A statistical method and A/B testing framework that attempts to determine the likely probability an expected outcome will occur.

In comparison to traditional hypothesis-based A/B testing, the Bayesian methodology enables an A/B test to run with a much smaller sample size over a much shorter time frame.

Because it reduces traffic and time constraints, some testers consider it the superior testing methodology.

However, the tradeoff is experimenters sacrifice some degree of statistical rigour because the results only provide a reasonably likely outcome of what is expected to occur -- rather than a proven outcome of what has occurred, based on the sample tested.

Many A/B testing platforms, including VWO Smarts Stat Mode and Google Optimize currently use a Bayesian testing methodology.