.avif)
MDE (Minimum Detectable Effect)
.webp)
The smallest difference in performance between two variants (e.g. conversion rate, revenue per visitor) that a statistical test can reliably detect, taking into account sample size, confidence level and statistical power. In CRO andA/B testing, the MDE is used to define the minimum amplitude of change that the test can identify as statistically significant.
Too high an MDE risks masking real but small gains, while too low an MDE requires large volumes of traffic to be detected with any certainty. It is therefore an essential parameter to be adjusted according to business stakes, available traffic volume and test objectives. Properly calibrated, it enables tests to be prioritized according to their potential impact, and avoids decisions based on random fluctuations.
Similar definitions
Statistical power
The probability that a test will detect a real effect, if such an effect exists. High power reduces the risk of false negatives.
Uplift
The increase in conversions or value generated by a specific action, often measured in A/B tests.
Mann-Whitney U Test
A non-parametric test used to compare two independent samples to determine whether they come from the same distribution.
p-value
A statistical measure that indicates the probability of observing the data collected (or something more extreme) under the null hypothesis.
Frequentist
A statistical approach that estimates model parameters based solely on observed data, without using any a priori information.
.svg){kind=small}
