I've heard Kent Beck talk about having the smallest amount of tests that give you confidence. Which he also gave as an answer here: http://stackoverflow.com/q

I've heard Kent Beck talk about having the smallest amount of tests that give you confidence. Which he also gave as an answer here: http://stackoverflow.com/questions/153234/how-deep-are-your-unit-tests/153565#153565 http://stackoverflow.com/questions/153234/how-deep-are-your-... But I know a lot of people in the early days of XP went to extremes, 100% code coverage, mutation tools for every condition to ensure unit tests broke in expected ways, etc. But they were more experiments in pushing the limits rather than things that gave productivity gains.