A/B Test Significance Calculator
Enter the visitors and conversions for your control and variant to see whether your A/B test result is statistically significant (95%+ confidence).
Control (A)
Conversion rate: 10.00%
Variant (B)
Conversion rate: 13.00%
96.5% confidence
Statistically significant — Variant B is up 30.0%
You can be 95%+ confident this result is not due to chance.
How to read the result
The calculator runs a chi-square test comparing your two variants and returns a confidencelevel. At 95% or higher, the difference is statistically significant — you can be confident the winner is real and not random noise. Below 95%, keep collecting data.
Before you trust a result
- Run the test across a full business cycle (usually 1–2 weeks), not just a few days.
- Don't stop the moment a variant looks ahead — early leads frequently reverse.
- Make sure both variants ran at the same time with randomly split traffic.
Want this calculated automatically as your test runs? SplitLab tracks significance in real time on your own domain — see the full guide.
FAQ
What is statistical significance in A/B testing?
Statistical significance is the probability that the difference between your variants is real and not due to random chance. A result at 95% confidence means there is only a 5% chance the difference is a fluke. Most teams treat 95%+ as the threshold for calling a winner.
How does this calculator work?
It runs a chi-square test on your two variants using conversions and visitors, then converts the result into a confidence percentage. At 95% or higher, the difference is considered statistically significant.
Why is my A/B test not significant yet?
Usually you need more data. Small samples, small differences between variants, or an early read all produce low confidence. Keep the test running across a full business cycle until you reach 95%+.