Persona walkthroughs with drop-off

The killer use-case lands: a persona walks through a real onboarding flow or prototype and shows exactly where it drops off. The walkthrough is artifact-first — built from real screenshots the persona’s host actually looks at — because reacting to real screens is more reliable than driving a live app.

• Define a flow from uploaded screenshots: an ordered, citable sequence of real screens.
• Each persona walks it screen by screen — honest friction on the canonical scale, a would-continue verdict per step, and the drop-off point with its reason.
• The segment funnel aggregates the cohort: entered, continued and dropped per step, drop reasons, the personas who bailed, and the biggest-drop-off headline.
• Walkthrough predictions carry likelihoods, so when your real funnel data arrives they score directly into the calibration loop.