My experience in software is similar. While companies go out of their way to hire smart people who regularly come up with improvements to how things are done,

My experience in software is similar. While companies go out of their way to hire smart people who regularly come up with improvements to how things are done, their efforts are easily thwarted by the marching morons of the status quo (which is obviously working so what's the problem?). Those who luck out and break out from the herd to implement those improvements end up getting promoted. But in the employer's defense, once these guys are in the driver's seat, it can frequently be hard to distinguish the signal from the smart guys from the rantings of particularly vocal morons. I think the take home lesson is to be very careful choosing your co-workers and manager. As for the subject of this thread, there are no magic bullets, right? Sometimes a plan is awesome when the target is clear. Other times it only gets in the way when one is doing a significant amount of research in parallel with the development. And everything in between...