Enter the visitors and conversions for each version. You get each conversion rate, the lift, and whether the difference is statistically significant or could still be chance.
Version A (control)
Version B (variant)
- Lift (B vs A)
- -
- Confidence the difference is real
- -
Reading the result
The test compares the two conversion rates (a two-sided two-proportion z-test). "Significant at 95%" means a difference this big would happen by chance less than 5% of the time. Run tests for full weeks and decide the sample size up front - stopping the moment a result looks good makes false wins likely.