Data quality

Abstract

Data are used in the delivery of many products and services, and so data quality is an important component of customers' perceptions of the quality of these products and services. The paper describes efforts initiated by AT&T to control and improve the quality of data it uses to operate its worldwide intelligent network, to conduct its day-to-day operations, and to manage its businesses smoothly. These efforts stem from the observation that it is extremely difficult to fix faulty data once they are in a database. Therefore attention must be directed at processes that introduce, modify, and transform data. Only when these processes have been put into a state of statistical control can sustainable improvements in data quality be expected. The report describes AT&T's four-part data quality improvement program: • • Develop the technical foundations for understanding and measuring data quality. Four objective dimensions of data quality (accuracy, completeness, consistency, and currency) are defined. • • Extend and apply process management techniques to information management. • • Extend and apply methods of statistical process control. In particular data tracking, a method to evaluate quantitatively the processes by which data are introduced into a database, is described. Data tracking provides ongoing process control and helps identify improvement opportunities. • • Develop and apply methods that help ensure data quality in a data-processing environment. Collectively, such methods are referred to as data engineering and are particularly useful in ensuring data consistency. Taken together, these efforts form the basis of a comprehensive, overall approach to improving data quality and sustaining the improvements.