Emergence of data and non-data team networks: an agent-based model

Research on the organizational impact of business analytics (BA) systems has mainly focused on the identification of analytics resources and capabilities in isolation and linking them directly or indirectly to business value. In contrast, achieving value from data assets in real organizational settings requires data teams to form connections with other organizational (non-data) teams so that the teams can work together and utilize complementary and shared resources. This paper focuses on the embeddedness of data and non-data teams within an organizational network, and it examines how the two teams connect and collectively influence organizational performance. For that, we built an agent-based computational model in which data and non-data teams solve organizational problems under dynamic organizational structures and team-level learning. In our model, data and non-data teams (i) simultaneously search for satisfying solutions over a complex space (i.e. an NK landscape), (ii) are initially connected to each other through a given network configuration, (iii) are endowed with learning capabilities (through a reinforcement learning algorithm), and (iv) update their links to other agents (i.e. create new connections or disconnect existing ones) according to their learning capabilities. Results reveal conditions under which performance differences are obtained, considering variations in the number of agents, space complexity, agents' screening capabilities and reinforcement learning