Structure
Site Structure
Once you know what you want to be known for, the next move is making that visible in your site architecture. Entities become pages. Pages become a structure AI can actually navigate.
Entities become architecture.
Structure is how you make expertise legible.
You can have great content and still be invisible to AI systems. Not because the content is bad, but because it’s scattered. There’s no clear architecture connecting your posts, services, and brand to the concepts you want to own.
AI systems don’t just read individual pages. They look for patterns across a site — repeated associations, consistent relationships, content that clusters around the same ideas.
If your content is all over the place, that pattern doesn’t form. The signal stays muddy.
Structure is not about design. It is about making your expertise easier to find, interpret, and retrieve.
This step takes the entities you defined in Step 01 and turns them into architecture. Each entity gets a home. Each home pulls related content together. The result is a site that tells AI exactly what you should be known for — not through keywords, but through structure.
What is a hub page?
A hub page is a central destination for a topic entity. It’s the page that says: this is what we know about this subject, and here’s everything we’ve written, built, or done around it.
It doesn’t have to be massive. It doesn’t need a fancy template or six weeks of development time. What it needs is:
That’s it. A blog category page can do this job if it’s treated intentionally. For a lot of teams, that’s the fastest path to building entity structure without touching dev resources.
A hub page gives a topic entity a permanent address on your site. Without it, the entity exists in your strategy but not in your structure.
Hub pages, category pages, collection pages.
These three page types do the same fundamental job — giving a topic entity a home — but they work at different scales and in different contexts.
Built around a primary entity — usually a service, product, or concept — and designed to be the central destination for everything related to that topic. This is where supporting content links back to.
Blog category pages are often underused. When treated intentionally — with a clear description, entity-aligned name, and content organized beneath them — they become one of the strongest structural signals on your site.
Collection pages gather content around a specific audience, location, or use case entity. Useful when you serve multiple distinct segments and want AI to understand each one as a separate, connected relationship.
You don’t need all three types for every entity. The right choice depends on your site’s existing structure and what’s fastest to implement without starting from scratch.
How to build it.
This doesn’t require a full site rebuild. It requires deliberate choices about what goes where — and a willingness to reorganize what you already have.
Create a hub page for each topic entity
Each entity you defined in Step 01 gets its own page. The entity name goes in the H1. The page explains what the topic covers and connects to related content below it.
If a blog category page already exists for this topic, start there. Add a description. Make sure the name matches the entity. Clean up what’s underneath it. That’s often enough to make the structural signal significantly stronger.
Start with what exists. Build from there.
Reorganize existing content under the right entities
Most sites already have the content. It’s just scattered across the wrong categories, misfiled, or not connected to anything.
Go through your existing posts and pages. Reassign them to the entity hubs they actually belong to. Update categories. Remove posts from topics they were loosely tagged to but don’t really support.
The goal is cleaner topical lanes — not more content, but better organization of what’s already there.
You don’t need more content. You need better organization.
One post, one primary entity
Every piece of content should have one primary entity it’s supporting. Not two. Not three. One.
If a post is trying to reinforce multiple concepts at once, the signal gets diluted. AI has a harder time understanding what that piece is about and what it connects to.
Clean lanes help AI understand what each piece of content is doing. That’s the whole point of building this structure in the first place.
One post. One primary entity. Clean topical lanes.
Structure makes everything else work better.
Site structure isn’t just an organizational exercise. It’s what makes the rest of the framework functional.
When your entities have hub pages, your schema markup has something to reference. When your content is organized into clean clusters, your internal linking has a clear logic. When your architecture is readable, AI systems can retrieve you more reliably.
Without structure, the signals you’re trying to build — through schema, through linking, through content — are working against a disorganized foundation. They can’t do their job properly.
Structure is the layer that makes entity reinforcement possible. Build it before you try to reinforce anything.
Topic entities, brand, services, audiences, and concepts. The step everything else is built on.
← Back
Turn your entities into hub pages and content architecture that AI can actually navigate.
Currently reading ↑
Schema markup as an internal knowledge graph. Strategic internal linking that connects the pieces.
Read this next →
Mention rate, N×k sampling, and how to measure AI visibility with numbers that actually mean something.
Explore →