Creatives Takeover

How to Validate a Startup Idea Before You Build

To validate a startup idea, pick one specific customer, confirm they already struggle with the problem, and get them to take a small action, such as booking a call, joining a waitlist or paying a deposit, before you build. Validation is evidence of behavior, not opinions about your idea.

By Javier Alonso, Founder & CEO. Updated .

Step by step

  1. Write down the riskiest assumption

    Every idea rests on a few beliefs. Usually the riskiest is that a particular group has this problem badly enough to change what they do. Write it as one sentence you could prove wrong, for example: "Freelance designers lose at least two hours a week chasing late invoices."

  2. Choose one customer segment

    Validation fails when the customer is "everyone". Pick the group with the sharpest pain that you can reach this week. Our ideal customer profile entry explains how to narrow it.

  3. Interview for past behavior

    Talk to 5 to 10 people in that segment. Ask about the last time the problem happened, what they did, what it cost and what they have already tried. Do not pitch. Use the questions in customer interview questions.

  4. Test the promise with a small action

    Put the outcome you promise on a simple page and ask for a real step: join a waitlist, book a call, reply to an email or pay a refundable deposit. The waitlist or MVP guide covers how to set this up.

  5. Score the evidence against a rule you set in advance

    Before you look at the results, write the threshold that would justify building, for example "8 of 10 interviewees describe the problem unprompted and 30% of qualified visitors join the waitlist". Then decide: build, narrow the segment, change the promise or stop.

What counts as validation

Validation is any evidence that a specific customer has a painful problem and will act to solve it. The strongest evidence is behavior that costs the customer something: time, money, reputation or effort. The weakest is a compliment.

A useful way to rank what you collect:

  • Strong: pre-orders, paid pilots, deposits, people switching from a tool they pay for, repeat usage of a rough version.
  • Medium: waitlist signups from people who match your customer profile, booked calls, people introducing you to colleagues with the same problem.
  • Weak: "I would use that", likes, survey answers about future behavior, interest from friends and family.

If most of your evidence sits in the weak row, you have not validated the idea yet, however positive it feels.

Validate the pain before the solution

Founders often test the solution first and the problem second, which is backwards. If nobody struggles with the problem, the quality of your solution does not matter. Ask how people handle the problem today. A painful workaround, such as a spreadsheet maintained every Friday, an assistant hired for the task or a tool bought and abandoned, is a strong sign the problem is real and that budget may exist.

Once the pain is clear, test whether your proposed outcome is the one they want. Only then does it make sense to test a specific product.

Example

A founder wants to build scheduling software for physiotherapy clinics. Instead of building, she calls 12 clinic managers. Nine describe double-bookings and no-shows without being prompted, and six already pay for a reminder service they dislike. She builds a one-page offer promising "fewer no-shows without a new booking system" and asks managers to book a 20-minute setup call. Seven of the twelve book. That is enough evidence to build a narrow first version for clinic managers, and not for the yoga studios and dentists she originally had in mind.

Common mistakes

  • Asking people if they like the idea. Most people are polite. Ask about their past instead.
  • Testing with the wrong people. Feedback from people who do not have the problem is noise.
  • Moving the goalposts. If you decide the threshold after seeing the results, you will always find a reason to build.
  • Treating a waitlist as product-market fit. Signups show interest in a promise. Product-market fit needs usage, retention or payment.

For your role

Founders
Run this before you commit months of work or raise money on the idea. The evidence also becomes the first slides of your pitch deck.
Mentors
Ask founders for their written decision rule and the evidence against it. It turns a vague catch-up into a concrete review.
Accelerators & incubators
Validation evidence is a better progress signal than attendance. Ask every team for interviews logged, the action they asked customers to take and the result.

Primary sources

Common questions

What is the fastest way to validate a startup idea?
Define one narrow customer, talk to 5 to 10 of them about the problem, then ask them to take a small step such as booking a call or joining a waitlist. You can do this in one or two weeks without building anything.
How many customer interviews do I need?
There is no fixed number, but patterns usually appear after 5 to 10 conversations with the same segment. Stop when new interviews mostly repeat what you have heard.
Does a waitlist prove product-market fit?
No. A waitlist shows interest in a promise. Product-market fit needs stronger evidence, such as repeat usage, retention or customers paying.
Can I validate an idea without a product?
Yes. Interviews, a landing page, a manual version of the service or a clickable demo can all produce real evidence before any code exists.

Put it into practice

Validate turns these steps into a structured plan for testing demand and logging the evidence.

Plan your validation

Keep learning