This calculator helps determine the minimum detectable effect (MDE) for your A/B tests based on traffic, conversion rate, and experiment duration.
See Alexey's Swing Size Calculator for earlier version
Key features:
- One-sided tests: Use when you only care about detecting improvements (recommended for most A/B tests)
- 95% power recommended: Higher power means you're less likely to miss real effects
- Both analysis methods: Shows results for both Frequentist (fixed sample) and Sequential (optional stopping) approaches
- MDE is shown as relative lift: The percentage increase/decrease relative to your baseline conversion rate
- Color coding: Green (<5% lift), Yellow (5-10% lift), Red (>10% lift)
- Experiment recommendations: Get specific test design suggestions optimized for your traffic and conversion rate
Note: Sequential testing requires larger MDEs due to more conservative statistical thresholds to account for multiple looks at the data.