setup
Scope like an MVP
Hone the pilot to one problem and one use case.
One problem
Skip the tour of their whole workflow. Pick the pain that maps to cost, risk, or revenue and can produce evidence in one to two months.
Write the edges
List what you will not do. Send new requests to a phase-two backlog.
Clock discipline
Aim for one to two months. The source session saw a pattern: the longer a pilot runs, the further it drifts from an enterprise deal.
Claims and patterns on this page are paraphrased from practitioner session notes and the Startup Bible (Pilots that convert). See Sources.