A/B Test Calculator
Determine if your two-tailed A/B test results are statistically significant.
Your calculation result will show up here.
Hypothesis: Two-sided
How does the A/B test calculator work?
Compare the conversion rates of two independent variants using their visitor and conversion counts. The tool reports each rate, relative lift, a p-value, and whether the difference meets your selected confidence threshold.
How to use this tool
Enter visitors and conversions for A and B, then choose a confidence level. The calculator uses a Pearson chi-square test on conversions and nonconversions with one degree of freedom. At 95% confidence, it calls a difference significant when p is below 0.05.
Example
If A has 100 conversions from 1,000 visitors and B has 120 from 1,000, their rates are 10% and 12%. That is a 20% relative lift, or a two-percentage-point absolute increase; the difference is not significant at 95% using this test.
Understanding your result
Use independent, randomly assigned visitors and no more than one counted conversion per visitor. Sparse expected counts can make this approximation unreliable. A nonsignificant result does not prove equality, and repeatedly checking until a test passes increases false-positive risk.
Related tools: Sample Size Calculator · Marketing Funnel Template.
Embed this calculator
Put this calculator on your website to increase visitor engagement and get more social shares!