Ideal Customer Profile Template: Build an ICP for Outbound (2026)

Most advice on building an ideal customer profile sounds the same. Pick an industry. Add a company size. Add a region. Fill in a template in five minutes. That is exactly why most ICPs are useless for outbound.

An ICP that says "B2B SaaS, 50 to 500 employees, North America" describes tens of thousands of companies. It doesn't tell you which ones to contact this week. It doesn't tell you why they would reply. It ends up on a kickoff slide, not in a search filter.

The problem isn't that teams lack an ICP. The problem is an ICP that describes a market instead of pointing at accounts.

And that market is never ready to buy all at once. According to LinkedIn's B2B Institute, drawing on research with the Ehrenberg-Bass Institute, roughly 95% of potential buyers are out of market at any given moment. Only about 5% are ready to buy. An ICP that ignores timing sends 95% of your outreach to people with no reason to answer.

This guide gives you a method and a fill-in ideal customer profile template built for outbound. Two pillars: fit criteria you can actually filter on and signals that tell you when to act.

What an ideal customer profile really is

An ideal customer profile describes the type of company that gets the most value from what you sell, and brings you the most value back. It defines an account, not a person.

ICP vs. persona vs. segment

These three terms get mixed up constantly. They do different jobs.

Define the ICP first. Personas come next, inside the accounts you've kept. Flip the order and you get well-written messages sent to the wrong companies.

Three tests for a good ICP

A good ICP passes three simple tests.

• The sorting test: two reps reading the profile classify the same company the same way.
• The filter test: every criterion maps to a filter in your search tool or CRM.
• The "no" test: the profile says clearly who you do not target.

If your ICP never lets you say no to a prospect, it isn't an ICP. It's a market description.

Start from your best customers, not your gut

Most ICPs are written from memory in a meeting room. They describe the customer the team wishes it had, not the one who actually buys.

The right starting point: your CRM

The most reliable method starts with data. HubSpot's ICP guide recommends beginning with your best existing customers in the CRM: the ones with high revenue, low churn and strong product adoption.

In practice, export the customers you closed over the last 12 to 24 months. Sort them into three groups:

• Best fits: good deal size, short cycle, low churn, easy to serve.
• Average fits: profitable, but expensive to sell or support.
• Mistakes: closed, then quickly lost, unprofitable or draining.

Build the ICP from the first group. Use the third group to write your exclusions.

Interview them, then look for patterns

Data tells you who buys. It rarely tells you why. The same HubSpot guide suggests interviewing at least ten of those strong customers: how they found you, what drove the purchase, what results they got.

The most useful question is usually the simplest: "What happened on your side right before you started looking for a solution?" The answers become your future signals. A new hire. A leadership change. Growth that broke an existing process.

The "anyone could be a customer" trap

An eight-person SEO agency can technically serve an online store, a SaaS startup and a law firm. That's not a reason to target all three.

The broader the ICP, the more generic the message. And generic messages are exactly what buyers ignore.

The criteria that matter for outbound

Not all criteria are equal. In outbound, a criterion only counts if you can observe it before you send the first message.

Firmographics: the foundation

These are the basics. They filter broadly, but they filter fast.

• Industry: specific, not "services". "Accounting firms" beats "finance".
• Headcount: a range that matches what you can actually deliver.
• Geography: where you can sell, deliver and invoice.
• Business model: B2B, B2C, marketplace, subscription, project-based.

Technographics and team structure: the fine filter

These separate accounts that look like your customers from accounts that really are like them.

• Tools in use: an existing CRM, a specific platform, or the absence of one.
• Existing team: a sales team, a marketing lead, an HR function.
• Maturity: structured processes or everything still to build.

HubSpot also lists the technology stack, the buying process and the buying committee structure among the things to document. At a small company, the committee is often two people. Knowing that changes your sequence.

Exclusion criteria

This is the part almost everyone skips. It saves the most time.

A usable ideal customer profile template

Here's the template I recommend. It fits on one page. Every row becomes a filter, a signal or a message angle.

The template

Weak vs. strong: what the difference looks like

Weak "Fast-growing tech companies that need to hire."

Strong "Tech startups with 20 to 200 employees in the UK or Ireland, no in-house tech recruiter, and either 3+ open developer roles or a funding round in the last 6 months."

The first version describes a market. The second one goes straight into a search tool and returns a short list.

One ICP, several segments

A single ICP can hold two or three segments, each with its own trigger and angle. In the example above: startups that just raised, and startups hiring without a recent round. Same need. Different urgency, different message.

A good ICP doesn't describe your dream customer. It tells you who to contact on Monday morning, and why.

