Master data quality barriers: an empirical investigation

Abstract

Purpose

The development of IT has enabled organizations to collect and store many times more data than they were able to just decades ago. This means that companies are now faced with managing huge amounts of data, which represents new challenges in ensuring high data quality. The purpose of this paper is to identify barriers to obtaining high master data quality.

Design/methodology/approach

This paper defines relevant master data quality barriers and investigates their mutual importance through organizing data quality barriers identified in literature into a framework for analysis of data quality. The importance of the different classes of data quality barriers is investigated by a large questionnaire study, including answers from 787 Danish manufacturing companies.

Findings

Based on a literature review, the paper identifies 12 master data quality barriers. The relevance and completeness of this classification is investigated by a large questionnaire study, which also clarifies the mutual importance of the defined barriers and the differences in importance in small, medium, and large companies.

Research limitations/implications

The defined classification of data quality barriers provides a point of departure for future research by pointing to relevant areas for investigation of data quality problems. The limitations of the study are that it focuses only on manufacturing companies and master data (i.e. not transaction data).

Practical implications

The classification of data quality barriers can give companies increased awareness of why they experience data quality problems. In addition, the paper suggests giving primary focus to organizational issues rather than perceiving poor data quality as an IT problem.

Originality/value

Compared to extant classifications of data quality barriers, the contribution of this paper represents a more detailed and complete picture of what the barriers are in relation to data quality. Furthermore, the presented classification has been investigated by a large questionnaire study, for which reason it is founded on a more solid empirical basis than existing classifications.

Citation

Haug, A., Stentoft Arlbjørn, J., Zachariassen, F. and Schlichter, J. (2013), "Master data quality barriers: an empirical investigation", Industrial Management & Data Systems, Vol. 113 No. 2, pp. 234-249. https://doi.org/10.1108/02635571311303550