Search This Blog

Thursday, January 30, 2014

Need for Statistics for Project Managers

“I am not good at statistics. So, I am not cut out to be a project manager,” said a potential PMP aspirant after attending a PMP boot camp. My heart slipped a few beats to regain my stand and found out from this potential PMP candidate managing dates in the timeline, evaluating project slips, looking at contracts for vendor management, and monitoring metrics demand high mathematical skills for a project manager to be successful.

Let us look at the core knowledge areas in Project Management Book of Knowledge that includes management of scope, time, cost, risk, quality, communication, procurement, integration, human resources, and stake holders. Most of the process groups within each one of these domain knowledge areas involve qualitative thinking, leadership, organization, negotiation, and people management skills. For instance, the project manager that asks for what percentage of a task is completed draws attention in today’s context because even if 80% of the task is completed, the remaining 20% of the task can take more time to complete. Understanding people’s commitments and winning consensus towards project goals is central to task completion, and therefore eventually project success.

Even when we focus on metrics, such as cost or schedule variance, schedule or cost performance index, or other earned value metrics like planned value and actual value, estimate at completion, etc., the mathematical formulas involve simple subtraction and proportions. Do these mean they are statistics? How much difficulty exists in comparing the milestone slips from baseline to actual? Simple office automation tools like Microsoft Excel can accomplish these computations effectively and if the proper use of Microsoft Project is used, then these metrics can be easily computed.

Now, let us turn our attention to Agile. Technically, project plans are not preferred in this setting as the teams are dedicated and self-managed. The project manager is not managing the tasks. Focus shifts more towards product features, benefits to business, etc. Neither the estimate gathering process like planning poker or metric computation like velocity planning is quantitative. Besides, good application lifecycle management (ALM) tools like SpiraTeam, Version One, or Rally allow such metric gathering more effectively. So, where are statistics coming in to play as a core skill of a project manager?

Let us not leave capital project selection where the management using steering or governing committee must make a conscious selection of products to invest or consume resources to work on. These ranking of projects use basic mathematical concepts like payback period, net present value, or internal rate of return. None of these methods call for detailed analysis of variance, kurtosis, skewness, or involve factor analysis.

Besides, if simple observations of ratio and proportion or central tendency analysis using mean, mode, or median indicate statistical expertise, then, how much time are a Project Manager spending on such analysis compared to communicating and managing stakeholder expectations to ensure project success? It is true that interpreting such data for preventive and corrective actions as well as escalating risks and issues require some understanding of these thoughts. So, is statistics a core skill for PMP aspirants? Readers, what do you think?