Data quality in practice

This chapter examines the definition of data quality in terms of utility and highlights how understanding the measure of data quality may depend on the context and domain in which the data is scrutinized. The development of a data quality improvement program and what it takes to get the ball rolling can be summed up in seven phases—gaining senior level endorsement, training in data quality, creating and enforcing a data ownership policy, building the economic model, performing a current state assessment, selecting a project for improvement, and implementing and deploying the project. Different contexts for data quality—namely, data quality in an operational context, data quality in the database world, data quality and the data warehouse, data mining, electronic data interchange, and the Internet are also discussed in the chapter. While there is a distinct need for data quality in each of these contexts, the actual work that is done with respect to data quality is understated is likely to grow in importance.