Minimum Detectable Effect
Definition
The MDE denotes the smallest effect size detectable under a fixed test design.
The smallest effect size that a test should reliably detect at a fixed significance level, power, and sample size.
Simply explained
How to check before a measurement how small a difference may be so that the planned procedure detects it sufficiently often.
Why it matters in practice
Helps to manage marketing decisions measurably, comparably, and economically.
Example from business
Before an A/B test, it is calculated whether an increase of two percentage points is realistically detectable.
Limits and pitfalls
- Confusing MDE with observed effect or failing to specify power and baseline rate.
Origin & history
MDE is part of statistical experimental design and power analysis.
Use in mid-sized companies
- Possible application: An analysis team sets significance level, power, baseline rate, and sample size before an A/B test. It checks whether a business-relevant effect is detectable with the planned scope.
Frequently asked questions
What does smallest detectable effect mean?
The smallest effect size that a test should reliably detect at a fixed significance level, power, and sample size.
Which mistakes should be avoided?
Confusing MDE with observed effect or failing to specify power and baseline rate.
Sources
Source checked on 10.10.2026. Editorial application examples are hypothetical; no evidence of real customer results.
Editorially reviewed on: