How to Validate a Startup Idea Before You Build

To validate a startup idea, test the riskiest assumptions behind it with real potential customers before you build: first that the problem exists and matters (through problem interviews), then that people will commit to your proposed solution (through a pre-sale, pilot or deposit). Validation is about collecting commitments, not compliments. If you cannot get a handful of people in a narrow segment to give you time, data or money for a solution that does not fully exist yet, building it will rarely change their minds.

What you are actually validating

"Is this a good idea?" is too vague to test. Break the idea into assumptions and test the ones that would kill the business if they were wrong.

Assumption	Question it answers	How to test it
Problem	Does this pain exist and is it frequent or costly?	Problem interviews
Customer	Who exactly has it, and can we reach them?	Interviews, outreach response rates
Solution	Does our approach solve it better than the workaround?	Mock-ups, concierge test, prototype
Willingness to pay	Will they pay, and roughly how much?	Pricing conversations, pre-sales
Channel	Can we acquire customers at a reasonable cost?	Small outreach or ad tests

Rank your assumptions by two criteria: how uncertain they are and how fatal they would be. Test the most uncertain and most fatal one first. For most early ideas, that is the problem itself.

Step 1: Define a narrow target segment

Validation fails most often because the segment is too broad. "Small businesses" or "freelancers" includes people with completely different problems, budgets and buying processes. Their answers will contradict each other, and you will not learn anything clear.

Narrow it down until you could find 50 of these people by name:

• Broad: restaurants
• Better: independent restaurants with their own delivery
• Narrow enough: independent restaurants in one city that run their own delivery drivers and take orders by phone

You can widen the segment later. Starting narrow makes patterns visible.

Step 2: Run customer discovery interviews

Customer discovery interviews are structured conversations about the person's current behaviour. You are not pitching. You are learning how they experience the problem today.

Rules for useful interviews

• Ask about specific past events, not hypothetical futures.
• Let them talk; aim to speak less than a third of the time.
• Do not mention your solution until the end, if at all.
• Dig into costs: time, money, errors, stress, missed opportunities.
• Take notes in their words. Their phrasing becomes your marketing copy later.

Customer discovery interview questions

1. Tell me about the last time [problem] happened. What exactly did you do?
2. How often does that happen?
3. What does it cost you when it happens, in time, money or anything else?
4. How are you dealing with it today? What tools or workarounds do you use?
5. What have you tried before that did not work? Why not?
6. Have you spent money trying to solve this? On what?
7. Who else is involved when this comes up? Who decides on tools or budget?
8. If you could wave a magic wand, what would change?
9. Is there anyone else I should talk to about this?

Questions 6 and 7 are especially telling. Someone who has already paid for a partial solution has proven the problem is worth money. Knowing the decision maker tells you whether your interviewee can actually buy.

Questions to avoid

• "Would you use a product that does X?" (Most people say yes to be polite.)
• "How much would you pay for this?" (People guess, often low or high at random.)
• "Do you think this is a good idea?" (Opinions are not evidence.)

Step 3: Analyse the interviews honestly

After each interview, fill in a short summary while the conversation is fresh. After 10 to 15 interviews, look for patterns.

A simple interview scorecard:

Interviewee	Has problem?	Frequency	Current workaround	Spent money on it?	Strength (1–5)
A	Yes	Weekly	Spreadsheet	No	3
B	Yes	Daily	Hired a freelancer	Yes	5
C	Rarely	Monthly	Ignores it	No	1
D	Yes	Weekly	Paid tool, unhappy	Yes	5

What to look for: a cluster of interviewees who rate high on frequency and have already spent money or significant time on a workaround. If that cluster shares characteristics (company size, role, industry), you have found your early adopters.

Warning signs: people describe the problem as "annoying" but do nothing about it; every interviewee has a different main pain; nobody can name a cost.

Step 4: Test a concrete offer

Once the problem looks real, test whether people will commit to your solution. The test should ask for something that costs them a little, so the result means something.

Validation tests from weak to strong

Test	What you ask for	Signal strength
Waitlist signup	Email address	Weak
Smoke test landing page	Click on "Buy" or "Start trial"	Weak to medium
Follow-up call booked	30 minutes of their time	Medium
Letter of intent	Written statement of intent to buy	Medium to strong
Pilot with start date	Time, data, internal effort	Strong
Pre-order or deposit	Money	Strongest

Example (hypothetical): a founder targeting independent restaurants offers a manually run delivery-scheduling service for €49 per month, with the first month paid upfront. She contacts 40 restaurants she interviewed or was introduced to. Twelve take a demo call and five pay for the first month. That is far stronger evidence than 300 email signups from a generic landing page, because the five paying restaurants are in the exact segment she plans to serve.

Set pass/fail criteria in advance

Write down before the test what result would count as success, and stick to it. For example: "At least 5 of 30 qualified prospects pay for a pilot within three weeks." Without a predefined threshold, it is tempting to reinterpret any result as encouraging.

Step 5: Check the economics roughly

An idea can pass every interview and still fail as a business if customers are expensive to acquire relative to what they pay. You do not need a full financial model at this stage, but a rough sanity check helps.

Take the price you tested, an assumed gross margin, a guessed monthly churn and a guessed cost to acquire one customer, and see whether lifetime value comfortably exceeds acquisition cost. The SaaS unit economics calculator does this in seconds, and our article on the LTV to CAC ratio explains how to read the result. Treat these numbers as rough estimates and revisit them when you have real data.

It also helps to estimate how long you can afford to keep testing. If validation will take three months, check that your savings or budget cover that period with the startup runway calculator.

Step 6: Decide: build, pivot or stop

After interviews and at least one commitment test, make a deliberate decision.

• Build if a clear segment has the problem, has spent money on it, and committed to your offer at a price that looks sustainable.
• Pivot if the problem is real but your solution, segment or price did not land. Change one variable and test again.
• Stop if the problem turned out to be minor or nobody would commit. Record what you learned; it often points to a better problem.

If you decide to build, keep the first version as small as possible. Our guide on how to build an MVP explains how to scope it around the assumption you still need to test.

Validation checklist

• One-sentence problem statement written
• Narrow target segment defined (could name 50 people)
• Riskiest assumptions listed and ranked
• 15+ problem interviews completed and summarised
• Pattern of early adopters identified
• Concrete offer tested with a commitment (pilot, LOI, pre-order)
• Pass/fail threshold set before the test
• Rough unit economics checked
• Build, pivot or stop decision documented

Common validation mistakes

• Interviewing friends and family. They want to support you, so their answers are biased. Talk to strangers in your segment.
• Pitching instead of listening. Once you pitch, people respond to your enthusiasm, not their problem.
• Counting signups as demand. Free actions are cheap. Ask for something that costs a little.
• Changing the criteria afterwards. If you did not hit your threshold, accept it and learn from it.
• Validating the easy assumption. Testing whether people like your design is pointless if you have not confirmed they have the problem.
• Stopping too early or too late. Three interviews are not enough; sixty interviews without a commitment test is procrastination.

How validation fits into the bigger picture

Validation is the second step in the sequence described in our guide on how to start a startup: choose a problem, validate demand, build an MVP, then scale what works. It is also the evidence that later makes fundraising easier. When investors ask why you believe customers want this, the best answer is a list of real commitments from a well-defined segment, not a market-size slide.