Most small teams have an ideal customer profile that reads “B2B companies that need marketing” or “growing businesses that want better sales”. That sentence describes half the internet, so the team still takes every call and wonders why the pipeline is full of people who never buy.
An ideal customer profile (ICP) is a one-page description of the companies that get the most value from what you sell, plus the events that make them buy now and the signs that rule a company out. Build it from your best recent wins, then record the fit on every company in Fluid CRM.
A small team needs one page and four blocks: firmographics (the company facts, such as industry, size and location), trigger events, pains and disqualifiers. Pains and disqualifiers do more work than firmographics, and disqualifiers are the block most teams leave blank.
The page is also only half the job. An ICP that lives in a Google Doc gets read once, while one that sits on every company in your startup CRM gets used every time someone decides whether a lead deserves a follow-up.
What an Ideal Customer Profile Is, and What It Is Not
An ideal customer profile describes a type of company, not a person. It says which businesses buy from you fastest, pay you well, stay longest and are easy to serve, and it says it in facts you can check before the first call.
Three terms get mixed up all the time, and the mix-up is why so many ICPs end up useless.
| Term | What it describes | Example |
|---|---|---|
| Ideal customer profile | The company that is your best fit | A 10 to 50 person B2B software company in the US that just hired its first marketer |
| Buyer persona | The person inside that company who signs or champions the deal | The new marketing lead, six weeks into the job, asked to show pipeline by the end of the quarter |
| Target market | Everyone you could sell to | Every B2B software company in North America |
The target market is everyone you could sell to, the ICP is the slice you go after first and the persona is the person you talk to there. You need the middle one most.
For a team of one to five people, a written buyer persona is optional. You talk to every buyer yourself, so you already know who they are. An ICP is not optional, because without it you can’t say no to the wrong company before you’ve spent three calls and a proposal on it.
The One-Page ICP Template (Copy This)
This worksheet is the whole template. Copy the table into a doc or a spreadsheet and fill in the third column from real deals, not from what you hope is true.
| Block | What to write | Example answer | How to check it on a new lead |
|---|---|---|---|
| Firmographics | Industry, headcount band, location, how they sell, what they spend on this problem today | B2B service firms, 5 to 40 people, US and UK, sell through sales calls, already pay a freelancer or tool for this | Their website, LinkedIn headcount, a two-minute look at their team page |
| Trigger events | What happened in the 90 days before your best customers bought | Lost their biggest client, hired a first salesperson, launched in a new market, a tool contract ending | A job post, a LinkedIn announcement or the first question on the call |
| Pains | The problem in the buyer’s own words, taken from calls and emails | “We’re busy but nothing new is coming in” or “Every month starts from zero” | Whether they say it back to you unprompted on the first call |
| Disqualifiers | At least three signs that a company looks right on paper but isn’t | Sells only to consumers, wants results in two weeks, nobody owns the project on their side | One question each, asked before you send a proposal |
If the table feels like too much, start with the one-sentence version and grow it into the table later:
“We work best with [type of company] of [size] in [place] who just [trigger event] and are stuck with [pain]. We say no when [disqualifier].”
There’s no download behind this, on purpose. Copy the table, because a template you typed yourself is a template you’ve read. If your leads live in Google Sheets or Excel today, add an “ICP fit” column to your copy of our free lead tracker template so every row gets checked the day it arrives.
How to write a negative ICP
Some teams call the disqualifier block a negative ICP: the companies you turn away even when they’re ready to pay. It deserves the same space as the rest of the page, because it’s the only block that saves you time on the first day.
Every disqualifier has to be checkable in a minute by anyone on your team. “Too small” fails that test, while “fewer than five employees and no one in sales” passes it, because a look at LinkedIn settles it.
Five ways an ICP goes wrong
- It’s too broad. “B2B companies with 10 to 500 employees” rules nobody out, so it changes nothing about who you call
- It’s a 15-field worksheet. It gets filled once and never again. Four blocks, a few lines each, is enough for a team under 10
- It lists persona facts. Age, hobbies and favorite podcasts describe a person, and you can’t check them on a company website before the call
- It has no disqualifiers. Without them the ICP only ever says yes
- It still describes last year’s offer. Your best customers change when your offer does, so redo the six steps below whenever you change what you sell or what you charge
How to Build Your ICP From Your Last 10 Wins
The best ICP comes from the customers you already have, not from a brainstorm. Six steps, about two hours the first time, and most of that is reading your own deals.
