Am I Building the Wrong Thing? A Diagnostic Framework

The first diagnostic split: is this a demand problem or a distribution problem? A demand problem means the market does not urgently need what you built, or does not need it in the form you built it. A distribution problem means the demand exists but you are not reaching it efficiently — the right customers cannot find you, or your sales process doesn't match how the buyer makes decisions.

These look identical from the inside. Both produce flat growth, disappointing conversion rates, and investor conversations that feel like they are missing something. The distinction matters enormously because the fix for each is completely different. Treating a distribution problem as a demand problem leads to product pivots that destroy real value. Treating a demand problem as a distribution problem leads to channel investment that accelerates the rate at which you reach people who don't want your product.

The clearest signal for a distribution problem: you have a small cluster of customers who love the product deeply — they use it frequently, would be upset if it disappeared, and have referred at least one other person without being asked. But you cannot find more customers like them through any consistent channel. The product is working; the go-to-market is not. In this case, study your best customers rather than rebuild what they are already using.

Question one: of your current customers, which three use the product most consistently and would be most upset if it disappeared? What do they have in common that is not on your ICP document — specifically, what trigger event made the problem urgent for them? Not a demographic characteristic but something that happened in their world in the 30 days before they started using your product.

Question two: why did your last three churned customers leave? Not what they said in the exit conversation — what you observed about their usage patterns before they left. Did usage decline gradually (suggesting the problem was solved or no longer urgent) or drop suddenly (suggesting a competing solution or a change in circumstances)? The pattern reveals whether churn is a product problem or a market targeting problem.

Question three: if you could only sell to one specific type of customer for the next 90 days — one trigger event, one role, one urgency level — which one would you choose based on current evidence? If you cannot answer this with a specific description rather than a demographic range, your ICP is not precise enough to be operationally useful, and that imprecision is costing you the clarity to answer the bigger question.

If the demand versus distribution test points consistently to demand — no cluster of deeply engaged users, no segment where retention is meaningfully better, no customers who would fight to keep the product — then the honest answer may be yes: not necessarily 'this idea is wrong,' but 'this version, for this segment, is not working.' These are different conclusions with different responses.

The most common mistake at this point is the incremental pivot: changing one thing while keeping everything else the same. Moving from B2B to B2C. Adding an integration a large customer requested. Repositioning the marketing copy without changing the product. These moves feel like action but rarely change the fundamental signal because they don't address the underlying mismatch between what was built and what a specific market urgently needs.

A genuine pivot requires identifying the new hypothesis specifically — which customer, which problem, which urgency level, which existing behaviour — and testing it at the smallest possible scale before committing resources. One customer who will pay before you build the full pivoted version is worth more than ten customers who say they would pay after you do. The hypothesis test must come before the build, not after.