Free to use, no account
A/B Test Calculator
Compare conversions in two independent groups. See each conversion rate, the percentage-point difference, relative lift, 95% confidence intervals and an approximate two-sided test.
How do I compare two A/B test conversion rates?
Enter visitors and conversions for each group. The calculator uses a pooled two-proportion z-test and Wilson/Newcombe 95% intervals. Plan your metric and sample size before a test; checking repeatedly can raise false-positive risk.
Frequently asked questions
How do I compare two A/B test conversion rates?
Enter visitors and conversions for each group. The calculator uses a pooled two-proportion z-test and Wilson/Newcombe 95% intervals. Plan your metric and sample size before a test; checking repeatedly can raise false-positive risk.
Are my experiment numbers uploaded?
No. The calculation runs in your browser; this tool does not upload or save the numbers.
What does a p-value below 0.05 mean?
Under the test assumptions, it means results this different would be uncommon if the true conversion rates were equal. It does not measure the chance that a version is best.
Does an inconclusive result mean both versions are equal?
No. It means this test did not provide enough evidence of a difference at the selected threshold. A larger or better-planned sample may be needed.
Should I stop a test as soon as it becomes significant?
Repeatedly checking and stopping early can increase false positives. Define the metric, sample size and test duration before starting.