Anytime-Valid Confidence Sequences in an Enterprise A/B Testing Platform

TL;DR

This paper introduces a new approach for A/B testing using anytime-valid confidence sequences, enabling continuous monitoring and stopping without inflating error rates, demonstrated on real and simulated data.

Contribution

The paper presents the adaptation and deployment of asymptotic confidence sequences in a commercial A/B testing platform, allowing for flexible, error-controlled experimentation.

Findings

Effective continuous monitoring with confidence sequences

Controlled error rates during data-dependent stopping

Successful deployment on real-world experiments

Abstract

A/B tests are the gold standard for evaluating digital experiences on the web. However, traditional "fixed-horizon" statistical methods are often incompatible with the needs of modern industry practitioners as they do not permit continuous monitoring of experiments. Frequent evaluation of fixed-horizon tests ("peeking") leads to inflated type-I error and can result in erroneous conclusions. We have released an experimentation service on the Adobe Experience Platform based on anytime-valid confidence sequences, allowing for continuous monitoring of the A/B test and data-dependent stopping. We demonstrate how we adapted and deployed asymptotic confidence sequences in a full featured A/B testing platform, describe how sample size calculations can be performed, and how alternate test statistics like "lift" can be analyzed. On both simulated data and thousands of real experiments, we show the desirable properties of using anytime-valid methods instead of traditional approaches.