Intelligent Enterprise

We all need to understand better how to establish high-quality, cost-effective information management strategies. As part of this effort, we need to identify methods and technologies that can automate and improve data processing.

Yes, data processing and other basic tenets of management information systems (MIS) from the 1980s are cycling back to the forefront of business and the desk of the CIO. The silos of data generated over the last decade continue to proliferate and to challenge organizations as they look to gain a consistent view of their operations, both historical and ongoing.

How can you improve your efforts to gain a single version of the truth? Many of you probably have some collection of tools to deal with data integration, data quality, database, message queues and the like. We still need to improve these areas to streamline the process of integrating data from across the enterprise. Of course, the task is not just moving some data from point to point any more; we have enormous volumes of data and have to perform transformations on them. This is complicated by the fact that usually the people who have domain knowledge of various data are not database developers.

To begin to address this challenge, maybe we should look again at a traditional method from the MIS days. Data discovery is one way to identify data that should be assimilated into a common definition for processing into other systems and integration with contemporary information management architectures. Data discovery is the ability to discover associations and meanings in data that can be used for processing into applications, databases and even data integration and master data management technologies.

Data discovery is important. It can help by preprocessing master data management and business intelligence data to reduce the time needed to associate data efficiently and increase the quality of the process. Part of maturing your information management efforts is to find methods and technologies that can help with tasks that are time-consuming, such as discovering relationships among data and proposing associations for mapping into a common data definition and repository. Data discovery can help eliminate a lot of labor-intensive steps. Just as importantly, it can be a standard workbench tool for data stewards supporting data governance and information management initiatives.

Let me know your thoughts.

Mark Smith is CEO And Senior Vice President of Research at Ventana Research. Write to him at mark.smith@ventanaresearch.com.