A/B Test Calculator
Calculate statistical significance for your A/B tests to make data-driven decisions.
Enter Test Data
Control (Variant A)
Conversion Rate: 0.00%
Variant B (Test)
Conversion Rate: 0.00%
Results
Statistical Significance
—
Not yet significant (need more data)
Relative Uplift
—
Variant B vs Control
Enter your test data to see results.
| Metric | Control A | Variant B |
|---|---|---|
| Visitors | 0 | 0 |
| Conversions | 0 | 0 |
| Conv. Rate | 0.00% | 0.00% |
How Much Traffic Do You Need?
The sample size needed depends on your baseline conversion rate and the minimum effect you want to detect.
| Baseline CVR | 5% Uplift | 10% Uplift | 20% Uplift |
|---|---|---|---|
| 1% | ~310K / variant | ~78K / variant | ~20K / variant |
| 2% | ~155K / variant | ~39K / variant | ~10K / variant |
| 5% | ~62K / variant | ~16K / variant | ~4K / variant |
| 10% | ~31K / variant | ~8K / variant | ~2K / variant |
Based on 95% confidence level and 80% statistical power. Actual requirements may vary.
A/B Testing Best Practices
Do:
- •Test one variable at a time
- •Wait for statistical significance
- •Run tests for full business cycles
- •Document all tests and learnings
Don't:
- •Stop tests early when you see a winner
- •Change test parameters mid-experiment
- •Ignore external factors (seasons, sales)
- •Run multiple tests on same audience
Related Tools
Want more context on testing?
The weekly briefing covers metric interpretation and decision framing for scaling.