To main content

Networking in a large-scale distributed agile project


Context: In large-scale distributed software projects the expertise may be scattered across multiple locations.

Goal: We describe and discuss a large-scale distributed agile project at Ericsson, a multinational telecommunications company headquartered in Sweden. The project is distributed across four development locations (one in Sweden, one in Korea and two in China) and employs 17 teams. In such a large scale environment the challenge is to have as few dependences between teams as possible, which is one reason why Ericsson introduced cross-functional feature teams -- teams that are capable of taking the full responsibility for implementing one entire feature. To support such teams when solving problems, ensure knowledge sharing within the project and safeguard the quality Ericsson introduced a new role -- Technical Area Responsible (TAR).

Method: We conducted extensive fieldwork for 9 months at two Ericsson sites in Sweden and China. We interviewed representatives from different roles in the organization, in addition to focus groups and a survey with seven teams.

Results: We describe the TAR role, and how the TARs communicate, coordinate and support the teams. Also architects support the teams, however not as closely as the TARs. We found that the TAR is usually a senior developer working halftime or fulltime in the role. We also present measures of the actual knowledge network of three Chinese and three Swedish teams and the TARs position in it.

Conclusions: TARs are central in the knowledge network and act as the boundary spanners between the teams and between the sites. We learned that availability of the TARs across sites is lower than that with local TARs. We also found that the size of a team's knowledge network depends on how long the team members have been working in the company. Finally we discuss the advantages and the challenges of introducing experts in key roles in large scale distributed agile development.


Academic chapter/article/Conference paper


  • Research Council of Norway (RCN) / 235359




  • Nils Brede Moe
  • Darja Šmite
  • Aivars Šāblis
  • Anne-Lie Börjesson
  • Pia Andréasson


  • SINTEF Digital / Software Engineering, Safety and Security
  • Blekinge Institute of Technology
  • Sweden




Association for Computing Machinery (ACM)


ESEM '14 2014 ACM-IEEE International Symposium on Empirical Software Engineering and Measurement, Torino, Italy — September 18 - 19, 2014



View this publication at Cristin