Free tools
A/B testing calculators
Four small, honest calculators for planning, reading and sanity-checking experiments. Instant results, the formula shown under each one, your inputs kept in the link, and no email wall.
Before the test
Sample-size and duration calculator
How many visitors you need to detect a given lift, and how many days that takes at your traffic. The same method as the planner inside the plugin.
You enter: baseline rate, lift to detect, confidence, visitors per day.You get: visitors per variant, in total, and the days to run.
Open the calculatorAfter the test
A/B test significance calculator
Is the difference real? A two-proportion z-test with a plain verdict, the two conversion rates, the uplift and the confidence, the same test Pro runs on live results.
You enter: visitors and conversions for A and B, your threshold.You get: a verdict, both rates, the uplift, confidence and p-value.
Open the calculatorDuring the test
Sample ratio mismatch checker
Is the traffic split what you set? A chi-squared check that tells you whether an imbalance is normal variation or a broken test you should not read.
You enter: visitors in each variant and the split you expected.You get: healthy, suspicious or broken, with the odds and the p-value.
Open the checkerAny time
Conversion rate calculator
Your conversion rate, revenue per visitor and average order value, plus what a given uplift would be worth per period and per year.
You enter: visitors, conversions, optionally revenue and an uplift.You get: the rate, revenue per visitor, order value, the value of the lift.
Open the calculatorHow they fit together: size the test before it starts, check the split while it runs, judge the result when it ends, and put a number on the lift any time.
Prefer this inside WordPress? The free plugin includes a sample-size & duration planner, and Pro computes significance on your live results automatically. New to testing? Start with the guides.
Do the math where the test runs
The free plugin plans your tests; Pro reads them for you - significance, Bayesian probability, and SRM checks.