Enter visitors and conversions for variant A and B. Get each conversion rate, the relative lift, a two-proportion z-test p-value, and a clear yes/no on significance at 95%.
Answers as you type · Formula shown, not hidden · Free, no sign-up
Control group sample size.
Orders, sign-ups or clicks on A.
Treatment group sample size.
Orders, sign-ups or clicks on B.
p = < 0.001. The lift of 50.00% clears the 95% bar — safe to ship B on conversion rate alone.
The calculator gives you the number instantly. Leave your email and we will look at your actual accounts and send back what we would change first.
Each variant’s conversion rate is conversions ÷ visitors. Relative lift is (rate B − rate A) ÷ rate A. Significance uses a two-proportion z-test: the rates are pooled, a standard error is formed from that pool, and z = (rate B − rate A) ÷ SE. The two-tailed p-value comes from the standard normal distribution; p under 0.05 is significant at 95%.
The test assumes independent visitors, one conversion per visitor, and that you decided the sample size before peeking. Sequential peeking or stopping early inflates false positives — run the planned sample, then read the p-value once.
Statistical significance is not practical significance. A 2% relative lift on a tiny baseline can clear p < 0.05 with enough traffic and still be too small to pay for the engineering or creative change. Pair this calculator with the sample-size calculator so you only power tests for lifts that matter.
Control: 100 conversions from 1,000 visitors (10%). Variant: 150 from 1,000 (15%). Relative lift is 50%. The pooled rate is 12.5%, the z-score is about 3.3, and p is well under 0.001 — significant at 95%, and at 99%.
Skip the typing. Connect Meta, Google, TikTok or LinkedIn and Sprites reads your actual spend, impressions, clicks and conversions, then shows where they sit against the ranges above.
Put the A/B Test Significance Calculator on your own site or blog. Paste this HTML where you want it to appear. It runs in the reader's browser and needs no account.
<iframe src="https://www.sprites.ai/embed/ab-test-significance-calculator" title="A/B Test Significance Calculator" width="100%" height="620" style="border:0;max-width:720px" loading="lazy"></iframe>
<p style="font-size:13px"><a href="https://www.sprites.ai/tools/ab-test-significance-calculator">A/B Test Significance Calculator</a> by Sprites</p>What people ask about the a/b test significance calculator.
See it on your own ad accounts in a short demo, or try every part free for 7 days.
Use a two-proportion z-test. Pool the conversion rates, compute the standard error, divide the rate difference by that error to get z, then convert |z| to a two-tailed p-value. p under 0.05 is significant at 95% confidence.
It means that if there were truly no difference, a gap at least this large would show up less than 5% of the time by chance. It is not a 95% chance that B is better.
The percent change from A to B: (rate B − rate A) ÷ rate A. Going from 10% to 15% is a 50% relative lift, or a 5 percentage-point absolute lift.
Repeated looks without a sequential method inflate the false-positive rate. Decide the sample size first with a sample-size calculator, then evaluate significance once that sample is in.
The rate is still defined (0%), but relative lift from a zero baseline is undefined, and a pooled rate of 0% or 100% leaves no variance for a z-test. Wait for more data.
Research, creatives and media buying, run by one agent.