Check whether the difference between two variants is statistically significant. Enter visitors and conversions for A and B to get uplift and confidence. Free, client-side.
Two-proportion z-test → confidence = 1 − p-value (two-tailed)
Clicked reads the ad pixels and cart you already have and shows real revenue, conversions and A/B TEST SIGNIFICANCE by source - live, from one line on your site. No cookie banner, no signup to try.
Before you call an A/B test a winner, you need to know the difference isn't just noise. This calculator runs a two-proportion z-test on your two variants and returns each conversion rate, the relative uplift, and the statistical confidence that B truly differs from A. As a rule, aim for 95% confidence (and a decent sample size) before acting - a big uplift on 30 visitors means nothing.
95% is the standard threshold - it means there's only a 5% chance the difference is random noise. For lower-risk decisions 90% may be enough; for high-stakes changes some teams want 99%. Below your threshold, keep the test running or collect more data.
It depends on your baseline conversion rate and the uplift you want to detect - smaller effects need much larger samples. As a rough guide, a few hundred conversions per variant is often needed to detect modest differences. Don't stop the test the moment it hits significance; ending early inflates false positives.
No. Significance says the difference is probably real; it doesn't say it's big enough to matter. A 0.1% uplift can be significant with a huge sample yet be commercially irrelevant. Weigh confidence and the size of the uplift together.
Money-first analytics for stores and anyone buying traffic. Know what pays, skip the cookie banner.