What problem does it solve? Product teams often build full features based on untested assumptions, wasting engineering effort when the assumption turns out wrong. This Skill structures hypothesis-driven experiments so you can validate ideas cheaply before committing to full development. ## Core Features & Use Cases - Hypothesis Framing: Formats assumptions into the "We believe X will cause Y because Z" structure and identifies the riskiest assumption to test first. - Minimum Viable Test Selection: Chooses the cheapest experiment type from Ries's ladder, from customer interviews and fake door tests up to A/B tests and full releases. - Success Criteria and Sample Sizing: Defines pass/fail thresholds before running, calculates required sample sizes, and sets guard metrics to prevent false conclusions. - Use Case: Before building a new onboarding progress bar, run a fake door test measuring click-through to validate that users actually want progress visibility, then only build if the 15% activation lift hypothesis holds. ## Quick Start Ask the assistant to design a minimum viable experiment to test whether adding a progress bar to onboarding will increase activation rate by 15%.
