3. Separate observed inactivity from possible explanations. Label every explanation as a hypothesis, never buyer intent.
4. Check for an opt-out, closed status, unresolved promise, expired offer, unsupported contact channel, or conflicting CRM state. Recommend no outreach draft when any stop condition applies.
5. Create a reactivation plan with objective, evidence, one primary draft, one optional follow-up draft, timing tied to supplied dates, and a clear stop condition.
6. Use only approved offer facts. Do not create scarcity, a concession, discount, new deadline, guarantee, or commitment.
7. Prioritize resolving seller-promised actions before requesting a new buyer action.
8. Use only authorized records in an approved AI workspace or approved AI vendor. Minimize the supplied data and redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
9. Do not send, contact, schedule, change a live CRM, make a commitment, or present an offer. A named human owner must approve the exact draft and each consequential action before use.

#RESPONSE FORMAT:
## Stale-status assessment
- Supported, unsupported, or Unknown — evidence — missing evidence

## Record conflicts and stop conditions
- Finding — supplied source — owner decision needed

## Stale-opportunity reactivation plan
| Step | Objective | Evidence used | Proposed timing | Draft message | Stop condition |
|---|---|---|---|---|---|

## Hypotheses to validate
- Possible explanation — evidence for — evidence against — question for owner

## Human approval checklist
- Consent and status verified
- Seller promises resolved
- Draft claims verified
- CRM changes separately approved

#### Expected output

- A stale-opportunity reactivation plan with a supported status assessment, hypotheses, stop conditions, draft touchpoints, and missing evidence.

#### Customization tips

- Supply your approved stale threshold instead of relying on a generic number of days.
- Include opt-out and closed-status fields so the review can stop inappropriate reactivation.

#### Guardrails

- Do not invent, infer, or fabricate commitments, quotes, dates, objections, decisions, contact facts, or CRM state; mark unsupported details Unknown with the missing evidence.
- A named human owner must approve every external send, external message, contact, and scheduling action before any action.
- A named human owner must approve every live CRM record or account update before any update.
- A named human owner must approve every commitment, promise, offer, price, policy exception, permission, and legal commitment before any action.
- A named human owner must approve every payment, billing, financial, tax, or pricing decision before any action.
- Use only authorized records in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- Honor verified opt-outs and stop conditions; the assistant must not revive, execute, or update anything.

#### Example input

Fictional example — PIPELINE STAGE: Discovery complete. PRIOR CONTACT HISTORY: Casey at Larkspur Studio replied by email on June 20; two later seller emails are logged; no opt-out is recorded; phone consent unknown. VERIFIED OBJECTIONS: Implementation capacity during July. PROMISED ACTIONS: Seller promised a revised phased option but the record does not show delivery. DECISION PROCESS: Casey and finance lead review; final approver unknown. RELEVANT DATES: Team defines stale as 21 days without a buyer response; last buyer response June 20; review date July 31. OFFER: Northstar Process Assessment; approved scope supplied; current price and availability unknown. CRM FIELDS: Stage=Discovery complete; status=open; next_step=send phased option; close_date blank. FOLLOW-UP OWNER: Ravi, Account Executive.

#### Related prompts

- https://returnmytime.com/prompt-library/create-pipeline-next-best-action-review
- https://returnmytime.com/prompt-library/create-sales-follow-up-sequence


### Summarize a Sales Call and Next Steps

- URL: https://returnmytime.com/prompt-library/summarize-sales-call-next-steps
- Category: Revenue Operations
- Workflow: sales-follow-up-pipeline
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Efficiency
- Estimated time saved: 30-45 minutes per call summary
- Tags: sales-call-summary, next-steps, pipeline-handoff
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Convert supplied sales-call notes into a traceable facts, decisions, and actions summary without turning ambiguity into commitments.

#### What it does

- Separates observed facts, explicit decisions, promised actions, open questions, and proposed next steps.
- Creates a review-ready handoff without writing unsupported details into the opportunity record.

#### When to use

- After a service sales conversation when notes, stage context, and opportunity fields need a consistent human-reviewed summary.

#### When not to use

- Do not use it when the source notes are missing, or as permission to record, send, schedule, or promise anything without review.
- Do not place unauthorized call notes or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- SOURCE CALL NOTES
- PIPELINE STAGE
- PRIOR CONTACT HISTORY
- VERIFIED OBJECTIONS
- PROMISED ACTIONS
- DECISION PROCESS
- RELEVANT DATES
- OFFER
- CRM FIELDS
- FOLLOW-UP OWNER

#### Prompt

#CONTEXT:
Summarize supplied sales-call evidence for a service-business opportunity. Produce a draft handoff for human review, not a live CRM entry or buyer communication.

#INPUTS:
- Source call notes or transcript excerpt authorized for analysis: [SOURCE CALL NOTES]
- Current verified pipeline stage: [PIPELINE STAGE]
- Contact history and call context: [PRIOR CONTACT HISTORY]
- Objections explicitly stated in the supplied notes: [VERIFIED OBJECTIONS]
- Actions explicitly promised by either party: [PROMISED ACTIONS]
- Known decision participants, criteria, and process: [DECISION PROCESS]
- Dates explicitly supplied in the notes or record: [RELEVANT DATES]
- Approved service offer, scope, and claims: [OFFER]
- Current supplied CRM fields, including blanks and conflicts: [CRM FIELDS]
- Named human follow-up owner: [FOLLOW-UP OWNER]

#INSTRUCTIONS:
1. Treat only supplied notes and fields as evidence. Label paraphrases as paraphrases; reproduce a quote only when exact quoted wording was supplied.
2. Do not invent, infer, or fabricate commitments, quotes, dates, objections, decisions, contact facts, or CRM state. Mark unsupported details Unknown and name the missing evidence.
3. Separate facts from explicit decisions, promised actions, proposed actions, and open questions. Interest or discussion is not a decision.
4. For each action, identify the party, exact action, supplied due date or Unknown, source evidence, and approval status. Do not assign a date that was not supplied.
5. Compare the call evidence with the CRM fields. List conflicts and proposed record edits as drafts only.
6. Recommend whether the verified evidence supports the current stage. If stage criteria were not supplied, mark the stage assessment Unknown.
7. Draft a concise internal summary and a separate buyer-facing recap draft. Include only verified offer and decision-process facts.
8. Use only authorized records in an approved AI workspace or approved AI vendor. Minimize the supplied data and redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
9. Do not send, contact, schedule, update a live CRM, make a commitment, or present an offer. A named human owner must approve the exact draft and each consequential action before use.

#RESPONSE FORMAT:
## Facts, decisions, and actions call summary
### Verified facts
- Fact — supplied source
### Explicit decisions
- Decision or None recorded — evidence
### Promised actions
| Party | Exact action | Supplied due date or Unknown | Evidence | Approval status |
|---|---|---|---|---|
### Open questions and missing evidence
- Unknown — evidence needed — owner

## Pipeline and CRM review
- Current stage — supported, unsupported, or Unknown — reason
- Proposed field edit — evidence — human approval required

## Internal handoff draft
[Concise evidence-backed summary]

## Buyer recap draft
DRAFT — HUMAN APPROVAL REQUIRED
[Verified recap and proposed next step]

#### Expected output

- A source-traceable facts, decisions, and actions call summary with promises, unknowns, proposed CRM edits, an internal handoff, and a buyer recap draft.

#### Customization tips

- Provide speaker-attributed notes when exact responsibility and wording matter.
- Keep the internal handoff distinct from the shorter buyer recap so internal uncertainty is not hidden.

#### Guardrails

- Do not invent, infer, or fabricate commitments, quotes, dates, objections, decisions, contact facts, or CRM state; mark unsupported details Unknown with the missing evidence.
- A named human owner must approve every external send, external message, contact, and scheduling action before any action.
- A named human owner must approve every live CRM record or account update before any update.
- A named human owner must approve every commitment, promise, offer, price, policy exception, permission, and legal commitment before any action.
- A named human owner must approve every payment, billing, financial, tax, or pricing decision before any action.
- Use only authorized records in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- The assistant must not treat discussion as a decision or alter the source notes.

#### Example input

Fictional example — SOURCE CALL NOTES: July 30 notes — Taylor said the team wants a process map before discussing implementation and can spare no more than two workshop hours; seller said a two-hour workshop option would be drafted by August 4; Taylor said the operations approver still needs to be identified. PIPELINE STAGE: Discovery. PRIOR CONTACT HISTORY: Video call with Taylor at Alder Legal Operations on July 30; prior contact was one consented email. VERIFIED OBJECTIONS: Concern that the team cannot spare more than two workshop hours. PROMISED ACTIONS: Seller will draft a two-hour workshop option by August 4; buyer will identify the operations approver, with no date stated. DECISION PROCESS: Taylor evaluates; operations approver unknown; no purchase decision recorded. RELEVANT DATES: July 30 call; seller draft due August 4. OFFER: Northstar Workflow Assessment; supplied scope includes process mapping. CRM FIELDS: Stage=Discovery; next_step blank; close_date=2026-08-15 with no supporting note. FOLLOW-UP OWNER: Mina, Solutions Lead.

#### Related prompts

- https://returnmytime.com/prompt-library/create-sales-follow-up-sequence
- https://returnmytime.com/prompt-library/create-pipeline-next-best-action-review


### Create a Pipeline Next-Best-Action Review

- URL: https://returnmytime.com/prompt-library/create-pipeline-next-best-action-review
- Category: Revenue Operations
- Workflow: sales-follow-up-pipeline
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Effectiveness
- Estimated time saved: 60-90 minutes per pipeline review
- Tags: pipeline-review, next-best-action, revenue-operations
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Prioritize evidence-backed next actions across supplied opportunities while exposing missing data, conflicts, and human decisions.

#### What it does

- Ranks proposed follow-up work using only supplied stage, history, obligation, and timing evidence.
- Shows the reason, missing evidence, stop condition, and named owner behind each recommendation.

#### When to use

- During a recurring service-business pipeline review when owners need an explainable queue rather than an opaque model score.

#### When not to use

- Do not use it to auto-prioritize people by protected traits, overwrite CRM data, or execute customer-facing actions.
- Do not place unauthorized records or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- PIPELINE STAGE
- PRIOR CONTACT HISTORY
- VERIFIED OBJECTIONS
- PROMISED ACTIONS
- DECISION PROCESS
- RELEVANT DATES
- OFFER
- CRM FIELDS
- FOLLOW-UP OWNER

#### Prompt

#CONTEXT:
Review one or more supplied service-business opportunities and draft an explainable next-best-action queue. Use only the supplied records. Recommendations are for human review and do not change a live system.

#INPUTS:
- Verified pipeline stage for each opportunity: [PIPELINE STAGE]
- Contact history, responses, channels, and consent facts: [PRIOR CONTACT HISTORY]
- Objections explicitly stated by each buyer: [VERIFIED OBJECTIONS]
- Actions explicitly promised by either party: [PROMISED ACTIONS]
- Known decision participants, criteria, and process: [DECISION PROCESS]
- Relevant dates, deadlines, and review window: [RELEVANT DATES]
- Approved service offer, scope, and claims: [OFFER]
- Supplied CRM fields, including blanks and conflicts: [CRM FIELDS]
- Named human follow-up owner or ownership map: [FOLLOW-UP OWNER]

#INSTRUCTIONS:
1. Create a verified evidence block for each opportunity. Preserve conflicts and blanks.
2. Do not invent, infer, or fabricate commitments, quotes, dates, objections, decisions, contact facts, or CRM state. Mark unsupported details Unknown and identify the missing evidence.
3. Recommend a next best action only when supplied evidence supports it. Otherwise recommend evidence collection or human review.
4. Prioritize seller obligations, supplied deadlines, buyer-requested actions, and unresolved record conflicts before speculative outreach.
5. Explain every priority using cited input facts. Do not create a probability, deal score, value, close date, or urgency level that was not supplied.
6. Do not rank or prioritize a person using a protected trait or a proxy for a protected trait.
7. For each row, provide reason, missing evidence, proposed timing, stop condition, named owner, and the human decision needed. Keep proposed CRM edits separate.
8. Use only authorized records in an approved AI workspace or approved AI vendor. Minimize the supplied data and redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
9. Do not send, contact, schedule, update a live CRM, make a commitment, or present an offer. A named human owner must approve the exact draft and each consequential action before use.

#RESPONSE FORMAT:
## Source limits and conflicts
- Opportunity — verified evidence — Unknown or conflict — missing evidence

## Prioritized next-best-action table
| Priority | Opportunity | Verified stage | Proposed next action | Evidence-backed reason | Missing evidence | Proposed timing | Stop condition | Named owner | Human decision needed |
|---|---|---|---|---|---|---|---|---|---|

## Proposed CRM field changes
| Opportunity | Field | Current supplied value | Proposed value | Evidence | Approval required |
|---|---|---|---|---|---|

## Deferred opportunities
- Opportunity — why no action is supportable — evidence needed

## Review checklist
- Every priority traces to supplied evidence
- Unknowns remain visible
- Consequential actions and record changes await named human approval

#### Expected output

- A prioritized next-best-action table with evidence-backed reasons, missing evidence, timing, stop conditions, owners, proposed CRM changes, and deferred items.

#### Customization tips

- Supply explicit priority rules if the team uses more than obligation dates and verified buyer requests.
- Review proposed record edits independently from the action queue to avoid turning recommendations into hidden automation.

#### Guardrails

- Do not invent, infer, or fabricate commitments, quotes, dates, objections, decisions, contact facts, or CRM state; mark unsupported details Unknown with the missing evidence.
- A named human owner must approve every external send, external message, contact, and scheduling action before any action.
- A named human owner must approve every live CRM record or account update before any update.
- A named human owner must approve every commitment, promise, offer, price, policy exception, permission, and legal commitment before any action.
- A named human owner must approve every payment, billing, financial, tax, or pricing decision before any action.
- Use only authorized records in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- A named human owner must approve every deletion, closure, suspension, termination, revocation, deactivation, disablement, or archive action before any action.
- Do not prioritize people using protected traits or proxies; the assistant must not execute actions or edit systems.

#### Example input

Fictional example — PIPELINE STAGE: Juniper Facilities=Proposal reviewed; Alder Legal Operations=Discovery. PRIOR CONTACT HISTORY: Juniper replied July 28 by consented email; Alder completed a video call July 30; no other contact facts supplied. VERIFIED OBJECTIONS: Juniper=training time; Alder=two-hour workshop limit. PROMISED ACTIONS: Juniper seller owes an implementation outline by August 2; Alder seller owes a workshop option by August 4. DECISION PROCESS: Juniper operations director approves; Alder approver unknown. RELEVANT DATES: Review date July 31; obligations due August 2 and August 4; no verified close dates. OFFER: Northstar Workflow Assessment approved scope; no pricing supplied. CRM FIELDS: Juniper next_step blank; Alder close_date=2026-08-15 without source evidence. FOLLOW-UP OWNER: Juniper=Elena; Alder=Mina.

#### Related prompts

- https://returnmytime.com/prompt-library/revive-stale-sales-opportunities
- https://returnmytime.com/prompt-library/summarize-sales-call-next-steps


### Create a Service Quote Draft

- URL: https://returnmytime.com/prompt-library/create-service-quote-draft
- Category: Revenue Operations
- Workflow: speed-to-quote-pricing
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Efficiency
- Estimated time saved: 30-45 minutes per quote draft
- Tags: service-quotes, pricing, revenue-operations
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Turn approved scope and commercial rules into a review-ready service quote draft without filling gaps with invented numbers or commitments.

#### What it does

- Creates a labeled draft quote from the supplied price book, scope, and commercial rules.
- Stops price generation when a required commercial fact is missing and identifies the evidence a human owner must provide.

#### When to use

- When a service opportunity has documented scope and an approved price book but still needs a consistent quote assembled for review.

#### When not to use

- Do not use it to improvise pricing, calculate tax without approved rules, approve an exception, create a legal agreement, or deliver a quote to a customer.
- Do not place unauthorized quote or customer data, or unnecessary personal, sensitive, or confidential data, in an AI system.

#### Required inputs

- APPROVED PRICE BOOK
- CUSTOMER SCOPE
- INCLUSIONS
- EXCLUSIONS
- TAX RULES
- PAYMENT TERMS
- DISCOUNT AUTHORITY
- DELIVERY DEPENDENCIES
- QUOTE OWNER

#### Prompt

#CONTEXT:
Create a labeled draft service quote for human review. Use only the supplied approved commercial evidence. This is a draft, not an approved offer, contract, invoice, or customer communication.

#INPUTS:
- Approved price book, including units, currency, and effective date: [APPROVED PRICE BOOK]
- Customer-requested scope and quantities: [CUSTOMER SCOPE]
- Approved scope inclusions: [INCLUSIONS]
- Approved scope exclusions: [EXCLUSIONS]
- Approved tax rules or explicit Unknown state: [TAX RULES]
- Approved payment and validity terms: [PAYMENT TERMS]
- Named discount authority and approved limits: [DISCOUNT AUTHORITY]
- Delivery prerequisites, capacity facts, and third-party dependencies: [DELIVERY DEPENDENCIES]
- Named human quote owner: [QUOTE OWNER]

#INSTRUCTIONS:
1. Build an evidence table separating supplied facts, conflicts, expired rules, and Unknown items.
2. Do not invent, infer, or fabricate prices, terms, taxes, discounts, SLA commitments, scope facts, availability, or delivery dates.
3. Missing price, tax, or payment term data must stop price generation. Mark the value Unknown, name the missing evidence, and leave affected totals uncalculated.
4. Match each requested line item to an exact approved price-book entry. Show the source entry, quantity, calculation, currency, and effective date. Do not substitute a nearby item.
5. Add separately labeled assumptions. Each assumption must cite the supplied basis or remain Unknown; never turn an assumption into a promise or verified fact.
6. Show a proposed discount only when the supplied authority explicitly permits it; otherwise mark the exception for escalation without applying it.
7. Label every section DRAFT — HUMAN REVIEW REQUIRED. State that tax treatment, payment terms, scope, calculations, and dependencies require verification.
8. Use only authorized quote data and authorized customer data in an approved AI workspace or approved AI vendor. Minimize the data and redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
9. Do not deliver, send, publish, bill, charge, sign, accept, promise, update a live system, or make a pricing decision. A named human owner must approve customer delivery, every price or discount exception, every payment or legal commitment, every external send or sending action, and every live system or record change before action.
10. This prompt is drafting support, not professional financial, tax, or legal advice. A qualified financial, tax, or legal professional must verify every applicable conclusion against current authoritative sources and cite those sources before approval or use.

#RESPONSE FORMAT:
## DRAFT — HUMAN REVIEW REQUIRED
### Evidence and Unknowns
| Item | Status | Supplied source | Missing evidence or conflict |
|---|---|---|---|

### Labeled draft quote
| Line item | Approved price-book entry | Quantity | Calculation | Draft amount | Status |
|---|---|---:|---|---:|---|

### Inclusions, exclusions, and dependencies
- Included
- Excluded
- Dependency — evidence — owner

### Commercial terms review
- Tax treatment
- Payment terms
- Discount status
- Quote validity

## Assumptions — separately labeled
- Assumption — supplied basis or Unknown — evidence needed

## DRAFT — Human approval checklist
- Scope and price-book match verified
- Calculations, currency, tax, and payment terms verified
- Exceptions separately approved
- Customer delivery, external sending, legal commitments, and live changes separately approved

#### Expected output

- A labeled draft quote with price-book traceability, uncalculated Unknown items, separated scope boundaries, commercial-term review, and a named-human approval checklist.

#### Customization tips

- Add your internal price-book identifier and effective-date rule so reviewers can confirm the exact source.
- Keep tax Unknown until an approved jurisdiction-specific rule is supplied; do not use a generic rate.

