Masterarbeit Franziska Tobisch

Rapid changes in technology, markets, and customer needs force organizations to respond fast to remain competitive. In today’s digitally-driven world, especially software development has been affected by these emerging requirements. The resulting demand for agility gave rise to the success of agile software development, which has lastingly coined existing software development practices. Inspired by the success of agile methods on a small scale, large organizations have started to apply agile methods in a larger context, known as large-scale agile software development. However, the arising complexity of adopting agile practices at scale can cause various challenges. One example is the monitoring, coordination, and steering of the multi-layered development process. Metrics and the benefits they yield can support organizations in mastering these challenges of adopting large-scale agile software development and can be the basis of continuous improvement. Still, introducing and successfully using metrics in this context is a challenging endeavour itself. Besides integrating metrics into the complex, individual structure of scaled agile environments, metrics involve several general pitfalls, such as their dependency on data quality. Despite the central role of metrics in traditional software development contexts and their relevance in agile software development, research investigating metrics in industrial large-scale agile software development and related advice in literature is still scarce. However, due to the challenges and lacking experience regarding metrics for agile software development in large organizations, there is a need for guidance. To fill this gap, this thesis provides a metric catalog for the domain of large-scale agile software development, containing a structured collection of metrics used in industry. Therefore, this thesis first analyzes existing research on metrics in large-scale agile software development, followed by a presentation of the results of an interview study with experts from different organizations on the usage of metrics in the industry. Then, the thesis introduces a metric management fact sheet and a metric catalog for the domain of scaled agile software development, which documents metrics identified in the study with the designed fact sheet, both aiming to support organizations in establishing metrics. Subsequently, these two solution artifacts of this thesis are evaluated through surveys with 14 participants who also took part in the interview study. Finally, the key findings are summarized, and an outlook on possible further research is provided. The results indicate a high approval of the designed artifacts and a strong need for context-specific guidance regarding the establishment of metrics in large-scale agile software development.