Academy / A/B Testing Fundamentals

Sample size and test duration

Work out how many visitors a test needs and how long it must run, from your baseline rate, the smallest lift worth detecting, significance and power.

• Convert a relative minimum detectable effect into an absolute difference and explain how it drives sample size
• Calculate sample size per variation and total duration in full weeks from daily traffic and the number of variations
• Decide what to do when the required duration is too long: bolder change, higher-traffic page or closer metric