My Software engineering Prof. Once wrote, those agile methods were first used by highly skilled teams that knew what they wanted. There has never been any evide

My Software engineering Prof. Once wrote, those agile methods were first used by highly skilled teams that knew what they wanted. There has never been any evidence for them to be a silver bullet. This is the case for pretty much anything new and cool. Big planning slows down experienced engineers, but can help the unskilled quite well... Plans serve (at least) two purposes. One is so that you know what you're doing. This doesn't have to be formal and can live partly inside your head. Experienced people probably do more of this, they just do it faster and don't bother writing as much of it out (because recalculating on-the-fly is faster and less error-prone than looking up last week's overly-detailed predictions). The other is so that other people know what you're doing and when you'll be done with it. For this the proper amount is whatever provides the most useful information (for all combinations of "most" and "useful"), without taking so much time as to cost more than it's worth.