How to Build an Ideal Customer Profile When You Have Zero Customer Data

Ask ten founders how they built their ideal customer profile, and nine will describe a person: age range, job title, company size, maybe a stock photo in their head. That's a buyer persona wearing an ICP's name tag. The founders who get to product-market fit faster tend to start somewhere else entirely — with the problem, not the person — and treat the whole thing as a hypothesis to test rather than a document to finish.

Start from the problem, not the person

When you have no customer data, the most stable starting point isn't "who" — it's "what painful, expensive, or urgent problem does this solve, and for whom is it most painful?" Rank potential segments across three axes:

• Severity — is this a vitamin or a painkiller for them?

• Frequency — do they hit this problem daily or weekly, or once a year?

• Cost of inaction — what does it cost them in money, time, or risk to not solve it?

The segment where all three are highest is your best early hypothesis, even with zero firmographic data behind it. An expense-tracking tool might look like it's for "small businesses," but the segment with the sharpest pain is finance teams closing books manually every month — that's frequency and cost of inaction stacked on top of each other. A solar installer might assume homeowners are the market, when the sharper pain sits with new-home developers who legally can't close a sale without hitting a solar mandate. Same framework, completely different-looking answer depending on who's actually bleeding.

Founder-market fit is a hypothesis, not a verdict

If you or your team came from a specific industry, role, or pain point, that lived experience is real data — just qualitative, not statistical. Advanced early-stage teams treat founder insight as a starting hypothesis to falsify, not a truth to confirm. The danger is stopping at "I used to have this problem" without ever testing whether it generalizes past your own experience.

Talk to 10 to 15 people before you touch demographics

Structured discovery interviews — not surveys — ask about a prospect's current workaround: what they've already tried, and what they've already paid for or built internally. People who have spent real money or effort solving a problem are a far stronger ICP signal than people who say a product sounds useful. That's the vitamin-versus-painkiller test made visible through behavior instead of opinion. Someone who's cobbled together three spreadsheets and a Zapier automation to solve a problem is telling you something a survey response never will.

Write the hypothesis down, then test it like a scientist

Write the ICP hypothesis down with explicit assumptions — "we believe X-sized companies in Y industry with Z trigger will convert because of [reasoning]." Then design your first 90 days of go-to-market as a test of that hypothesis, tracking which assumptions held and which didn't. Compare who actually converts and engages against who you predicted, and revise. This is closer to a scientific method loop than a marketing exercise, and it's the single biggest differentiator between teams that find PMF-adjacent ICP fast and teams that spend a year marketing to a guess.

Borrow signal you don't have yet: proxy data sources

You don't need your own customer data to start narrowing the field:

• Competitor customer lists — G2 and Capterra reviewers, case studies, and LinkedIn followers of competitors reverse-engineer who's already buying in the category.

• Job postings — companies hiring for a role your product supports are self-identifying a need and a budget line for it.

• Community and forum activity — find where the pain is actually being discussed (Reddit, niche Slack or Discord communities, industry forums) and see who's posting.

Don't let the ICP be perfect — close what you can close first

It's rare to have a fully-formed picture of your ideal customer before you've won any customers at all.

"For companies that are pre-revenue, I would always advise to close the deals you can close, and get a batch of customers that are paying you before you spend much time thinking about your ICP."

Your earliest sales aren't wasted even if they don't fit the eventual profile — they're your first pattern to study. The point of the framework above isn't to delay selling until the ICP is perfect. It's to make sure that once you do have customers, you're looking at the right signal: not who resembles you, but who was already in enough pain to act.

Seeing it play out end to end

We walked through this exact process in detail with a real-feeling case study — an installer who almost built its whole GTM strategy around the wrong segment. Read how a solar company found its actual ICP →