Developing a federated data governance framework: evolving from minimum viable structures towards governance maturity

The importance of data governance has been increasingly recognized within organizations in recent years. Data is an important business asset across multiple organizational functions, and its effective utilization as a strategic asset can lead to sustainable competitive advantage. As an organization’s data landscape grows, data can become fragmented across business units, applications, legacy systems, and databases. Data fragmentation can lead to a decrease in data quality, negatively affecting business processes, reporting, and decision-making. Furthermore, data legislation (e.g. EU Data Act, GDPR) is increasingly created to ensure that organizations collect and utilize data in a compliant way. While data governance is widely recognized as an organization-dependent concept, existing data governance literature and frameworks are often aimed towards companies with high organizational maturity and resource availability. Most literature remains on a conceptual level, without providing practical direction on data governance development within the organizational environment. These factors combined make the existing data governance frameworks difficult to implement in organizations at early stages of governance development. This thesis addresses this gap by designing a data governance framework for a global organization with a large data ecosystem, but low governance maturity and limited resources. The thesis utilizes Design Science Research (DSR) methodology, that is used to develop and evaluate a four-stage federated data governance framework that enables gradual progression from a minimum viable governance structure to a fully mature governance model. The framework incorporates modern best governance practices from the academic knowledge base, incorporating principles of agility and data mesh. Organizational applicability is ensured through empirical data collection, including interviews, workshops, and company documentation. The DSR methodology involves multiple iterative design cycles and evaluations, ensuring academic rigor and organizational applicability. The findings confirm that the proposed framework is both feasible and implementable, effectively addressing the organization’s real-world governance challenges. The staged, incremental approach reduces the threshold for governance adaptation and reduces complexity. From a scientific perspective, the thesis contributes to existing data governance knowledge base by operationalizing academic principles and best practices in a real-world context. Further, it illustrates the DSR process that can be useful to and replicated by other researchers and practitioners. While the model and research process can be applied by other organizations and researchers, further research is needed to evaluate the implementation of the framework as well as its generalizability.