1. List your 10 most recent customers
Open your CRM or your spreadsheet and list the 10 customers who signed most recently. Add up to five lost deals that looked perfect on paper, because those show where your firmographics are lying to you.
Ten is enough to see a pattern and few enough to finish in one sitting. If you have 40 customers, use the most recent 10, since your offer has probably changed since the early ones.
2. Mark each one on four things
Give every customer a simple yes or no on four questions: was the margin good, did it close faster than your average, was the work easy to deliver and would you take this exact customer again tomorrow?
The fourth question matters most. A client who paid well but drained your team every week is not your ideal customer, and writing them into the ICP means you’ll go find five more of them.
3. Keep the top five and write down what they share
Take the five customers with the most yeses and compare them side by side. Only write down what at least four of the five have in common: the industry, the headcount band, the way they sell, the tool they used before you.
Can you name what your five best clients had in common before they signed? If the honest answer is “they all had budget”, keep digging, because budget is a result of fit, not a description of it.
4. Find the trigger event
Ask each of the five, or check your notes, what changed in the 90 days before they contacted you. A new hire, a lost client, a funding round, a price rise from their current vendor.
This is the block that tells you when to reach out, not only who. I’d ask “what made you start looking now?” on every first call, and our discovery call structure has a place for it near the start.
5. Write the disqualifiers from your worst three
Now look at the three worst customers on your list and the losses that looked right. What do they share that your best five don’t? That’s your negative ICP, written with the one-minute test above.
6. Test it on your next 20 leads
Run every new lead through the template for the next few weeks and note which ones fit. Then look at which of them turned into real conversations and fix whatever block was wrong. Once fit is clear, our lead scoring model adds two intent questions to decide who gets called first.
An ICP is a guess you keep checking. The first version is usually too broad on firmographics and too thin on disqualifiers.
What I saw at Fenixtal
At Fenixtal, my B2B outbound agency, we booked 1,250+ sales calls, and every campaign started with research before a single email went out. For a podcast production agency, we researched the market on the client’s own criteria, found the relevant decision-makers in every target company and prioritized a few signals that showed a company was ready to buy.
That client went from two to five low-quality leads a month to 101 sales calls and 1.2M€ in pipeline over 12 months, with companies including Nordea and Rovio. In my experience the list and the timing did more of that work than the wording of any email.
Another problem I saw often in the agencies we spoke to was no clear offer. They did a bit of everything for everyone, so no ICP could narrow anything down.
When you have fewer than five customers
Write the template anyway, as a guess, and label it a guess. Base it on the people who replied fastest and asked the best questions, then test it the same way. Our guide to founder-led sales covers how to sell while the guess is still forming.
ICP Examples: An Agency, a Consultancy and a B2B SaaS
The first three examples below are illustrations, not real companies. Each one shows the four blocks filled in for a different kind of small B2B seller, so you can see how specific a good answer is.
Example 1: a paid social agency (illustration)
A six-person agency that runs LinkedIn and Meta ads for B2B companies.
- Firmographics: B2B software and service companies, 20 to 150 employees, US, already spending at least $5,000 a month on ads, with an in-house marketer but no paid specialist
- Trigger events: a new head of marketing in the last 90 days, a funding round, a freelancer who just left
- Pains: “We spend on ads every month and can’t tell which ones bring sales calls”
- Disqualifiers: monthly ad spend under $2,000, sells only to consumers, wants a one-month trial project, the founder approves every ad by hand
Example 2: an operations consultancy (illustration)
Two partners who fix order-to-delivery processes for small manufacturers.
