Identifying information about a document is stored in a database in the system through a process known as indexing, so that the document image can be retrieved. This information can consist of any information that will uniquely identify a particular document, for example:

• Document date
• Identifying number (such as invoice, employee number, college course catalog number, etc.)
• Customer name

The imaging system links this information to the graphics file that was created by the scanner, so when a retrieval request is entered, the system knows exactly which image to display, and where it is stored.

With manual data entry, all index values are entered by the user. Some data entry can be saved by matching a single identifier against a host database to retrieve the remaining index values.

However, when the document imaging system features automatic recognition, a bar code on the document can be read and its value matched against the host database, eliminating virtually all manual keying of index data!