Large organizations make use of data to improve decision-making. However, traditional data architectures cannot keep up with the scale and complexity of data. Data mesh addresses these problems by decentralizing data management. In this data architecture, data is part of a data product, which contains data, metadata, code, interfaces, and infrastructure.
In this research, we employ Design Science Research to design a methodology for developing and maintaining data products within a data mesh architecture. In this study, we conducted a gray literature review, which enabled us to gain insight into existing data and software methodologies, as well as the development of data products in practice. This knowledge was used to design our methodology. Furthermore, we utilized the FEDS evaluation framework to conduct a formative and summative evaluation of the methodology.
The primary contribution of this study is the Development Methodology for Data Products within a Data Mesh Architecture (DPDM-DMA). This methodology was perceived as useful and easy to use by experts in the field. To our knowledge, the DPDM-DMA is the first structured methodology that focuses on developing data products within a data mesh architecture. This work lays the foundation for creating high-quality, sharable data products. Further research can give insights into how this can be applied in various sectors.