This ideal customer profile template turns "our best customers" into a scored worksheet that sales, marketing and your website can all use. It has five parts: weighted fit criteria, a scoring rule, A/B/C tiers, an anti-ICP list of disqualifiers, and a method for checking the whole thing against deals you have actually won and lost. Copy the tables into a spreadsheet, or download the template as CSV.
For the thinking behind an ICP (and how it differs from a buyer persona), read our guide on how to build an ideal customer profile. This page is the worksheet.
What Most Ideal Customer Profile Templates Get Wrong
Most ICP templates are a list of boxes: industry, size, revenue, pains. They describe a company, but they do not let you rank two real accounts against each other. Our view: an ICP is only useful if it produces a number or a tier for any account in your CRM, and if that tier predicts win rate. That is why this template has weights and a validation step. A profile that has never been checked against closed-lost deals is a hypothesis, not an ICP.
Part 1: Fit Criteria and Weights
Pick five to eight criteria you can actually get data for. Give each a weight so the weights add up to 100. Define what counts as a strong and a partial fit before you score anything.
| Criterion | Weight | Strong fit (full points) | Partial fit (half points) | No fit (0) | Data source |
|---|---|---|---|---|---|
| Industry | CRM, enrichment, NAICS mapping | ||||
| Company size (employees) | Enrichment | ||||
| Region / country | CRM, IP country | ||||
| Technology in use | Technographic data, sales discovery | ||||
| Team or function exists | LinkedIn, job posts | ||||
| Trigger event (last 6 months) | News, hiring, funding data | ||||
| Budget signal | Funding stage, revenue band | ||||
| Total | 100 | ||||
Rules for setting weights:
- Start from differences between won and lost deals, not from opinion. If 80% of won deals use a specific CRM and lost deals are split evenly, technology deserves a heavy weight.
- No single criterion above 30. Otherwise one field decides the tier and you have built a filter, not a score.
- Only include criteria you can fill for at least most accounts. A criterion that is blank for most of your CRM adds noise. For field definitions, see the glossary entries for firmographic data and technographics.
Part 2: Scoring Rule and Tiers
Each criterion scores its full weight for a strong fit, half its weight for a partial fit and 0 for no fit. Add them up for a score between 0 and 100.
| Tier | Score | Sales motion | Website treatment |
|---|---|---|---|
| A | 75 to 100 | Named-account outreach, senior rep | Personalized hero, matching proof, direct meeting CTA |
| B | 50 to 74 | Inbound follow-up, nurture | Segment-level personalization (industry or size) |
| C | Below 50 | Self-serve or no active pursuit | Default site |
| Excluded | Any anti-ICP rule hit | Do not pursue | Default site, no sales alerts |
Part 3: Anti-ICP (Disqualifiers)
The anti-ICP is a short list of conditions that exclude an account regardless of score. It is often more useful than the ICP itself, because it stops sales time going to deals that look good and never close or churn fast.
| Disqualifier | Evidence that justifies it | Applies to |
|---|---|---|
| Example: fewer than 10 employees | Example: most such customers churned within the first year | Outbound and paid targeting |
| Example: regulated sector requiring on-premise hosting | Example: product is cloud only | All motions |
| Example: existing contract with a competitor that runs another 2+ years | Example: sales cycles stalled at procurement | Outbound |
Part 4: Filled Example (Fictional Company)
Example only. "Example Co" is a fictional company selling scheduling software to field service businesses (installers, maintenance contractors). All numbers below are illustrative.
