Glossary/Measurement & Attribution/Minimum Detectable Effect (MDE)
Measurement & Attribution
Minimum Detectable Effect (MDE)
The smallest lift a test is designed to reliably detect, given its sample size, baseline rate, significance level and statistical power.
By Brandon Mina·CEO, BrandPilot AI·Updated November 2026
Short answer
The minimum detectable effect (MDE) is the smallest lift a test is designed to reliably detect, given its sample size, baseline rate, significance level and statistical power. Setting it before launch determines how long a test must run. A smaller MDE requires more traffic, while an unrealistically large one risks missing genuine improvements.