MISSION

GTM engineering that turns closed-won calls into pipeline

Your closed-won calls explain why customers bought. This mission turns that into ICP signals, lookalike prospect lists and outreach that references real buying triggers.

Inspired by

Chris Pisarski

View original post →View LinkedIn profile →

July 28, 2026

Mission type

Account & Deal Intelligence

Best for

Sales
Operations / RevOps

Output

Scored lookalike accounts and first-touch drafts

System used

REQUEST A MISSION

Tell us the meeting and what should happen after it. More detail means a sharper flight plan.

Request a mission

What it does

This mission adds call context to a GTM engineering stack. When a deal closes, AI reads the Fathom calls behind it and pulls out the buying trigger in the customer's own words. It enriches the account with firmographic data, finds companies that share the same traits and scores them against your ideal customer profile.

When a past champion changes jobs, it flags the move to the original rep with a drafted first touch that references the work you did together. Every list and message is grounded in why real customers bought.

Who it's for

GTM engineers, RevOps leads and sales teams already working in Claude Code or similar tools with their CRM and enrichment data connected.

The flight plan

Build it once, then let closed-won deals and champion job changes trigger it.

  1. Connect Fathom, your CRM, an enrichment tool and your outreach tool to Claude or ChatGPT.
  2. Write your ICP as a short document the AI scores against.
  3. Run the starter prompt on recent closed-won deals to find shared buying triggers and traits.
  4. Score lookalike accounts and review the top of the list.
  5. Draft first touches, and send only after a person reviews them.

RULES

Pull buying triggers from the customer's words, not the rep's notes.

Score against a written ICP, not a vague description.

Test the scoring on accounts you already know before trusting it.

Keep a person reviewing every message before it sends.

Log which signals lead to meetings so the scoring improves.

What good looks like

A good lookalike list says why each account fits, based on what real customers said. A mediocre one is a list of companies in the same industry.

GOOD

"Northwind (Series B, hiring six SDRs). Matches the trigger from four closed-won deals: 'We doubled the sales team and lost track of what reps promise.' First touch references how a similar team fixed that in their first month."

MEDIOCRE

"Northwind is a SaaS company in your target industry. Reach out about our product."

STARTER PROMPT

⧉ Copy

Pull my closed-won deals from the last 90 days from [CRM]. For each, use Fathom to find the sales calls and pull out the buying trigger in the customer's own words, plus any sign of the pain they were feeling.

Group the deals by shared triggers and traits, and compare them with my ICP document. Then score this list of target accounts against those patterns: [list or enrichment export].

For the top 10, explain why each fits and draft a first-touch email that references the matching trigger. Don't send anything.

FAQs

Questions? We’ve got you covered.

How do I use sales call recordings in GTM engineering?

Use them as the source for why customers buy. Pulling buying triggers from closed-won calls gives your ICP scoring and outreach real evidence, and connecting Fathom to Claude or ChatGPT alongside your CRM and enrichment tools lets AI do it at scale.

Do I need Claude Code for this?

The original build ran in Claude Code with several connectors, but you can run the research and scoring steps in Claude or ChatGPT with Fathom and your CRM connected. Enrichment and outreach tools add the automation.

Which Fathom plan do I need to run this skill in ChatGPT or Claude?

Any Fathom plan, including Free. The skill reads your calls through Fathom's MCP integration with ChatGPT and Claude, which is on every plan.

What is GTM engineering?

GTM engineering is the practice of building automated systems for go-to-market work, like prospecting, enrichment, scoring and outreach, usually by connecting data sources and AI tools. It replaces manual research with repeatable workflows.

New missions, straight to your inbox

A new mission launches every 1–2 weeks – real ways Fathom users are turning meetings into content, intelligence, and outcomes nobody expects. Subscribe and we'll send the new ones as they go live.

Your meetings already have the answer

Tell us what mission you're trying to run – we'll help you
chart the flight plan.