Return My TimeBook a Discovery Call

Revenue Operations

Create a Sales Follow-Up Sequence

Turn verified opportunity history into a stage-specific follow-up sequence that preserves unknowns and routes every draft through human review.

By Nick Spisak

Quick facts

Best for
Owners · Founder-operators · Sales leaders
Prompt type
Efficiency
Expected result
A stage-specific multistep follow-up sequence with evidence-backed drafts, timing, stop conditions, unknowns, and a human approval checklist.
Time saved
45-60 minutes per sequence draft
Required inputs
PIPELINE STAGE · PRIOR CONTACT HISTORY · VERIFIED OBJECTIONS · PROMISED ACTIONS · DECISION PROCESS · RELEVANT DATES · OFFER · CRM FIELDS · FOLLOW-UP OWNER
Works with
ChatGPT + Claude

What this prompt 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.

The 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

Input checklist

  • PIPELINE STAGE
  • PRIOR CONTACT HISTORY
  • VERIFIED OBJECTIONS
  • PROMISED ACTIONS
  • DECISION PROCESS
  • RELEVANT DATES
  • OFFER
  • CRM FIELDS
  • FOLLOW-UP OWNER

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.

Expected output structure

  • A stage-specific multistep follow-up sequence with evidence-backed drafts, timing, stop conditions, unknowns, and a human approval checklist.

Customize this prompt

  • 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.

KEEP BUILDING

Related prompts

NEXT STEP

Next step

Map where revenue is leaking.