Glossary

Minimum detectable effect (MDE)

The smallest true effect an experiment is designed to reliably distinguish from no effect, given its sample size and power.

Also called: minimum detectable lift

The minimum detectable effect (MDE) is the smallest true difference between a treatment and a control that an experiment has a stated chance of detecting as statistically significant, given its planned sample size and statistical power. It is set before the experiment runs, as an input to sizing it, not a result read off after the fact.

Holding variance, significance threshold and power fixed, the required sample size scales roughly as n ∝ 1/MDE²: halving the effect a team wants to detect roughly quadruples the sample needed. A test with too few users has a large MDE by default, meaning it can only reliably catch big swings and will systematically miss the small but real improvements common in mature products.

Teams size experiments by first choosing an MDE deemed practically meaningful, then computing the sample size and run time A/B testing needs to hit it at a target power (commonly 80%) and significance level (commonly 5%). The common mistake is confusing the MDE with the effect a team expects or hopes for: building around too small an MDE demands more traffic and time than most products can spare, while too large an MDE will report "no significant difference" for real, smaller effects. A reported result should always be read alongside the MDE and sample size the experiment assumed.

Last reviewed September 22, 2026

In the index now

Related terms

Statistical power The probability that a hypothesis test correctly detects a real effect, given that one truly exists. Sample size The number of observations in a study or test, which determines how precisely and reliably an effect can be measured. A/B testing A controlled experiment that compares two variants to measure which one produces a better outcome. Effect size A standardized measure of the magnitude of a difference or relationship, independent of sample size. Statistical significance A threshold judgment that an observed effect is unlikely to be due to chance alone, given the data and test used.

Related guides

How to read an A/B test result Significance, power, confidence intervals, sample ratio mismatch and peeking — what a dashboard's green checkmark does and does not tell you. How to set up an experimentation program From your first A/B test to a durable program — the plumbing, the statistics rules, the review ritual, and the tools that support each stage.