Understanding Sample Ratio Mismatch in A/B Testing
Sample Ratio Mismatch (SRM) is a critical issue in A/B testing that occurs when the actual ratio of users assigned to different variants differs significantly from the expected ratio. This can indicate problems with your randomization process, data collection, or implementation.
Learn how to detect SRM using chi-squared tests, understand why it matters for experiment validity, and discover common causes and solutions. Try the SRM Checker tool to analyze your own experiments.