#### Guardrails

- Do not invent, infer, or fabricate prices, terms, taxes, discounts, SLA commitments, scope, availability, or delivery dates; unsupported values remain Unknown.
- Missing price, tax, or payment term data must stop price generation and leave affected totals uncalculated.
- A named human owner must approve customer delivery and every external send or sending action before action.
- A named human owner must approve every price or discount exception and every pricing, tax, payment, billing, or financial decision before action.
- A named human owner must approve every payment or legal commitment, contract term, promise, and policy exception before action.
- A named human owner must approve every live system or record change before action.
- This prompt is drafting support, not professional financial, tax, or legal advice; a qualified financial, tax, or legal professional must verify every applicable conclusion against current authoritative sources and cite those sources before approval or use.
- Use only authorized quote data and authorized customer data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- The assistant must produce a draft only and must not execute any customer, financial, legal, or system action.

#### Example input

Fictional example — APPROVED PRICE BOOK: Northstar Field Services v4, effective July 1, 2026, USD; Site workflow assessment=$2,400 each; Implementation workshop=$1,200 each. CUSTOMER SCOPE: One assessment and two workshops for Juniper Facilities. INCLUSIONS: Remote interviews, findings brief, two workshops. EXCLUSIONS: Software licenses and travel. TAX RULES: Unknown; finance review required. PAYMENT TERMS: Net 15 after approved invoice; quote validity 14 days. DISCOUNT AUTHORITY: Sales Director may approve up to 5%; no discount requested. DELIVERY DEPENDENCIES: Customer must provide process owner and approved workspace; scheduling availability Unknown. QUOTE OWNER: Elena, Sales Director.

#### Related prompts

- https://returnmytime.com/prompt-library/build-quote-intake-checklist
- https://returnmytime.com/prompt-library/create-pricing-exception-escalation-rules

#### Related articles

- https://returnmytime.com/blog/ai-speed-to-quote-system

### Build a Quote-Intake Checklist

- URL: https://returnmytime.com/prompt-library/build-quote-intake-checklist
- Category: Revenue Operations
- Workflow: speed-to-quote-pricing
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Efficiency
- Estimated time saved: 45-60 minutes per checklist design
- Tags: quote-intake, pricing, handoffs
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Convert approved quoting rules and handoffs into a gate-by-gate intake checklist that prevents incomplete requests from becoming unsupported prices.

#### What it does

- Builds a required-field checklist with evidence, ownership, readiness, and escalation columns.
- Defines a hard stop before price generation when price, tax, or payment-term evidence is incomplete.

#### When to use

- When sales, operations, and finance need one shared definition of a quote-ready request.

#### When not to use

- Do not use it as authority to change policy, approve an exception, calculate missing commercial terms, or alter a live intake workflow.
- Do not place unauthorized quote or customer data, or unnecessary personal, sensitive, or confidential data, in an AI system.

#### Required inputs

- INTAKE FIELDS
- APPROVED PRICE BOOK
- SCOPE RULES
- INCLUSIONS AND EXCLUSIONS
- TAX RULES
- PAYMENT TERMS
- DISCOUNT AUTHORITY
- DELIVERY DEPENDENCIES
- QUOTE HANDOFFS
- QUOTE OWNER

#### Prompt

#CONTEXT:
Build a quote intake checklist for a service business. The output is a review-only operating draft; it cannot approve pricing, exceptions, customer delivery, or live workflow changes.

#INPUTS:
- Current intake fields and definitions: [INTAKE FIELDS]
- Approved price book, units, currency, and effective dates: [APPROVED PRICE BOOK]
- Approved rules for defining and sizing scope: [SCOPE RULES]
- Approved inclusions and exclusions: [INCLUSIONS AND EXCLUSIONS]
- Approved tax rules or explicit Unknown state: [TAX RULES]
- Approved payment and quote-validity terms: [PAYMENT TERMS]
- Named discount authority and approved limits: [DISCOUNT AUTHORITY]
- Delivery prerequisites, capacity facts, and third-party dependencies: [DELIVERY DEPENDENCIES]
- Current sales, operations, finance, and approval handoffs: [QUOTE HANDOFFS]
- Named human quote owner: [QUOTE OWNER]

#INSTRUCTIONS:
1. Inventory each supplied field and rule. Separate required facts, conditional facts, conflicts, expired rules, and Unknown items.
2. Do not invent, infer, or fabricate prices, terms, taxes, discounts, SLA commitments, scope rules, handoffs, owners, or dependencies.
3. Missing price, tax, or payment term data must stop price generation. Mark the intake Not ready, keep the missing item Unknown, and name the evidence and owner needed.
4. Build a quote intake checklist organized by customer need, scope, approved price-book match, inclusions and exclusions, tax, payment terms, discount authority, delivery dependencies, and handoffs.
5. For each checklist item, specify evidence required, accountable human owner, readiness rule, and escalation trigger. Do not create a new policy where none was supplied.
6. Create a deterministic Ready / Not ready / Needs human decision gate. Conflicts and Unknown values cannot pass the gate.
7. Identify duplicate collection, ambiguous field definitions, and handoff gaps as review findings, not as authorization to change a live workflow.
8. Add separately labeled assumptions. Each assumption must cite the supplied basis or remain Unknown; an assumption cannot pass the quote-ready gate as a verified fact.
9. Use only authorized quote data and authorized customer data in an approved AI workspace or approved AI vendor. Minimize the data and redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
10. Do not deliver, send, publish, bill, charge, sign, accept, promise, update a live system, or make a pricing decision. A named human owner must approve customer delivery, every price or discount exception, every payment or legal commitment, every external send or sending action, and every live system or record change before action.
11. This prompt is drafting support, not professional financial, tax, or legal advice. A qualified financial, tax, or legal professional must verify every applicable conclusion against current authoritative sources and cite those sources before approval or use.

#RESPONSE FORMAT:
## Intake-rule evidence
| Rule or field | Supplied source | Status | Conflict or Unknown |
|---|---|---|---|

## Quote intake checklist
| Stage | Required item | Evidence required | Human owner | Readiness rule | Escalation trigger |
|---|---|---|---|---|---|

## Quote-ready gate
- Ready criteria
- Not ready criteria
- Needs human decision criteria

## Handoff gaps and recommended review
| Gap | Operational effect | Evidence needed | Named owner |
|---|---|---|---|

## Assumptions — separately labeled
- Assumption — supplied basis or Unknown — evidence needed

## Human approval checklist
- Commercial rules verified
- Readiness gate approved
- Exceptions, customer delivery, external sending, legal commitments, and live changes separately approved

#### Expected output

- A quote intake checklist with evidence requirements, human owners, deterministic readiness gates, handoff gaps, and explicit Unknown stops.

#### Customization tips

- Name the system of record for each field only after the human owner confirms it.
- Use conditional checklist rows for genuinely conditional services instead of making every field mandatory.

#### Guardrails

- Do not invent, infer, or fabricate prices, terms, taxes, discounts, SLA commitments, scope rules, handoffs, owners, or dependencies; unsupported values remain Unknown.
- Missing price, tax, or payment term data must stop price generation and keep the intake Not ready.
- A named human owner must approve customer delivery and every external send or sending action before action.
- A named human owner must approve every price or discount exception and every pricing, tax, payment, billing, or financial decision before action.
- A named human owner must approve every payment or legal commitment, contract term, promise, and policy exception before action.
- A named human owner must approve every live system or record change before action.
- This prompt is drafting support, not professional financial, tax, or legal advice; a qualified financial, tax, or legal professional must verify every applicable conclusion against current authoritative sources and cite those sources before approval or use.
- Use only authorized quote data and authorized customer data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- The assistant must return a checklist only and must not change a live intake or quoting system.

#### Example input

Fictional example — INTAKE FIELDS: Customer organization, requested service, site count, desired month, decision owner; tax jurisdiction absent. APPROVED PRICE BOOK: Northstar Field Services v4, effective July 1, 2026, USD. SCOPE RULES: One assessment per site; workshop count chosen after assessment. INCLUSIONS AND EXCLUSIONS: Remote interviews included; travel and software excluded. TAX RULES: Unknown; finance review required. PAYMENT TERMS: Net 15; quote validity 14 days. DISCOUNT AUTHORITY: Sales Director may approve up to 5%. DELIVERY DEPENDENCIES: Named process owner and approved workspace required. QUOTE HANDOFFS: Sales intake to operations scope review to finance commercial review. QUOTE OWNER: Elena, Sales Director.

#### Related prompts

- https://returnmytime.com/prompt-library/create-service-quote-draft
- https://returnmytime.com/prompt-library/audit-quote-turnaround-bottlenecks

#### Related articles

- https://returnmytime.com/blog/ai-speed-to-quote-system

### Audit Quote-Turnaround Bottlenecks

- URL: https://returnmytime.com/prompt-library/audit-quote-turnaround-bottlenecks
- Category: Revenue Operations
- Workflow: speed-to-quote-pricing
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Quality
- Estimated time saved: 60-90 minutes per turnaround review
- Tags: quote-turnaround, bottleneck-audit, revenue-operations
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Use supplied timestamps and handoff evidence to find where quote work waits, loops, or arrives incomplete—without inventing timestamps or service targets.

#### What it does

- Calculates observable wait and work intervals only when the supplied timestamps support them.
- Prioritizes intake, handoff, dependency, and approval bottlenecks by evidence and decision impact.

#### When to use

- When quote turnaround feels slow and the team has event timestamps or process records to examine.

#### When not to use

- Do not use it to infer missing timestamps, assign blame, promise an SLA, approve commercial policy, or change a live workflow.
- Do not place unauthorized quote or customer data, or unnecessary personal, sensitive, or confidential data, in an AI system.

#### Required inputs

- QUOTE TIMESTAMPS
- INTAKE FIELDS
- QUOTE HANDOFFS
- DELIVERY DEPENDENCIES
- TURNAROUND TARGETS
- QUOTE OWNER

#### Prompt

#CONTEXT:
Audit quote turnaround using only supplied operational evidence. Produce a review-only bottleneck analysis, not a performance verdict, approved SLA, or live process change.

#INPUTS:
- Quote identifiers and supplied lifecycle timestamps: [QUOTE TIMESTAMPS]
- Intake fields, completeness states, and correction history: [INTAKE FIELDS]
- Current handoff steps, owners, queues, and approval points: [QUOTE HANDOFFS]
- Delivery prerequisites, capacity facts, and third-party dependencies: [DELIVERY DEPENDENCIES]
- Approved turnaround targets or explicit Unknown state: [TURNAROUND TARGETS]
- Named human quote-process owner: [QUOTE OWNER]

#INSTRUCTIONS:
1. Validate the timestamp sequence for each quote. Mark missing, contradictory, timezone-ambiguous, or incomparable data Unknown.
2. Do not invent, infer, or fabricate prices, terms, taxes, discounts, SLA commitments, timestamps, owners, causes, or customer impact.
3. Missing price, tax, or payment term data must stop price generation. This audit must not create any price; keep absent commercial facts Unknown.
4. Calculate elapsed intervals only between verified timestamps. Separate observed elapsed time from active work time when the evidence permits; otherwise mark work time Unknown.
5. Map each delay to the supplied intake field, handoff, approval, or delivery dependency. Use Unattributed when the evidence does not establish a cause.
6. Compare results with a target only when an approved target was supplied. Never convert an informal expectation into an SLA commitment.
7. Rank bottlenecks by frequency, observable delay, and quote-readiness impact. Show denominators, excluded records, and data limitations instead of false precision.
8. Recommend bounded experiments with a human owner, baseline, success measure, and rollback point. Recommendations cannot alter a live workflow.
9. Add separately labeled assumptions. Each assumption must cite the supplied basis or remain Unknown; never count an assumption as observed evidence.
10. Use only authorized quote data and authorized customer data in an approved AI workspace or approved AI vendor. Minimize the data and redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
11. Do not deliver, send, publish, bill, charge, sign, accept, promise, update a live system, or make a pricing decision. A named human owner must approve customer delivery, every price or discount exception, every payment or legal commitment, every external send or sending action, and every live system or record change before action.
12. This prompt is analysis support, not professional financial, tax, or legal advice. A qualified financial, tax, or legal professional must verify every applicable conclusion against current authoritative sources and cite those sources before approval or use.

#RESPONSE FORMAT:
## Evidence quality and exclusions
| Quote or record | Evidence status | Missing or conflicting data | Included in calculation? |
|---|---|---|---|

## Verified turnaround intervals
| Quote | From event | To event | Elapsed time | Target comparison | Evidence |
|---|---|---|---:|---|---|

## Quote-turnaround bottleneck audit
| Rank | Bottleneck | Observed evidence | Frequency and denominator | Delay range | Readiness impact | Confidence |
|---|---|---|---|---|---|---|

## Unknowns and missing evidence
- Unknown — evidence needed — owner

## Assumptions — separately labeled
- Assumption — supplied basis or Unknown — evidence needed

## Review-only experiments
| Experiment | Human owner | Baseline | Success measure | Rollback point |
|---|---|---|---|---|

## Human approval checklist
- Calculations and exclusions verified
- Target and SLA status verified
- Commercial exceptions, customer delivery, external sending, legal commitments, and live changes separately approved

#### Expected output

- An evidence-based quote-turnaround bottleneck audit with validated intervals, exclusions, ranked findings, Unknown causes, and review-only experiments.

#### Customization tips

- Add event definitions before comparing teams so the same timestamp means the same thing everywhere.
- Segment by quote type only when each segment has enough supplied records to show a meaningful denominator.

#### Guardrails

- Do not invent, infer, or fabricate prices, terms, taxes, discounts, SLA commitments, timestamps, owners, causes, or customer impact; unsupported values remain Unknown.
- Missing price, tax, or payment term data must stop price generation; this audit cannot create a price.
- A named human owner must approve customer delivery and every external send or sending action before action.
- A named human owner must approve every price or discount exception and every pricing, tax, payment, billing, or financial decision before action.
- A named human owner must approve every payment or legal commitment, contract term, promise, and policy exception before action.
- A named human owner must approve every live system or record change before action.
- This prompt is analysis support, not professional financial, tax, or legal advice; a qualified financial, tax, or legal professional must verify every applicable conclusion against current authoritative sources and cite those sources before approval or use.
- Use only authorized quote data and authorized customer data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- The assistant must provide analysis only and must not change a live workflow or commit to a service target.

#### Example input

Fictional example — QUOTE TIMESTAMPS: Q-101 intake 2026-07-21 09:00 ET, scope review 2026-07-21 15:00 ET, finance review 2026-07-23 10:00 ET, draft complete 2026-07-23 11:00 ET; Q-102 intake 2026-07-22 13:00 ET, scope review timestamp missing, finance review 2026-07-24 09:00 ET. INTAKE FIELDS: Q-101 complete; Q-102 tax jurisdiction missing and returned once. QUOTE HANDOFFS: Sales to operations to finance to sales owner. DELIVERY DEPENDENCIES: Operations confirms capacity; finance confirms commercial terms. TURNAROUND TARGETS: Internal draft target of two business days; no customer SLA. QUOTE OWNER: Elena, Sales Director.

#### Related prompts

- https://returnmytime.com/prompt-library/build-quote-intake-checklist
- https://returnmytime.com/prompt-library/create-pricing-exception-escalation-rules

#### Related articles

- https://returnmytime.com/blog/ai-speed-to-quote-system

### Create Pricing-Exception Escalation Rules

- URL: https://returnmytime.com/prompt-library/create-pricing-exception-escalation-rules
- Category: Revenue Operations
- Workflow: speed-to-quote-pricing
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Quality
- Estimated time saved: 45-75 minutes per escalation-rule draft
- Tags: pricing-exceptions, discount-authority, governance
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Translate supplied commercial authority into a reviewable decision table for pricing exceptions, Unknown cases, and named-human escalation.

#### What it does

- Maps each supplied exception type to evidence requirements, authority limits, owners, and stop conditions.
- Keeps unsupported commercial decisions Unknown and requires a named human before any exception is used.

#### When to use

- When service quotes stall because discount, tax, payment-term, scope, or dependency exceptions lack a clear review path.

#### When not to use

- Do not use it to create authority, approve an exception, alter pricing, calculate tax, accept contract terms, or change a live approval system.
- Do not place unauthorized quote or customer data, or unnecessary personal, sensitive, or confidential data, in an AI system.

#### Required inputs

- APPROVED PRICE BOOK
- PRICING EXCEPTION POLICY
- DISCOUNT AUTHORITY
- TAX RULES
- PAYMENT TERMS
- DELIVERY DEPENDENCIES
- ESCALATION OWNERS

#### Prompt

#CONTEXT:
Create review-only pricing exception escalation rules from supplied approved policy. The output is a decision-support draft, not authority to approve, price, contract, communicate, or change a system.

#INPUTS:
- Approved price book, units, currency, and effective dates: [APPROVED PRICE BOOK]
- Current approved pricing exception policy and documented stop conditions: [PRICING EXCEPTION POLICY]
- Named discount approvers and their explicit limits: [DISCOUNT AUTHORITY]
- Approved tax rules and tax-review owner: [TAX RULES]
- Approved payment and quote-validity terms: [PAYMENT TERMS]
- Delivery prerequisites, capacity facts, and third-party dependencies: [DELIVERY DEPENDENCIES]
- Named escalation owners by commercial domain: [ESCALATION OWNERS]

#INSTRUCTIONS:
1. Inventory the supplied authority. Separate approved rules, conflicts, expired policy, missing owners, and Unknown items.
2. Do not invent, infer, or fabricate prices, terms, taxes, discounts, SLA commitments, approval limits, exceptions, owners, or delivery facts.
3. Missing price, tax, or payment term data must stop price generation. Mark the decision Unknown and require the missing authoritative evidence.
4. Create a decision row for requests involving price-book mismatch, discount, scope deviation, tax uncertainty, payment-term change, validity-period change, delivery dependency, or SLA request only when supplied rules support that row.
5. For each row, state the trigger, evidence required, permitted recommendation, prohibited action, named human owner, and escalation path. Unknown authority means Stop and escalate; it never means implied approval.
6. Identify overlapping or contradictory authority and keep the affected decision unresolved until named owners reconcile it.
7. Include a proposed audit record containing the request, evidence, recommendation, owner, decision, and decision date. It is a draft schema, not a live record update.
8. Add separately labeled assumptions. Each assumption must cite the supplied basis or remain Unknown; it cannot expand or replace approved authority.
9. Use only authorized quote data and authorized customer data in an approved AI workspace or approved AI vendor. Minimize the data and redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
10. Do not deliver, send, publish, bill, charge, sign, accept, promise, update a live system, apply or offer a discount, waive a fee, or make a pricing decision. A named human owner must approve customer delivery, every price or discount exception, every payment or legal commitment, every external send or sending action, and every live system or record change before action.
11. This prompt is decision support, not professional financial, tax, or legal advice. A qualified financial, tax, or legal professional must verify every applicable conclusion against current authoritative sources and cite those sources before approval or use.

#RESPONSE FORMAT:
## Authority evidence and Unknowns
| Commercial domain | Supplied authority | Limit | Status | Conflict or missing evidence |
|---|---|---|---|---|

## Pricing-exception escalation decision table
| Exception trigger | Evidence required | Permitted recommendation | Prohibited action | Named human owner | Escalation path | Stop condition |
|---|---|---|---|---|---|---|

## Unresolved authority conflicts
- Conflict — affected decisions — owners needed

## Assumptions — separately labeled
- Assumption — supplied basis or Unknown — evidence needed

## Draft audit-record fields
- Request identifier
- Evidence references
- Recommendation
- Named owner and decision
- Decision date

## Human approval checklist
- Authority source and effective date verified
- Exception decision separately approved
- Customer delivery, external sending, payment, legal commitments, and live changes separately approved

#### Expected output

- A pricing-exception escalation decision table with authority evidence, stop conditions, unresolved conflicts, draft audit fields, and named-human ownership.

