Skip to content

A/B test significance calculator

Enter the visitors and conversions from a finished test. The calculator shows whether the difference between two versions is real, or whether it needs more traffic before you can tell.

Confidence level
Hypothesis
Variant Visitors Conversions Conversion rate Remove

Enter your test data

Visitors and conversions for the control and at least one variant. Results update as you type.

How much better or worse each variant really is

How these numbers are calculated
Significance test

Two-proportion z-test against the control, with unpooled standard errors so the interval and the verdict cannot disagree.

SE_a = sqrt( p_a (1 - p_a) / n_a ) SE_b = sqrt( p_b (1 - p_b) / n_b ) SE_diff = sqrt( SE_a^2 + SE_b^2 ) z = ( p_b - p_a ) / SE_diff
Confidence interval

Calculated on the absolute difference, then divided by the control rate to read in relative terms.

CI = ( p_b - p_a ) +/- z_crit x SE_diff
Observed power

The probability of detecting an effect the size of the one measured, at this sample size and confidence level.

power = phi( |z| - z_crit )
Multiple comparisons

Sidak correction across k comparisons against the control, holding the test-wide false positive rate at the level selected.

alpha_adj = 1 - (1 - alpha)^(1 / k)
Sample ratio mismatch

Chi-square goodness of fit on the visitor counts against an even split, flagged below p = 0.001.

X^2 = sum( (observed - expected)^2 / expected ), df = k - 1
Limits
  • Fixed-horizon test. Set the sample size in advance and evaluate once. Stopping at the first significant reading inflates false positives beyond the stated level.
  • Binary outcomes only. Revenue per visitor, average order value and items per order are skewed by large orders and need different methods.
  • The normal approximation needs roughly 5 conversions and 5 non-conversions per variant as a floor.
  • Significance is not effect size. A significant result can still be too small to be worth shipping.
  • Sizing a test before you launch it is a different calculation, on the sample size calculator.

Get started
with Wisepops

Book a demo to get a personalized audit of your traffic.

Help