A/B Test Calculator: check whether a difference is significant
Compare two conversion rates using a two-tailed two-proportion z-test, with uplift, p-value and a sample-size estimate.
How to use
- Enter visitors and conversions for A, your control.
- Enter visitors and conversions for B, your variant.
- Read the conversion rates, uplift and significance verdict.
- Check the sample-size estimate against your planned test before making a decision.
The estimate assumes a 10% relative uplift. Plan sample size and duration in advance; do not stop early because a daily check looks significant.
No sign-up needed Your work stays in your browser Copy or export to keep your work
–Conv. rate A
–Conv. rate B
–Relative uplift
–Z-score
–P-value
–Confidence
Enter your test data
Add visitors and conversions for both variants to see whether the result is statistically significant.
Two-proportion z-test, two-tailed, 95% confidence level. A common rule of thumb: to detect a 10% relative uplift with 80% power you need roughly – visitors per variant at these baseline rates.