Turning the ICP into a prospect list

An ICP is only worth something if it produces a list. This is where most ICPs quietly die.

Map every criterion to a filter

Go through the profile line by line. For each criterion, write down the matching filter in your tool: industry, headcount, region, job title, time in role. Anything that maps to no filter becomes a qualification point you check by hand.

For job titles, a well-built Boolean search in LinkedIn Sales Navigator saves a lot of time. CTO OR "Chief Technology Officer" OR "VP Engineering" covers the variants without pulling in off-target profiles.

Enrich only what passes the first filters

Some criteria aren't visible on LinkedIn: tech stack, whether a team exists, revenue. That's where enrichment comes in. Our guide to B2B lead enrichment covers the methods and the rules to follow.

Rule of thumb: only enrich accounts that already pass your firmographic filters. Enriching a raw list is expensive and doesn't make it better.

Run a 20-account sanity check before launch

Before you send a single message, pick 20 random accounts from the list. Open each company page. Ask one question: "Would I genuinely want this customer?"

• More than 15 yes: the filters work. Launch.
• 10 to 15: one criterion is letting noise through. Find it.
• Fewer than 10: the ICP wasn't translated properly. Go back to the template.

It takes half an hour. It saves weeks of sequences sent to accounts that would never have closed. In Reakly, the acquisition engine lets you build and adjust these filters before a campaign goes live.

Time saved is selling time

A tight ICP isn't a marketing exercise. It's a capacity question. Salesforce's 2026 State of Sales report, based on a survey of 4,050 sales professionals in August and September 2025, found that reps spend only 40% of their week actually selling. The rest goes to data entry, prospecting and other non-selling work.

The same report says 47% of reps call cold outreach one of the worst parts of the job, and 47% say their team lacks the bandwidth to do it. Every off-target account you remove from the list is time handed back to real conversations.

Add timing: fit alone isn't enough

An account can match your ICP perfectly and not need you for two years. That's the limit of every static ICP.

Fit and intent are two separate axes

Fit tells you whether an account could buy. Intent tells you whether it could buy now. You need both.

This is where the B2B Institute's 95-5 rule really bites. If 95% of your ICP is out of market, the question isn't just "who?" but "who, right now?"

Signals to attach to your ICP

The triggers you uncovered in customer interviews become your signals. The most common in B2B:

• Funding: budget is available and growth pressure is on.
• Hiring: a job post reveals a need, sometimes exactly the one you solve.
• New role: a new decision-maker reviews tools and vendors in their first months.
• Engagement: a comment, a profile visit, a reaction to a post.

We broke down each signal and the message angle that fits it in our guide to B2B buying signals. In Reakly, intent signal detection cross-checks those signals against your ICP criteria, so only accounts with both fit and a trigger rise to the top.

Weak vs. strong: a message built on the ICP

Weak "Hi, we help tech companies hire faster. Do you have time for a call?"

Strong "Hi Claire, I noticed three backend roles open on your careers page since the June round. Without an in-house tech recruiter, that's usually where time-to-hire starts to slip. Is that what you're seeing?"

The second message isn't longer. It's written for one specific segment of the ICP, at the right moment.

The mistakes that make an ICP useless

The same mistakes show up in almost every ICP we see in the field.

• Writing it without data: it describes the dream customer, not the real one.
• Making it too broad: "growing SMBs" filters nothing.
• Skipping exclusions: the team burns hours on accounts that were never going to close.
• Confusing ICP and persona: the message is polished, but it goes to the wrong companies.
• Not wiring it into your tools: an ICP that only lives in a doc shapes no list.
• Never updating it: the market, the offer and the customers change. The ICP has to follow.

Review it every quarter

HubSpot recommends reviewing the ICP quarterly and treating it as a living document. In practice, a one-hour quarterly review is enough: which segments closed, which stalled, which new signals showed up in won deals.

Measure the ICP in your CRM

An ICP is validated by results, not by consensus. Tag each account with its segment in your prospecting CRM, then compare by segment:

A segment that replies a lot but never buys isn't a good segment. It's a curious one.

Conclusion: judge your ICP by the list it produces

Start from your best customers, not your wishes. Ask them what triggered the purchase. Write criteria you can type into a tool. Add exclusions. Tie each segment to a signal and an angle. Measure by segment in the CRM. Review every quarter.

A vague ICP fills lists. A sharp ICP fills calendars.

---

If you want to turn your ICP into accounts worth contacting this week, Reakly matches your targeting criteria with buying signals (funding, hiring, new roles, engagement), runs LinkedIn and email sequences on accounts that have both fit and timing, and tracks every conversation in the CRM.