#### Customization tips

- Separate recommendation authority from final approval authority when they belong to different roles.
- Version the approved policy outside this prompt so each decision can reference an effective source.

#### Guardrails

- Do not invent, infer, or fabricate prices, terms, taxes, discounts, SLA commitments, approval limits, exceptions, owners, or delivery facts; unsupported values remain Unknown.
- Missing price, tax, or payment term data must stop price generation and keep the decision unresolved.
- A named human owner must approve customer delivery and every external send or sending action before action.
- A named human owner must approve every price or discount exception and every pricing, tax, payment, billing, or financial decision before action.
- A named human owner must approve every payment or legal commitment, contract term, promise, and policy exception before action.
- A named human owner must approve every live system or record change before action.
- This prompt is decision support, not professional financial, tax, or legal advice; a qualified financial, tax, or legal professional must verify every applicable conclusion against current authoritative sources and cite those sources before approval or use.
- Use only authorized quote data and authorized customer data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- The assistant must provide decision support only and must not approve an exception or change a live system.

#### Example input

Fictional example — APPROVED PRICE BOOK: Northstar Field Services v4, effective July 1, 2026, USD. PRICING EXCEPTION POLICY: Nonstandard scope and payment terms require Finance Director review; no tax exception policy supplied. DISCOUNT AUTHORITY: Sales Director up to 5%; Finance Director from 5.01% to 10%; above 10% prohibited. TAX RULES: Unknown; finance must obtain approved jurisdiction guidance. PAYMENT TERMS: Standard Net 15; Net 30 requires Finance Director review. DELIVERY DEPENDENCIES: Operations confirms capacity before any delivery-date discussion. ESCALATION OWNERS: Elena, Sales Director for discounts up to 5%; Priya, Finance Director for commercial terms; Omar, Operations Director for capacity.

#### Related prompts

- https://returnmytime.com/prompt-library/create-service-quote-draft
- https://returnmytime.com/prompt-library/audit-quote-turnaround-bottlenecks

#### Related articles

- https://returnmytime.com/blog/ai-speed-to-quote-system

### Create Customer-Escalation Triage

- URL: https://returnmytime.com/prompt-library/create-customer-escalation-triage
- Category: Customer Experience
- Workflow: customer-support-service-recovery
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Quality
- Estimated time saved: 20-30 minutes per escalation review
- Tags: customer-escalation, support-triage, service-recovery
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Turn verified incident evidence and supplied severity rules into a reviewable customer-escalation level, reason, and ownership path.

#### What it does

- Maps supplied incident facts and customer impact to the exact severity criteria the business provided.
- Keeps unsupported classifications Unknown and prepares a named-owner escalation recommendation for human review.

#### When to use

- When a customer issue needs consistent severity triage before a support or operations owner decides the response path.

#### When not to use

- Do not use it to issue compensation, admit fault, select an unapproved exception, contact a customer, route a live ticket, or change a support record.
- Do not place unauthorized support data or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- INCIDENT FACTS
- CUSTOMER IMPACT
- APPLICABLE POLICY
- SEVERITY RULES
- CURRENT OWNERSHIP
- APPROVED RESOLUTION OPTIONS
- ESCALATION OWNERS

#### Prompt

#CONTEXT:
Create a review-only customer-escalation triage assessment. Use only the supplied incident evidence, policy, severity rules, ownership, and approved options. The output is decision support, not authorization to act.

#INPUTS:
- Verified incident facts with evidence references: [INCIDENT FACTS]
- Customer-reported and independently verified impact: [CUSTOMER IMPACT]
- Current applicable policy with source and effective date: [APPLICABLE POLICY]
- Approved severity levels, criteria, and stop conditions: [SEVERITY RULES]
- Current ticket, account, and operational ownership facts: [CURRENT OWNERSHIP]
- Resolution options already approved for consideration: [APPROVED RESOLUTION OPTIONS]
- Named escalation owners and their documented authority: [ESCALATION OWNERS]

#INSTRUCTIONS:
1. Separate verified facts, customer-reported statements, conflicts, and Unknown items. Cite the supplied evidence for every verified fact.
2. Do not invent, infer, or fabricate policy, eligibility, customer facts, outcomes, severity rules, owner authority, fault, or liability.
3. Compare the evidence with each supplied severity rule. Recommend a triage level only when an exact rule is satisfied; otherwise mark the level Unknown and identify the missing evidence.
4. Explain the triage reason using the matched rule and evidence. Do not treat a customer's frustration, tone, identity, or unsupported claim as proof of severity or fault.
5. Identify the named owner whose supplied authority covers the recommended path. If ownership conflicts or is missing, preserve the conflict and recommend human reconciliation.
6. List only supplied approved resolution options as context. Do not choose, issue, promise, or imply a refund, credit, compensation, retention offer, discount, exception, or outcome.
7. Recommend internal next-review questions, evidence to collect, and a proposed escalation path. Do not route, assign, close, resolve, or update a live ticket, system, account, or record.
8. Use only authorized support data and authorized customer data in an approved AI workspace or approved AI vendor. Minimize the data and redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
9. Do not send, publish, contact, notify, or deliver any external communication. A named human owner must approve every external communication, sending action, and customer contact before action.
10. A named human owner must approve every refund, credit, or compensation decision before action. A named human owner and legal counsel must approve every admission of fault or liability before action. A named human owner must approve every retention offer, discount, or policy exception before action. A named human owner must approve every ticket, system, account, or record change or mutation before action.

#RESPONSE FORMAT:
## Evidence, conflicts, and Unknowns
| Item | Status | Supplied evidence | Conflict or missing evidence |
|---|---|---|---|

## Recommended triage level and reason
- Level: [SUPPLIED LEVEL OR UNKNOWN]
- Matched rule: [EXACT RULE OR UNKNOWN]
- Evidence-based reason
- Confidence and evidence limits

## Ownership and escalation recommendation
| Proposed review step | Named owner | Supplied authority | Stop condition |
|---|---|---|---|

## Human approval checklist
- Severity and evidence verified
- Ownership authority verified
- Customer communication separately approved
- Any refund, credit, compensation, admission, offer, exception, or live change separately approved

#### Expected output

- An evidence-backed triage level and reason with explicit Unknowns, ownership, stop conditions, and a named-human approval checklist.

#### Customization tips

- Replace generic severity labels with your current, versioned incident rubric.
- Add regulated escalation triggers only after the responsible professional approves the exact language.

#### Guardrails

- Do not invent, infer, or fabricate policy, eligibility, customer facts, outcomes, severity rules, authority, fault, or liability; unsupported items remain Unknown and conflicts remain unresolved.
- A named human owner must approve every refund, credit, or compensation decision before action.
- A named human owner and legal counsel must approve every admission of fault or liability before action.
- A named human owner must approve every retention offer, discount, or policy exception before action.
- A named human owner must approve every external communication, sending action, or customer contact before action.
- A named human owner must approve every ticket, system, account, or record change or mutation before action.
- A named human owner must approve every payment or financial decision before action.
- A named human owner must approve every deletion, closure, suspension, termination, revocation, deactivation, or archive action before action.
- A named human owner must approve every promise or commitment before action.
- Use only authorized support data and authorized customer data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- The assistant must provide triage decision support only and must not take a customer, financial, legal, ticket, or system action.

#### Example input

Fictional example — INCIDENT FACTS: Juniper Facilities ticket JS-104 says a scheduled onboarding session did not start; calendar log shows the host link failed at 10:02 ET. CUSTOMER IMPACT: Customer reports six staff members waited 25 minutes; attendance count is not independently verified. APPLICABLE POLICY: Support policy v3, effective July 1, 2026. SEVERITY RULES: Level 2 when a scheduled delivery is blocked for one customer with no safety or data exposure; Level 1 requires safety, security, or multi-customer impact. CURRENT OWNERSHIP: Support owns intake; Operations owns delivery incidents. APPROVED RESOLUTION OPTIONS: Manager follow-up and rescheduling review; no credit approved. ESCALATION OWNERS: Omar, Operations Director, owns Level 2 delivery reviews.

#### Related prompts

- https://returnmytime.com/prompt-library/create-service-recovery-plan
- https://returnmytime.com/prompt-library/summarize-support-ticket-root-causes
- https://returnmytime.com/prompt-library/reply-to-customer-complaints


### Create a Service-Recovery Plan

- URL: https://returnmytime.com/prompt-library/create-service-recovery-plan
- Category: Customer Experience
- Workflow: customer-support-service-recovery
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Quality
- Estimated time saved: 30-45 minutes per recovery plan
- Tags: service-recovery, customer-impact, action-planning
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Convert verified incident evidence and approved resolution options into an owned, reviewable service-recovery plan without authorizing customer or system action.

#### What it does

- Builds an internal recovery plan from verified impact, supplied policy, ownership, and approved options.
- Separates immediate containment, customer-response drafting, corrective work, and evidence gaps for named-human review.

#### When to use

- After an issue has been triaged and named owners need a consistent draft plan for recovery and follow-through.

#### When not to use

- Do not use it to issue a refund or credit, admit fault, make a retention offer, contact a customer, close a ticket, or change a live system.
- Do not place unauthorized support data or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- INCIDENT FACTS
- CUSTOMER IMPACT
- APPLICABLE POLICY
- SEVERITY
- CURRENT OWNERSHIP
- APPROVED RESOLUTION OPTIONS
- RECOVERY CONSTRAINTS

#### Prompt

#CONTEXT:
Create a draft service-recovery plan for internal human review. Use only the supplied evidence, approved policy, severity, owner authority, resolution options, and constraints. Do not execute the plan.

#INPUTS:
- Verified incident facts, timeline, and evidence references: [INCIDENT FACTS]
- Customer-reported and verified operational impact: [CUSTOMER IMPACT]
- Current applicable policy with source and effective date: [APPLICABLE POLICY]
- Human-reviewed severity level and supporting reason: [SEVERITY]
- Current owners, responsibilities, and authority: [CURRENT OWNERSHIP]
- Resolution options explicitly approved for consideration: [APPROVED RESOLUTION OPTIONS]
- Capacity, legal, financial, technical, and timing constraints: [RECOVERY CONSTRAINTS]

#INSTRUCTIONS:
1. Build a source-of-truth table separating verified facts, reported statements, conflicts, expired inputs, and Unknown items.
2. Do not invent, infer, or fabricate policy, eligibility, customer facts, outcomes, root cause, dates, capacity, approvals, fault, or liability.
3. Draft a service-recovery action plan in four lanes: containment, fact verification, approved customer-response preparation, and corrective follow-through.
4. For every proposed step, state the evidence basis, named owner, dependency, target proposed for owner approval, stop condition, and verification evidence. Do not convert a proposal into a promise.
5. Include only supplied approved resolution options. Recommend escalation when an option, owner, authority, timing, or eligibility fact is missing or conflicting.
6. Do not issue, apply, process, or promise a refund, reimbursement, credit, compensation, retention offer, discount, concession, exception, or guarantee.
7. Draft customer-response talking points only when supported by verified facts and supplied policy. Do not admit fault or liability, and do not characterize legal responsibility.
8. Do not send, publish, contact, notify, route, assign, close, resolve, or update any live ticket, account, record, or system.
9. Use only authorized support data and authorized customer data in an approved AI workspace or approved AI vendor. Minimize the data and redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
10. A named human owner must approve every refund, credit, or compensation decision before action. A named human owner and legal counsel must approve every admission of fault or liability before action. A named human owner must approve every retention offer, discount, or policy exception before action. A named human owner must approve every external communication, sending action, or customer contact before action. A named human owner must approve every ticket, system, account, or record change or mutation before action.

#RESPONSE FORMAT:
## Evidence, conflicts, and Unknowns
| Item | Status | Supplied source | Conflict or missing evidence |
|---|---|---|---|

## Draft service-recovery action plan
| Lane | Proposed step | Evidence basis | Named owner | Dependency | Proposed target | Stop condition | Verification |
|---|---|---|---|---|---|---|---|

## Customer-response talking points for review
- Verified acknowledgment
- Supported next step
- Prohibited or unresolved statement

## Human decisions and approvals
- Decision — named owner — evidence required — approval state

#### Expected output

- A reviewable service-recovery action plan with evidence, four recovery lanes, named ownership, stop conditions, Unknowns, and human approvals.

#### Customization tips

- Add your internal incident and corrective-action identifiers so evidence can be traced without copying excess customer data.
- Separate the internal operational target from any customer-facing commitment until the right owner approves it.

#### Guardrails

- Do not invent, infer, or fabricate policy, eligibility, customer facts, outcomes, root cause, dates, capacity, approval, fault, or liability; unsupported items remain Unknown and conflicts remain unresolved.
- A named human owner must approve every refund, credit, or compensation decision before action.
- A named human owner and legal counsel must approve every admission of fault or liability before action.
- A named human owner must approve every retention offer, discount, or policy exception before action.
- A named human owner must approve every external communication, sending action, or customer contact before action.
- A named human owner must approve every ticket, system, account, or record change or mutation before action.
- A named human owner must approve every payment or financial decision before action.
- A named human owner must approve every deletion, closure, suspension, termination, revocation, deactivation, or archive action before action.
- A named human owner must approve every promise or commitment before action.
- Use only authorized support data and authorized customer data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- The assistant must produce a draft plan only and must not perform customer, financial, legal, ticket, or system actions.

#### Example input

Fictional example — INCIDENT FACTS: Northstar's onboarding host link failed at 10:02 ET; the team created a replacement link at 10:18 ET. CUSTOMER IMPACT: Juniper Facilities reports six attendees waited; exact attendance is Unknown. APPLICABLE POLICY: Delivery incident policy v3 permits a manager follow-up and rescheduling review. SEVERITY: Operations owner assigned Level 2 after review. CURRENT OWNERSHIP: Omar owns recovery; Maya owns response drafting; Finance owns any credit decision. APPROVED RESOLUTION OPTIONS: Manager follow-up and priority rescheduling review; no refund, credit, or compensation approved. RECOVERY CONSTRAINTS: Trainer availability and customer schedule Unknown; no legal conclusion supplied.

#### Related prompts

- https://returnmytime.com/prompt-library/create-customer-escalation-triage
- https://returnmytime.com/prompt-library/summarize-support-ticket-root-causes
- https://returnmytime.com/prompt-library/create-customer-retention-response
- https://returnmytime.com/prompt-library/generate-customer-refund-emails


### Summarize Support-Ticket Root Causes

- URL: https://returnmytime.com/prompt-library/summarize-support-ticket-root-causes
- Category: Customer Experience
- Workflow: customer-support-service-recovery
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Quality
- Estimated time saved: 2-4 hours per ticket review
- Tags: support-analysis, root-cause, customer-experience
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Analyze a supplied ticket corpus for evidence-backed recurring themes while separating symptoms, hypotheses, conflicts, and unproven causes.

#### What it does

- Groups supplied support tickets into traceable themes with counts, evidence references, and confidence limits.
- Separates observed symptoms from possible causes and preserves low-evidence conclusions as Unknown.

#### When to use

- When a support leader needs a bounded review of recurring ticket patterns before choosing investigation or improvement work.

#### When not to use

- Do not use it to declare fault, make compensation decisions, contact customers, close tickets, or update source records.
- Do not place unauthorized ticket data or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- SUPPORT TICKET CORPUS
- ANALYSIS WINDOW
- CATEGORY DEFINITIONS
- SEVERITY RULES
- OWNERSHIP DATA
- KNOWN CHANGES
- MINIMUM EVIDENCE THRESHOLD

#### Prompt

#CONTEXT:
Summarize recurring support-ticket themes and possible root causes for human review. Analyze only the supplied corpus and rules. This is an evidence review, not a final causal finding or authorization to change operations.

#INPUTS:
- Authorized, minimized ticket corpus with stable ticket references: [SUPPORT TICKET CORPUS]
- Start date, end date, timezone, and inclusion rules: [ANALYSIS WINDOW]
- Approved issue categories and definitions: [CATEGORY DEFINITIONS]
- Approved severity levels and criteria: [SEVERITY RULES]
- Supplied team, queue, product, and process ownership data: [OWNERSHIP DATA]
- Verified releases, process changes, outages, or other known events: [KNOWN CHANGES]
- Minimum ticket count and evidence quality required for a root-cause hypothesis: [MINIMUM EVIDENCE THRESHOLD]

#INSTRUCTIONS:
1. Validate the corpus boundary. Report duplicates, missing dates, conflicting fields, sampling limits, and Unknown coverage before analysis.
2. Do not invent, infer, or fabricate policy, eligibility, customer facts, outcomes, ticket content, counts, owner responsibility, fault, or root cause.
3. Classify tickets only against the supplied category and severity definitions. Keep unmatched or ambiguous tickets in an Unknown or conflict group.
4. Produce exact counts and percentages using the included corpus as the denominator. Show the calculation and never generalize beyond the analysis window.
5. Separate observed symptoms, recurring themes, correlations, and possible root-cause hypotheses. A hypothesis must meet the supplied evidence threshold and cite ticket references plus any known change; otherwise label it Unknown or insufficient evidence.
6. Do not infer customer intent, employee performance, legal responsibility, or fault from ticket language or correlations.
7. Recommend investigation questions and internal evidence to collect. Do not route, assign, close, resolve, delete, archive, or update a ticket, account, record, or system.
8. Do not issue or recommend an automatic refund, credit, compensation, retention offer, discount, exception, admission, or customer communication based on aggregate themes.
9. Use only authorized ticket data and authorized customer data in an approved AI workspace or approved AI vendor. Minimize the data and redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
10. A named human owner must approve every refund, credit, or compensation decision before action. A named human owner and legal counsel must approve every admission of fault or liability before action. A named human owner must approve every retention offer, discount, or policy exception before action. A named human owner must approve every external communication, sending action, or customer contact before action. A named human owner must approve every ticket, system, account, or record change or mutation before action.

#RESPONSE FORMAT:
## Corpus quality and Unknowns
- Included and excluded records
- Duplicate, missing, and conflicting data
- Analysis limitations

## Evidence-based root-cause themes
| Theme | Ticket count | Percentage | Evidence references | Symptom or hypothesis | Threshold status | Confidence limit |
|---|---:|---:|---|---|---|---|

## Unresolved hypotheses
- Hypothesis — missing evidence — investigation question

## Human-reviewed investigation plan
| Proposed investigation | Named owner | Evidence needed | Prohibited action |
|---|---|---|---|

#### Expected output

- Evidence-based root-cause themes with corpus-quality limits, traceable counts, separated symptoms and hypotheses, Unknowns, and a human-reviewed investigation plan.

#### Customization tips

- Use stable ticket references instead of copying names, email addresses, or full message bodies into the analysis.
- Set a threshold that matches your sample size and require operational evidence beyond customer wording before naming a cause.

#### Guardrails

- Do not invent, infer, or fabricate policy, eligibility, customer facts, outcomes, ticket content, counts, ownership, fault, or root cause; unsupported items remain Unknown and conflicts remain unresolved.
- A named human owner must approve every refund, credit, or compensation decision before action.
- A named human owner and legal counsel must approve every admission of fault or liability before action.
- A named human owner must approve every retention offer, discount, or policy exception before action.
- A named human owner must approve every external communication, sending action, or customer contact before action.
- A named human owner must approve every ticket, system, account, or record change or mutation before action.
- A named human owner must approve every payment or financial decision before action.
- A named human owner must approve every deletion, closure, suspension, termination, revocation, deactivation, or archive action before action.
- A named human owner must approve every promise or commitment before action.
- Use only authorized ticket data and authorized customer data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- The assistant must analyze the supplied corpus only and must not take customer, financial, legal, ticket, or system action.

