Vertically Scaling Agile : A Multiple-Case Study

• Main Authors: Östman, Nicklas, Lindström, Rasmus
• Format: Others
• Language: English
• Published: Linköpings universitet, Programvara och system 2017
• Subjects: Agile development; Decision-making; Agility; Scaling Agile; Agile management; Computer and Information Sciences; Data- och informationsvetenskap
• Online Access: http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-136458

The conceptual framework of agile software development is an ever-growing movement in the software industry. However, recent studies have shown that large, less software-focused companies, where software development is primarily used for in-house IT-solutions, struggle with giving up traditional command-control type of management. This hits hard on some of the most important principles of agile software development and in many cases this phenomenon has inevitably led to large gaps between development teams and more managerial parts of the organization. This thesis has aimed to study this gap and investigate how it affects software development teams’ ability to carry out their work. By comparing three software teams that were internally highly similar but with varying external conditions, impact on the teams’ behaviour based on their different environments was studied. The study was carried out using a multiple-case study approach with primary data sources consisting of survey gathered data from all team members and interviews with a subset of the team members. The results gathered from this study suggest that agile development teams are extremely dependent on a well-functioning interface to business related parts of an organization. Regarding teams’ ability to make decisions and being agile in their way of working, the results primarily isolate impediments with roots in an unwillingness to adhere to and lack of understanding of agile principles. In this thesis, our gathered results were also correlated with a modern framework called Flow in order to confirm its relevance regarding analyzing software development teams in large-scale environments.