A/B Test Significance Calculator
Enter your control and variant numbers to find out whether your result is a real, statistically significant difference — or just noise you'd expect from randomness alone.
Your test numbers
This uses a two-proportion z-test, the same method behind most CRO and A/B testing tools. It compares your two conversion rates against the pooled variance you'd expect from random chance, producing a z-score and a p-value — the probability of seeing a difference this large (or larger) if there were actually no real difference between the two variants. If the p-value is below your chosen significance threshold (5% for 95% confidence, 1% for 99%), the result is considered statistically significant.
Before giving a significance verdict, this calculator requires at least 30 visitors, 5 conversions and 5 non-conversions in each variant. Smaller samples or more extreme splits receive a warning because the z-test approximation is not reliable enough for an action-ready decision.
A significant result doesn't guarantee the effect is exactly the size you measured — that's what the confidence interval on the difference is for. It also doesn't tell you the test was run correctly: it still assumes visitors were randomly split, the test ran for at least one full business cycle, and you didn't stop it the moment it looked good.