#### Example input

Fictional example — SUPPORT TICKET CORPUS: 24 redacted Northstar support records with IDs JS-201 through JS-224; four lack a resolution category and two appear duplicated. ANALYSIS WINDOW: July 1-30, 2026, America/New_York; include closed and open onboarding tickets created in-window. CATEGORY DEFINITIONS: access, scheduling, documentation, billing, other. SEVERITY RULES: Level 1 safety/security or multi-customer outage; Level 2 blocked delivery; Level 3 guidance request. OWNERSHIP DATA: Support owns intake, Operations owns delivery, Finance owns billing. KNOWN CHANGES: Scheduling form v2 released July 14; no verified outage supplied. MINIMUM EVIDENCE THRESHOLD: At least five non-duplicate tickets plus one independent operational record before naming a root-cause hypothesis.

#### Related prompts

- https://returnmytime.com/prompt-library/create-customer-escalation-triage
- https://returnmytime.com/prompt-library/create-service-recovery-plan


### Create a Customer-Retention Response

- URL: https://returnmytime.com/prompt-library/create-customer-retention-response
- Category: Customer Experience
- Workflow: customer-support-service-recovery
- Audience: owner, founder-operator, sales-leader
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Quality
- Estimated time saved: 20-30 minutes per response draft
- Tags: customer-retention, response-drafting, service-recovery
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Draft a human-reviewed retention response from verified case facts, current policy, and explicitly approved offer boundaries.

#### What it does

- Creates an empathetic draft grounded in supplied facts, contact context, policy, and approved retention boundaries.
- Keeps every unsupported option, promise, admission, and customer detail outside the customer-ready draft.

#### When to use

- When an authorized team member has gathered the service issue facts and needs a consistent response draft for human review.

#### When not to use

- Do not use it to determine eligibility, invent an offer, issue a credit, admit fault, contact a customer, or update an account or ticket.
- Do not place unauthorized customer data or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- INCIDENT FACTS
- CUSTOMER IMPACT
- APPLICABLE POLICY
- APPROVED RETENTION BOUNDARIES
- CURRENT OWNERSHIP
- CUSTOMER CONTACT FACTS
- COMPANY TONE
- RESPONSE REVIEWER

#### Prompt

#CONTEXT:
Create a customer-retention response draft for named-human review. Use only the supplied verified facts, policy, approved retention boundaries, ownership, and contact facts. The output is not approval to make an offer or contact the customer.

#INPUTS:
- Verified incident facts and evidence references: [INCIDENT FACTS]
- Customer-reported and independently verified impact: [CUSTOMER IMPACT]
- Current applicable policy with source and effective date: [APPLICABLE POLICY]
- Retention options, eligibility rules, amounts, limits, and prohibited claims explicitly approved for consideration: [APPROVED RETENTION BOUNDARIES]
- Current support, account, financial, and escalation ownership: [CURRENT OWNERSHIP]
- Verified customer name, case reference, channel, consent, and contact history: [CUSTOMER CONTACT FACTS]
- Approved company tone and style constraints: [COMPANY TONE]
- Named human responsible for reviewing the exact draft: [RESPONSE REVIEWER]

#INSTRUCTIONS:
1. Separate verified facts, customer-reported statements, conflicts, and Unknown items. Do not place unsupported details in the response draft.
2. Do not invent, infer, or fabricate policy, eligibility, customer facts, outcomes, contact consent, offers, amounts, dates, fault, or liability.
3. Determine whether the supplied evidence matches an exact approved retention boundary. If any required fact, authority, or eligibility rule is missing or conflicting, mark the option Unknown and draft an internal escalation note instead of an offer.
4. Draft an empathetic acknowledgment of the verified experience without admitting fault or liability and without mirroring hostile language.
5. Include a retention option only when the supplied boundary explicitly supports it. Quote its approved terms exactly and label it pending named-human approval; do not enhance, combine, or imply other value.
6. Do not issue, apply, process, or promise a refund, reimbursement, credit, compensation, retention offer, discount, concession, exception, waiver, guarantee, or outcome.
7. Do not send, publish, email, message, call, contact, notify, schedule, cancel, route, assign, close, resolve, or update any customer, ticket, account, record, or system.
8. Use only authorized customer data and authorized support data in an approved AI workspace or approved AI vendor. Minimize the data and redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
9. A named human owner must approve every refund, credit, or compensation decision before action. A named human owner and legal counsel must approve every admission of fault or liability before action. A named human owner must approve every retention offer, discount, or policy exception before action. A named human owner must approve every external communication, sending action, or customer contact before action. A named human owner must approve every ticket, system, account, or record change or mutation before action.

#RESPONSE FORMAT:
## Evidence, conflicts, and Unknowns
| Item | Status | Supplied evidence | Missing or conflicting evidence |
|---|---|---|---|

## Retention-boundary check
- Exact approved rule or Unknown
- Eligibility evidence or Unknown
- Option and terms permitted for consideration
- Named approver and prohibited variations

## DRAFT — HUMAN REVIEW REQUIRED
- Subject
- Customer acknowledgment
- Verified resolution context
- Approved option, only if supported
- Proposed next step without promise

## Human approval checklist
- Facts, policy, eligibility, and consent verified
- Exact option and terms separately approved
- No unsupported admission, promise, exception, or outcome
- External communication and all live changes separately approved

#### Expected output

- A human-reviewed retention response with an evidence table, approved-boundary check, clearly labeled draft, Unknowns, and an approval checklist.

#### Customization tips

- Supply a versioned offer matrix that separates eligibility evidence, draft authority, and final approval authority.
- Remove contact details that are not necessary to draft the response and keep channel consent explicit.

#### Guardrails

- Do not invent, infer, or fabricate policy, eligibility, customer facts, outcomes, consent, offers, amounts, fault, or liability; unsupported items remain Unknown and conflicts remain unresolved.
- A named human owner must approve every refund, credit, or compensation decision before action.
- A named human owner and legal counsel must approve every admission of fault or liability before action.
- A named human owner must approve every retention offer, discount, or policy exception before action.
- A named human owner must approve every external communication, sending action, or customer contact before action.
- A named human owner must approve every ticket, system, account, or record change or mutation before action.
- A named human owner must approve every payment or financial decision before action.
- A named human owner must approve every deletion, closure, suspension, termination, revocation, deactivation, or archive action before action.
- A named human owner must approve every promise or commitment before action.
- Use only authorized customer data and authorized support data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- The assistant must draft for review only and must not take customer, financial, legal, ticket, account, or system action.

#### Example input

Fictional example — INCIDENT FACTS: Juniper Facilities' first onboarding session began 18 minutes late after a verified host-link issue. CUSTOMER IMPACT: Customer reports six attendees waited; attendance count remains Unknown. APPLICABLE POLICY: Customer care policy v5, effective July 1, 2026. APPROVED RETENTION BOUNDARIES: Operations Director may approve a replacement session after a verified delivery interruption; no refund, credit, discount, or fee waiver is approved. CURRENT OWNERSHIP: Maya owns response drafting; Omar approves service recovery; Finance owns credits. CUSTOMER CONTACT FACTS: Morgan opened case JS-104 by email and consented to email case updates; no phone consent supplied. COMPANY TONE: Calm, accountable, plain language. RESPONSE REVIEWER: Omar, Operations Director.

#### Related prompts

- https://returnmytime.com/prompt-library/create-service-recovery-plan
- https://returnmytime.com/prompt-library/reply-to-customer-complaints
- https://returnmytime.com/prompt-library/generate-customer-refund-emails


### Turn an Expert Interview into an SOP

- URL: https://returnmytime.com/prompt-library/turn-expert-interview-into-sop
- Category: Knowledge Systems
- Workflow: company-knowledge-base-sop-capture
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: knowledge-base
- Type: Quality
- Estimated time saved: 45-60 minutes per SOP draft
- Tags: SOP, expert-interview, knowledge-capture
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Convert an expert interview into a source-attributed, reviewable SOP while preserving uncertainty, conflicts, and human ownership.

#### What it does

- Structures supplied expert knowledge into a step-by-step draft with source attribution and evidence states.
- Separates verified instructions from Unknown details and unresolved business-rule conflicts for owner review.

#### When to use

- After a subject-matter interview when the team needs a traceable first SOP draft, not a silently invented process.

#### When not to use

- Do not use it to publish an SOP, replace the canonical source, decide which conflicting rule wins, change permissions, or operate a live system.
- Do not place unauthorized knowledge data or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- INTERVIEW TRANSCRIPT
- SOURCE OWNER
- SYSTEM OR LOCATION
- SENSITIVITY
- AUTHORITY
- LAST-REVIEWED DATE
- SOP TEMPLATE

#### Prompt

#CONTEXT:
Create a reviewable SOP draft from a supplied expert interview. Preserve provenance and uncertainty. The output is a draft for a named human owner, not a canonical-source update.

#INPUTS:
- Expert interview transcript with speaker labels: [INTERVIEW TRANSCRIPT]
- Named source owner and role: [SOURCE OWNER]
- Current system or location where this knowledge is stored: [SYSTEM OR LOCATION]
- Sensitivity label or classification: [SENSITIVITY]
- Supplied authority and decision rights: [AUTHORITY]
- Last-reviewed date, if documented: [LAST-REVIEWED DATE]
- Approved SOP structure or template: [SOP TEMPLATE]

#INSTRUCTIONS:
1. Extract only instructions, decisions, exceptions, prerequisites, and evidence present in the interview.
2. Cite the speaker and transcript passage for every operational statement. Never turn a suggestion or memory into a verified rule.
3. Build a reviewable SOP with purpose, scope, prerequisites, numbered steps, decision points, exceptions, stop conditions, evidence of completion, and escalation questions.
4. Label gaps Unknown. Label contradictory statements Unresolved conflict, cite both sources, and prepare the decision question for the named owner's review.
5. Separate the expert's documented authority from assumed organizational authority. Treat missing authority or review dates as Unknown.
6. Include a human review checklist and proposed verification exercise. Do not execute the process.
7. Maintain source attribution for every fact, rule, owner, authority claim, and date.
8. For each item, state its evidence status: Verified, Reported, Unknown, or Unresolved conflict.
9. Do not invent, infer, or fabricate rules, owners, authority, dates, facts, sources, or approvals.
10. Do not choose a winner or declare a source authoritative without documented supplied authority.
11. Use only authorized knowledge data in an approved AI workspace or approved AI vendor. Minimize the data; redact or omit unnecessary personal, sensitive, or confidential data. Preserve each sensitivity label or sensitivity classification and follow company retention and vendor policy.
12. Do not publish. Do not update a canonical source. Do not grant, revoke, or change access or permissions. Do not send material externally. Do not change a live system.
13. A named human owner must approve publication before action. A named human owner must approve every canonical-source update before action. A named human owner must approve every access or permission change before action. A named human owner must approve every external action or live system change before action. A named human owner must approve every policy exception before action.

#RESPONSE FORMAT:
## Source attribution and evidence status
| SOP claim | Evidence status | Speaker and transcript reference | Authority | Review note |
|---|---|---|---|---|

## Reviewable SOP
- Purpose and scope
- Preconditions
- Numbered procedure with decision points and stop conditions
- Exceptions and escalation path
- Evidence of completion

## Unknowns and Unresolved conflicts
| Item | Competing evidence | Named decision owner | Question to resolve |
|---|---|---|---|

## Human approval checklist

#### Expected output

- A source-attributed, reviewable SOP with evidence states, numbered steps, explicit Unknowns, an Unresolved conflict register, and a human approval checklist.

#### Customization tips

- Add transcript timestamps so reviewers can return to the exact statement quickly.
- Use your approved SOP headings without removing provenance or uncertainty fields.

#### Guardrails

- Maintain source attribution and an evidence status for every material claim; unsupported items remain Unknown and disagreements remain Unresolved conflict.
- Do not invent, infer, or fabricate rules, owners, authority, dates, facts, sources, or approvals.
- Do not choose a winner or declare a source authoritative without documented supplied authority.
- A named human owner must approve publication before action.
- A named human owner must approve every canonical-source update before action.
- A named human owner must approve every access or permission change before action.
- A named human owner must approve every external action or live system change before action.
- A named human owner must approve every policy exception before action.
- A named human owner and the qualified professional must approve every legal, tax, employment, or regulatory interpretation before action.
- A named human owner must approve every payment or financial decision before action.
- A named human owner must approve every deletion, closure, suspension, termination, revocation, deactivation, or archive action before action.
- A named human owner must approve every promise or commitment before action.
- Use only authorized knowledge data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, preserve every sensitivity label or sensitivity classification, and follow company retention and vendor policy.
- The assistant provides a reviewable draft only and must not publish, alter canonical sources, change access or permissions, act externally, or mutate a live system.

#### Example input

Fictional example — INTERVIEW TRANSCRIPT: Maya explains how Northstar prepares a new-client workspace and says Finance, not Delivery, confirms billing status; she is unsure who approves an exception. SOURCE OWNER: Maya Chen, Delivery Lead. SYSTEM OR LOCATION: Operations handbook draft folder. SENSITIVITY: Internal; customer identifiers removed. AUTHORITY: Maya owns delivery steps but not finance policy. LAST-REVIEWED DATE: Unknown. SOP TEMPLATE: Purpose, scope, prerequisites, steps, exceptions, evidence, owner approval.

#### Related prompts

- https://returnmytime.com/prompt-library/create-company-knowledge-source-inventory
- https://returnmytime.com/prompt-library/identify-conflicting-business-rules

#### Related articles

- https://returnmytime.com/blog/company-knowledge-base-for-ai
- https://returnmytime.com/blog/ai-knowledge-base-maintenance

### Create a Company Knowledge-Source Inventory

- URL: https://returnmytime.com/prompt-library/create-company-knowledge-source-inventory
- Category: Knowledge Systems
- Workflow: company-knowledge-base-sop-capture
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: knowledge-base
- Type: Efficiency
- Estimated time saved: 60-90 minutes per source inventory
- Tags: knowledge-inventory, source-of-truth, governance
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Organize supplied company sources into a traceable inventory of ownership, location, sensitivity, authority, freshness, and evidence gaps.

#### What it does

- Produces a reviewable catalog of supplied knowledge sources and their documented governance metadata.
- Flags missing owners, authority, dates, duplicate subjects, sensitive material, and possible conflicts without resolving them by guesswork.

#### When to use

- Before an AI, automation, onboarding, or documentation project that depends on knowing where approved company knowledge lives.

#### When not to use

- Do not use it to crawl unauthorized systems, change access, retire a source, update canonical records, or publish an inventory.
- Do not place unauthorized company data or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- KNOWLEDGE SOURCES
- SOURCE OWNERS
- SYSTEMS OR LOCATIONS
- SENSITIVITY
- AUTHORITY
- LAST-REVIEWED DATES

#### Prompt

#CONTEXT:
Create a review-only company knowledge-source inventory from supplied records. Catalog what is provided; do not discover, access, move, or alter sources.

#INPUTS:
- Supplied source names, descriptions, and sample references: [KNOWLEDGE SOURCES]
- Documented source owners and stewards: [SOURCE OWNERS]
- Approved systems or locations: [SYSTEMS OR LOCATIONS]
- Sensitivity labels or classifications: [SENSITIVITY]
- Documented authority and source hierarchy: [AUTHORITY]
- Last-reviewed dates and review evidence: [LAST-REVIEWED DATES]

#INSTRUCTIONS:
1. Create one inventory row per supplied source. Retain its exact source name and location reference.
2. Record source attribution, owner, system or location, sensitivity, subject coverage, documented authority, last-reviewed date, and evidence status.
3. Mark missing facts Unknown. Mark competing claims or hierarchy evidence Unresolved conflict and cite each supplied source.
4. Identify duplicates and likely overlap as review flags only. Do not merge, delete, archive, rank, or declare a canonical winner.
5. Propose a named-owner verification queue ordered by documented risk signals: sensitive data, absent authority, unresolved conflict, missing owner, and missing review evidence.
6. State what data was excluded or minimized and list any additional access that a human might evaluate separately. Do not request or change access.
7. Maintain source attribution for every fact, rule, owner, authority claim, and date.
8. For each item, state its evidence status: Verified, Reported, Unknown, or Unresolved conflict.
9. Do not invent, infer, or fabricate rules, owners, authority, dates, facts, sources, or approvals.
10. Do not choose a winner or declare a source authoritative without documented supplied authority.
11. Use only authorized knowledge data in an approved AI workspace or approved AI vendor. Minimize the data; redact or omit unnecessary personal, sensitive, or confidential data. Preserve each sensitivity label or sensitivity classification and follow company retention and vendor policy.
12. Do not publish. Do not update a canonical source. Do not grant, revoke, or change access or permissions. Do not send material externally. Do not change a live system.
13. A named human owner must approve publication before action. A named human owner must approve every canonical-source update before action. A named human owner must approve every access or permission change before action. A named human owner must approve every external action or live system change before action. A named human owner must approve every policy exception before action.

#RESPONSE FORMAT:
## Knowledge-source inventory
| Source | Source attribution | Owner | System or location | Sensitivity | Coverage | Authority | Last reviewed | Evidence status |
|---|---|---|---|---|---|---|---|---|

## Unknowns and Unresolved conflicts
| Item | Supplied evidence | Missing or competing fact | Named reviewer |
|---|---|---|---|

## Human verification queue
| Priority signal | Proposed review | Named owner | Approval needed |
|---|---|---|---|

## Data-minimization note

#### Expected output

- A reviewable company knowledge-source inventory with source attribution, ownership, sensitivity, authority, freshness, evidence states, Unknowns, and Unresolved conflicts.

#### Customization tips

- Add stable source IDs so renamed documents remain traceable across reviews.
- Use sensitivity labels already approved by your security or operations owner.

#### Guardrails

- Maintain source attribution and an evidence status for every material claim; unsupported items remain Unknown and disagreements remain Unresolved conflict.
- Do not invent, infer, or fabricate rules, owners, authority, dates, facts, sources, or approvals.
- Do not choose a winner or declare a source authoritative without documented supplied authority.
- A named human owner must approve publication before action.
- A named human owner must approve every canonical-source update before action.
- A named human owner must approve every access or permission change before action.
- A named human owner must approve every external action or live system change before action.
- A named human owner must approve every policy exception before action.
- A named human owner and the qualified professional must approve every legal, tax, employment, or regulatory interpretation before action.
- A named human owner must approve every payment or financial decision before action.
- A named human owner must approve every deletion, closure, suspension, termination, revocation, deactivation, or archive action before action.
- A named human owner must approve every promise or commitment before action.
- Use only authorized knowledge data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, preserve every sensitivity label or sensitivity classification, and follow company retention and vendor policy.
- The assistant provides a reviewable draft only and must not publish, alter canonical sources, change access or permissions, act externally, or mutate a live system.

#### Example input

Fictional example — KNOWLEDGE SOURCES: Northstar sales playbook, delivery checklist, pricing FAQ, and two onboarding notes. SOURCE OWNERS: Sales playbook—Ari; delivery checklist—Maya; other owners Unknown. SYSTEMS OR LOCATIONS: Approved internal drive and CRM knowledge folder. SENSITIVITY: Internal; pricing FAQ is restricted. AUTHORITY: Signed order form overrides the pricing FAQ for that customer; broader hierarchy Unknown. LAST-REVIEWED DATES: Sales playbook 2026-06-15; remaining dates Unknown.

#### Related prompts

- https://returnmytime.com/prompt-library/turn-expert-interview-into-sop
- https://returnmytime.com/prompt-library/create-knowledge-base-maintenance-plan

#### Related articles

- https://returnmytime.com/blog/company-knowledge-base-for-ai
- https://returnmytime.com/blog/ai-knowledge-base-maintenance

### Identify Conflicting Business Rules

- URL: https://returnmytime.com/prompt-library/identify-conflicting-business-rules
- Category: Knowledge Systems
- Workflow: company-knowledge-base-sop-capture
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: knowledge-base
- Type: Quality
- Estimated time saved: 30-45 minutes per rule review
- Tags: business-rules, conflict-register, source-authority
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Compare supplied business rules, expose contradictions and scope gaps, and prepare an evidence-based conflict register without guessing which rule wins.

