A/B Test Significance Calculator

Plug in conversions and visitors per variant. Get statistical significance, lift, and a clear verdict on whether to ship.

Variant A (Control)

Variant B (Test)

Result

—
confidence level

How to read the result

  • 95%+ confidence: ship the winner.
  • 90–95%: trending — keep testing.
  • Below 90%: not significant. Run longer or larger.

Sample size matters

The most common A/B testing mistake is calling a winner too early. With 200 visitors per variant, you can’t reliably detect a 10% lift. Aim for 1,000+ conversions per variant before declaring.

Need a CRO testing program? →