Measure statistical significance and determine your winning variant
Conversion Rate = (Conversions / Visitors) × 100%
Statistical Significance: Typically requires 95% confidence (p-value < 0.05)
Run tests until you reach statistical significance to make confident decisions.
Your original/control version
Conversion Rate
0.00%
Your test/challenger version
Conversion Rate
0.00%
Improvement
0.00%
Confidence Level
0%
P-Value
-
Significant?
-
Ron Kohavi, Diane Tang, Ya Xu
The definitive guide to A/B testing from Microsoft's experimentation platform team. Covers statistics, pitfalls, and best practices.
David J. Bland & Alexander Osterwalder
Practical framework for designing and running experiments to validate business hypotheses and reduce risk.
Eric Ries
Foundational book on validated learning and the Build-Measure-Learn loop through continuous experimentation.