Define your ICP from your 10 best customers
Reverse-engineers a tight, one-page ICP from your actual closed-won history instead of aspirational guesswork. It forces narrow ranges, separates signal from coincidence, and flags when your customer base actually contains two different ICPs — the insight most founders miss until their outbound numbers tell them the hard way.
You are a GTM strategist who has built ideal customer profiles for over 50 B2B companies. You know the difference between a real ICP and a wish list: a real ICP is reverse-engineered from closed-won evidence, not from who the founder hopes will buy. You also know the classic failure mode — an ICP so broad it excludes nobody ('companies with 10-10,000 employees that value efficiency') — and you refuse to produce one.
I'm going to give you data on my 10 best customers. Your job is to find the pattern and produce a one-page ICP document with these sections:
1. FIRMOGRAPHICS: industry, employee range, revenue range, geography. Tight ranges, not everything.
2. STRUCTURAL SIGNALS: what these companies have in common structurally — team makeup, tech stack, business model, growth stage.
3. PAIN PATTERN: the shared underlying problem, stated in the customer's words, not mine.
4. BUYING TRIGGERS: the events that made them ready to buy (new hire, funding, failed alternative, growth wall).
5. THE ONE-SENTENCE ICP: 'Companies that ___, with ___, who are experiencing ___.'
6. WHAT TO IGNORE: attributes that look important in my data but are probably coincidence, with your reasoning.
Be opinionated. If my 10 customers split into two distinct patterns, say so and profile both rather than averaging them into mush.
Before you write anything, interview me. Ask me these questions ONE AT A TIME, waiting for my answer each time:
1. What does your company sell, and what does it cost?
2. List your 10 best customers — for each: industry, headcount, what they bought, deal size, and why you consider them 'best'.
3. What made these deals close relatively easily, from what you saw?
4. Are there customers you deliberately excluded from this list? Why?
Once you have my answers, produce the ICP document. If my customer list is thinner than 10, work with what I have but flag which conclusions are shaky.How to use it
- 1
Copy the prompt into Claude, ChatGPT, or any LLM.
- 2
Before you start, pull your CRM export or just list your 10 best customers with deal size and industry — 'best' meaning profitable, low-drama, and likely to renew.
- 3
Answer the interview honestly, especially question 4 — the customers you excluded say as much about your ICP as the ones you kept.
- 4
Challenge the output: if a firmographic range feels too broad, ask the model to tighten it and defend the cut.
- 5
Save the one-sentence ICP — it feeds directly into the list-building and messaging prompts in this category.
Best practices
Pick 'best' customers by profit and retention, not logo prestige — the recognizable brand that took 9 months to close and haggles at renewal is not your ICP.
If the model says you have two ICPs, believe it. Run separate campaigns rather than writing one email that speaks to neither.
Include at least deal size and sales cycle length per customer — those two numbers do more pattern-finding work than any other field.
Rerun this quarterly with your latest wins; an ICP is a living document, not a founding myth.
Example: what this looks like in practice
A founder of a 12-person compliance software company lists her 10 best customers: eight are financial services firms between 50 and 200 employees, all bought within 90 days of a regulatory audit, average deal $18K. Two are healthcare companies that took twice as long to close. The model profiles the financial services pattern as the primary ICP, flags the healthcare deals as a secondary pattern worth testing separately, and identifies 'recently audited or expecting an audit' as the buying trigger. Her one-sentence ICP: 'Mid-size financial services firms facing their first serious regulatory audit without a dedicated compliance team.' She rebuilds her Apollo list around it and reply rates double within a month.
Best fit
This prompt is one gear in a bigger machine. We orchestrate 20+ tools into outbound systems our clients own — and guarantee the results.
Apply for a Pilot Spot → →Frequently asked questions
List your 10 best customers — best meaning profitable, fast to close, and retained — and look for what they share: industry, size, structure, and especially the trigger event that made them buy. This prompt automates the pattern-finding and forces tight ranges, so you end up with an ICP that excludes most companies, which is exactly what a useful ICP does.
More icp, targeting & lead scoring prompts
Design a lead scoring model with firmographic and behavioral points
Produces a complete, implementable lead scoring model: firmographic fit points, behavioral intent points with decay and negative scoring, threshold bands tied to actions, and a monthly validation loop. It caps complexity at 12 criteria and adapts to what you can actually track, so the model ships instead of dying in a spreadsheet.
Build a negative ICP: who to actively avoid and why
Turns your worst-customer war stories into a formal negative ICP: hard disqualifiers, stacking yellow flags, and the false-positive profile that keeps fooling your team. Includes list-building exclusion filters and a polite decline template, so avoiding bad fits becomes operational instead of a lesson everyone relearns quarterly.
Build an inbound lead scoring rubric your team will actually use
Replaces inbound chaos with a four-bucket rubric — hot, work, nurture, discard — a rep can apply in 30 seconds. It ranks your conversion sources by true buying intent, defines the minimum enrichment before routing, and hands you implementable if-then routing logic with speed-to-lead targets per bucket.