Free A/B test significance calculator
Check whether your A/B test result is statistically significant. Enter visitors and conversions for two versions. Free, no signup.
Result
What it tests
The calculator runs a two-sided two-proportion z-test. It asks: if A and B really converted at the same rate, how likely is a gap this big? That probability is the p-value. Below 0.05 (95% confidence) is the usual bar for calling a winner.
Avoid common mistakes
- Don't stop early. Checking daily and stopping the first time it looks significant inflates false wins. Decide the sample size first.
- Test one change. If two things differ you won't know which one mattered.
- Small samples mislead. A handful of conversions per version says very little.
Not sure what to test? Find your weakest step with a funnel; our guides show how.
Questions
What confidence level should I use?
95% is the common default. Use a stricter level for changes that are hard to undo.
Does it work for small numbers?
It uses a normal approximation, which is unreliable with very few conversions (under about 10 per version). Treat those results as a hint only.
Want to see this for your own site?
agentlytics counts visitors, conversions and campaigns without cookies, and your AI agent can read it.
Start free See the live demo