Chapter 2: Test Data Generation

Incorrect test data leads to incorrect test results. 

Here's an example: if every virtual user sends the same username, adds the same product, or searches for the same keyword, the server will cache the result after the first request and use it for responding to subsequent ones. Your response times will look awesome. But your production system will not behave the same way, and the test will tell you nothing useful because you were not testing real life usage of the system under test.

Generating realistic, diverse test data at scale has historically been one of the most complex, time consuming (and boring) parts of test preparation. In some cases, coming up with proper test data takes more time and effort than building a test plan and analyzing the results. AI changes that considerably.