Ideal customer profile template
Most ICPs are written in a workshop: the leadership team describes the customers they wish they had, and the result is broad enough that every account in the CRM qualifies. An ICP is only useful if it excludes accounts, and the only defensible way to decide what to exclude is to look at who you actually win, who you lose, and who leaves. This template starts from that data, turns it into firmographic and technographic criteria, and ends with a scoring model that sales and marketing can apply to any account.
Download it
What you get
- A segment analysis table that puts win rate, deal size, and 12-month churn side by side
- A firmographic and technographic criteria table with the evidence behind each criterion
- A 100-point account scoring model with a worked example that adds up
- Tier thresholds and an exclusion list so the ICP says no as well as yes
- A checklist for pulling the data and a review cadence to keep the ICP current
Who it's for
- RevOps leads rebuilding an ICP that no longer matches who buys
- Marketing ops teams deciding which accounts get budget
- Sales leaders drawing territories and target account lists
What's inside
7 sections, each listed with its own size. The first is previewed below; the remaining 6 come with the download, with blank rows ready to fill in.
| Section | Size | Previewed |
|---|---|---|
| Data pull | 8-point checklist | Below |
| Segment analysis (example data) | 7 columns, 3 worked example rows | In the download |
| Firmographic and technographic criteria | 5 columns, 6 worked example rows | In the download |
| Account scoring model (100 points) | 5 columns, 5 worked example rows | In the download |
| Scoring worked example | 8 columns, 3 worked example rows | In the download |
| ICP statement and exclusions | 5 fields to complete | In the download |
| Why opinion-based ICPs fail | Guidance notes | In the download |
Preview of section 1
Data pull
Gather this before any workshop. Most of it comes from the CRM; churn usually comes from finance or CS.
- All closed-won opportunities from the last eight quarters, one row per deal
- All closed-lost opportunities from the same period, with loss reason
- Account employee count, industry, and region at the time of the deal, not today
How to use it
1. Pull two years of closed deals and churn
Export every closed-won and closed-lost opportunity from the last eight quarters, with account size, industry, region, source, and deal size. Join churn and downgrade data from finance or CS. Opinion comes in later, as a sense check, not as the starting point.
2. Compare segments on three numbers together
Win rate, median deal size, and 12-month logo churn. A segment with a high win rate and high churn is a trap: you close it easily and lose it within a year. The ICP is where all three are acceptable at once.
3. Turn patterns into criteria someone can check
Each criterion must be something a rep or an enrichment tool can verify before the first call: employee count, industry, CRM in use, number of sales reps, hiring activity. 'Innovative culture' is not a criterion.
4. Score, tier, and write the exclusions
Weight each criterion by how strongly it separated winners from losers in your data, not by how important it feels. Set tier thresholds, then write down the accounts you will not pursue. If nothing is excluded, you do not have an ICP.
5. Re-run the analysis twice a year
Products, pricing, and markets move. Re-pull the data every six months and after any major packaging change, and compare the new segment table with the old one before changing the score.
Frequently asked questions
What should an ideal customer profile include?
Firmographic criteria (size, industry, region), technographic criteria (tools in use), trigger events, the primary buyer persona, a scoring model, and an explicit list of exclusions. Each criterion should be backed by your own win, loss, and churn data.
What is the difference between an ICP and a buyer persona?
An ICP describes the company that is a good fit: size, industry, tools, triggers. A buyer persona describes the person inside that company you sell to. You need both, but the ICP comes first because it decides which accounts are worth a persona's time.
How do you build an ICP from data?
Export closed-won and closed-lost deals from the last two years, join 12-month churn, and compare segments on win rate, median deal size, and churn together. The segments that are acceptable on all three become the ICP, and the criteria that separated them become the scoring model.
How often should you update your ICP?
Re-run the analysis every six months and after any pricing, packaging, or product change that could shift who buys. Keep the old segment table so you can see what moved.