Algorithm-assisted team formation in a software development business setting

Abstract

Many companies see a need in computationally-assisted team formation when their employee database keeps expanding. Most of these team formation algorithms are top-down which might result in approaching the team formation process as a mere computational prob- lem. By including the human component in the form of personality based information such as the Drivers test, we fill this gap. This study investigates the impact of driver-based team formation on teamwork within a business context, measured by work quality, well-being, and collabora- tion quality. Balanced teams are contrasted with unbalanced teams. The findings indicate that balanced teams outperform unbalanced teams in terms of work quality and collaboration qual- ity. The results indicate that balanced groups have a better objective score, an increased flexibil- ity among team members, enhanced decision-making abilities, reduced interruptions, are more vocal and positive collaboration, and have a improved collaboration efficiency. However, self- reported well-being does not differ significantly between the two team types, and unbalanced teams exhibit more positive interactions overall. The observed higher collaboration efficiency in balanced groups aligns with established theories of team formation stages, with unbalanced teams exhibiting a greater tendency for conflict and balanced teams achieving consensus more swiftly in problem-solving approaches and leadership dynamics. This study sheds light on the influence of personal drivers on collaboration within corporate environments and proposes the extension of the algorithm to incorporate job match, interpersonal compatibility, and expectation alignment. The study’s significance lies in its potential for organizations to strategically form higher-performing teams from existing employees, without the need for additional hiring, lead- ing to improved work quality and productivity. Leveraging the unique dataset provided by the Drivers test, this study fills a research gap and expands the understanding of how driver-based team formation impacts team dynamics and performance. Overall, this research contributes to the development of more efficient team formation algorithms, and advances the utilization of the Drivers test as a valuable tool in team composition studies.