Checking a test every day manufactures winners
A difference that looks significant one day can vanish the next. What peeking really costs, measured on simulated A/A tests.
Further reading
The traps that skew how a test gets read, and how to avoid them.
All resourcesA difference that looks significant one day can vanish the next. What peeking really costs, measured on simulated A/A tests.
Further reading
On average, the winner of a test overstates its effect. How to correct for it before you promise a number to anyone.
Read articleCreate your workspace, or explore the demo first with data from a fictional store.