The idea behind it
Work strategy down, not keyword up.
Traditional SEO starts with keywords. You find the terms people are searching for, optimize your content around them, and hope Google agrees you’re relevant.
That still matters. But AI Search works differently. AI systems aren’t just matching keywords to pages — they’re building a picture of who is expert in what. They’re drawing on signals about entities, relationships, and recognition to decide whose voice gets amplified in a response.
Which means the starting point needs to shift. Not “what keywords should I target?” but “what do I actually want to be known for?”
Once you’ve answered that, the rest of the framework is about making that answer legible to AI — and then measuring whether it’s working.
The core principle
Work strategy down, not keyword up.
Start with what you want to be recognized for. Then build everything else — content, structure, schema, measurement — around that clarity. It’s a different order of operations, and it makes a significant difference in how useful the output is.
What do you want to be known for?
Map your expertise. Define your entities. Find the visibility opportunities worth going after.
This is the most important stage — and the one most people skip. Not because it’s hard, but because it feels like you should already know the answer. You probably do. But “knowing it” and “having it clearly defined in a way that’s useful for AI visibility” are two different things.
Discover is about getting specific. Not “we do SEO consulting” but the precise topics, expertise areas, and entity relationships that should be associated with your brand in AI responses. What should AI systems reliably reach for when someone asks about your category?
That clarity becomes the foundation for everything else. Without it, you’re optimizing for the wrong things from the start.
What this stage covers
-
Topic entities
The specific subjects, categories, and concepts you want to be recognized for — named and defined precisely.
-
Expertise mapping
What depth of knowledge should AI systems associate with your brand? In what contexts? For what audiences?
-
Entity relationships
How your topics, people, brand, and expertise connect to each other — and to the broader knowledge graph.
-
Visibility opportunities
Where are the gaps between what you’re currently recognized for and what you should be recognized for?
Can AI systems actually understand it?
Make your expertise legible. Build the architecture that AI can parse, follow, and trust.
You can have deep expertise and still be invisible to AI if that expertise isn’t structured in a way AI systems can understand. This is where a lot of brands run into problems — great content, terrible signal quality.
Structure is about making your expertise legible. That means your information architecture should make clear relationships between topics. Your schema markup should tell AI systems who you are, what you do, and what you’re associated with. Your content should be organized in a way that AI can confidently extract and cite.
The question isn’t whether your content is good. It’s whether AI can understand it well enough to use it.
What this stage covers
-
Information architecture
How your site is organized — and whether that organization makes it easy for AI to understand topic relationships and depth of expertise.
-
Schema markup
Structured data that tells AI systems explicitly who you are, what you’re expert in, and how your entities relate to each other. Not all schema is equally useful — this is about the types that actually move the needle.
-
Entity relationships
How clearly your brand, your people, your topics, and your expertise connect — both within your site and in relation to external entities AI already recognizes.
-
Content structure
How individual pieces of content are organized so AI can extract clear, attributable information rather than guessing at meaning from unstructured prose.
Are the right signals strengthening recognition?
Build the external proof. Get mentioned, cited, and referenced in the places that matter.
AI systems don’t just learn from your website. They learn from the broader web — from who mentions you, who cites you, who references you in what contexts. That external signal is a significant part of how AI builds confidence in who you are and what you’re known for.
Reinforce is about building those signals deliberately. Not chasing links in the old SEO sense, but showing up in the right conversations, in the right communities, with the right people — so that when AI systems encounter references to your expertise, they’re consistent and credible.
This is also where speaking, writing, and community participation pay off in ways that are often underestimated.
What this stage covers
-
Mentions & citations
Being named, quoted, and referenced by other sources — particularly in contexts relevant to your expertise and topic entities.
-
External references
Other sources linking to, citing, or summarizing your work — adding weight to the external signal about your expertise.
-
Community participation
Showing up in the communities, conversations, and publications where your topic entities are discussed — so your name becomes associated with those topics through repeated, consistent presence.
-
External validation
Conference appearances, podcast interviews, guest contributions — the signals that tell AI systems other people consider you a credible voice on your topics.
Is recognition actually increasing?
Track mention rate, entity presence, and visibility over time — with data you can actually trust.
This is where Signals Over Noise — the measurement method — connects directly to the framework. You’ve done the Discover, Structure, and Reinforce work. Now you need to know whether it’s working.
And to know that, you need reliable data. Not a handful of prompts and a screenshot. A structured sampling approach with enough queries and runs to see patterns — and a mention rate score you can compare over time.
The goal isn’t a perfect score. The goal is a number that moves in the right direction. That means you set a baseline, you make changes, you measure again, and you use the delta to understand what’s working and what isn’t.
What this stage covers
-
Mention rate tracking
How often your brand appears across structured AI query sampling — the core metric for AI visibility. Trackable, comparable, and benchmarkable over time.
-
Entity presence
Whether AI systems are correctly associating your brand with the right topics, expertise areas, and entities — not just whether you appear, but how you appear.
-
Visibility benchmarking
Setting a baseline, then measuring against it. This is how you know whether the Discover, Structure, and Reinforce work is actually paying off.
-
Structured sampling
Using the Signals Over Noise method — persona modeling, entity-based query sets, defined sample sizes — to get data you can actually trust and act on.
How it fits together
Each stage makes the next one more effective.
Discover
Clarity on what you want to be known for
Structure
Architecture that makes it legible to AI
Reinforce
External signals that build AI confidence
Measure
Data that tells you whether it’s working
Discover without Structure means you know what you want to be known for but haven’t made it legible. Structure without Reinforce means you’ve built a clear signal internally but haven’t backed it up externally. Reinforce without Measure means you’re doing the work but have no idea whether it’s paying off.
The framework works as a system. Each stage makes the next one more meaningful — and Measure is what closes the loop and tells you what to do next.
Dig deeper
Where to go from here
The framework is the map. These are the tools and the territory.