Skip to main content
Free Calculator

Minimum Detectable Effect (MDE) Calculator

Find out the smallest conversion rate change your A/B test can reliably detect given your available sample size and test parameters.

Available visitors per variation

Your current conversion rate

Confidence level (typically 95%)

Probability of detecting a real effect (typically 80%)

Minimum Detectable Effect

Absolute MDE

0.90%

percentage points

Relative MDE

18.0%

minimum detectable lift

Detectable Range

4.10% – 5.90%

conversion rate range

Your test can detect conversion rate changes of at least ±0.90% from the 5.00% baseline (a 18.0% relative lift) with 95% confidence and 80% power.

Methodology

This calculator determines the minimum detectable effect (MDE) given your available sample size, baseline conversion rate, significance level, and statistical power. It is the inverse of the sample size calculation. Where the sample size formula solves for n given MDE, this calculator solves for MDE given n, using an iterative approach. The underlying equation is the same two-proportion z-test formula: n = ( Z_alpha/2 * sqrt(2 * p_bar * (1 - p_bar)) + Z_beta * sqrt(p1*(1-p1) + p2*(1-p2)) )^2 / (p2 - p1)^2 Where p2 - p1 = MDE (absolute). The calculator iteratively finds the MDE value that makes the required n equal to your available sample size. The result is shown as: - Absolute MDE: the raw conversion rate difference (e.g., 0.50 percentage points) - Relative MDE: the percentage change relative to baseline (e.g., 10.0%) - Detectable range: the conversion rate range your test can distinguish from the baseline This helps you understand whether a test is worth running given your traffic constraints.

Frequently Asked Questions

What is minimum detectable effect (MDE)?▾
The minimum detectable effect is the smallest difference in conversion rates that your test is powered to detect with statistical confidence. If the true effect of your change is smaller than the MDE, your test is unlikely to produce a significant result — even if the change actually works. Think of it as your test's resolution limit.
Why should I calculate MDE before running a test?▾
Knowing your MDE helps you decide whether a test is worth running. If your MDE is 20% but you only expect a 5% lift, you are very unlikely to get a significant result with your available traffic. You would either need to increase traffic, extend the test duration, or focus on a higher-impact change.
How does MDE relate to sample size?▾
MDE and sample size are inversely related. More traffic (larger sample) allows you to detect smaller effects. If you double your sample size, your MDE decreases (roughly by a factor of 1/sqrt(2), or about 30%). This is why high-traffic sites can test smaller optimizations while low-traffic sites need to focus on larger changes.
What is a good MDE for my site?▾
This depends on your traffic volume and what constitutes a meaningful business impact. High-traffic sites (100k+ monthly visitors) can often achieve MDEs of 2-5%. Medium-traffic sites (10k-100k) typically see MDEs of 5-15%. Low-traffic sites (under 10k) may only be able to detect effects of 15%+ and should focus on testing bold, high-impact changes.
What is the difference between absolute and relative MDE?▾
Absolute MDE is the raw percentage point difference (e.g., going from 5% to 5.5% is a 0.5 percentage point absolute MDE). Relative MDE is the percentage change relative to the baseline (e.g., 0.5 / 5 = 10% relative MDE). This calculator shows both so you can assess the detectable effect in the way that makes most sense for your context.

Related Calculators

A/B Test Sample Size CalculatorA/B Test Significance CalculatorRevenue Impact CalculatorA/B Test Duration Calculator

Updated for 2026. Built by GrowthLayer.