We often have conversations with our clients about risk management and regression testing because these days with development cycles so short, we rarely have time for full regression. So we have to choose wisely where to spend our time; manual testing, automated testing, exploratory testing, platform compatibility, etc. It’s not an easy task in developing a test strategy in terms of where to put our effort so that we can minimize risk while achieving the level of confidence that we want that the release is ‘defect-free’. I started thinking about definitions of priority and risk in a test case context. Which test cases come first and when? Major release, minor release, patch release? I developed this rudimentary matrix trying to think out loud about the issue. Of course the colors between the cells would be contiguous shading and would not be that distinct. Red would be for tests that need to be executed with every release no matter what, where as yellow, only full regression. These are just some initial ideas. If you have any comments or feedback, please let me know.
Regression Testing Strategy Diagram