Enter the visitors/clicks and conversions for each version. We calculate statistical confidence and tell you which version to choose, or how much more data you need.
How many people per version and how many days to run before you can draw a conclusion
Enter people and conversions
for both versions
* Uses a two-tailed two-proportion z-test. Results are accurate when people are split randomly, versions run at the same time, and you don't stop the test the moment you see a good result.
Decide with data, not gut feel
An A/B test statistical significance calculator that tells you whether the difference between 2 ads, creatives, landing pages or messages is big enough to trust, and helps you plan how many people and days you need before concluding.
Test the right way to get results you can trust
It's the confidence that the difference you see isn't due to chance. For example, 95% confidence means that if the two versions were actually the same, there'd be less than a 5% chance of seeing a difference this large by luck.
At least until you reach the number of people the tool recommends, and for at least 7 days to cover both weekday and weekend behavior. Don't stop as soon as one version pulls ahead, because early results often swing.
Change one thing at a time, such as only the image or only the headline, so you know what improved results. If you change several things at once, you'll know which version is better but not why.
Yes. Use clicks (or reach) as the number of people and results as each version's conversions. This tool helps double-check whether the result the platform announces is significant enough.
Use them together to cover every part of your marketing plan