Assumption Testing & Prototyping

Validating the value of a customer problem before anyone designs a response to it — how urgent it is, how widely it is felt, and whether anyone would pay to be rid of it. The work runs from choosing problems off the Canonical Problem Register, or taking a one-off problem on an intake form, and writing each as a falsifiable hypothesis with a named buyer, a number and a threshold, through experiments whose pass and fail lines lock before the first person is approached, until every result is read against its line and assembled into a Problem Value Validation Record.

Its purpose is to tell a team where the evidence stands on each problem without deciding for it what is worth solving — usability and feasibility belong to a solution and are deferred, and a hypothesis the evidence does not support is reported as a kill signal, not a kill decision. A round that lands reads every result against a line drawn before the tests ran; one that does not moves its lines once the answers arrive, reads a warm meeting as a commitment, and averages a buying group’s answers into a price.