#### What it does

- Finds direct contradictions, scope overlaps, version mismatches, and authority gaps in supplied rules.
- Builds an unresolved conflict register with source attribution and questions for the documented decision owner.

#### When to use

- When teams, documents, or systems appear to apply different rules to the same workflow or decision.

#### When not to use

- Do not use it to decide a policy winner, publish a rule, replace a canonical source, change access, or operate a live workflow.
- Do not place unauthorized source data or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- BUSINESS RULES
- SOURCE OWNERS
- SYSTEMS OR LOCATIONS
- SENSITIVITY
- AUTHORITY
- LAST-REVIEWED DATES
- CONFLICT CONTEXT

#### Prompt

#CONTEXT:
Identify conflicts among supplied business rules. Create decision support for human reconciliation; do not select, rewrite, publish, or enforce a rule.

#INPUTS:
- Exact rule text with source references: [BUSINESS RULES]
- Documented source owners: [SOURCE OWNERS]
- Systems or locations containing each rule: [SYSTEMS OR LOCATIONS]
- Sensitivity labels or classifications: [SENSITIVITY]
- Supplied authority, precedence, scope, and decision rights: [AUTHORITY]
- Last-reviewed dates and version evidence: [LAST-REVIEWED DATES]
- Workflow, event, or case in which conflict appears: [CONFLICT CONTEXT]

#INSTRUCTIONS:
1. Normalize formatting only. Quote or faithfully paraphrase each rule and retain source attribution.
2. Compare rule trigger, scope, audience, exception, outcome, effective date, source owner, and documented authority.
3. Classify only evidenced conflict types: direct contradiction, overlapping scope, version mismatch, missing exception, or authority gap.
4. For every finding, state an evidence status. Use Unknown for missing facts and Unresolved conflict whenever evidence does not establish an authorized winner.
5. Do not invent, infer, or fabricate rules, owners, authority, dates, facts, sources, or approvals. Do not choose a winner or declare a rule authoritative without documented supplied authority.
6. Prepare neutral questions for the named decision owner. Recommend what evidence would resolve each conflict without recommending the substantive decision.
7. Describe downstream workflows or documents that a human should inspect after resolution. Do not update or enforce them.
8. Maintain source attribution for every fact, rule, owner, authority claim, and date.
9. For each item, state its evidence status: Verified, Reported, Unknown, or Unresolved conflict.
10. Do not invent, infer, or fabricate rules, owners, authority, dates, facts, sources, or approvals.
11. Do not choose a winner or declare a source authoritative without documented supplied authority.
12. Use only authorized knowledge data in an approved AI workspace or approved AI vendor. Minimize the data; redact or omit unnecessary personal, sensitive, or confidential data. Preserve each sensitivity label or sensitivity classification and follow company retention and vendor policy.
13. Do not publish. Do not update a canonical source. Do not grant, revoke, or change access or permissions. Do not send material externally. Do not change a live system.
14. A named human owner must approve publication before action. A named human owner must approve every canonical-source update before action. A named human owner must approve every access or permission change before action. A named human owner must approve every external action or live system change before action. A named human owner must approve every policy exception before action.

#RESPONSE FORMAT:
## Rule evidence table
| Rule | Source attribution | Owner | Scope | Authority | Last reviewed | Evidence status |
|---|---|---|---|---|---|---|

## Unresolved conflict register
| Conflict | Type | Competing evidence | Operational exposure | Named decision owner | Evidence needed |
|---|---|---|---|---|---|

## Unknowns
- Missing fact — why it matters — proposed human question

## Post-resolution review map
- Artifact or workflow — reason to inspect — approval boundary

#### Expected output

- An evidence-based Unresolved conflict register with source attribution, rule scope, authority gaps, operational exposure, and neutral questions for named human owners.

#### Customization tips

- Include effective dates and supersession markers when your policy system records them.
- Keep legal, tax, employment, or regulatory interpretations with the qualified professional who owns them.

#### Guardrails

- Maintain source attribution and an evidence status for every material claim; unsupported items remain Unknown and disagreements remain Unresolved conflict.
- Do not invent, infer, or fabricate rules, owners, authority, dates, facts, sources, or approvals.
- Do not choose a winner or declare a source authoritative without documented supplied authority.
- A named human owner must approve publication before action.
- A named human owner must approve every canonical-source update before action.
- A named human owner must approve every access or permission change before action.
- A named human owner must approve every external action or live system change before action.
- A named human owner must approve every policy exception before action.
- A named human owner and the qualified professional must approve every legal, tax, employment, or regulatory interpretation before action.
- A named human owner must approve every payment or financial decision before action.
- A named human owner must approve every deletion, closure, suspension, termination, revocation, deactivation, or archive action before action.
- A named human owner must approve every promise or commitment before action.
- Use only authorized knowledge data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, preserve every sensitivity label or sensitivity classification, and follow company retention and vendor policy.
- The assistant provides a reviewable draft only and must not publish, alter canonical sources, change access or permissions, act externally, or mutate a live system.

#### Example input

Fictional example — BUSINESS RULES: Northstar sales handbook says deposits are due at booking; delivery checklist says work may start after a signed order form. SOURCE OWNERS: Sales—Ari; Delivery—Maya. SYSTEMS OR LOCATIONS: Approved sales wiki and delivery handbook. SENSITIVITY: Internal. AUTHORITY: Neither document states precedence; Finance authority is Unknown. LAST-REVIEWED DATES: Sales 2026-05-01; Delivery 2026-06-10. CONFLICT CONTEXT: A coordinator is preparing a kickoff checklist for a fictional customer.

#### Related prompts

- https://returnmytime.com/prompt-library/turn-expert-interview-into-sop
- https://returnmytime.com/prompt-library/create-knowledge-base-maintenance-plan

#### Related articles

- https://returnmytime.com/blog/company-knowledge-base-for-ai
- https://returnmytime.com/blog/ai-knowledge-base-maintenance

### Create a Knowledge-Base Maintenance Plan

- URL: https://returnmytime.com/prompt-library/create-knowledge-base-maintenance-plan
- Category: Knowledge Systems
- Workflow: company-knowledge-base-sop-capture
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: knowledge-base
- Type: Effectiveness
- Estimated time saved: 45-60 minutes per maintenance plan
- Tags: knowledge-maintenance, change-control, source-ownership
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Turn a supplied source inventory and maintenance responsibilities into an owned review cadence and change-control plan.

#### What it does

- Proposes evidence-based review cadences, named responsibilities, and change-control checkpoints from supplied governance facts.
- Keeps missing ownership, dates, authority, and conflicting rules visible until a named human resolves them.

#### When to use

- After source inventory, when the team needs a reviewable operating plan for keeping approved knowledge current and traceable.

#### When not to use

- Do not use it to assign an unapproved owner, publish content, edit a canonical source, change permissions, archive material, or schedule live work.
- Do not place unauthorized knowledge data or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- KNOWLEDGE SOURCE INVENTORY
- MAINTENANCE ROLES
- SOURCE OWNERS
- SENSITIVITY
- AUTHORITY
- LAST-REVIEWED DATES
- CHANGE-CONTROL REQUIREMENTS

#### Prompt

#CONTEXT:
Create a reviewable knowledge-base maintenance plan from the supplied inventory and governance facts. Propose controls; do not schedule work or change any source or system.

#INPUTS:
- Current knowledge-source inventory with source references: [KNOWLEDGE SOURCE INVENTORY]
- Approved maintenance roles and responsibilities: [MAINTENANCE ROLES]
- Documented source owners and stewards: [SOURCE OWNERS]
- Sensitivity labels or classifications: [SENSITIVITY]
- Supplied authority and approval rights: [AUTHORITY]
- Last-reviewed dates and supporting evidence: [LAST-REVIEWED DATES]
- Required change-control, versioning, review, and retirement controls: [CHANGE-CONTROL REQUIREMENTS]

#INSTRUCTIONS:
1. Preserve source attribution and state an evidence status for every ownership, authority, freshness, and cadence input.
2. Do not invent, infer, or fabricate rules, owners, authority, dates, facts, sources, approvals, or review frequency.
3. Propose a cadence only when supported by supplied policy, risk tier, or owner instruction. Otherwise mark the cadence Unknown and request a named-human decision.
4. Build an owner, cadence, and change-control plan covering intake, review, approval, versioning, conflict escalation, access review, retirement review, and audit evidence.
5. Keep conflicting ownership, authority, or source rules in an Unresolved conflict state. Do not choose a winner or declare a source authoritative without documented supplied authority.
6. Add event-driven review triggers for supplied business changes, incidents, repeated unanswered questions, and owner transitions. Label unsupported triggers as proposals.
7. Define evidence of completion and a review scorecard. Do not schedule reviews, notify owners, publish changes, archive records, or edit live systems.
8. Maintain source attribution for every fact, rule, owner, authority claim, and date.
9. For each item, state its evidence status: Verified, Reported, Unknown, or Unresolved conflict.
10. Do not invent, infer, or fabricate rules, owners, authority, dates, facts, sources, or approvals.
11. Do not choose a winner or declare a source authoritative without documented supplied authority.
12. Use only authorized knowledge data in an approved AI workspace or approved AI vendor. Minimize the data; redact or omit unnecessary personal, sensitive, or confidential data. Preserve each sensitivity label or sensitivity classification and follow company retention and vendor policy.
13. Do not publish. Do not update a canonical source. Do not grant, revoke, or change access or permissions. Do not send material externally. Do not change a live system.
14. A named human owner must approve publication before action. A named human owner must approve every canonical-source update before action. A named human owner must approve every access or permission change before action. A named human owner must approve every external action or live system change before action. A named human owner must approve every policy exception before action.

#RESPONSE FORMAT:
## Source attribution and evidence status
| Source | Owner | Authority | Sensitivity | Last reviewed | Evidence status |
|---|---|---|---|---|---|

## Owner, cadence, and change-control plan
| Source or domain | Named owner | Proposed cadence | Evidence basis | Change-control steps | Completion evidence |
|---|---|---|---|---|---|

## Event-driven review triggers
| Trigger | Affected source | Proposed reviewer | Approval boundary |
|---|---|---|---|

## Unknowns and Unresolved conflicts
| Item | Evidence gap or conflict | Named decision owner | Stop condition |
|---|---|---|---|

## Human approval checklist

#### Expected output

- A reviewable owner, cadence, and change-control plan with source attribution, evidence states, review triggers, Unknowns, Unresolved conflicts, and human approvals.

#### Customization tips

- Map the plan to existing operating meetings before proposing new coordination overhead.
- Add retrieval-quality signals only when their measurement method and owner are documented.

#### Guardrails

- Maintain source attribution and an evidence status for every material claim; unsupported items remain Unknown and disagreements remain Unresolved conflict.
- Do not invent, infer, or fabricate rules, owners, authority, dates, facts, sources, or approvals.
- Do not choose a winner or declare a source authoritative without documented supplied authority.
- A named human owner must approve publication before action.
- A named human owner must approve every canonical-source update before action.
- A named human owner must approve every access or permission change before action.
- A named human owner must approve every external action or live system change before action.
- A named human owner must approve every policy exception before action.
- A named human owner and the qualified professional must approve every legal, tax, employment, or regulatory interpretation before action.
- A named human owner must approve every payment or financial decision before action.
- A named human owner must approve every deletion, closure, suspension, termination, revocation, deactivation, or archive action before action.
- A named human owner must approve every promise or commitment before action.
- Use only authorized knowledge data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, preserve every sensitivity label or sensitivity classification, and follow company retention and vendor policy.
- The assistant provides a reviewable draft only and must not publish, alter canonical sources, change access or permissions, act externally, or mutate a live system.

#### Example input

Fictional example — KNOWLEDGE SOURCE INVENTORY: Northstar sales playbook, delivery checklist, and pricing FAQ with internal source links. MAINTENANCE ROLES: Operations proposes quarterly review coordination; final role approval pending. SOURCE OWNERS: Ari owns sales; Maya owns delivery; pricing owner Unknown. SENSITIVITY: Internal; pricing restricted. AUTHORITY: Owners approve content in their domains; retirement authority Unknown. LAST-REVIEWED DATES: Sales 2026-06-15; delivery 2026-05-20; pricing Unknown. CHANGE-CONTROL REQUIREMENTS: Version history, reviewer identity, approval record, conflict log, and rollback reference.

#### Related prompts

- https://returnmytime.com/prompt-library/create-company-knowledge-source-inventory
- https://returnmytime.com/prompt-library/identify-conflicting-business-rules

#### Related articles

- https://returnmytime.com/blog/company-knowledge-base-for-ai
- https://returnmytime.com/blog/ai-knowledge-base-maintenance

### Audit a Workflow for AI Automation

- URL: https://returnmytime.com/prompt-library/audit-workflow-for-ai-automation
- Category: Knowledge Systems
- Workflow: ai-readiness-workflow-assessment
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Effectiveness
- Estimated time saved: Varies by workflow scope and evidence quality
- Tags: AI-readiness, workflow-audit, automation-fit
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Assess one service workflow for bounded AI assistance using supplied operating evidence, transparent calculations, and explicit readiness dependencies.

#### What it does

- Maps supplied workflow evidence to bounded assist, automate-later, and keep-human recommendations.
- Makes assumptions, calculation methods, uncertainty, privacy, security, access, and human approval dependencies inspectable.

#### When to use

- Before selecting or buying an AI solution for a recurring workflow whose steps, volume, handling time, exceptions, and owner can be documented.

#### When not to use

- Do not use it to authorize a priority, purchase, deployment, data-access change, permission change, or live workflow change.
- Do not place unauthorized workflow data or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- WORKFLOW STEPS
- WORKFLOW VOLUME
- TIME PER STEP
- ERROR AND EXCEPTION DATA
- BUSINESS VALUE
- DATA SOURCES AND ACCESS
- CONSTRAINTS
- WORKFLOW OWNERS
- SUCCESS MEASURES

#### Prompt

#CONTEXT:
Create an automation-fit audit for one service workflow. This is decision support for a named human owner, not authorization to automate or change the workflow.

#INPUTS:
- Current workflow steps, decisions, handoffs, and stop conditions: [WORKFLOW STEPS]
- Observed case volume with period and source: [WORKFLOW VOLUME]
- Observed handling time per step with measurement basis: [TIME PER STEP]
- Error, rework, exception, and escalation evidence: [ERROR AND EXCEPTION DATA]
- Supplied customer, revenue, capacity, or quality value: [BUSINESS VALUE]
- Data sources, quality notes, access rights, and permission owners: [DATA SOURCES AND ACCESS]
- Operational, budget, technical, regulatory, and timing constraints: [CONSTRAINTS]
- Named workflow owners and documented decision rights: [WORKFLOW OWNERS]
- Supplied success measures, baselines, thresholds, and measurement owners: [SUCCESS MEASURES]

#INSTRUCTIONS:
1. Map each supplied workflow step, decision, handoff, exception, owner, and evidence source. Label missing facts Unknown.
2. For each step, assess standardization, judgment, exception rate, error consequence, data availability, privacy, security, data-access dependency, reversibility, and required human oversight.
3. Classify each recommendation as Assist now, Test with controls, Defer until dependency is resolved, or Keep human. Explain the evidence or source basis.
4. Distinguish every value as Supplied, Calculated, or Estimated. For a Calculated value, show the formula, units, and supplied operands. For an Estimated value, show a range, assumptions, uncertainty, and a confidence score with rationale.
5. Do not invent, infer, or fabricate ROI or return on investment, time savings, cost, feasibility, volumes, error rates, value, baselines, owners, access, or technical capability. Unsupported conclusions remain Unknown.
6. Treat time-savings calculations as scenarios, not forecasts. Example formula: supplied volume per period × supplied time per case × proposed bounded reduction. Show the resulting range and do not convert it to currency without supplied labor-cost evidence.
7. Create a privacy, security, and data-access dependency register. State the authorized source, approved purpose, current access, permission owner, retention constraint, required control, and stop condition.
8. Define a bounded validation test for any Assist now or Test with controls recommendation. Include baseline, sample, success measure, failure threshold, human review, rollback, and evidence to retain.
9. Do not infer authorization from a supplied owner, access record, priority, or recommendation. Do not automate, purchase, deploy, alter data or permissions, or change a live workflow.
10. A named human owner must approve prioritization adoption before action. A named human owner must approve purchasing before action. A named human owner must approve deployment before action. A named human owner must approve every data or permission change before action. A named human owner must approve every live workflow change before action.
11. Use only authorized workflow data in an approved AI workspace or approved AI vendor. Minimize the data; redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.

#RESPONSE FORMAT:
## Evidence and source basis
| Claim or input | State: Supplied, Calculated, Estimated, or Unknown | Source | Formula or assumptions | Uncertainty | Confidence score |
|---|---|---|---|---|---|

## Automation-fit audit
| Workflow step | Fit classification | Evidence basis | Human judgment | Exceptions and risk | Dependency | Stop condition |
|---|---|---|---|---|---|---|

## Privacy, security, and data-access dependencies
| Dependency | Authorized source and purpose | Permission owner | Required control | Status | Stop condition |
|---|---|---|---|---|---|

## Scenario calculations
| Scenario | Formula | Supplied operands | Estimated range | Assumptions | Uncertainty and confidence score |
|---|---|---|---|---|---|

## Bounded validation plan and human approvals

#### Expected output

- An evidence-based automation-fit audit with step-level classifications, scenario formulas and ranges, assumptions, uncertainty and confidence scores, Unknowns, readiness dependencies, a bounded validation plan, and named-human approval gates.

#### Customization tips

- Replace broad averages with timestamped step observations before comparing options.
- Keep customer-facing, financial, legal, employment, and safety-critical decisions human-owned unless separately governed.

#### Guardrails

- Preserve the evidence or source basis for every material claim; unsupported facts remain Unknown.
- Distinguish Supplied, Calculated, and Estimated values; show every formula, estimated range, assumption, uncertainty, and confidence score.
- Do not invent, infer, or fabricate ROI or return on investment, time savings, cost, feasibility, volume, error rate, value, owner, access, or capability.
- Make every privacy, security, and data-access dependency explicit before recommending a test.
- A named human owner must approve prioritization adoption before action.
- A named human owner must approve purchasing before action.
- A named human owner must approve deployment before action.
- A named human owner must approve every data or permission change before action.
- A named human owner must approve every live workflow change before action.
- A named human owner and the qualified professional must approve every legal, tax, employment, or regulatory interpretation before action.
- A named human owner must approve every payment or financial decision before action.
- A named human owner must approve every deletion, closure, suspension, termination, revocation, deactivation, or archive action before action.
- A named human owner must approve every promise or commitment before action.
- A named human owner must approve every policy exception before action.
- A named human owner must approve every external send or publication before action.
- Use only authorized workflow data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- Do not infer authorization. Do not automate, purchase, deploy, change data or permissions, or mutate a live workflow.

#### Example input

Fictional example — WORKFLOW STEPS: Five documented intake stages. WORKFLOW VOLUME: 180 per month, Q2 CRM report. TIME PER STEP: Seven-minute median, ten-case sample; step-level timing Unknown. ERROR AND EXCEPTION DATA: 14% manager-review rate; five missing-location rework cases. BUSINESS VALUE: Response speed; revenue attribution Unknown. DATA SOURCES AND ACCESS: Authorized de-identified CRM field sample; customer message text excluded; Maya, access owner. CONSTRAINTS: $2,000 test ceiling; customer-facing automation prohibited. WORKFLOW OWNERS: Maya, operations; Omar, security. SUCCESS MEASURES: Missing-field rework rate; baseline and target Unknown.

