Auto-Allocation powered by Multi-Armed Bandits (MAB) lets Convert automatically send more traffic to better-performing variations while your test is running.

Instead of fixed splits, MAB adapts in real time—helping you learn faster, reduce wasted traffic, and drive more conversions during the experiment. Works with both Frequentist and Bayesian stats and supports strategies like Thompson Sampling, Epsilon-Greedy, and UCB.

Once enabled, traffic allocation is fully automatic, with clear reporting to show how it evolves over time.

Learn more: https://support.convert.com/hc/en-us/articles/multi-armed-bandit-mab-auto-allocation-in-convert