Skip to content
Peekbird

Free tool

How long should your A/B test run?

Enter your conversion rate and traffic: see how many visitors a test needs to spot a real difference, and how many days that takes.

Your test needs

…

How to read it

The number is how many visitors each version needs so that, if the change really moves your conversion rate by that much, the test is 80% likely to show it, with 95% confidence that it isn't luck. It's sized as Peekbird sizes its tests: for results you can check as often as you like, which takes about a fifth more visitors than a test you look at only once, at the end. How sure the calls are

Run at least a full week, even when the numbers arrive sooner, so weekdays and weekends both count. And don't stop early because one version is ahead: early leads swing a lot.

Too long? Test a bolder change (a bigger difference shows sooner), test on a page with more traffic, or use a goal closer to the change, such as adding to the cart instead of buying.

Peekbird does this sum for every test it runs, and says “too early to tell” until the numbers are in. Start free.

See what your visitors see.

Start free in a few minutes: one snippet, every feature. Move to a bigger plan when you need more tracked visits.