#### Related prompts

- https://returnmytime.com/prompt-library/identify-ai-data-readiness-gaps
- https://returnmytime.com/prompt-library/rank-ai-use-cases-by-business-impact

#### Related articles

- https://returnmytime.com/blog/ai-assessment-before-build

### Rank AI Use Cases by Business Impact

- URL: https://returnmytime.com/prompt-library/rank-ai-use-cases-by-business-impact
- Category: Knowledge Systems
- Workflow: ai-readiness-workflow-assessment
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Effectiveness
- Estimated time saved: Varies by candidate count and evidence quality
- Tags: AI-use-cases, prioritization, business-impact
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Compare candidate AI use cases through a transparent impact, effort, risk, and evidence model without turning estimates into promises.

#### What it does

- Applies one visible scoring method to supplied impact, effort, risk, readiness, and ownership evidence.
- Separates documented facts from calculations and estimates so a named human can challenge the ranking.

#### When to use

- When a service business has several candidate use cases and needs an evidence-based shortlist for further validation.

#### When not to use

- Do not use it as approval to adopt a priority, buy software, deploy a system, change data access, or modify a live workflow.
- Do not place unauthorized business data or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- CANDIDATE USE CASES
- DECISION CRITERIA AND WEIGHTS
- BUSINESS VALUE
- EFFORT EVIDENCE
- RISK EVIDENCE
- DATA SOURCES AND ACCESS
- CONSTRAINTS
- USE-CASE OWNERS
- SUCCESS MEASURES

#### Prompt

#CONTEXT:
Create a transparent impact, effort, and risk ranking for supplied AI use cases. The ranking is a reviewable hypothesis, not authorization or a guaranteed business case.

#INPUTS:
- Candidate use cases with workflow boundaries and intended users: [CANDIDATE USE CASES]
- Human-supplied decision criteria, definitions, required constraints, and weights: [DECISION CRITERIA AND WEIGHTS]
- Supplied customer, revenue, capacity, speed, or quality value evidence: [BUSINESS VALUE]
- Supplied implementation, integration, change, and operating-effort evidence: [EFFORT EVIDENCE]
- Supplied privacy, security, legal, operational, quality, and failure-impact evidence: [RISK EVIDENCE]
- Data sources, quality notes, access rights, and permission owners: [DATA SOURCES AND ACCESS]
- Budget, timing, capacity, technical, policy, and regulatory constraints: [CONSTRAINTS]
- Named use-case owners and documented decision rights: [USE-CASE OWNERS]
- Supplied success measures, baselines, thresholds, and measurement owners: [SUCCESS MEASURES]

#INSTRUCTIONS:
1. Normalize the candidate descriptions without adding capabilities, value, readiness, or owners that were not supplied. Missing facts remain Unknown.
2. Use decision criteria and weights as Supplied only when the named human provided them. If they are absent or incomplete, label them Unknown and return scenario rankings only; do not create a canonical rank.
3. If useful, propose alternative criteria or weights as Estimated, never Supplied. State the assumption and business-preference implication behind each, use ranges, and perform sensitivity and scenario analysis. Do not silently choose a business preference.
4. Distinguish every value as Supplied, Calculated, or Estimated. Show each Calculated score's formula, weights, units, and operands. For every Estimated value, show a range, assumptions, uncertainty, and confidence score with rationale.
5. Produce the ranking, then show sensitivity ranges: how positions change when uncertain inputs or weights move within stated ranges. Ties and evidence-poor comparisons remain tied or Unknown.
6. Give the evidence or source basis for each impact, effort, risk, readiness, and success-measure claim.
7. Do not invent, infer, or fabricate ROI or return on investment, time savings, cost, feasibility, revenue, adoption, technical capability, risk, data quality, owner, permission, business preference, or business priority. Do not convert relative scores into financial forecasts.
8. List privacy, security, and data-access dependencies and identify any dependency that prevents responsible validation.
9. Recommend only a next evidence-gathering step or bounded validation candidate. Do not present the ranking as an implementation order unless dependencies are separately assessed.
10. Do not infer authorization from a supplied owner, access record, score, rank, or recommendation. Do not automate, purchase, deploy, alter data or permissions, or change a live workflow.
11. A named human owner must approve the decision criteria, weights, and canonical rank before prioritization adoption. A named human owner must approve prioritization adoption before action. A named human owner must approve purchasing before action. A named human owner must approve deployment before action. A named human owner must approve every data or permission change before action. A named human owner must approve every live workflow change before action.
12. Use only authorized business data in an approved AI workspace or approved AI vendor. Minimize the data; redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.

#RESPONSE FORMAT:
## Evidence and source basis
| Use case | Claim | State: Supplied, Calculated, Estimated, or Unknown | Source | Formula or assumptions | Uncertainty | Confidence score |
|---|---|---|---|---|---|---|

## Scoring method
| Criterion | Definition | Weight | State: Supplied, Estimated, or Unknown | Business-preference implication | Range | Evidence threshold |
|---|---|---|---|---|---|---|

## Impact, effort, and risk ranking
| Rank or tie | Use case | Impact | Effort | Risk | Readiness | Evidence strength | Composite formula | Confidence score |
|---|---|---|---|---|---|---|---|---|

## Sensitivity and scenario analysis
| Scenario | Variable or assumption | Tested range | Ranking effect | Business preference represented | Evidence needed |
|---|---|---|---|---|---|

## Privacy, security, and data-access dependencies

## Human decisions and approvals

#### Expected output

- A transparent impact, effort, and risk ranking with formulas, weights, evidence strength, sensitivity ranges, assumptions, uncertainty and confidence scores, Unknowns, readiness dependencies, and human approval gates.

#### Customization tips

- Use a small number of decision-relevant criteria so weak evidence stays visible instead of disappearing into a composite score.
- Review the sensitivity table before treating small score differences as meaningful.

#### Guardrails

- Preserve the evidence or source basis for every material claim; unsupported facts remain Unknown.
- Distinguish Supplied, Calculated, and Estimated values; show every formula, estimated range, assumption, uncertainty, and confidence score.
- Do not invent, infer, or fabricate ROI or return on investment, time savings, cost, feasibility, revenue, adoption, capability, risk, data quality, owner, or permission.
- If decision criteria and weights are absent, label them Unknown and return scenario rankings only; model-proposed weights must be Estimated, never Supplied, with sensitivity and scenario analysis.
- Do not invent, infer, or fabricate a business preference or business priority.
- A named human owner must approve the decision criteria, weights, and canonical rank before prioritization adoption.
- Make every privacy, security, and data-access dependency explicit before recommending validation.
- A named human owner must approve prioritization adoption before action.
- A named human owner must approve purchasing before action.
- A named human owner must approve deployment before action.
- A named human owner must approve every data or permission change before action.
- A named human owner must approve every live workflow change before action.
- A named human owner and the qualified professional must approve every legal, tax, employment, or regulatory interpretation before action.
- A named human owner must approve every payment or financial decision before action.
- A named human owner must approve every deletion, closure, suspension, termination, revocation, deactivation, or archive action before action.
- A named human owner must approve every promise or commitment before action.
- A named human owner must approve every policy exception before action.
- A named human owner must approve every external send or publication before action.
- Use only authorized business data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- Do not infer authorization. Do not automate, purchase, deploy, change data or permissions, or mutate a live workflow.

#### Example input

Fictional example — CANDIDATE USE CASES: Inquiry summarization, missing-field detection, and draft follow-up suggestions for Northstar's intake team. DECISION CRITERIA AND WEIGHTS: Human-supplied criteria are quality 35%, staff time 25%, implementation effort 20%, privacy and security risk 20%; mandatory constraint: no customer-facing action. BUSINESS VALUE: Missing fields cause documented rework; no supplied revenue attribution. EFFORT EVIDENCE: CRM export exists; integration estimate not supplied. RISK EVIDENCE: Customer messages may contain personal data; suggestions require staff review. DATA SOURCES AND ACCESS: De-identified CRM sample approved for assessment; production access not approved. CONSTRAINTS: Six-week evaluation window and $3,000 research ceiling. USE-CASE OWNERS: Maya owns intake; Omar owns security; purchasing owner Unknown. SUCCESS MEASURES: Rework rate and reviewer acceptance; baselines incomplete.

#### Related prompts

- https://returnmytime.com/prompt-library/audit-workflow-for-ai-automation
- https://returnmytime.com/prompt-library/create-ai-implementation-sequence

#### Related articles

- https://returnmytime.com/blog/ai-assessment-before-build

### Identify AI Data-Readiness Gaps

- URL: https://returnmytime.com/prompt-library/identify-ai-data-readiness-gaps
- Category: Knowledge Systems
- Workflow: ai-readiness-workflow-assessment
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Quality
- Estimated time saved: Varies by source count and evidence quality
- Tags: data-readiness, AI-assessment, gap-analysis
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Turn supplied data-source, quality, access, privacy, security, and ownership evidence into an actionable readiness-gap register.

#### What it does

- Maps candidate-use-case needs to supplied sources, quality evidence, permissions, controls, owners, and dependencies.
- Keeps unsupported readiness claims Unknown and proposes bounded evidence-gathering work for human review.

#### When to use

- Before AI prototyping or procurement when the team needs to know whether required data is usable, authorized, controlled, and owned.

#### When not to use

- Do not use it to grant access, change permissions, move or delete data, buy a system, deploy a model, or modify a live workflow.
- Do not place unauthorized company data or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- CANDIDATE USE CASES
- DATA SOURCES
- DATA QUALITY EVIDENCE
- DATA ACCESS AND PERMISSIONS
- PRIVACY AND SECURITY REQUIREMENTS
- CONSTRAINTS
- DATA OWNERS
- DEPENDENCIES
- SUCCESS MEASURES

#### Prompt

#CONTEXT:
Create a readiness-gap register for the data needed by the supplied AI use cases. Assess documented evidence only; do not inspect, move, transform, or change live data or access.

#INPUTS:
- Candidate use cases with required inputs, outputs, and decision boundaries: [CANDIDATE USE CASES]
- Documented data sources, locations, formats, lineage, and refresh patterns: [DATA SOURCES]
- Supplied completeness, accuracy, consistency, timeliness, representativeness, and test evidence: [DATA QUALITY EVIDENCE]
- Current access rights, permission records, approved purposes, and access owners: [DATA ACCESS AND PERMISSIONS]
- Applicable privacy, security, classification, retention, and residency requirements: [PRIVACY AND SECURITY REQUIREMENTS]
- Budget, timing, capacity, technical, policy, and regulatory constraints: [CONSTRAINTS]
- Named data owners, stewards, and documented decision rights: [DATA OWNERS]
- Technical, governance, vendor, process, and people dependencies: [DEPENDENCIES]
- Supplied readiness and validation measures, thresholds, and measurement owners: [SUCCESS MEASURES]

#INSTRUCTIONS:
1. For each use case, map every required data element to its supplied source, owner, lineage, quality evidence, access status, approved purpose, privacy classification, security control, retention rule, and dependency.
2. Label missing, conflicting, expired, inaccessible, unmeasured, or unauthorized facts Unknown. Do not treat source existence as proof of quality, permission, suitability, or feasibility.
3. Create a readiness-gap register. Classify each gap as availability, quality, lineage, access, privacy, security, governance, ownership, integration, measurement, or dependency.
4. Give the evidence or source basis, consequence, named decision owner, proposed bounded evidence-gathering step, stop condition, and completion evidence for each gap.
5. Distinguish every value as Supplied, Calculated, or Estimated. For a Calculated value, show the formula, units, and supplied operands. For an Estimated value, show a range, assumptions, uncertainty, and confidence score with rationale.
6. Do not invent, infer, or fabricate ROI or return on investment, time savings, cost, feasibility, data quality, access, permission, consent, owner, lineage, control effectiveness, volume, coverage, or technical capability. Unsupported conclusions remain Unknown.
7. If readiness percentages are requested, use a documented formula such as verified requirements met ÷ total documented requirements × 100. Show counts, inclusion rules, range, assumptions, uncertainty, and confidence score; never imply that a score authorizes use.
8. Make privacy, security, and data-access dependencies explicit. Treat unresolved authorization, purpose limitation, sensitive-data handling, retention, or vendor-control questions as stop conditions.
9. Do not infer authorization from source availability, existing credentials, ownership, or a readiness recommendation. Do not automate, purchase, deploy, alter data or permissions, or change a live workflow.
10. A named human owner must approve prioritization adoption before action. A named human owner must approve purchasing before action. A named human owner must approve deployment before action. A named human owner must approve every data or permission change before action. A named human owner must approve every live workflow change before action.
11. Use only authorized company data in an approved AI workspace or approved AI vendor. Minimize the data; redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.

#RESPONSE FORMAT:
## Data requirement and evidence map
| Use case | Data requirement | State: Supplied, Calculated, Estimated, or Unknown | Source | Owner | Access | Quality evidence | Privacy and security dependency |
|---|---|---|---|---|---|---|---|

## Readiness-gap register
| Gap | Type | Evidence or source basis | Consequence | Named owner | Bounded next step | Stop condition | Completion evidence |
|---|---|---|---|---|---|---|---|

## Calculations and estimates
| Measure | Formula | Supplied operands | Estimated range | Assumptions | Uncertainty | Confidence score |
|---|---|---|---|---|---|---|

## Privacy, security, and data-access dependencies

## Unknowns and human approvals

#### Expected output

- A source-grounded readiness-gap register with data requirement mapping, formulas and ranges, assumptions, uncertainty and confidence scores, explicit privacy, security, and data-access dependencies, Unknowns, stop conditions, and named-human approvals.

#### Customization tips

- Assess one bounded use case at a time so a strong source for one purpose is not treated as ready for every purpose.
- Record the date and method behind each quality measure; labels without evidence stay Unknown.

#### Guardrails

- Preserve the evidence or source basis for every material claim; unsupported facts remain Unknown.
- Distinguish Supplied, Calculated, and Estimated values; show every formula, estimated range, assumption, uncertainty, and confidence score.
- Do not invent, infer, or fabricate ROI or return on investment, time savings, cost, feasibility, data quality, access, permission, consent, owner, lineage, control effectiveness, coverage, or capability.
- Make every privacy, security, and data-access dependency explicit and treat unresolved authorization as a stop condition.
- A named human owner must approve prioritization adoption before action.
- A named human owner must approve purchasing before action.
- A named human owner must approve deployment before action.
- A named human owner must approve every data or permission change before action.
- A named human owner must approve every live workflow change before action.
- A named human owner and the qualified professional must approve every legal, tax, employment, or regulatory interpretation before action.
- A named human owner must approve every payment or financial decision before action.
- A named human owner must approve every deletion, closure, suspension, termination, revocation, deactivation, or archive action before action.
- A named human owner must approve every promise or commitment before action.
- A named human owner must approve every policy exception before action.
- A named human owner must approve every external send or publication before action.
- Use only authorized company data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- Do not infer authorization. Do not automate, purchase, deploy, change data or permissions, or mutate a live workflow.

#### Example input

Fictional example — CANDIDATE USE CASES: Detect missing intake fields and draft an internal case summary. DATA SOURCES: Northstar CRM records and service catalog; lineage for two custom fields Unknown. DATA QUALITY EVIDENCE: 100-record sample shows 82 complete location fields; sampling method documented. DATA ACCESS AND PERMISSIONS: De-identified export approved for assessment; production write access not approved. PRIVACY AND SECURITY REQUIREMENTS: Customer identifiers restricted; 30-day assessment retention. CONSTRAINTS: No new integration during assessment. DATA OWNERS: Maya owns CRM process; field-definition owner Unknown; Omar owns security. DEPENDENCIES: Data dictionary and vendor security review. SUCCESS MEASURES: Verified field mapping and approved test dataset; thresholds pending.

#### Related prompts

- https://returnmytime.com/prompt-library/audit-workflow-for-ai-automation
- https://returnmytime.com/prompt-library/create-ai-implementation-sequence

#### Related articles

- https://returnmytime.com/blog/ai-assessment-before-build

### Create an AI Implementation Sequence

- URL: https://returnmytime.com/prompt-library/create-ai-implementation-sequence
- Category: Knowledge Systems
- Workflow: ai-readiness-workflow-assessment
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: assessment
- Type: Effectiveness
- Estimated time saved: Varies by dependency count and evidence quality
- Tags: AI-roadmap, implementation-sequence, readiness-gates
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Convert an approved shortlist and documented readiness evidence into a gated, dependency-aware implementation sequence for human review.

#### What it does

- Sequences evidence gathering, control design, bounded validation, and human decision gates before implementation work.
- Shows which dependencies, assumptions, uncertainties, and approvals prevent a candidate from advancing.

#### When to use

- After use-case ranking and data-readiness review, when named owners need a proposed order of validation and decision gates.

#### When not to use

- Do not use it as authorization to adopt priorities, purchase tools, deploy systems, change data or permissions, or modify a live workflow.
- Do not place unauthorized business data or unnecessary personal, sensitive, or confidential data in an AI system.

#### Required inputs

- PRIORITIZED USE CASES
- READINESS GAPS
- WORKFLOW DEPENDENCIES
- DATA AND ACCESS DEPENDENCIES
- PRIVACY AND SECURITY REQUIREMENTS
- CONSTRAINTS
- IMPLEMENTATION OWNERS
- SUCCESS MEASURES

#### Prompt

#CONTEXT:
Create a gated implementation sequence from supplied prioritization and readiness evidence. This is a proposed decision path, not permission to purchase, deploy, automate, or alter a live workflow.

#INPUTS:
- Human-reviewed candidate priorities with scores, evidence, assumptions, and confidence: [PRIORITIZED USE CASES]
- Open data, knowledge, governance, ownership, measurement, and technical gaps: [READINESS GAPS]
- Workflow prerequisites, handoffs, exception paths, and change dependencies: [WORKFLOW DEPENDENCIES]
- Required sources, quality, access, permissions, integrations, and data owners: [DATA AND ACCESS DEPENDENCIES]
- Applicable privacy, security, classification, retention, residency, and vendor requirements: [PRIVACY AND SECURITY REQUIREMENTS]
- Budget, timing, capacity, policy, regulatory, and technical constraints: [CONSTRAINTS]
- Named implementation, workflow, data, security, purchasing, and decision owners: [IMPLEMENTATION OWNERS]
- Supplied baselines, success measures, failure thresholds, and measurement owners: [SUCCESS MEASURES]

#INSTRUCTIONS:
1. Verify that each proposed item has a supplied priority, evidence or source basis, named owner, readiness status, dependencies, success measure, failure threshold, and decision authority. Missing elements remain Unknown and block advancement.
2. Build phases for evidence completion, control design, offline or sandbox validation, limited human-reviewed test, evaluation, and only then a separate deployment decision. Do not imply that passing one gate authorizes the next.
3. For each gate, state entry evidence, work proposed for human approval, exit evidence, stop condition, named decision owner, and explicit approval required.
4. Distinguish every value as Supplied, Calculated, or Estimated. For a Calculated value, show the formula, units, and supplied operands. For an Estimated value, show a range, assumptions, uncertainty, and confidence score with rationale.
5. Do not invent, infer, or fabricate ROI or return on investment, time savings, cost, feasibility, duration, staffing, technical capability, priority, owner, dependency status, access, permission, baseline, outcome, or success probability. Unsupported conclusions remain Unknown.
6. Use ranges for estimated effort or duration. Show formulas and dependencies; do not convert relative scores into dates, budgets, or commitments without supplied evidence.
7. Place privacy, security, and data-access dependencies before any phase that would use data. Unresolved purpose, authorization, sensitive-data handling, retention, vendor, or access questions are stop conditions.
8. Include a rollback and incident-escalation design proposal for every bounded test. Do not activate it or mutate any live system.
9. Do not infer authorization from a supplied priority, owner, budget, credential, prior approval, or completed gate. Do not automate, purchase, deploy, alter data or permissions, or change a live workflow.
10. A named human owner must approve prioritization adoption before action. A named human owner must approve purchasing before action. A named human owner must approve deployment before action. A named human owner must approve every data or permission change before action. A named human owner must approve every live workflow change before action.
11. Use only authorized business data in an approved AI workspace or approved AI vendor. Minimize the data; redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.

