Sample Size Calculator
Your current conversion rate (e.g., 5% means 5 conversions per 100 visitors)
Relative improvement you want to detect (e.g., 20% means detecting a 1% absolute lift from 5%)
Confidence level for your test
Probability of detecting a true effect
Daily visitors to estimate test duration
A/B Testing Tips
- Run tests for at least 1-2 weeks to account for weekly patterns
- Don't stop tests early - wait for statistical significance
- Test one variable at a time for clear insights
- Split traffic evenly (50/50) between variations
- Larger samples give more reliable results
- 95% confidence level is industry standard