// tool — client-side, no data leaves your browser
A/B Significance Calculator
Enter the visitors and conversions for each variant. This runs a two-proportion z-test and tells you whether the difference is statistically significant — or just noise. Nothing is sent to a server.
- RATE A
- —
- RATE B
- —
- REL. LIFT
- —
- P-VALUE
- —
- Z-SCORE
- —
- CONFIDENCE
- —
- 95% CI ON DIFF
- —
How the test works
The calculator compares two proportions — conversion rate A and conversion rate B — using a two-proportion z-test. It pools the two samples to estimate a common rate under the null hypothesis (that A and B are equal), computes the standard error, and derives a z-score. The two-tailed p-value is the probability of seeing a difference this large if the variants were truly identical. Below your chosen α, the result is flagged significant.
A caveat this newsroom insists on: statistical significance is not business significance. A tiny, real lift can be significant with a large enough sample and still not be worth shipping. And significance testing assumes the experiment ran cleanly — no peeking, stable traffic, one test at a time. See our methodology on reporting uncertainty honestly.