Investigating the Adoption of Metrics in Large-Scale Agile Software Development

Many organizations have started using agile methods to develop software in small projects. The success of agile methods on a small scale inspired organizations to scale them to larger contexts. However, adopting agile practices at scale can complicate the monitoring, coordination, and steering of the multi-layered development process. Metrics can address this challenge but are controversial since their implementation is a challenge in itself. Hitherto, research on large-scale agile development lacks publications investigating the adoption of metrics. We conducted an expert interview study to explore their (i) reasons for adopting metrics (ii), what metrics they use on the team, program, and portfolio level (iii), and the most occurring metrics in the expert organizations. Our results show that metrics are mainly used for transparency, improvement, and controlling. Most metrics occur on the program level. Finally, we identified story point estimation, velocity, and sprint burn-down, as the most established metrics.