All tools
Calculators
A/B Test Sample Size Calculator
How many users do you need per variant for the test to actually conclude? Standard difference-in-proportions math, no AI.
Free · no signup required · runs entirely in your browser
%
Absolute: a change in percentage points (5% → 6% = 1pp). Relative: a % change of the baseline (5% × 20% = 6%).
Two-tailed is the default — you don't yet know whether your variant is better or worse. One-tailed only when you have a strong prior the change can't backfire.
Sample size per variant8,158
Total sample (control + variant)16,316
Estimated test duration17 days
Verdict
Standard test
Test fits in a 1-4 week window. Most A/B tests should run at least 2 weeks regardless of sample size to absorb day-of-week effects.