The approach I use to evaluate an Information Technology (IT) project's probability of failure is very much a heuristic one. My Leading Indicators are designed to steer the thinking through the issues rather than provide specific results. Thus the approach is a subjective review rather than any clearly defined objective approach.
There are many sources of information that point to specific examples of how projects fail, ranging from lack of stakeholder engagement, to poor requirements definition, to constant requests for change. In my view many of these are symptoms of more fundamental underlying problems. Therefore, from my experience the following are the items that, if addressed during a recovery effort, provide the basis of a project turnaround.
The easiest leading indicator of increased risk of project failure relates to the size of the project as reflected by the budget; the schedule; the team size; and number of teams involved. Indeed, risk of project failure seems to double as the project:
Statistics from the Standish Group's Chaos Report seem to validate this by reporting that small projects are far more successful than larger projects.