Metrics for the evaluation of data quality - Design and practical use

In recent years data quality (DQ) has gained more and more importance in theory and practice due to an extended use of data warehouse systems, management information systems and a higher relevance of customer relationship management. This refers to the fact that for decision makers the benefit of data heavily depends on completeness, correctness and timeliness for example. The growing relevance of DQ revealed the need for adequate measurement because quantifying DQ (e.g. of a data base) is essential for planning DQ measures in an economic manner. The article analyzed how DQ criteria can be quantified in a goal-oriented and economic manner. The aim was to develop new metrics for the DQ criteria correctness and timeliness. These metrics proposed enable an objective and automated measurement. In contrast to existing approaches the metrics were designed according to important requirements like feasibility and interpretability. In cooperation with a major German mobile services provider, the developed metrics were applied and they turned out to be appropriate for practical problems.