Most people think of AB tests as a way to randomize users into groups and test different product designs for them. Actually, it doesn't need to be users. We also ran tests on stores, cities and other entities. At the same time we may test different pricing strategies, motivation systems and routing algorithms. Versatile AB Platform should support all randomization units applicable to your business model.

As the company grows more and more teams ran their specific AB tests and may be affected by the tests from other departments. This is when AB tests calendar is needed. It provides observability and keeps the whole company on-track with what tests are going right now.

We found that Data Analysts spend 50% to 80% of their time on analyzing the results of AB tests. This makes tests expensive and time consuming. We at Tez Data believe that most of this work could be eliminated by automated results calculation. This way most of the conclusions could be made right after the end of the test and Data Analysts are required only to check the overall validity of the experiment.

If you are using AB tests in your company you might have heard that many of them turned our to be inconclusive. Our team may help you to increase the sensitivity of your tests applying advanced mathematical techniques. This way you will have more results to make decisions and drive the growth of your business.