- Firmographics: manufacturers with 30 to 200 employees and $5M to $40M in revenue, Midwest US, running on spreadsheets or an old ERP, family or founder owned
- Trigger events: a second-generation owner taking over, a failed ERP project, a big customer threatening to leave over late deliveries
- Pains: “We grow every year and lose money on half the orders”
- Disqualifiers: private-equity owned with a corporate operations team, wants software picked rather than processes fixed, no one on site who can own the changes
Example 3: a small B2B SaaS (illustration)
A four-person team selling booking and reminder software to physical therapy clinics.
- Firmographics: independent clinics with 3 to 15 therapists, US, front desk staff answering phones, using paper or a basic calendar for bookings
- Trigger events: opening a second location, a front desk hire leaving, a spike in no-shows the owner just noticed
- Pains: “Our front desk spends half the day on the phone moving appointments”
- Disqualifiers: part of a hospital network, needs a full electronic health record, a single therapist with no front desk
A real one: Fluid CRM’s own ICP
Fluid CRM fits small B2B teams who sell, meaning they move deals through stages toward a yes or a no. If most of your work is project delivery or account management, Fluid CRM is not the tool for it, and an honest ICP says so.
How to Use Your ICP in a CRM Pipeline
An ICP pays off only when it’s recorded where the team already works, on the company itself, the day a lead arrives. Four habits make that happen without anyone opening the ICP doc.
Record fit and trigger on the company
An ICP describes companies, so the company record is where fit belongs. Add two custom fields to companies: “ICP fit”, a text field holding yes, partly or no, and “Trigger”, a short note on what just happened at that company.
In Fluid CRM custom fields work on deals, contacts and companies, so you record the fit once, on the company, not again on every deal. Every contact you know there sits under the company too. Nicolas Virtsonis, a verified customer who runs a specialist B2B marketing agency, said this about that view:
“I will touch up on the nested contacts within the companies / contacts view which is extremely practical, especially for ABM strategies and complex sales.”
Label the deals you take anyway
Add a label such as “Outside ICP” to any deal that hits a disqualifier but that you’re taking anyway. Sometimes that’s the right call, a quiet month or a big name you want on your client list, and the label keeps the decision on the record so you can see later whether it was worth it.
Ask the ICP questions in your lead form
If leads come through a form on your site, ask two or three of the firmographic questions there: company size, industry, what they use today. With Fluid CRM’s lead capture forms, the answers land in the new deal’s notes, so you can check fit before the first reply.
Close misfits early, with a reason
When a lead clearly hits a disqualifier, mark it lost the same day and pick a lost reason such as “Not a fit”. Every misfit left open is another follow-up someone does for nothing.
Fluid CRM asks for a reason when you mark a deal lost, and the workspace owner can rename one of the reasons to “Not a fit” if it isn’t on the list. Reports shows lost reasons by date range, so a rising “Not a fit” count tells you your outreach or your form is letting the wrong companies in. Our weekly sales report template puts that lost-reason count in front of the team.
Once a quarter, open your recent wins, read the fit and trigger on each company and rewrite the template where the real customers disagree with it.
Ideal Customer Profile FAQ
An ICP describes the company that fits you best, such as its industry, size and the event that makes it buy. A buyer persona describes the person inside that company who signs or champions the deal. Small teams need the ICP first and can often skip a written persona.
One, or two at most if you sell two clearly different offers. Every extra ICP splits your outreach, your case studies and your attention. If you have three, you probably have one real ICP and two hopes.
Review it every quarter against your recent wins and rewrite it whenever you change your offer or your price. A quick check takes 20 minutes. A full rebuild from your last 10 wins takes about two hours.
Yes, but treat it as a guess and label it that way. Base it on the people who replied fastest and asked the best questions, then test it on your next 20 leads. Rewrite it after your first five customers sign.
Write Your ICP Before the Next 10 Leads Arrive
Block out two hours this week, list your last 10 customers and fill in the four blocks. Then add the fit and trigger fields to your startup CRM so every new company gets judged against the ICP on day one, not after the third call.
If you want a steady flow of companies to check against it, our pipeline generation strategy is the outbound process I used at Fenixtal, and our guide on how to get clients covers the weekly routine for working alone. Fluid CRM keeps the fit, the labels and the lost reasons in one plan.
Start your 7-day free trial here, no credit card required.