Lost and Found: Content Modeling for Large Digital Presences

The keys! Not in my pockets. Not in the drawer. Not on the table. Where are they?

It’s bad enough to have to turn everything upside down when your house is, sort of, tidy. It takes longer, and drives you crazier, when the house is a mess. Chaotic, disorganized places are stressful.

The same is true for websites. Managing badly structured web content is hard. When content authors are stuck with an unusable editorial interface or with content types that don’t make sense, they start working around the CMS. At some point, there are so many workarounds that training and documentation become confusing too, resulting in a desperately reactive approach to web content management (WCM).

Content modeling—when done right—is meant to prevent this mess from appearing in the first place. It is about organizing content in meaningful ways, so that there’s a place for everything in your digital home. 

Content modeling is the process of organizing content assets and relationships between them in order to enable logical, efficient, and sustainable content management practices using chosen technology platforms.

According to Deane Barker, author of Web Content Management, content modeling is defined by the following:

  • Content types
  • Content attributes
  • Relationships between them

Content modeling takes root in data modeling, a mature software engineering practice used to translate business requirements into database design. However, in content modeling, the freedom to design databases in any custom way you like just isn’t there. Between the business requirements and the database design, there is a CMS that imposes conceptual and technical limitations on how the content can be organized. This additional CMS layer influences the way content modeling is done and the terminology that is used in the process. To create an effective content model, it’s helpful to have a broad knowledge of the CMS principles, so that you’re led—but not blinded—by the chosen CMS and its philosophy.

Both data modeling and content modeling cannot be done well by software developers alone. Input from business stakeholders is essential to make the technical design meaningful and aligned to business goals. In fact, you can spot a content model done by a techie by a mile—it’s usually an over-engineered, cover-all-eventualities, logical structure that is beautiful and horrifying in equal measure. The real world isn’t like that.


The Basics of Content Modeling

On a small scale, content modeling is easy. Content types used for the University of Leeds’ conferences template are pages and widgets. There are hardly any relationships between the content types to speak of—except that a widget can be contained within a page.

So far, so good—but this conference website is a tiny part of a much bigger collection of university events, which are displayed in different areas of the university’s digital presence, depending on their date (past or present), type (workshop, seminar, or conference), and intended audience (undergraduates, prospective students, academics, and so on). In this larger context, sticking to our original set of content types just won’t cut it. To be able to connect our conference website with other parts of the university’s digital presence, we need to look at it from a bird’s-eye view.


High-Level Content Modeling Considerations

  1. Establish priorities—Not all content is created equal. Some content is supporting key customer journeys, and some content is supporting somebody’s whim. Some content attracts hundreds of thousands of visitors, and some content attracts analytics spambots from Russia. Identifying content that sells and content that makes a promise is crucial. This priority content is your living room. It must be tidy, otherwise no one will want to come to your digital home.
  2. Identify enterprise-wide content—This is the content that’s used by multiple departments and areas of business. It needs to exist in a central place as a single point of truth. It cuts through the silos and has to make peace with other systems that are in place.
  3. Future-proof—For long-term success, there needs to be an easy way of identifying where to slot the new stuff. A sitemap provides a good enough level of categorization for a small website, but for a large digital presence—in which new websites get created every day—it’s helpful to have an even higher-level overview of the digital presence. Categorize different websites by brands, impact, technical complexity, purpose, and so on.


Strive for the Balance

Extremes are easy. In more than a decade spent implementing and managing CMSs, I’ve seen both understructured and overstructured content. Too little structure and you end up with webpages resembling Word documents from hell, with big red headings and garish fonts that hurt your eyes. Too much structure and content editors are losing the will to live because they are entering the web content, sentence by sentence, into the tiny boxes. Generally speaking, marketers tend to value flexibility over structure, whereas software developers prefer modularity and stability. To create an effective, sustainable content model, strive for balance.

Similar to how a floorplan helps in deciding where to put your furniture and belongings during a house move, content modeling is useful in making decisions about where to place your web content. It helps to clarify, or sometimes even establish, business requirements. It is often influenced by the chosen CMS, its philosophy, and its terminology.

Content modeling doesn’t have to take the form of a flashy presentation or a glorified document. Documenting the content model is useful, but the thinking and planning behind it is just as important.