Minimum Detectable Effect (MDE)

Definition

MDE is the smallest effect size you want your experiment to reliably detect. Smaller MDE requires much larger samples.

Example

If baseline conversion is 2%, an MDE of 0.3 points means detecting a lift to 2.3%.

How to use it

• Choose an MDE that is both realistic and action-worthy.
• Use absolute percentage points for conversion rates to avoid confusion.
• Align MDE with the cost of acting on a change.
• Revisit MDE as baseline rates change over time.
• Define MDE before the test so you can size the sample correctly.
• Use separate MDE targets for critical funnel steps if volume differs.

Common mistakes

• Setting MDE so low that the sample size is unattainable.
• Using relative percent when stakeholders expect absolute points.
• Choosing MDE after seeing early results.
• Using a generic MDE across very different funnels or segments.

Why this matters

This term matters because it affects how you interpret performance and make budget decisions. If you use inconsistent definitions or windows, ROAS/CPA can look "better" while profit gets worse.

Practical checklist

• Write a 1-line definition for "Minimum Detectable Effect (MDE)" that your team will use consistently.
• Keep the time window consistent (weekly/monthly/quarterly) when comparing trends.
• Segment results (channel/plan/cohort) before drawing big conclusions from blended averages.
• Use a calculator that references this term (e.g., A/B Test Sample Size Calculator) to sanity-check assumptions.
• Read the related guide (e.g., A/B test sample size: how to plan conversion experiments) for context and common pitfalls.

Where to use this on MetricKit