STAREAST 2009: Measuring the Value of Testing

Value is based on objectives, so why do we test? We test to find defects effectively, gain confidence in the software, and assess risk. So, the value of testing should be measured based on test effectiveness, confidence validation, and reduced system risk. In terms of testing effectiveness, the most useful metric is defect detection percentage (DDP)-the ratio of defects found in testing divided by the total number of defects found by testing and users in production. Dorothy Graham explains when to use this metric and outlines the choices, problems, and benefits of using DDP. In addition, Dorothy describes confidence and risk metrics that will tell you if you are going in the right direction-or not. She explains how to take costs into account to assess the return on investment for testing and outlines a simple way to bring home the message about the value of testing in terms of what the organization can save.