record standardization

In data management, record standardization is a process in which files are first parsed (assigned to appropriate fields in a database record) and then translated to a common format. For example, if an original record had the customer's name and address as "Bob Jones, VP Acme. Co., 15 S. Main St, Brooklyn" the standardized record might read "Bob Jones, Vice President, Acme Corporation, 15 South Main Street, Brooklyn, New York". Data in a database or data warehouse often lacks consistency simply because there are a lot of ways to say the same thing. Standardizing the record ensures that when a query is run for a particular field, accurate results will be returned.

There are many vendor products available to assist in record standardization. They work by comparing data in a particular field to a look-up library and can be configured either to suggest a change to the record (much like spell-check in Microsoft Word) or automatically make the change as the data is parsed. Most record standardization products allow the customer to tailor the look-up library to the customer's own needs.

This was last updated in September 2005

Dig Deeper on SQL Server Database Modeling and Design