What, precisely, do we mean by ‘operational data quality’? It is not a single thing. That is, it can be deployed to support a variety of different requirements. Any definition of operational data quality must therefore be broad and generic, but the key point is that it is being used within processes that are themselves operational. In other words, within processes that are active in real-time or near real-time. Examples would include order entry into transaction processing systems (whether via eCommerce sites or more directly), stock receipts within just-in-time manufacturing environments, real-time evaluation of online insurance claims, and many others. In addition, operational data quality can be important indirectly: for example, in maintaining a 360o view of the customer, where that view is used operationally by other business processes. Thus, we might redefine operational data quality as ‘data quality for operational processes’ where these processes involve such things as transaction processing, operational business intelligence, real-time data monitoring, event-driven processes, and so on. It is difficult to imagine any business organisation that could not profit from the use of operational data quality in at least one, and more likely several, areas.
This paper is organised into two parts. First, we will discuss the importance of operational data quality in various scenarios and use cases. Second, we will consider the technical requirements needed to support operational data quality and the sorts of features required from a suitable data quality product.
To download this paper you must be logged in.