What I'd be interested in knowing is exactly what caused the differences between the initial project estimates and the eventual outcomes.

It could be that the people running the projects are all completely useless and that the projects are all a shambles.

On the other hand, it could be that the projects are run fine and it is the crapness of the original estimates that give rise to the discrepancies.

Let's not forget that a key way of getting large projects authorised is to deliberately underestimate the initial costs & timescales in the knowledge that by the time the customers find out what it's really going to cost, it'll be too late and they can't can the project for fear of looking foolish.

In addition to crap projects and crap estimates, we have the issue of crap customers, i.e. the ones that change their minds every 10 minutes and act surprised when it turns out that this affects the project costs. How much of these DWP overruns were caused by change control items?

There's a lot more to all of this than meets the eye. Perhaps you professional journalists out there could find out and tell us.

The problem with large projects arises when the client keeps changing the original specification. I have known large projects that were done on time and to the estimate because nobody changed anything.This was much to the annoyance of the contracting company, who put in a low bid and lost money, as they was hoping to recoup money on alterations to the contract.

See this: DWP fires Siemens from late project

http://www.kablenet.com/kd.nsf/FrontpageRSS/CC0B1CA7888BE4C88025740A003DFA07!OpenDocument