When You Should Prefer “Thompson Sampling” Over A/B Tests

Imagine you have two ads to choose from: the red one and the blue one. Of course, you would like to show your users the ad with the highest click rate. <img alt="" src="https://miro.medium.com/v2/resize:fit:770/1*nvzDOazRkwhCuGT03NJTZw.png" style="height:290px; width:700px" /> What to show the user, the red ad or the blue ad? [Image by Author] But how do you find out which ad has the highest click rate? The most common approach to answering this question is doing an A/B test. This implies setting apart some users and showing the first ad to half of them and the second ad to the other half. Finally, you can compute the click rate of each alternative and select the best one. <a href="https://towardsdatascience.com/when-you-should-prefer-thompson-sampling-over-a-b-tests-5e789b480458">Learn More</a>