I'm listening to the "Vision" audio downloaded from someones BLOG reference (I forgot). I heard Clark speak about project failure rates - NOT on time, on budget. If "on time, on budget" is the measure of project success no wonder the failure rate is so high.
No project in the aerospace, major construction, petrochem or any other domain I've worked in has an "ending budget" the same as the starting budget. Rebudgeting or "re baselining" is part of the project management process. So judging projects by the metric "did we come in on the original budget?" then we're doomed. Change control applies to budget, features, duration.
What is needed in these discussion - project failure - is to assess the "success criteria" upfront in the project charter. How much over the initial baseline are we willing to accept WITHOUT change in features?
I'm not a fan of the Iron Triangle, but this is a good analogy. How far can the sides of the triangle stretch before we consider it a problem? I now the Standish numbers are used everywhere to rationalize the use of new and innovative PM or development techniques. But those numbers, when used in the absence of a context, business domain, and scalar "trip points" are meaningless.
We as project professionals need to speak up about this misuse of statistics and establish a better set of success criteria for projects.
