In short: Validating a startup idea means finding evidence that a specific group of people already spends money or meaningful time solving a problem, before you build anything. In practice that is seven to ten conversations with people who have the problem, one written summary of what they currently do instead, and a clear answer to what they would have to stop using to adopt yours.
Validation is not asking people if they like it
The most common failure is a conversation that produces encouragement rather than evidence. If you describe your idea and ask whether it sounds useful, almost everyone says yes, because saying no to an enthusiastic person is socially expensive and costs them nothing to avoid.
Useful validation asks about the past, not the future. What did you do the last time this happened? How long did it take? What did you use? What did that cost? Past behaviour is a fact; future intention is a guess wearing a fact's clothing.
The seven-day sequence
- Day 1 — Write down who specifically has this problem. Not 'small businesses' but 'the office manager at a four-chair dental practice'. If you cannot name a role, you cannot find ten of them.
- Day 2 — Find fifteen of those people. LinkedIn, industry forums, local associations, or the comment sections of the software they already complain about.
- Day 3-5 — Have seven to ten conversations. Twenty minutes each. Ask what they did last time, not what they would do next time.
- Day 6 — Write one page summarising what they currently do instead. If you cannot describe the incumbent workflow in detail, you have not learned enough.
- Day 7 — Decide. Do at least half of them describe the same painful workflow, and can you name what they would stop using?
The questions that actually work
- Walk me through the last time you had to do this. — Produces the real workflow, including the parts people forget to mention.
- What did you try before that? — Surfaces the graveyard of solutions they have already rejected, and why.
- How much time did it take, and who did it? — Converts vague pain into a number you can price against.
- What happens if it just does not get done? — Distinguishes an annoyance from a real cost. If the answer is 'nothing much', the demand score is optimistic.
- Who else would need to approve buying something for this? — Reveals the buying committee before it ambushes you.
Signals that should stop you
- Everyone agrees it is a problem but nobody has ever looked for a solution. Latent pain is much harder to monetise than active pain.
- The workflow varies so much between people that there is no shared job to automate.
- The person with the problem cannot authorise spending, and the person who can does not feel it.
- The incumbent is free and good enough, and switching costs are high.
- Every conversation ends with a feature request for a different product.
What validation cannot tell you
Validation reduces the chance of building something nobody wants. It does not tell you whether you can reach those people affordably, whether you can build it well, or whether the market is large enough to matter. Those are separate questions, and each one has killed companies that validated perfectly.