Market validation is not asking friends whether they like your idea. It is finding out whether people already spend time, money or attention on the problem — and whether they will change what they do because of you. This sequence takes about a week and costs nothing but nerve.
Turn your idea into the single sentence that must be true for it to work: 'Freelance bookkeepers lose more than two hours a week reconciling receipts, and they would pay to stop.' If it cannot be false, it is not an assumption.
You have many assumptions. Rank them by two things: how badly it hurts if it is wrong, and how little you know today. Test the top one first — not the easiest one.
Forums, comparison searches, existing spreadsheets, competing tools with paying users, the workaround people built themselves. Existing behaviour is stronger evidence than any stated intention.
Ask about the last time it happened, what it cost, and what they already tried. Never describe your solution in the first half of the call.
A pre-order, a deposit, a calendar slot, an intro to their boss. Compliments are free; commitment is data.
| Assumption | Test | What counts as a pass |
|---|---|---|
| The problem is frequent enough to matter | 5 problem interviews about the last occurrence | 4 of 5 describe it happening in the past month, unprompted |
| They already pay to solve it | Ask what they spend today, including workarounds and staff time | At least half name a real current cost |
| They will switch to something new | Offer a paid pilot or a pre-order at real price | One stranger (not a friend) commits money or a signed slot |
| You can reach them repeatedly | One channel test: post, cold email, or an ad to a landing page | A measurable response above your channel's baseline |
| The economics can work | Sketch price minus delivery cost on ten customers | A positive number without assuming scale |
A failed validation test is not a dead idea — it is a narrower one. Most 'no' answers tell you the problem is real but you aimed at the wrong segment, the wrong moment, or the wrong price. Rewrite the assumption with what you learned and run it again. Kill the idea only when the same assumption fails across three different segments.
Market validation is the process of collecting evidence that a specific group of people has a problem they already spend time or money on, and that they will change their behaviour for your solution. It is evidence about behaviour, not opinions about an idea.
One riskiest assumption should be testable in under a week. If a test takes a month, it is not a test — it is a build.
Five per segment is usually enough to spot a pattern. If the first five contradict each other, your segment is too broad rather than your idea being wrong.
You can gather signals from search demand, competitor pricing and existing workarounds, and those matter. But none of them tell you why people behave that way, which is exactly what you need to build the right thing.
Market research describes a market that exists. Idea validation tests whether your specific bet inside that market is true, using the cheapest possible experiment.