A/B Test Sample Size Calculator
Plan your split test before you launch it: how many visitors each variant needs to reliably detect the uplift you care about.
The smallest improvement worth detecting: from 5.00% to 5.50% conversion.
Visitors per variant
31,243
Total visitors
62,486
Estimated duration
32 days
Whole test weeks
5 weeks
Run the test for whole weeks where possible, and don't stop early just because the result looks significant - peeking inflates false positives.
How it works
Enter your baseline conversion rate, the minimum uplift you want to detect and your desired significance and power. The calculator tells you how many visitors you need per variant and how long the test will take at your current traffic.
Basic guide
Enter your baseline conversion rate, the smallest uplift you want the test to detect, weekly traffic and your desired confidence and statistical power.
The calculator estimates visitors required per variant, total visitors and the approximate test duration.
Example
If your current conversion rate is 5% and you want to reliably detect an uplift to 5.5%, enter:
Baseline conversion: 5% Minimum detectable uplift: 10%
Then add your available weekly visitors and preferred confidence and power settings.
The calculator estimates how much traffic the test needs before you launch it.