π A/B Test Significance Calculator
Enter control and variant data to automatically calculate conversion rates, uplift, Z-score, P-value, and statistical significance. Supports 90%/95%/99% confidence levels. Runs entirely in your browser.
Experiment Data
Control
Variant
Quick examples:
Control Conversion Rate
β
Variant Conversion Rate
β
Relative Uplift
β
Absolute Uplift
β
| Z-Score | β |
| P-value | β |
| Confidence Interval (95%) | β |
| Statistical Power | β |
About A/B Test Significance
A/B test significance testing determines whether the difference between your control and variant is caused by a real effect rather than random variation. This tool uses the two-proportion Z-test to evaluate the statistical significance of conversion rate differences between two groups.
How to Use
- Enter data: Fill in the total visitors and conversions for both the control and variant groups
- Choose confidence level: 90% (quick tests), 95% (standard), or 99% (high-stakes)
- Review results: Conversion rate comparison, uplift, Z-score, P-value, and significance verdict
Calculation Method
This calculator uses the two-proportion Z-test with the following formulas:
- Pooled conversion rate:
p = (xβ + xβ) / (nβ + nβ) - Standard error:
SE = β(pΓ(1-p)Γ(1/nβ + 1/nβ)) - Z-score:
Z = (pβ - pβ) / SE - P-value:
P = 2Γ(1 - Ξ¦(|Z|))(two-tailed)
Frequently Asked Questions
- What does a significant result mean? β At the chosen confidence level, the variant's uplift is unlikely (probability below the threshold) to be random noise.
- What if the result is not significant? β Your sample size may be too small, or there may be no real difference. Try collecting more data and retesting.
- Is my data safe? β All calculations run locally in your browser. Nothing is uploaded to any server.