Assign Experiment Variants at Scale in Online Controlled Experiments

TL;DR

This paper introduces a novel, fast, and statistically sound algorithm for assigning experiment variants at scale in online A/B testing, ensuring unbiased and independent randomization across numerous concurrent experiments.

Contribution

The paper presents a new assignment algorithm that is computationally efficient, maintains independence between experiments, and is suitable for large-scale online A/B testing environments.

Findings

Algorithm is computationally fast and scalable.

Assignments are unbiased and independent.

Validated through statistical tests.

Abstract

Online controlled experiments (A/B tests) have become the gold standard for learning the impact of new product features in technology companies. Randomization enables the inference of causality from an A/B test. The randomized assignment maps end users to experiment buckets and balances user characteristics between the groups. Therefore, experiments can attribute any outcome differences between the experiment groups to the product feature under experiment. Technology companies run A/B tests at scale -- hundreds if not thousands of A/B tests concurrently, each with millions of users. The large scale poses unique challenges to randomization. First, the randomized assignment must be fast since the experiment service receives hundreds of thousands of queries per second. Second, the variant assignments must be independent between experiments. Third, the assignment must be consistent when users revisit or an experiment enrolls more users. We present a novel assignment algorithm and statistical tests to validate the randomized assignments. Our results demonstrate that not only is this algorithm computationally fast but also satisfies the statistical requirements -- unbiased and independent.