By Mohit “Mo” Sahgal
In this third installment, I explore the use of factors, and sizing of the same to improve work effort estimates. Oftentimes, even on complex projects which should require formal estimation, no work effort estimates are produced. In other cases, no factors, inappropriate or incorrect factors, too many or too few factors, are used to size the work effort. In all cases, at least one or more factors can be found to rationalize the work effort. In other words, avoiding the definition of factors is just a poor excuse for justifying guessing. Why would you guess? The answer is simple: you should never guess.
Factors should size the work effort for all of the tasks in the methodology required to deliver the solution (objective). How are factors defined? Factors are based on the work products or deliverables for each task. Each task must have an outcome (product); otherwise, why perform the task? Common outcomes include, number of requirements, number of use cases, number of processes, number of tables to be designed and implemented, number of reports, number of operational procedures, number of programs, number of users to be trained, number of critical data elements, and so on. I think you get the point. Only in very rare circumstances is it possible that a factor to size the work effort is elusive. Principle: All tasks have pre-defined outcomes; all outcomes can be quantified; therefore, the work effort for all outcomes can be quantified. How many factors do you need? It depends on the tasks, and associated deliverables, that comprise the approach or methodology. If one or more factors cannot be found for a particular task, it is highly probable that the task must be decomposed into smaller units of activity for which the work effort can be quantified. In the rare event that no factors can be found, then make appropriate assumptions.
Derived factors, or factors based on factors, are acceptable. For most projects, complex factors and factor interrelationships are not required. How sophisticated does the sizing calculations need to be? According to Steve McConnell, “Complex formulas sometimes do more harm than good”[1]. McConnell is right – select an approach that works, and use it consistently. Just don’t guess, and avoid creating more complete and accurate work effort estimates.
1] McConnell, Steve, 2006, “Software Estimation: Demystifying the Black Art”, Microsoft Press
Mr. Mohit “Mo” Sahgal is the VP of Analytics of Paradigm Technology. He is an accomplished Senior Executive with decades of management consulting expertise. Mo has held various leadership positions including Senior Executive at Accenture, Partner at IBM, Partner at Capco, and Executive Director at Ernst & Young LLP. Mr. Sahgal LinkedIn profile can be found here.
Paradigm Technology is a strategic consulting company serving the utilities, banking, airline, manufacturing and high-tech verticals. We utilize innovative business and technology solutions to help clients enable their digital transformation programs, and improve their Analytics, Cloud, Master Data Management, and Project Leadership solution delivery. Paradigm is ready to support you in your business journey. For more information about Paradigm Technology, email [email protected] or visit us at www.pt-corp.com.