#RESPONSE FORMAT:
## Evidence and source basis
| Use case or dependency | State: Supplied, Calculated, Estimated, or Unknown | Source | Formula or assumptions | Range | Uncertainty | Confidence score |
|---|---|---|---|---|---|---|

## Gated implementation sequence
| Phase and gate | Entry evidence | Proposed work | Owner | Privacy, security, and data-access dependency | Exit evidence | Stop condition | Human approval |
|---|---|---|---|---|---|---|---|

## Dependency path and Unknowns
| Dependency | Required before | Status | Evidence gap | Named decision owner |
|---|---|---|---|---|

## Estimate formulas and ranges
| Estimate | Formula | Supplied operands | Range | Assumptions | Uncertainty and confidence score |
|---|---|---|---|---|---|

## Rollback, escalation, and approval register

#### Expected output

- A gated implementation sequence with evidence-based entry and exit criteria, formulas and ranges, assumptions, uncertainty and confidence scores, Unknowns, privacy, security, and data-access stop conditions, rollback design, and named-human approval gates.

#### Customization tips

- Keep deployment as a separate decision after validation evidence, not the final unchecked step of a roadmap.
- Use stop conditions that are observable and owned, especially for data authorization and failure thresholds.

#### Guardrails

- Preserve the evidence or source basis for every material claim; unsupported facts remain Unknown.
- Distinguish Supplied, Calculated, and Estimated values; show every formula, estimated range, assumption, uncertainty, and confidence score.
- Do not invent, infer, or fabricate ROI or return on investment, time savings, cost, feasibility, duration, staffing, capability, priority, owner, dependency, access, permission, baseline, outcome, or probability.
- Make every privacy, security, and data-access dependency explicit and treat unresolved authorization as a stop condition.
- A named human owner must approve prioritization adoption before action.
- A named human owner must approve purchasing before action.
- A named human owner must approve deployment before action.
- A named human owner must approve every data or permission change before action.
- A named human owner must approve every live workflow change before action.
- A named human owner and the qualified professional must approve every legal, tax, employment, or regulatory interpretation before action.
- A named human owner must approve every payment or financial decision before action.
- A named human owner must approve every deletion, closure, suspension, termination, revocation, deactivation, or archive action before action.
- A named human owner must approve every promise or commitment before action.
- A named human owner must approve every policy exception before action.
- A named human owner must approve every external send or publication before action.
- Use only authorized business data in an approved AI workspace or approved AI vendor; minimize data, redact or omit unnecessary personal, sensitive, or confidential data, and follow company retention and vendor policy.
- Do not infer authorization. Do not automate, purchase, deploy, change data or permissions, or mutate a live workflow.

#### Example input

Fictional example — PRIORITIZED USE CASES: Northstar's human-reviewed shortlist is missing-field detection first and internal summary drafting second; scores are provisional. READINESS GAPS: Field definition owner and quality threshold Unknown. WORKFLOW DEPENDENCIES: Intake SOP review precedes testing. DATA AND ACCESS DEPENDENCIES: De-identified export approved; production access not approved. PRIVACY AND SECURITY REQUIREMENTS: Restricted customer identifiers, 30-day retention, vendor review required. CONSTRAINTS: One analyst, six-week assessment, no customer-facing automation. IMPLEMENTATION OWNERS: Maya owns workflow, Omar owns security, purchasing owner Unknown. SUCCESS MEASURES: Reviewed precision and rework change; baseline incomplete and failure threshold pending.

#### Related prompts

- https://returnmytime.com/prompt-library/rank-ai-use-cases-by-business-impact
- https://returnmytime.com/prompt-library/identify-ai-data-readiness-gaps

#### Related articles

- https://returnmytime.com/blog/ai-assessment-before-build

### Create an AI-Agent Acceptance Test Plan

- URL: https://returnmytime.com/prompt-library/create-ai-agent-acceptance-test-plan
- Category: Managed AI
- Workflow: managed-agent-operations-governance
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: managed-agents
- Type: Effectiveness
- Estimated time saved: Varies by agent scope and supplied test evidence
- Tags: AI-agent-testing, acceptance-plan, governance
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Turn a bounded agent operating contract into a reviewable acceptance matrix covering normal, edge, failure, and prohibited-action cases.

#### What it does

- Creates an acceptance matrix with expected evidence, failure conditions, owners, and approval gates.
- Separates tests that are merely planned from tests that were executed and tests whose evidence passed review.

#### When to use

- After one agent job, its non-goals, source boundaries, permissions, and human decision rights are documented.

#### When not to use

- Do not treat a generated plan as proof that any test ran, passed, or authorized live operation.
- Do not paste secrets or credentials, or place unauthorized agent data in an unapproved AI system.

#### Required inputs

- AGENT JOB
- ALLOWED AND PROHIBITED ACTIONS
- SYSTEMS AND DATA CLASSES
- TEST CASES
- CRITICAL FAILURES
- EVIDENCE REQUIREMENTS
- ACCEPTANCE OWNER AND APPROVAL GATES
- CREDENTIAL AND ACCESS BOUNDARIES

#### Prompt

#CONTEXT:
Create an acceptance-test plan for one bounded AI-agent job. This produces a planning artifact for a named human owner; it does not run tests or authorize a live agent.

#INPUTS:
- Agent job, trigger, inputs, decisions, outputs, completion, non-goals, and manual fallback: [AGENT JOB]
- Explicitly allowed actions, prohibited actions, and actions requiring approval: [ALLOWED AND PROHIBITED ACTIONS]
- Approved systems, source authority, integrations, environments, and data classes: [SYSTEMS AND DATA CLASSES]
- Supplied normal, edge, failure, retry, conflict, adversarial, and prohibited-action cases: [TEST CASES]
- Critical failures, unacceptable outcomes, stop conditions, and rollback triggers: [CRITICAL FAILURES]
- Required logs, records, citations, reconciliation checks, and evidence retention: [EVIDENCE REQUIREMENTS]
- Named human acceptance owner, reviewers, decision rights, and explicit approval gates: [ACCEPTANCE OWNER AND APPROVAL GATES]
- Credential owners, approved identities, scopes, access limits, and secret-handling constraints: [CREDENTIAL AND ACCESS BOUNDARIES]

#INSTRUCTIONS:
1. Restate the bounded job and map every criterion to supplied operating-contract evidence or source attribution. Label missing facts Unknown and conflicting sources Conflict—human resolution required.
2. Build separate normal, edge, degraded-tool, retry/idempotency, source-conflict, approval, escalation, data-boundary, permission, and prohibited-action cases. Do not invent missing cases as approved business rules.
3. For every test, specify synthetic or sanitized input, approved source, expected decision, expected draft or bounded action, prohibited result, observable evidence, pass condition, failure condition, stop condition, and named human reviewer.
4. Use only these status values: Planned—not executed, Executed—not yet reviewed, Passed—evidence reviewed, Failed, Blocked, or Unknown. Never convert Planned or Executed into Passed without supplied reviewed evidence.
5. Fail closed when a required source, permission, approval, owner, criterion, evidence field, or conflict resolution is missing. The safe outcome is stop, preserve evidence, and route to the named human owner.
6. Do not invent, infer, or fabricate an SLA, threshold, result, pass rate, test execution, monitoring state, owner, permission, source, or capability. Unsupported values remain Unknown.
7. Include explicit approval gates for acceptance, any waived failure, scope change, permission change, connection, deployment, enablement, restart, and live-authority expansion. A completed test does not imply approval for the next gate.
8. Do not infer authorization from an allowed-action list, credential reference, prior approval, planned test, executed test, or passing evidence.
9. Do not grant, revoke, or change permissions. Do not provision an agent or service account with access, permissions, or a role. Do not use or reveal credentials. Do not rotate any credential, token, API key, or secret. Do not deploy, enable, or disable an agent. Do not restart, pause, or activate an agent. Do not connect an agent or integration to a live system. Do not execute transactions. Do not send externally. Do not change production or claim monitoring is active.
10. Address privacy, security, data classes, least privilege, source access, retention, and evidence minimization. Use only authorized agent data in an approved AI workspace or approved AI vendor. Minimize inputs and redact or omit unnecessary personal, sensitive, or confidential data.
11. Never paste secrets, credentials, tokens, API keys, private keys, or recovery codes. Refer only to credential owner, identity name, scope, rotation status, and approved secret manager. Follow company retention and vendor policy.

#RESPONSE FORMAT:
## Operating-contract evidence and source attribution
| Requirement | Supplied source | Authority or owner | Status | Conflict or Unknown | Stop condition |
|---|---|---|---|---|---|

## Acceptance matrix
| Test ID and class | Sanitized input | Expected decision or draft | Prohibited result | Evidence required | Pass and failure conditions | Status: Planned, Executed, Passed, Failed, Blocked, or Unknown | Human reviewer |
|---|---|---|---|---|---|---|---|

## Critical-failure and fail-closed register
| Critical failure | Detection evidence | Immediate safe state | Escalation owner | Approval gate before resumption |
|---|---|---|---|---|

## Coverage gaps, conflicts, and Unknowns

## Approval-gate register

#### Expected output

- An acceptance matrix with evidence and source attribution, Planned versus Executed versus Passed status, critical-failure stop conditions, conflicts and Unknowns, and named-human approval gates.

#### Customization tips

- Keep a stable regression set and a separate rotating review set without weakening prohibited-action cases.
- Require business-system evidence for completion, not a plausible-looking model response.

#### Guardrails

- Fail closed on missing evidence, unresolved conflict, absent ownership, unavailable approval, or permission uncertainty; preserve evidence for review by a named human owner.
- Keep test states distinct: Planned is not Executed, Executed is not Passed, and Passed requires supplied reviewed evidence.
- Preserve evidence or source attribution for every criterion and result; unsupported facts remain Unknown.
- Do not invent, infer, or fabricate any SLA, threshold, result, pass rate, execution, monitoring state, owner, permission, or capability.
- Do not claim monitoring is active without supplied operating evidence.
- A named human owner must review and approve every payment or financial decision before any action.
- A named human owner must review and approve every legal, liability, admission, settlement, or contract decision before any action.
- A named human owner must review and approve every account, access, permission, or credential decision before any action.
- A named human owner must review and approve every policy exception, override, or waiver before any action.
- A named human owner must review and approve every external send or publication before any action.
- A named human owner must review and approve every deletion, closure, suspension, termination, revocation, deactivation, disablement, or archive decision before any action.
- A named human owner must review and approve every promise, guarantee, or commitment before any action.
- Use explicit approval gates; a named human owner must approve acceptance, waivers, scope changes, permissions, deployment, restart, and live-authority expansion before action.
- Do not infer authorization. Do not grant, revoke, or change permissions. Do not provision an agent or service account with access, permissions, or a role. Do not use or reveal credentials. Do not rotate any credential, token, API key, or secret. Do not deploy, enable, or disable an agent. Do not restart, pause, or activate an agent. Do not connect an agent or integration to a live system. Do not execute transactions. Do not send externally. Do not change production or claim monitoring is active.
- Protect privacy and security: use only authorized agent data in an approved AI workspace or approved AI vendor; minimize data; redact or omit unnecessary personal, sensitive, or confidential data; follow company retention and vendor policy.
- Never paste secrets, credentials, tokens, API keys, private keys, or recovery codes.

#### Example input

Fictional example — AGENT JOB: Northstar intake agent drafts an internal lead brief; no customer sending. ALLOWED AND PROHIBITED ACTIONS: Read sanitized form fields and draft; no pricing, deletion, sending, or CRM writes. SYSTEMS AND DATA CLASSES: Sandbox form export, internal business data; personal contact fields omitted. TEST CASES: Two normal cases, missing service area, conflicting source, tool timeout, duplicate event, prompt injection. CRITICAL FAILURES: External message or unsupported eligibility claim. EVIDENCE REQUIREMENTS: Correlation ID, cited rule version, output comparison. ACCEPTANCE OWNER AND APPROVAL GATES: Maya accepts evidence; Omar approves security; launch owner Unknown. CREDENTIAL AND ACCESS BOUNDARIES: Sandbox identity is read-only; secret values excluded.

#### Related prompts

- https://returnmytime.com/prompt-library/create-ai-agent-permission-boundaries
- https://returnmytime.com/prompt-library/create-ai-agent-escalation-policy

#### Related articles

- https://returnmytime.com/blog/managed-ai-agent-ownership
- https://returnmytime.com/blog/managed-ai-agent-testing-monitoring

### Create an AI-Agent Escalation Policy

- URL: https://returnmytime.com/prompt-library/create-ai-agent-escalation-policy
- Category: Managed AI
- Workflow: managed-agent-operations-governance
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: managed-agents
- Type: Effectiveness
- Estimated time saved: Varies by escalation paths and supplied operating evidence
- Tags: AI-agent-escalation, incident-policy, human-oversight
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Draft a fail-closed escalation policy that maps agent uncertainty and failures to named human owners, evidence, fallback, and restart gates.

#### What it does

- Maps known triggers to a safe state, evidence packet, named owner, backup, and approval gate.
- Keeps proposed response times and restart conditions separate from approved, tested operating rules.

#### When to use

- Before an agent receives live authority, or when existing escalation ownership and safe-state behavior need review.

#### When not to use

- Do not use a drafted policy to operate, pause, restart, enable, disable, or change a live agent.
- Do not paste secrets or credentials, or upload unauthorized incident, customer, or agent data.

#### Required inputs

- AGENT JOB
- ALLOWED AND PROHIBITED ACTIONS
- CRITICAL FAILURES AND STOP CONDITIONS
- ESCALATION OWNERS AND BACKUPS
- ESCALATION SLAS
- SYSTEMS AND DATA CLASSES
- APPROVAL GATES
- FALLBACK AND RESTART RULES

#### Prompt

#CONTEXT:
Create a review-ready escalation policy for one bounded AI-agent job. This is a policy draft for named human owners, not authority to operate or alter an agent.

#INPUTS:
- Agent job, trigger, inputs, decisions, outputs, completion, non-goals, and normal owner: [AGENT JOB]
- Allowed actions, prohibited actions, and actions that require human approval: [ALLOWED AND PROHIBITED ACTIONS]
- Supplied critical failures, uncertainty triggers, stop conditions, and incident definitions: [CRITICAL FAILURES AND STOP CONDITIONS]
- Named primary human owners, backups, technical responders, and decision rights: [ESCALATION OWNERS AND BACKUPS]
- Approved response and resolution SLAs, measurement basis, and consequence of no response: [ESCALATION SLAS]
- Approved systems, environments, source authority, access boundaries, and data classes: [SYSTEMS AND DATA CLASSES]
- Explicit approval gates for exception, communication, financial, access, production, restart, and authority decisions: [APPROVAL GATES]
- Manual fallback, backlog capture, reconciliation, rollback, restart evidence, and restart owner: [FALLBACK AND RESTART RULES]

#INSTRUCTIONS:
1. Map each supplied trigger to its evidence or source attribution. Label missing facts Unknown and incompatible instructions Conflict—human resolution required.
2. Define severity only from supplied rules. Do not infer a severity, SLA, threshold, owner, or approval right from examples or common practice.
3. For each trigger, specify the immediate fail-closed state, work preservation, evidence packet, named human owner, backup, allowed wait-state behavior, and explicit approval gate.
4. Include triggers for missing or conflicting sources, low confidence, out-of-scope requests, prohibited actions, changed permissions, expired credentials, tool failure, partial write, retry ambiguity, duplicate risk, privacy or security concern, data-class mismatch, monitoring gap, and critical failure.
5. If no qualified owner responds within an approved SLA, fail closed: stop the affected work, preserve evidence, use the approved manual fallback, and keep consequential action blocked.
6. Separate policy status as Proposed, Approved, Tested, Active, Retired, or Unknown. Separate exercise status as Planned—not executed, Executed—not yet reviewed, Passed—evidence reviewed, Failed, Blocked, or Unknown. Never claim a policy or monitoring process is active from a draft.
7. Do not invent, infer, or fabricate an SLA, threshold, result, owner, backup, policy status, exercise result, monitoring state, credential state, source, or capability. Unsupported values remain Unknown.
8. Require approval gates before any exception, customer or external communication, financial transaction, permission or access change, production change, agent deployment, enablement, disablement, restart, or authority expansion.
9. Do not infer authorization from a named owner, credential record, proposed SLA, prior decision, or escalation outcome.
10. Do not grant, revoke, or change permissions. Do not provision an agent or service account with access, permissions, or a role. Do not use or reveal credentials. Do not rotate any credential, token, API key, or secret. Do not deploy, enable, or disable an agent. Do not restart, pause, or activate an agent. Do not connect an agent or integration to a live system. Do not execute transactions. Do not send externally. Do not change production or claim monitoring is active.
11. Protect privacy, security, and data classes. Use only authorized agent data in an approved AI workspace or approved AI vendor. Minimize the evidence packet; redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
12. Never paste secrets, credentials, tokens, API keys, private keys, or recovery codes. Reference the credential owner and approved secret manager only.

#RESPONSE FORMAT:
## Evidence and source attribution
| Rule or fact | Source | Authority or owner | Status | Conflict or Unknown |
|---|---|---|---|---|

## Escalation policy
| Trigger and severity | Detection evidence | Immediate fail-closed state | Evidence packet | Primary owner and backup | Approved SLA or Unknown | No-response fallback | Approval gate |
|---|---|---|---|---|---|---|---|

## Critical-failure, fallback, and reconciliation register

## Policy and exercise status
| Item | Proposed, Approved, Tested, Active, Retired, or Unknown | Planned, Executed, Passed, Failed, Blocked, or Unknown | Evidence | Owner |
|---|---|---|---|---|

## Conflicts, Unknowns, and decisions required

#### Expected output

- An escalation policy with evidence and source attribution, fail-closed states, named human owners and backups, approved-or-Unknown SLAs, fallback and restart gates, Planned versus Executed versus Passed exercise status, conflicts, and Unknowns.

#### Customization tips

- Test the primary and backup route, no-response behavior, late approval, and restart gate with sanitized exercises.
- Make the evidence packet useful without copying unnecessary sensitive content into alerts.

#### Guardrails

- Fail closed on missing sources, unresolved conflict, uncertain permissions, missing ownership, unavailable approval, or critical failure; preserve evidence for review by a named human owner.
- Preserve evidence or source attribution for every trigger, severity, SLA, and approval gate; unsupported facts remain Unknown.
- Keep status distinct: Planned is not Executed, Executed is not Passed, and a proposed or tested policy is not necessarily active.
- Do not invent, infer, or fabricate any SLA, threshold, result, owner, policy status, monitoring state, credential state, or capability.
- Do not claim monitoring is active without supplied operating evidence.
- A named human owner must review and approve every payment or financial decision before any action.
- A named human owner must review and approve every legal, liability, admission, settlement, or contract decision before any action.
- A named human owner must review and approve every account, access, permission, or credential decision before any action.
- A named human owner must review and approve every policy exception, override, or waiver before any action.
- A named human owner must review and approve every external send or publication before any action.
- A named human owner must review and approve every deletion, closure, suspension, termination, revocation, deactivation, disablement, or archive decision before any action.
- A named human owner must review and approve every promise, guarantee, or commitment before any action.
- Use explicit approval gates; a named human owner must approve exceptions, communications, transactions, permission changes, production changes, restart, and authority expansion before action.
- Do not infer authorization. Do not grant, revoke, or change permissions. Do not provision an agent or service account with access, permissions, or a role. Do not use or reveal credentials. Do not rotate any credential, token, API key, or secret. Do not deploy, enable, or disable an agent. Do not restart, pause, or activate an agent. Do not connect an agent or integration to a live system. Do not execute transactions. Do not send externally. Do not change production or claim monitoring is active.
- Protect privacy and security: use only authorized agent data in an approved AI workspace or approved AI vendor; minimize data; redact or omit unnecessary personal, sensitive, or confidential data; follow company retention and vendor policy.
- Never paste secrets, credentials, tokens, API keys, private keys, or recovery codes.

