A/B Test Significance Calculator
Is the difference between two variants real, or just noise?
Variant A (control)
Variant B
How this works
This runs a two-proportion z-test. It compares each variant's conversion rate (positives ÷ responses), pools the variance, and computes a z-score and a two-tailed p-value. A result is called significant at 95% confidence when the p-value is below 0.05. "Conversions" can be any yes/no signal: repurchase intent, top-two-box likers, "would recommend," and so on.
A caution for small tests
With small samples, a big-looking difference can still be noise. If the calculator says "not significant," you usually need more responses, not a different conclusion. Use the sample size calculator to plan the count before you run.
Run a real head-to-head
TryMe can test your product blind against a competitor or your current formula, with verified skin-matched testers.
See how it works