It’s easy to say that a new website should be built around structured content rather than pages. The harder question is: what does that actually mean when you start building it?
That’s a question we’re working through as the University of Leicester’s University Content Management (UCM) programme moves further into delivery.
Traditionally, university websites have often treated the page as the basic unit of content. An editor creates a page, adds everything somebody might need, publishes it and maintains it. The same information can then appear somewhere else, owned by somebody else – and gradually we acquire several versions of the same thing. Messy.
We’re trying to break that cycle.
Take a researcher profile. Rather than thinking first about what the page should contain, we’re defining the information and relationships behind it. Who is this person? What is their expertise? What research are they connected to? Which publications, projects, outputs and stories relate to them?
Crucially, we’re also asking where that information should originate.
Our emerging research model, for example, proposes that publications shouldn’t be manually maintained again in the CMS. They should come from an appropriate authoritative University source and connect to the researcher. The same principle allows people, projects, research outputs and stories to be related through an agreed taxonomy rather than editors continually creating pages to recreate those connections.
Changing the questions we ask
This changes the design conversation.
Instead of beginning with “What should this page look like?”, we’re increasingly starting with:
- What is this content?
- Who needs it and why?
- Who owns it?
- Where is its authoritative source?
- Which individual fields need to be structured?
- How does it relate to other content?
- Where else might we want to use it?
Only then do we decide how it should be presented.
That order is important. If you design the page first, its content can quickly become bound to that particular presentation. If you model the information first, you create more opportunities to reuse it.
A research project, for example, doesn’t necessarily need to exist only on a research project page. Once represented as structured content, its relationships with researchers, themes, outputs and related stories can become part of how users discover and navigate research across the wider site.
And thinking beyond today’s website is important too. Structured, clearly described and connected information gives search engines and emerging AI-driven discovery services more explicit context about what our content means and how different things relate. Remember, GEO and AEO will (sooner than we may think) become part of the ecosystem in the same way as an SEO is – so content design, structure and governance will persist as pillars of our strategy.
The difficult bit happens earlier
There is a trade-off.
Content modelling forces conversations that page-based publishing can sometimes avoid. Teams need to agree terminology, ownership, relationships, sources of truth and what should happen when information changes.
That can make the work harder upfront. But resolving those questions during modelling is preferable to leaving hundreds or thousands of individual pages and their editors to answer them differently later.
It also changes what reimagination of content and user journeys means for us.
We’re not treating the remaining estate simply as pages waiting to be transferred into the new headless environment. We’re using content mapping, user needs and our emerging models to help determine what the information is, whether it should exist, where it should come from and how it should connect before deciding how it should be presented.
That’s a subtle shift, but an important one.
Replacing a CMS gives you newer technology. Rethinking the content model gives you the opportunity to build a better digital service.
