Original Post Date: Monday, November 28, 2011 

Check out this blog post  on project estimation.  The author discusses the practice of ‘padding’ effort estimates and how destructive this practice can be to healthy project management.  She suggests that project team members, rather than padding their individual efforts at a task level, should collaborate with project management in order to produce a good solid project plan with sufficient contingency reserves.  This allows for the project plan to reflect the most likely case but contains a safety net for those cases where the stuff that was unknown at the time of project planning disrupts the project.  The management should assign these reserves based on their degree of confidence that they understand potential unknowns and the criticality of the success of the project.

This is a great idea but how is management supposed to determine the right amount of reserves for a particular project.  A risk analysis exercise is a great way to do this since it provides an analytical way to translate uncertainty about the knowns and unknowns into a probability distribution that represents all the potential effort possibilities for the project each assigned a confidence level.  So rather than adding a 30% reserve the project manager who wants an 90% certainty that their project will finish on time and budget can develop a plan based on the most likely estimate (let’s say the confidence level on the most likely estimate is at the 70% confidence level) and establish a reserve equal to the difference between the 70% estimate and the 90% estimate.  This gives project management a means to credibly quantify their legitimate uncertainties into a sensible and defendable contingency reserve.

How do you handle the uncertainties associated with unknowns when you plan projects?