A/B Test Significance Calculator
Find out whether your A/B test result is statistically significant
Frequently Asked Questions
This uses a two-proportion z-test. It pools both variants' conversion rates, computes the pooled standard error, then divides the difference between the two rates by that standard error to get a z-score, which converts to a confidence level.
95% is the standard threshold in most industries, meaning there's only a 5% chance the observed difference happened by random chance if there were truly no difference. Some teams use 90% for lower-stakes decisions or 99% for high-stakes ones.
A higher rate alone doesn't prove a real effect — with small sample sizes, random variation alone can easily produce a difference that looks meaningful but isn't statistically reliable. You need either a bigger difference or more traffic to reach significance.
No — stopping early the moment you cross the significance threshold inflates your false-positive rate ('peeking'). Decide your sample size in advance and let the test run to completion before checking for significance.
How the z-test works
Pooled rate p = (conversions A + conversions B) ÷ (visitors A + visitors B). Standard error = √[p × (1 − p) × (1/visitors A + 1/visitors B)]. Z-score = (rate B − rate A) ÷ standard error. A z-score of ±1.96 or beyond corresponds to 95% confidence.