Indexing - Applying Metadata to Records and Information within the Organization

‘Indexing’ is the process of capturing relevant metadata associated with your records.  Some of the metadata is used to index the records to make retrieval easier; some of the metadata is used for later management of those records.  So capturing the most appropriate metadata to enable easier retrieval and management is important.

You will need to develop your policy and guidelines for capturing metadata.  For example, is it best for your organization to capture metadata for all documents (recognizing that only some will become records later) or should you just capture metadata for those documents that you declare as records.  You will also need to define what metadata elements should be captured for different types of records.  You will want to create a metadata model for your organization.

The process of capturing the metadata can be manual or automatic and the metadata itself can be captured from information from a variety of sources.  For example, when using standard desktop applications such as Microsoft Office, the electronic records management (ERM) system may capture useful information about the document from the ‘document properties’.

Other sources of metadata are:

• the classification scheme for retention information,
• the ERM system itself for metadata like the ‘unique record number’ and
• the underlying operating system for information such as date and time of capture.

The actual amount and type of metadata required will be dependent on your organization’s business needs.

Many of your records will be evidence of an important business activity or transaction, so it is important that you capture relevant metadata relating to:

• the people involved in the activity or transaction
• the nature of the activity or transaction itself
• the outcome of the activity or transaction
• reference to any other important related records

If your organization has not been using an ERM system, it is highly likely that you and the rest of the people in your organization have not been capturing metadata.  Therefore, asking them to start manually capturing metadata as you move to ERM will not be popular and could undermine the success of the ERM Project.  Consequently, you will need to restrict the amount of manual metadata capture to an absolute minimum, ideally no more than one element.  Indexing should be as automated as possible. 

A crucial part of the indexing process is to provide an audit trail of what has happened to a record over its lifecycle.  Therefore, as a minimum during the indexing process, the following metadata should be captured:

• the Unique Identifier – this is usually a unique number (or alphanumeric string) assigned to the record by the ERM system
• the date and time of capture of the record
• the title of the record
• the author of the record.  This may be one person or an organization such as a company or a team (sometimes called a “corporate author).

Those of us who have worked in records management have applied labels to folder and boxes, inventoried file cabinets and drawers and created content listings of our holdings.  All of this represents applying metadata to the records and information we manage.  We have been successful in the past and through an ERM system we will be even better and more thorough.

By Carl E. Weise