Free calculator

Is your A/B test result real, or just noise?

Enter the visitor and conversion numbers from your two ad variants and this tool works out whether the difference between them is statistically significant, or small enough to have happened by chance. It's built for anyone comparing ad creative, landing pages or email subject lines who wants a clearer answer than gut feel, though like any calculator it can only be as good as the numbers you put in.

How-To Guide

Understanding your A/B test results before you act on them

The A/B Test Significance Calculator works out whether the difference between two versions of an ad, landing page or email is likely real or just noise from a small sample. This guide is for marketers and business owners running paid campaigns or on-site tests who want to know when a result is worth acting on. Like any calculator, it can only work with the numbers you give it, so a small sample or an early reading can still point you the wrong way.

Level: Beginner
Best for: Marketers and business owners running paid ad or landing page tests
Screenshot of the A/B Test Significance Calculator with fields for visitors and conversions per variant
  1. Enter your test numbers

    Add the number of visitors and conversions for each variant you're comparing.

  2. Read the significance result

    See whether the gap between your variants is statistically significant or could plausibly be down to chance.

  3. Decide what to do next

    Use the result to decide whether to declare a winner, keep testing, or run the test for longer before drawing conclusions.

A/B Test Significance Calculator

Enter the visitor and conversion counts for your control and variant. The calculator works out each conversion rate, the uplift, and whether the difference is statistically significant at the 95% confidence level.

Control
Variant
Free Tools

See whether your test result is real, or just noise

The calculator checks whether the difference between two versions of an ad, landing page or email is large enough to trust, using standard significance testing. It's built for marketers and business owners running their own tests without a data team to check the sums. Like any statistical tool, it works from whatever numbers you enter, so a small sample or a test stopped too early can still produce a confident-looking answer.

Free to use, no account needed, works on any device