WHAT YOU CAN’T MEASURE - YOU CAN’T IMPROVE - The role of maturity models to improve data governance

Abstract

Background and purpose: As a consequence of the growing power of data, there is a need for companies to maximise the value derived from it. However, to maximise the value derived from data, it needs to be available, secure, relevant, and of high quality, which can be assured by data governance. In addition, data governance has become crucial for companies to meet legal requirements and to be competitive. The increasing need for data governance puts pressure on organisations to control how they work with data and thus a need to improve. To understand how an organisation works today and what can be improved, a maturity model can be used. However, available data governance maturity models do not only miss out on aspects within data governance but also on how to use the model. Thus, the purpose of this study is to explore how a maturity model can support organisations in improving data governance. The model is practically contributing as a tool for companies to assess their current level of maturity and to identify potential improvements. Methodology: A qualitative research strategy has been used throughout this study. After investigating existing literature, workshops with data governance experts were conducted. Based on the findings from literature and workshops, aspects important when creating the model could be identified and the TMT Data Governance Maturity model was created. To test the validity of the model and to determine what to take into consideration when using the model, it was applied to a case company where semi-structured interviews with employees were conducted. The findings from the interviews were analysed by comparing the answers to the levels in the model, using a thematic approach. The levels of maturity were then determined based on the average of all respondents' answers. By comparing the assigned levels with the higher levels, actions for how to improve were identified and relevant improvement areas could thereafter be defined. Main Findings: Based on the theoretical framework and workshops 13 elements were identified as crucial for data governance maturity models: Strategy & Approach, Leadership, Structure, Progress Measure, Knowledge & Change Management, Rules, Data Quality, Data Security & Privacy, Data Lifecycle Management, Metadata Management, Master Data Management, Business Intelligence, and Adherence. The research also showed that an important aspect of maturity models is interview questions reflecting the elements and some sort of measurement, which resulted in five levels being defined: Unaware, Ad Hoc, Proactive, Managed, and Optimised. When testing the model, one finding was that the model always needs to be adapted to each specific organisation before use to be of value, since all companies are unique. If adapting the model to be in line with the characteristics of the organisation, the current maturity level could be determined and thereby also what is needed to reach the higher levels by identification of the gap. However, the result from using the maturity model only works as guidance for what could be improved since the reality usually is more complex than assigning an organisation a level on a scale.