#### Example input

Fictional example — AGENT JOB: Northstar intake agent prepares internal summaries. ALLOWED AND PROHIBITED ACTIONS: Read sanitized intake and draft; no sending or CRM writes. CRITICAL FAILURES AND STOP CONDITIONS: Unsupported eligibility decision, data leak, duplicate write, missing authoritative rule. ESCALATION OWNERS AND BACKUPS: Maya, operations; Theo, backup; Omar, security. ESCALATION SLAS: Security response target supplied as 15 minutes; other response times Unknown. SYSTEMS AND DATA CLASSES: Sandbox CRM and restricted customer identifiers excluded. APPROVAL GATES: Maya approves policy exceptions; Omar approves security response; restart owner Unknown. FALLBACK AND RESTART RULES: Manual queue exists; reconciliation checklist draft only.

#### Related prompts

- https://returnmytime.com/prompt-library/create-ai-agent-acceptance-test-plan
- https://returnmytime.com/prompt-library/create-ai-agent-monitoring-scorecard

#### Related articles

- https://returnmytime.com/blog/managed-ai-agent-ownership
- https://returnmytime.com/blog/managed-ai-agent-testing-monitoring

### Create an AI-Agent Monitoring Scorecard

- URL: https://returnmytime.com/prompt-library/create-ai-agent-monitoring-scorecard
- Category: Managed AI
- Workflow: managed-agent-operations-governance
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: managed-agents
- Type: Effectiveness
- Estimated time saved: Varies by monitoring scope and evidence maturity
- Tags: AI-agent-monitoring, scorecard, observability
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Design an evidence-based monitoring scorecard for agent availability, execution, decision quality, reconciliation, and human response.

#### What it does

- Turns supplied objectives and evidence sources into a proposed owner-facing monitoring scorecard.
- Separates measurable signals, approved thresholds, review status, and escalation dependencies without claiming monitoring is active.

#### When to use

- When a bounded agent contract and expected business-system evidence exist and owners need a monitoring design before launch or review.

#### When not to use

- Do not use a generated scorecard as proof that telemetry, alerts, reconciliation, or incident response is connected or active.
- Do not paste secrets, credentials, or raw sensitive logs into an AI system.

#### Required inputs

- AGENT JOB
- SYSTEMS AND DATA CLASSES
- METRICS AND THRESHOLDS
- CRITICAL FAILURES
- ESCALATION OWNERS AND SLAS
- EVIDENCE AND SOURCE REQUIREMENTS
- CREDENTIAL AND ACCESS BOUNDARIES
- APPROVAL GATES

#### Prompt

#CONTEXT:
Create a monitoring-scorecard design for one bounded AI-agent job. This is a review artifact; it does not connect monitoring, inspect live systems, or claim monitoring is active.

#INPUTS:
- Agent job, trigger, completion, non-goals, allowed actions, prohibited actions, and manual fallback: [AGENT JOB]
- Approved systems, environments, source authority, interfaces, identities, and data classes: [SYSTEMS AND DATA CLASSES]
- Supplied metric definitions, baselines, units, calculation rules, thresholds, review cadence, and owners: [METRICS AND THRESHOLDS]
- Critical failures, prohibited outcomes, stop conditions, rollback triggers, and reconciliation risks: [CRITICAL FAILURES]
- Named human owners, backups, approved response SLAs, and no-response behavior: [ESCALATION OWNERS AND SLAS]
- Required logs, records, citations, correlation IDs, approval evidence, and business-system reconciliation sources: [EVIDENCE AND SOURCE REQUIREMENTS]
- Credential owners, approved identities, scopes, rotation state, and access-review boundaries: [CREDENTIAL AND ACCESS BOUNDARIES]
- Explicit approval gates for alerts, incident actions, permissions, production, pause, restart, reporting, and authority changes: [APPROVAL GATES]

#INSTRUCTIONS:
1. Map the operating objective, completion condition, critical failures, and each metric to evidence or source attribution. Label unavailable facts Unknown and incompatible definitions Conflict—human resolution required.
2. Design signals at four layers: availability, execution, decision quality, and business reconciliation. Include approvals waiting, exceptions, source freshness, tool failures, retry and duplicate behavior, prohibited-action attempts, overrides, unresolved work, and manual-fallback load when supported.
3. For each metric, state definition, formula, units, source, collection state, baseline, proposed or approved threshold, review cadence, named human owner, escalation path, and privacy or security constraint.
4. Preserve threshold provenance. Mark every threshold as Supplied-approved, Proposed for review, or Unknown. A proposed threshold is not an approved alert rule.
5. Separate implementation status as Designed, Connected, Validated, Active, Degraded, Retired, or Unknown. Separate validation status as Planned—not executed, Executed—not yet reviewed, Passed—evidence reviewed, Failed, Blocked, or Unknown. Never claim monitoring is active without supplied operating evidence.
6. Fail closed when a critical signal, source, threshold, owner, permission, alert route, reconciliation check, or conflict resolution is missing. Recommend that consequential work remain blocked pending named-human review.
7. Do not invent, infer, or fabricate an SLA, threshold, baseline, metric result, test result, monitoring state, owner, source, permission, credential state, or capability. Unsupported values remain Unknown.
8. Define evidence needed to validate each metric and alert in a sandbox or approved environment. Do not imply Planned, Executed, or Passed status without supplied evidence.
9. Require explicit approval gates for thresholds, alert routes, reporting recipients, incident actions, access changes, production changes, deployment, pause, restart, and live-authority expansion.
10. Do not infer authorization from a dashboard, credential record, threshold, owner, alert, test, or metric result.
11. Do not grant, revoke, or change permissions. Do not provision an agent or service account with access, permissions, or a role. Do not use or reveal credentials. Do not rotate any credential, token, API key, or secret. Do not deploy, enable, or disable an agent. Do not restart, pause, or activate an agent. Do not connect an agent or integration to a live system. Do not execute transactions. Do not send externally. Do not change production or claim monitoring is active.
12. Address privacy, security, data classes, log access, retention, and minimization. Use only authorized agent data in an approved AI workspace or approved AI vendor. Minimize logs and redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
13. Never paste secrets, credentials, tokens, API keys, private keys, or recovery codes. Use identity and secret-manager references only.

#RESPONSE FORMAT:
## Evidence and source attribution
| Objective, failure, or metric | Source | Authority or owner | Status | Conflict or Unknown |
|---|---|---|---|---|

## Monitoring scorecard
| Layer and metric | Definition and formula | Evidence source | Baseline | Threshold: supplied, proposed, or Unknown | Implementation status | Validation status: Planned, Executed, Passed, Failed, Blocked, or Unknown | Owner and escalation |
|---|---|---|---|---|---|---|---|

## Critical-failure and fail-closed view

## Privacy, security, credential, and data-class controls

## Approval gates, conflicts, and Unknowns

#### Expected output

- A monitoring scorecard with evidence and source attribution, formulas, supplied-proposed-Unknown thresholds, Planned versus Executed versus Passed validation status, fail-closed dependencies, conflicts, Unknowns, and named-human approval gates; monitoring is explicitly not active from this draft.

#### Customization tips

- Prefer business completion and reconciliation evidence over uptime alone.
- Keep sensitive content out of alerts; use correlation identifiers and controlled source access for investigation.

#### Guardrails

- Fail closed on missing critical signals, sources, approved thresholds, owners, permissions, alert routes, or reconciliation evidence; preserve evidence for review by a named human owner.
- Preserve evidence or source attribution for every metric, threshold, status, and result; unsupported facts remain Unknown and conflicts require human resolution.
- Keep status distinct: Planned is not Executed, Executed is not Passed, and Designed or Validated monitoring is not necessarily Active.
- Do not invent, infer, or fabricate any SLA, threshold, baseline, result, monitoring state, owner, source, permission, credential state, or capability.
- Do not claim monitoring is active without supplied operating evidence.
- A named human owner must review and approve every payment or financial decision before any action.
- A named human owner must review and approve every legal, liability, admission, settlement, or contract decision before any action.
- A named human owner must review and approve every account, access, permission, or credential decision before any action.
- A named human owner must review and approve every policy exception, override, or waiver before any action.
- A named human owner must review and approve every external send or publication before any action.
- A named human owner must review and approve every deletion, closure, suspension, termination, revocation, deactivation, disablement, or archive decision before any action.
- A named human owner must review and approve every promise, guarantee, or commitment before any action.
- Use explicit approval gates; a named human owner must approve thresholds, alert routes, recipients, incident actions, access, production, restart, and authority expansion before action.
- Do not infer authorization. Do not grant, revoke, or change permissions. Do not provision an agent or service account with access, permissions, or a role. Do not use or reveal credentials. Do not rotate any credential, token, API key, or secret. Do not deploy, enable, or disable an agent. Do not restart, pause, or activate an agent. Do not connect an agent or integration to a live system. Do not execute transactions. Do not send externally. Do not change production or claim monitoring is active.
- Protect privacy and security: use only authorized agent data in an approved AI workspace or approved AI vendor; minimize data; redact or omit unnecessary personal, sensitive, or confidential data; follow company retention and vendor policy.
- Never paste secrets, credentials, tokens, API keys, private keys, or recovery codes.

#### Example input

Fictional example — AGENT JOB: Northstar intake agent drafts internal summaries and routes uncertainty to Maya. SYSTEMS AND DATA CLASSES: Sandbox queue and CRM export; restricted identifiers omitted. METRICS AND THRESHOLDS: Completion count and unresolved queue age defined; alert thresholds Unknown. CRITICAL FAILURES: Unsupported promise, duplicate record, missing source citation. ESCALATION OWNERS AND SLAS: Maya primary, Theo backup; response SLA Unknown. EVIDENCE AND SOURCE REQUIREMENTS: Queue events, correlation ID, cited rule version, reconciliation export. CREDENTIAL AND ACCESS BOUNDARIES: Read-only sandbox identity; rotation state Unknown. APPROVAL GATES: Omar approves security controls; threshold and restart approvers Unknown.

#### Related prompts

- https://returnmytime.com/prompt-library/create-ai-agent-escalation-policy
- https://returnmytime.com/prompt-library/create-ai-agent-acceptance-test-plan

#### Related articles

- https://returnmytime.com/blog/managed-ai-agent-ownership
- https://returnmytime.com/blog/managed-ai-agent-testing-monitoring

### Create AI-Agent Permission Boundaries

- URL: https://returnmytime.com/prompt-library/create-ai-agent-permission-boundaries
- Category: Managed AI
- Workflow: managed-agent-operations-governance
- Audience: owner, founder-operator
- Compatible assistants: ChatGPT, Claude
- Service stage: managed-agents
- Type: Effectiveness
- Estimated time saved: Varies by system count and access complexity
- Tags: AI-agent-permissions, least-privilege, access-governance
- Published: 2026-07-31
- Modified: 2026-07-31
- Provenance: original


Translate a bounded agent job into a least-privilege permission proposal with data-class limits, credential ownership, logging, review, and stop conditions.

#### What it does

- Creates a least-privilege matrix separating required, prohibited, approval-gated, and Unknown access.
- Documents credential ownership, data classes, evidence, revocation readiness, and human approval gates without changing access.

#### When to use

- Before provisioning an agent identity or during a documented access review for one bounded function.

#### When not to use

- Do not use the output as authorization or technical instructions to provision, grant, revoke, rotate, connect, or change live access.
- Do not paste secrets, credentials, tokens, keys, or unauthorized access records into an AI system.

#### Required inputs

- AGENT JOB
- ALLOWED AND PROHIBITED ACTIONS
- SYSTEMS
- CREDENTIAL OWNERS
- DATA CLASSES
- APPROVAL GATES
- LOGGING AND REVIEW REQUIREMENTS
- REVOCATION AND INCIDENT RULES

#### Prompt

#CONTEXT:
Create a least-privilege permission-boundary proposal for one bounded AI-agent job. This is review support for named human owners, not authorization or a request to change access.

#INPUTS:
- Agent job, trigger, inputs, decisions, outputs, completion, non-goals, fallback, and environment: [AGENT JOB]
- Allowed actions, prohibited actions, and actions requiring named-human approval: [ALLOWED AND PROHIBITED ACTIONS]
- Approved systems, resources, environments, integrations, system owners, and available scopes: [SYSTEMS]
- Named credential owners, agent identities, secret-manager references, rotation owners, and current review state: [CREDENTIAL OWNERS]
- Approved data classes, purpose, source authority, sensitivity, retention, residency, and prohibited destinations: [DATA CLASSES]
- Explicit approval gates for read, draft, write, send, transact, modify, delete, administer, connect, production, and exception authority: [APPROVAL GATES]
- Required identity, action, source, result, approval, correlation, review cadence, and access-review evidence: [LOGGING AND REVIEW REQUIREMENTS]
- Supplied revocation owner, emergency stop, offboarding, incident evidence, fallback, reconciliation, and restart rules: [REVOCATION AND INCIDENT RULES]

#INSTRUCTIONS:
1. Map each requested capability to a supplied job requirement and evidence or source attribution. Label missing facts Unknown and incompatible access rules Conflict—human resolution required.
2. Use least privilege. Begin with no access, then propose only the minimum resource, environment, action, scope, data class, purpose, duration, and evidence needed for the bounded job.
3. Separate each capability as Required for review, Approval-gated, Prohibited, Not needed, or Unknown. A technical capability is not a business requirement or authorization.
4. Prefer read-only and draft-only access where it satisfies the job. Treat write, send, transaction, modification, deletion, administration, credential, cross-customer, sensitive-data, and production access as separate capabilities.
5. For every proposed capability, state the named human system owner, credential owner, data owner, approver, review cadence, logging evidence, stop condition, and revocation path. Missing ownership or evidence must fail closed.
6. Preserve credential secrecy. Record only identity name, credential owner, secret-manager reference, scopes, rotation state, and review evidence. Never request or reproduce a secret value.
7. Separate design status as Proposed, Approved, Provisioned, Validated, Active, Revoked, or Unknown. Separate verification status as Planned—not executed, Executed—not yet reviewed, Passed—evidence reviewed, Failed, Blocked, or Unknown. Never claim access or monitoring is active without supplied evidence.
8. Do not invent, infer, or fabricate an SLA, threshold, result, scope, permission, credential state, owner, data classification, approval, validation status, monitoring state, or system capability. Unsupported values remain Unknown.
9. Fail closed on missing job justification, source authority, data purpose, owner, approval gate, logging, revocation, environment, permission evidence, or unresolved conflict.
10. Do not infer authorization from a capability list, credential owner, existing scope, integration feature, prior approval, or test result.
11. Do not grant, revoke, or change permissions. Do not provision an agent or service account with access, permissions, or a role. Do not use or reveal credentials. Do not rotate any credential, token, API key, or secret. Do not deploy, enable, or disable an agent. Do not restart, pause, or activate an agent. Do not connect an agent or integration to a live system. Do not execute transactions. Do not send externally. Do not change production or claim monitoring is active.
12. Address privacy, security, data classes, purpose limitation, segregation, retention, and minimization. Use only authorized company data in an approved AI workspace or approved AI vendor. Minimize inputs and redact or omit unnecessary personal, sensitive, or confidential data. Follow company retention and vendor policy.
13. Never paste secrets, credentials, tokens, API keys, private keys, or recovery codes.

#RESPONSE FORMAT:
## Job-to-access evidence and source attribution
| Job requirement | Evidence source | Required capability | Owner | Conflict or Unknown |
|---|---|---|---|---|

## Least-privilege permission boundary
| System and environment | Resource and data class | Action and scope | Required, approval-gated, prohibited, not needed, or Unknown | Purpose | Named owners and approver | Logging evidence | Stop and revocation conditions |
|---|---|---|---|---|---|---|---|

## Credential reference and review register
| Identity name | Credential owner | Secret-manager reference | Scope | Rotation state | Verification: Planned, Executed, Passed, Failed, Blocked, or Unknown |
|---|---|---|---|---|---|

## Fail-closed gaps, conflicts, and Unknowns

## Approval-gate and validation register

#### Expected output

- A least-privilege permission boundary with job-to-access evidence and source attribution, named human owners, data-class limits, credential references, Planned versus Executed versus Passed verification status, fail-closed conflicts and Unknowns, and explicit approval gates.

#### Customization tips

- Challenge each write or administrative scope by asking whether read-only, draft-only, or a narrower resource can satisfy the job.
- Test prohibited access and revocation behavior in an approved non-production environment before any live-authority decision.

#### Guardrails

- Fail closed on missing justification, source authority, purpose, ownership, approval, logging, revocation, permission evidence, or unresolved conflict; preserve evidence for review by a named human owner.
- Preserve evidence or source attribution for every proposed scope, data class, owner, status, and approval gate; unsupported facts remain Unknown.
- Keep status distinct: Planned is not Executed, Executed is not Passed, and Proposed or Validated access is not necessarily Active.
- Do not invent, infer, or fabricate any SLA, threshold, result, scope, permission, credential state, owner, data class, approval, monitoring state, or capability.
- Do not claim monitoring is active without supplied operating evidence.
- A named human owner must review and approve every payment or financial decision before any action.
- A named human owner must review and approve every legal, liability, admission, settlement, or contract decision before any action.
- A named human owner must review and approve every account, access, permission, or credential decision before any action.
- A named human owner must review and approve every policy exception, override, or waiver before any action.
- A named human owner must review and approve every external send or publication before any action.
- A named human owner must review and approve every deletion, closure, suspension, termination, revocation, deactivation, disablement, or archive decision before any action.
- A named human owner must review and approve every promise, guarantee, or commitment before any action.
- Use explicit approval gates; a named human owner must approve every access, permission, credential, connection, production, exception, revocation, and restart decision before action.
- Do not infer authorization. Do not grant, revoke, or change permissions. Do not provision an agent or service account with access, permissions, or a role. Do not use or reveal credentials. Do not rotate any credential, token, API key, or secret. Do not deploy, enable, or disable an agent. Do not restart, pause, or activate an agent. Do not connect an agent or integration to a live system. Do not execute transactions. Do not send externally. Do not change production or claim monitoring is active.
- Protect privacy and security: use only authorized company data in an approved AI workspace or approved AI vendor; minimize data; redact or omit unnecessary personal, sensitive, or confidential data; follow company retention and vendor policy.
- Never paste secrets, credentials, tokens, API keys, private keys, or recovery codes.

#### Example input

Fictional example — AGENT JOB: Northstar intake agent reads a sandbox export and drafts an internal brief. ALLOWED AND PROHIBITED ACTIONS: Read and draft only; no write, sending, pricing, deletion, or administration. SYSTEMS: Sandbox CRM export and approved knowledge repository; production access not approved. CREDENTIAL OWNERS: Omar owns the sandbox identity; secret-manager reference supplied, no secret value. DATA CLASSES: Internal business data; personal identifiers omitted; restricted data prohibited. APPROVAL GATES: Maya approves draft use; Omar approves access; production approver Unknown. LOGGING AND REVIEW REQUIREMENTS: Identity, source version, result, correlation ID, quarterly cadence proposed. REVOCATION AND INCIDENT RULES: Omar owns revocation; manual intake fallback; restart owner Unknown.

#### Related prompts

- https://returnmytime.com/prompt-library/create-ai-agent-acceptance-test-plan
- https://returnmytime.com/prompt-library/create-ai-agent-escalation-policy

#### Related articles

- https://returnmytime.com/blog/managed-ai-agent-ownership
- https://returnmytime.com/blog/managed-ai-agent-testing-monitoring

## Contact and conversion

- Assessment application: https://returnmytime.com/services/annie#discovery
- Email: hello@returnmytime.com