§The second whitepaper argued that discovery is judgment. This one is about the work that judgment is exercised on, and it is written differently: as a manual, in the second person, because you are going to do this yourself. There is no department to hand it to.

§A sprint, here, is not a week. Jake Knapp's five-day design sprint gave the word its modern shape, and it is a fine shape for a team that already has a problem and needs a decision by Friday. But the most common state of an early team is earlier than that: you have a hunch, a few conversations, and a strong urge to build. What you need is not a calendar. It is a sequence of questions, each of which produces a document that the next question can argue with.

§ 0.1Why insight comes before evidence#

§Teams that skip the first half test the wrong thing well. They build a clean prototype, run tidy sessions, and learn with great precision that nobody wanted the thing they never asked about. The first half exists to make sure the question you carry into the second half is a question a customer would recognize.

§Teams that never leave the first half have the opposite problem. They know the customer intimately and have shipped nothing, because every conversation revealed one more nuance. The second half exists to force a commitment: this is what we think, here is the cheapest way to find out, here is what we found.

§ 0.2The one rule this whitepaper insists on#

§Every chapter ends with the same heading: what you leave with. That is not a stylistic tic. It is the rule.

§The documents are small. A research plan is half a page. A problem statement is five lines. A tension map is three columns. The point is not the paperwork. It is that writing forces a decision that talking allows you to defer, and the sprint is a machine for making deferred decisions visible.

§ 0.3Condensed beats incomplete#

§Almost nobody has the time for the full version of what follows. That is expected. Every chapter tells you what the step is for, so you can do the smallest version that still answers the question. Three expert conversations done properly beat ten done as a pitch. A landing page tested with forty strangers beats a prototype tested with your co-founder.

§ 0.4What is here and what is not#

§Part I earns insight: a hierarchy of sources, desk research, expert conversations, community listening, customer conversations, synthesis, and a problem statement in the customer's words. Part II tests evidence: options before choices, four tests before building, a ladder of fidelity, the prototype as a question, testing with strangers, and the written position that hands off to strategy. Part III adapts the whole thing to your situation, catalogues how it fails, and closes with cases, including a 2026 field sprint in which four people with no design or engineering training produced two navigable prototypes in two days and then watched six of six strangers fail the feature the team liked most.

§Why discovery matters is the second whitepaper. What to do with the written position is the fourth. Whether the thing closes as a business is the fifth. Experiments after launch belong to the sixth. This one is the field.