Free tool
A/B test sample size and duration calculator
Find out how many visitors your test needs and how long to run it, before you launch. Change any value and the result updates.
More than 6 weeks: try a bolder change, a page with more traffic, or a metric closer to the change.
Standard two-proportion formula (two-sided). With more than two variants, a Bonferroni correction is applied.
How to use it
- Current conversion rate: the rate of the page or step you're testing, not your site-wide average. A product page that converts 3% to add-to-cart has a baseline of 3.
- Minimum detectable effect: the smallest relative uplift worth detecting. 10% means going from 3.0% to 3.3%. Smaller effects need far more traffic.
- Daily visitors: visitors who will actually enter the test each day, across all variants.
- Confidence and power: 95% confidence and 80% power are the usual defaults. Raise them only for high-risk decisions.
How to read the result
The duration is rounded up to full weeks, because people behave differently on weekdays and weekends. Run the test for that whole period, even if it looks significant earlier. Stopping early is one of the most common reasons "winning" tests don't hold up.
If the calculator says more than six weeks, the test is probably not worth running as planned. Test a bolder change, pick a page with more traffic, or measure a metric closer to the change.
Want the full reasoning? Read how long you should run an A/B test.
Want help running tests that give clear answers?
In a free 30-minute call we look at your site and traffic together, and I tell you what I'd test first.
Book a free discovery call