Developing a ‘Data Quality Management Maturity Model’ (DQM3) based on Critical Success Factors

Abstract

Aims. In existing scientific literature, there is no maturity model for the data quality management (DQM) domain that provides sufficient supporting materials for an organization to evaluate their current status of DQM capability without being assisted by third-parties or certified professional (i.e., performing a self-assessment) and includes all Critical Success Factors (CSFs) for DQM. The CSFs for DQM are recently identified in two publications, while the current models were developed three to fifteen years ago. The existing DQM maturity models are also not applicable to measure the maturity level of a business chain. Therefore, the purpose of this study is twofold: (1) to develop a maturity model that allows an organization to perform a self-assessment in which the DQM maturity level of a business chain is measured, and (2) to apply and validate this maturity model by performing a pretest and a case study at Achmea.

Methodology. This study was structured according to the design cycle of Wieringa, supported by the development cycle of Mettler. Various research methods were used throughout this study. In the first phase of the design cycle, problem investigation, a literature study was conducted to confirm the gaps in the current scientific literature. In addition, an unstructured interview was performed and company reports of Achmea were studied to define the problem statement of this study. In the second phase, treatment design (supported by the ‘define scope’ and ‘design model’ tasks of the development cycle), the maturity model was developed based on an extensive literature study. In the third and final phase of the design cycle, treatment validation (supported by the ‘evaluate design’ task of the development cycle), a pretest and a case study was performed at Achmea to apply and validate the maturity model. Finally, the design mutability of the maturity model was contemplated in the final phase of the development cycle: reflect evolution.

Results. The treatment design phase led to the creation of the Data Quality Management Maturity Model (DQM3), consisting of two models: a three-layered domain reference model and an assessment model. The domain reference model is populated with identified and merged CSFs for (corporate) DQM and represents this domain in thirteen domain components and thirty-one domain sub-components. The assessment model consists of an assessment instrument (questionnaire implemented in Qualtrics Survey Software), five maturity levels, and defines how these maturity levels are assigned to the components of the domain reference model.

Conclusion. This study shows that the designed maturity model performs well in practice: both in the pretest and the case study. Only some minor improvements were made to the formulation of some questions. No additions have been made to the domain reference model. Future research is needed to refine and validate the domain reference model and assessment model of the DQM3 to demonstrate generalizability in other industries and organizations of various sizes.