MDE (Minimum Detectable Effect)
The smallest improvement a test is statistically capable of detecting.
Minimum Detectable Effect (MDE) is the smallest lift an A/B test can reliably detect given its sample size and variance. MDE is a function of traffic volume, baseline conversion rate, and statistical power — typically set at 80%.
Context
MDE is calculated before the test runs so teams know whether the test can realistically detect the improvements they're hoping for. If MDE is 15% and you're trying to detect a 5% lift, the test cannot produce a meaningful answer regardless of how long it runs.
A common mistake is running long tests on low-traffic pages expecting meaningful results. A page with 2,000 sessions per week converting at 3% has an MDE of roughly 30% over 4 weeks — which means only massive lifts (hero rewrites, pricing page overhauls) will produce reliable signal.
To detect a 10% lift on a 5% baseline at 95% confidence and 80% power, you need approximately 15,800 conversions total (7,900 per variant). On a page converting at 5% with 20K weekly sessions, that's 8 weeks of test duration.
MDE scales with the square root of sample size. Doubling traffic doesn't double sensitivity; it only multiplies it by ~1.41. This is why high-traffic pages can detect small improvements and low-traffic pages can't.