MVP (minimum viable product)

The point is the test, not the product. If a landing page and twenty conversations answer the question, that is a better MVP than three months of code.

The common failure is building a small version of the wrong thing. An MVP should be minimal in scope and maximal in the strength of the signal it produces.

Signals are not equal, and they rank in one order: money, then a scheduled call, then an email address, then a like. Twenty conversations that produce three prepayments of $200 answer the question better than a thousand sign-ups, because a prepayment is the only one of those where somebody kept a promise that cost them something.

Write the falsifying number before you build: if fewer than ten of fifty people we speak to will pay a deposit, we stop. An MVP without a threshold agreed in advance always produces encouraging results, because a founder can find encouragement in anything - and the point of the exercise was to be able to be wrong.