I set fire to millions of Expedia's dollars and all I got was this talk
summary
Team builds a flights‑first package flow after paper prototypes and an AB test, writes a 109‑page product spec, runs a premortem, and launches a year later from a monolithic codebase. Customers distrust the combined price — a usability lab participant refuses to believe a $1,000 package saving — so conversion stays flat: nothing happened. Small AB subtests (ff21) increase conversion locally. Recommendation: log failures, publish evidence, create psychological safety to learn.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:03Avoiding failure talk
- 5:54Failed Expedia project
- 15:37Flawed product launch
- 19:10Launch that changed nothing
- 22:20Road to acceptance
- 25:14Team after failure
- 27:43Lessons from failure
- 31:37Publish failure stories
- 33:25Aviation failure sharing
- 34:50Culture that rewards failure
- 37:21Safety in failure
- 39:08Q&A
related talks
40mPresentation
44mOptimize Innovation Results with Discover, Build, Measure, Learn
39mEscaping the Build Trap
55mSurviving your First Big Customer as a Startup
50mAsaf Shochet Avida, Eitan Noy: Why My Amazing Test Automation Project Crumbled
55mKeynote: Experimenting in a Sociotechnical System in Software Engineering
38mFrom scratch to production in three months: the AI-first delivery model
36mA/B Testing in Production