Agile data management in NAV: A Case Study

TL;DR

This paper presents a case study of a Norwegian public sector organization transitioning from centralized to distributed data management, highlighting benefits and challenges of adopting data mesh architecture for agile data handling.

Contribution

It provides empirical insights into the practical transition to data mesh architecture in a public sector context, an area with limited prior research.

Findings

Distributed data management offers increased agility.

Transition challenges include organizational and technical hurdles.

Benefits include improved flexibility and responsiveness.

Abstract

To satisfy the need for analytical data in the development of digital services, many organizations use data warehouse, and, more recently, data lake architectures. These architectures have traditionally been accompanied by centralized organizational models, where a single team or department has been responsible for gathering, transforming, and giving access to analytical data. However, such centralized models presuppose stability and are incompatible with agile software development where applications and databases are continuously updated. To achieve more agile forms of data management, some organizations have therefore begun to experiment with distributed data management models such as 'data meshes'. Research on this topic is however limited. In this paper, we report findings from a case study of a public sector organization in Norway that has begun the transition from centralized to distributed data management, outlining both the benefits and challenges of a distributed approach.