Hyland Software's Document Management Blog

Getting meta about metadata: Connecting content to business transactions, part 1

// August 20th, 2010 // Document Management, Enterprise content management // Eric Proegler

When we think about keeping track of information, we usually start with when we save or capture it. We capture data all the time, such as when we upload photos of kids, a vacation view, a restaurant dish we tried, or some other reason for bragging on Facebook.

To find a particular photo, our mental database uses information about what we’re looking for. We can then skim thumbnails for the tiny green soccer jersey, white sandy beach, or monstrous bacon cheeseburger that we remember from the picture we’re trying to find.

At work, there are way too many documents, or too much content, to rely on manual scans supported by mental hints. We need more. To capture, preserve, and protect the content we care about, we can store it in an enterprise content management (ECM) solution. Capturing and storing content is where ECM starts.

However, the real value provided by ECM isn’t from storing the content; the value is provided by the ability to quickly and easily retrieve exactly the content you need. Usually, the first sorting of content is into content types, such as invoices. This creates a bucket of content that looks even more alike, though.

To find specific invoices, we need to capture more than just the content; we also need to capture index information that helps identify the content. This metadata, or data about data, is how ECM allows users to search for the content they need.

Some types of content have just a few things worth indexing. Other types of content have many attributes that could be useful to store. Different people in different roles have different reasons to find and use the same content. This means they are likely to use different types of metadata to locate the content they need.

When the ECM solution allows capture of all the useful metadata for an invoice, all of the ECM systems’ users can find the content they are looking for, with the information they use in their roles.

Users should be able to search for invoices by any of the following criteria: 

• Vendor name
• Vendor number
• Amount due
• Currency type
• Requestor
• Approver
• Account number
• Purchase order number
• Order date
• Shipping date
• Approved date
• Purchase order number
• Any other piece information associated with the invoice that you would like to capture
• Any combination of these terms, using connectors like AND, OR, not equals, greater and lesser than, and so on

When making decisions about what metadata to capture, consider that data captured today may be used tomorrow in ways we probably can’t think of now. As requirements and other systems change, different metadata will be available or needed for capture.

For future flexibility, you should consider capturing any potentially useful metadata when we ingest the content, capturing the set of metadata appropriate for each type of content, and updating what you capture when new data sources become available.

The question is not if metadata configurations will change. The questions are how often, and what will it cost to change them.

 Not every ECM solution makes this easy, or even possible. Some suites:

• Limit the number of metadata items
• Limit the number of entries per metadata type that can be captured for a piece of content, such as multiple purchase order numbers on the same invoice
• Limit the diversity of metadata configurations that can be captured
• Don’t allow reuse of metadata across content types
• Don’t support reusable groupings of metadata
• Require expensive code refactoring to make any changes to metadata capture configurations

When evaluating ECM solutions, flexibility and maintainability of metadata configuration should be a page one question. Your ECM users need to be able to find content in whatever manner supports the work they are doing, but it goes even farther than that.

Your business’ bottom line will benefit even more when metadata can be used to instrument business processes, automate processing, aggregate grouping for research and analytics and to help manage work.

We’ll talk more about what else can be done with metadata in part two of this post, coming soon.

Let me know what you think about all of this, and feel free to send along your pictures of unusual, massive, unusually massive or massively unusual food.

Read other posts by Eric Proegler