A/B Testing
Create

A/B testing guide

A/B Testing Sample Size: How Many Visitors Do You Need?

Sample-size planning protects you from tests that can never answer the question you are asking. The most important inputs are the baseline conversion rate and the smallest improvement that would change your decision.

Baseline conversion rate

A page converting at 2% behaves differently from one converting at 40%. Use recent representative data for the metric you will actually evaluate.

Minimum detectable effect

Choose the smallest lift worth acting on. Smaller effects require more observations, sometimes dramatically more.

Confidence and power

A common planning setup uses a 5% significance level and 80% power. These choices control false-positive and false-negative risk.

Plan before exposure

Choose the metric, effect and stopping rule before viewing variant outcomes. Post-hoc changes make results harder to interpret.

Use the number operationally

Translate required sample into days using eligible daily traffic. If the duration is unrealistic, reconsider the experimental question before launch.

Related tools and guides

Calculate sample sizeEstimate durationHow long should a test run?

Have two options to compare?

Create a focused preference test and collect evidence before the decision gets expensive.

Create a survey