[simon]

That's easy ... the worst thing about most project management software is the project manager (PM) that uses it. :-)

Now that I have gotten 23 years of corporate IT cynicism out of my system I'll try to be more sensible. I have only rarely used PM software, so I really cannot speak to it, but I have been inflicted with bad PMs for the vast majority of my time.

My observation is that it is the way that PMs approach the process of PM, not the software that makes the difference. As long as projects start out with a due date and a cool name, it doesn't matter what the software is or does or doesn't do, because the project is doomed already. Who needs estimates when the completion date has already been carved into stone?

The in-house methodology states that requirements will be gathered and estimates will be collected and they'll be put into the approved software and then they will begin to suffer from bit rot because they were bad to start with and will never be revisited. The requirements are always incomplete because there are whole aspects of software development that are never considered, especially testing and creating administration capabilities. The estimates are always wrong because projects are rarely enough like previous ones to make prior experience useful, programmers are by nature optimistic and project managers are people pleasers. (I've seen my careful estimates cut in half right in front of me.) And don't even get me started on the "green shift" that PMs apply as they report up the food chain!

My favorite way of describing most IT projects is "measure with a micrometer, mark with chalk, cut with an axe!"