A/B Test Significance Calculator
Check statistical significance from impressions and conversions
도구를 불러오는 중…
🔒 Everything runs 100% in your browser. Your files and input are never uploaded to any server.
This calculator tells you whether your A/B test winner is real or just noise. Enter impressions and conversions for variant A and B, and it runs a two-proportion z-test to instantly show the p-value and significance status, all computed right in your browser with nothing sent to a server. It's built for marketers, growth teams, and product managers who need a quick read on experiment results.
How to use
- Choose your target confidence level (90%, 95%, or 99%).
- Enter the impressions and conversions for Variant A (Control).
- Enter the impressions and conversions for Variant B (Variation).
- Read the significance verdict, p-value, conversion rates, and relative uplift shown at the top.
FAQ
- Does a low p-value always mean B wins?
- No. The tool only declares a 'significant winner' when B's conversion rate is higher than A's AND the p-value is below the threshold for your chosen confidence level. A low p-value where B underperforms A won't be marked as a win.
- What if the result says 'not significant'?
- It usually means your sample size is too small or the real difference is minor. Keep running the test to collect more data, then recalculate.
- Is my data sent anywhere?
- No. All calculations happen instantly in your browser; nothing is stored or transmitted.