Globalization has long since found its way into software engineering. Companies transfer part of their development activities to low wage countries in order to achieve a better cost position, get access to local markets and react to the prevailing lack of specialized workforce. However, Global Software Engineering (GSE) has also brought new challenges: Geographic separation, different time zones, and culture and language barriers make the collaboration of team members more diffcult and often lead to quality problems, project delays, and cost overruns. Project management metrics to monitor cost, time, and quality characteristics of a development project, are state of the practice today. However, these metrics often reveal problems in GSE projects too late, since they measure the symptoms – exceeded cost, missed milestones or poor quality – rather than the root causes of the problems.
In this paper, we report on a study of new project management metrics based on measurement of communication and collaboration, which provide project managers with an early warning system. We introduce a collaboration-based measurement model – a set of “GSE metrics” – and describe case studies of its application in multiple projects. In the observed projects, the GSE metrics were helpful to identify and resolve collaboration problems. Despite the subjective nature of the metrics, the results showed the GSE metrics to be useful and to accurately reject the reality. The effort for the application of the GSE metrics turned out to be a “low investment” with a high “return on invest”.