Market validation vs. market diagnosis

Market validation vs. market diagnosis. The two get used almost interchangeably, but they answer different questions, at different points, using different evidence. Most of the advice a founder reads is about the first one. Almost none of it is about the second, which is usually the one they actually need.

Market validation: before you build

Validation happens before the product exists in any sellable form. The question is should this exist at all: is there a real problem, does anyone care enough to pay, is the intent behind the interest strong enough to survive contact with a price tag. The evidence is necessarily indirect: problem interviews, waitlists, willingness-to-pay signals, adjacent buying behavior. It's the right toolkit for the question it's answering.

Its failure mode is specific: friendly signals. People are polite in interviews. A waitlist signup costs nothing. Validation is structurally vulnerable to hearing what it wants to hear, because nothing in the process has forced a real buying decision yet.

Market diagnosis: after you launch

Diagnosis starts where validation ends: after the product exists, is reachable, and has had a genuine chance to sell. The question changes completely: not should this exist, but it exists, and it isn't moving. Why. The evidence is no longer indirect. It's the live shape of demand, how crowded the field actually is, where execution is leaking. These are things you can only read once there's a real product in a real market to read.

This is the discipline almost no one runs honestly. Everyone validates before building. It's the default first step. Diagnosing after a launch underperforms is rarer, because it asks the founder to look directly at a result that already happened instead of a hypothesis that hasn't been tested yet.

Why founders keep re-validating instead

Validation is comforting even in its failure: a lukewarm signal can always be read as "needs more time" or "needs more marketing." Diagnosis is uncomfortable because it's about a decision already made, with money and months already spent on it. So the common pattern is a founder who keeps re-running validation-shaped thinking (more interviews, more surveys, more landing-page tests) on a product that has already launched and already has a real answer sitting in its actual sales data. They're asking if again, when the honest next question is why not.

Which one you need

If the product doesn't exist yet, or exists only as a prototype no one outside the building process has paid for, validation is the right tool. If it's live, reachable, and underperforming, more validation won't answer the question that's actually open. That question is a diagnosis. And it needs evidence from the market as it stands today, not more hypothetical interest in a version of it that hasn't launched.