Identifying trouble patterns in complex IT services engagements

What causes complex information technology services engagements to fail? The root cause for the failure may derive from several sources, including unclear requirements, poor project management practices, underestimation of technical complexity, not accounting for the overhead in using global labor arbitrage, and rapid scope or schedule changes. In this paper, we explore how quantifiable measures of project progress gathered at several important stages of the life cycle of a project can aid in early identification of troubled projects. Specifically, we define derived statistical measures that can be used to predict the eventual outcome of a project. These predictions are used to initiate accelerated risk mitigation plans. In addition to the discussion surrounding the prediction models, we also discuss the importance of the “packaging” and presentation of the statistical data that result in the successful use of our system. Our work has been adopted and used in a project health management system that is deployed worldwide by IBM Global Business Services.