Bayesian AB Testing

<p>Randomized experiments, a.k.a.&nbsp;<strong>AB tests</strong>, are the established standard in the industry to estimate causal effects. Randomly assigning the treatment (new product, feature, UI, &hellip;) to a subset of the population (users, patients, customers, &hellip;) we ensure that, on average, the difference in outcomes (revenue, visits, clicks, &hellip;) can be attributed to the treatment. Established companies like&nbsp;<a href="https://partner.booking.com/en-gb/click-magazine/industry-perspectives/role-experimentation-bookingcom" rel="noopener ugc nofollow" target="_blank">Booking.com</a>&nbsp;report constantly running thousands of AB tests at the same time. And newer growing companies like&nbsp;<a href="https://blog.duolingo.com/improving-duolingo-one-experiment-at-a-time/" rel="noopener ugc nofollow" target="_blank">Duolingo</a>&nbsp;attribute a large chunk of their success to their culture of experimentation at scale.</p> <p><a href="https://towardsdatascience.com/bayesian-ab-testing-ed45cc8c964d"><strong>Read More</strong></a></p>
Tags: Bayesian