Smaller effects need much larger samples: halving the MDE roughly quadruples the traffic required. Choose the MDE from what would change a business decision, then calculate whether you have the traffic to detect it in a reasonable time.
Example
At a 2% baseline, detecting a 10% relative uplift with 80% power at 95% confidence needs roughly 80,000 visitors per variant.
Check whether a test result is real, or plan how many visitors and days a test needs before you start.
Related terms
Part of the growth marketing glossary by Shiva Sharma.