A/B Test Sample-Size Estimator
Uses z-test formula to compute needed samples for given baseline & lift
Test Parameters
Current conversion rate of your control
Minimum relative improvement to detect
Confidence level (1 - Type I error rate)
Probability of detecting true effect
💡 Pro Tips for Lazy Marketers
- • Use 95% significance and 80% power as standard settings
- • Smaller lifts require much larger sample sizes to detect
- • Don't peek at results early - wait for full sample size
- • Consider running tests for full business cycles (weeks)
- • Factor in seasonality and external events when planning test duration