| Criterion | Weight | Strong fit | Partial fit | No fit |
|---|---|---|---|---|
| Industry | 25 | Building equipment contractors (NAICS 2382) | Other specialty trade contractors (NAICS 238, rest) | Anything else |
| Company size | 20 | 50 to 500 employees | 20 to 49 or 501 to 1,000 | Under 20 or over 1,000 |
| Field technicians | 20 | 15 or more technicians | 5 to 14 | Fewer than 5 |
| Technology in use | 15 | Uses a CRM we integrate with | Uses a different CRM | Spreadsheets only |
| Region | 10 | Nordics and DACH | Rest of EU | Outside EU |
| Trigger event | 10 | Hiring dispatchers or opening a new branch | Hiring technicians | None seen |
| Total | 100 | |||
Scoring two Example accounts:
| Criterion | Account 1: HVAC contractor, 140 staff, 40 technicians, integrated CRM, Germany, hiring a dispatcher | Account 2: electrical contractor, 35 staff, 8 technicians, spreadsheets, France, no trigger |
|---|---|---|
| Industry (25) | Strong: 25 | Strong: 25 |
| Company size (20) | Strong: 20 | Partial: 10 |
| Field technicians (20) | Strong: 20 | Partial: 10 |
| Technology (15) | Strong: 15 | No fit: 0 |
| Region (10) | Strong: 10 | Partial: 5 |
| Trigger (10) | Strong: 10 | No fit: 0 |
| Score and tier | 100, Tier A | 50, Tier B |
Both are in the right industry. Only the score shows that Account 1 deserves a senior rep this week and Account 2 belongs in nurture.
Part 5: Validate Against Won and Lost Deals
Do this before anyone uses the tiers. It takes an afternoon with a CRM export.
- Export closed deals. All closed-won and closed-lost opportunities from the last 12 to 24 months, with the fields your criteria need. Exclude deals lost for reasons unrelated to fit only if you can say so consistently (for example "no decision" stays in; "duplicate record" goes out).
- Score every deal with the account data as it was when the deal was created, if your CRM keeps history. Today's employee count is a reasonable fallback.
- Calculate win rate by tier. Won divided by (won plus lost) for A, B and C.
- Check the separation. Tier A should win clearly more often than B, and B more often than C. If two tiers win at similar rates, the weights are not doing their job.
- Look at the misfits. Read the Tier A losses and the Tier C wins one by one. They tell you which criterion is missing or overweighted.
- Adjust and re-score once. Change weights or definitions, re-run steps 2 to 4, and stop. Tuning repeatedly to the same history overfits it.
- Set a review date. Re-validate every six months with the new closed deals.
Example result for Example Co, 200 closed opportunities scored retroactively:
| Tier | Opportunities | Won | Win rate |
|---|---|---|---|
| A | 60 | 21 | 35% |
| B | 80 | 14 | 17.5% |
| C | 60 | 3 | 5% |
The tiers separate clearly: an A-tier opportunity wins twice as often as a B and seven times as often as a C in this Example. If your real numbers look flat across tiers, do not roll the ICP out yet. With small samples, treat the differences as directional: 3 wins out of 60 can move a lot with a handful of deals.
Using the ICP on Your Website
Once the tiers validate, they become segments. Tier A accounts can trigger alerts and account-level pages, Tier B gets industry or size personalization, Tier C gets the default site. The segmentation strategy guide covers how to turn tiers into website segments, and the lead scoring model template adds behavioural scoring on top of this fit score. To see which tiered accounts are visiting, you need company-level data from visitor identification. For the account-level scoring that combines fit and intent, see our post on the account scoring model.
Frequently Asked Questions
What should an ideal customer profile template include?
Fit criteria you can get data for (industry, size, region, technology, triggers), a weight for each, definitions of strong and partial fit, a scoring rule that produces tiers, an anti-ICP list of disqualifiers, and a validation step against won and lost deals. Without weights and validation, an ICP describes customers but cannot rank accounts.
How is an ICP different from a buyer persona?
An ICP describes the company that is a good fit: industry, size, technology, situation. A buyer persona describes a person inside that company: role, goals, objections. You choose accounts with the ICP and shape messaging with personas. Most B2B teams need one ICP with tiers and two or three personas per tier.
How many fit criteria should an ICP have?
Five to eight. Fewer than five usually means one or two fields dominate and the tiers behave like a simple filter. More than eight usually means some criteria are blank for most accounts, which adds noise. Keep only criteria that differ between your won and lost deals and that you can fill reliably.
How often should I update my ideal customer profile?
Re-validate every six months against new closed deals, and immediately after a major change such as a new product line, a new market or a pricing change. Adjust weights once per review rather than continuously, so the tiers stay stable enough for sales to trust them.