Using scenarios for measurement-based project control

Many software development organizations still lack support for obtaining intellectual con-trol over their development processes. This includes determining the performance of their processes and the quality of their products as well as the ability to react in a way that in-tended project and product goals can be fulfilled. Systematic support for detecting and react-ing to critical project states in order to achieve planned goals is usually missing. One means to institutionalize project control, systematic quality assurance, and management support on the basis of measurement and explicit models is the establishment of so-called Software Pro-ject Control Centres. From a research perspective, the set-up of such control centres is a challenging task: The metrics to be used typically need to be tailored to business goals and to the development context; appropriate interpretation and control techniques need to be com-bined to aggregate the information; different stakeholders require quite different visualizations for decision support. From a practical viewpoint, the deployment of a control centre is challenging because it needs to be tailored and integrated into a specific development envi-ronment that is often characterized by multi-disciplinary, distributed development. This requires careful analysis of usage scenarios for such control centres and smooth integration into the management and quality assurance processes of an organization. This article will give an introduction to basic control centre concepts and discuss their usage, exemplified by several typical scenarios from industry. Each scenario is described in the context of a related business goal, which should be achieved using control centres. The business goal is systematically mapped to selected measurement goals, and metrics are derived using the GQM paradigm. For interpreting and visualizing the measurement data, a Visualization Catena is defined, which formally describes what kind of data analysis has to be applied and how the data should be visualized. This way, statements about achieving the meas-urement and business goals can be made