Low-powered tests produce inconclusive results or, worse, lucky “wins” that do not replicate. Skewed metrics such as revenue per user need much larger samples than conversion metrics to reach the same power.
Definition
Statistical power.
Also called: Power
The probability that an experiment detects a true effect of at least the MDE — commonly set at 80% or 90%.
Seeing this number move and not sure what it means for your game?
Bring it to a first call →