A/B Test - Sample Size Calculator

How many subjects are needed for an A/B test?

%

The basic conversion to which you want to compare the results.

%

Smallest effect to be detected in (1-β)% of cases.


(0% - 0%)

(0% - 0%)
80%

Assumed probability of detecting a minimal effect, if any.

95%

The probability that the difference we are looking for is within the range we are examining.

This is how many users you need per test variant.

The above calculator is used to calculate the number of users per A / B test version.

If you want to use the calculator:

  • enter the base conversion for the tested element

  • define the minimum detectable effect that you want to detect with the A / B test,

  • decide if you want to measure the minimum detectable effect in percentage or percentage points,

  • determine the statistical power and confidence level. If you don’t know what these parameters are, leave the default values.

On the right side (or at the bottom in the mobile version) you will see the result calculated according to statistical formulas. The result shows the minimum number of users needed per experiment variant.