A/B test significance calculator
Check whether variant B really converts differently from A, and how many visitors you need.
How it works
A two-proportion z-test compares the conversion rates, and the p-value is the chance of a gap this big if there were no real difference.
The sample size is what each variant needs to detect the smallest lift you care about, at your confidence and 80% power.
z = (p_B − p_A) ÷ √(p(1 − p)(1/n_A + 1/n_B))Questions
Can I stop as soon as it’s significant?
Stopping early inflates false positives. Decide the sample size first and stop when you reach it.
Why do I need so many visitors?
Small lifts on low conversion rates are hard to tell from noise.