As the demand for content grows, we need better tools to help us structure it. Content models are an effective way of keeping a multi-disciplined project team aligned in their understanding of structured content. A firm and shared grasp on the structure and semantics of content, leads to more innovative ways to adapt it downstream.
Because, as Ann Mulhay, ex-CEO of Xerox succinctly puts it:
Unstructured content is stupid and old-fashioned. It’s costly, complex and does not generate a competitive advantage.
Content modeling is the process of creating content models that describe structured content.
How does this affect you?
Everyday we switch effortlessly between television, print and digital. However, companies wanting to deliver that “unifying customer experience”, relevant content must also flow seamlessly between these mediums. That’s the new norm but an old problem. We’re still witnessing print brands struggling to adapt their content to digital platforms. Even experienced digital players are backtracking and unpicking content trapped within HTML that needs to be repurposed and delivered down mobile channels.
Ann Rockley in her excellent book on (Managing Enterprise Content) uses the term information products. Depending upon what is valuable to your business, an information product can be a large as a magazine or as small as a page of a car manual. What’s important is that it can be dynamically assembled from well known content parts across multiple media formats and tailored to the consumer’s context.
Let’s go with an example. Consider a French tourist visiting the Tower of London with an iPad to hand and flicking through their favourite American magazine:
- The tourist reads it in French, in either portrait or landscape mode, getting a Tower of London exclusive.
- The publisher up-sells similar titles and delivers the breaking news France.
- The advertiser wants relevant eyeballs to push its targeted offers from as local restaurants, gift shops, etc.
This requires content to be adaptable. Adaptable content must be structured, where it’s broken down into intelligent parts with well understood relationships between them. Structured content can be disassembled and reassembled to create new information products on -the-fly, based upon context. The above magazine is fluid, using dynamically assembled adaptable content, based upon customer and their context, to deliver the relevant experiences.
A content model is a representation of structured content.
What is NOT a content model?
It’s not a site map depicting a top-down view of web site pages. It’s not a navigational scheme for describing how users get from A to B. And it’s not page types outlining how content should be laid out on a page.
When working with content models, forget about pages and web-sites altogether. Maybe you start there. But if you really want to set your content free, for it to used in contexts beyond your control and/or imagination, it needs to be meaningful. To achieve that, model content outside of the confines of its initial context, that typically being web site pages.
What is a content model?
A content model is a representation of the types of content and their inter-relationships. For example, a car dealership may have content types for Vehicle, Dealer, and Manufacturer because these are first class citizens in the business domain. A restaurant have Recipes, Chef, Menu and Venue. The aim of the content model is to identify these content types, model their relationships and provide a shared semantic understanding of content from sponsor to developer.
Content models appear to be the black sheep of information architecture. They are seldom done. The perception is that they are difficult and depend upon less understood things such as metadata and taxonomy. In fact, content modeling (or content modelling) is a highly creative activity and freakishly awkward to get right. But difficult, no. They are only a few new tools to learn. However, as both an art and a science, content modeling is much akin to software design, such that armed with these basic tools:
It’s simple to make things complex, but complex to make them simple.
Why is content modelling so hard?
Content modeling is hard because there are many different ways to achieve that same result, but one way will be more elegant than another. You just have to find that way. If I set you an exercise to go to the Amazon and produce a content model and then compare notes, guaranteed our content models would be different. And yes, we could caveat the exercise with, “there is no right or wrong answer”. That’s also true. But there will be a winner. One will definitely feel better. Be more elegant. Be able to flex. Be useful and more meaningful in other contexts. Be readable. Be adaptable to unforeseen changes. Be simple, but no simpler. And have all the expected itilities around maintainability, scalability, and so on. To borrow from Erin Kissane’s book on Elements of Content Strategy, a good content model should represent content that is:
…appropriate for your business, for your users, and for its context.
Granularity is another important thing to gauge when content modeling. Just how low do you go? When modeling a car dealership, we agree that a Car is a content type. Do we break out a new content type called Car Part and model the resulting complex relationships, where:
- A Car has a Seat
- The Main Dashboard contains many other smaller Dashboards (e.g. Fuel Gauge, Media Player, Mileage)
- Engine is linked to a specific manufacturer.
All these decisions today impact the content model for tomorrow. But decide we must based upon the planned and unplanned use of content in future unknown contexts. Anyone excited by that?
What do content models look like?
A content model may be a diagram on a whiteboard, a pile of cards describing your information products and services, an excel spreadsheet, or a fancy content modelling tool used by geographically distributed teams. All or none of them may be appropriate for your particular situation. What is important is that there is agreement on what the content is and the way it’s communicated between the cross-disciplined teams. That is what the content model should bring to the table. The means to define, describe, disseminate and discuss content.
There are no hard or fast rules for what a content model should look like. However, there are guidelines and best practices that we can beg, borrow and steal from other disciplines such as data modelling and object modelling. These folks have invested decades of experience into these areas. Let’s reuse. Hopefully, with a lot of help from other interested folks, we can turn content modeling into a recognised and valued upstream activity within information architecture.
