The Entity Mention Graph
Backlinks decay. Entity co-mentions compound. How AI engines score which entities belong together and why sub-DR-20 brands should optimize for co-occurrence instead of PageRank.
Why this matters
AI engines infer relationships between entities by watching who mentions whom. A brand that co-occurs with its topic cluster across third-party sources is treated as canonical. A brand that only self-declares is treated as marketing copy.
In this cluster
Cluster context
This article sits inside AI Visibility Engineering.
Entity graphs, schema architecture, and citation mechanics for sub-DR-20 sites competing on AI citations, not SERP rank.
SEO optimizes for rank. Answer engines optimize for citation-worthiness. This cluster is the engineering playbook for the second game, sized for operators, not enterprise SEO teams.
How AI Answer Engines Decide Which Sources to Cite: 6 Factors
How answer engines like ChatGPT and Perplexity decide which sources to cite: the 6 AEO factors, plus citation-integrity and attribution guidance.
Entity Optimization for Brands in AI Search
Rank is a single-page game. Entity coherence is the compounding game. How sub-DR-20 brands engineer a Person + Organization graph that AI search engines actually cite.
Schema.org for Answer Engines, the 40 Properties That Matter
A tactical guide to the Schema.org properties answer engines actually read. Which fields move citation decisions, which are noise, and how sub-DR-20 operators compress a full JSON-LD graph into the forty that matter.
The backlink graph has been the currency of SEO for twenty years. The entity mention graph is something else, and for sub-DR-20 brands competing on AI citations, it is the graph that matters. The framework this post sits inside is Answer Engine Optimization Explained; this post is the entity-mention-graph slice of the broader AEO methodology.
This post is the short version of a long thesis. Engines that cite, Perplexity, ChatGPT, Google AI Overviews, do not rely on PageRank alone. They build an internal graph of which entities tend to show up together, and they treat dense co-occurrence as evidence of relationship. A brand whose name appears near its topic across a dozen independent sources is treated as canonical. A brand that only declares the relationship on its own site is treated as marketing copy.
§1, Co-occurrence is the hidden signal
Engines read the web differently than a backlink crawler. They read it as a corpus of sentences, paragraphs, and contexts. When two entities consistently appear in the same sentence or paragraph across many sources, the engine infers a structural relationship between them. This is co-occurrence. It is not link equity. It is closer to mutual information.
§2, Why sub-DR-20 brands actually have an advantage here
Mention engineering is labor-intensive and unglamorous. Enterprise SEO teams optimize for backlinks because it is measurable and delegable. Smaller brands can move faster on the mention graph because the work is personal. A founder on a podcast. A hand-written GitHub README. A well-placed comment on a technical thread. These seed co-mentions in a way that a backlink campaign cannot.
§3, Five seeding tactics that compound
Podcast guest spots where the host names the entity and the topic in the same sentence. GitHub repos that include a Person schema in the README. Hacker News threads where the entity is referenced in a technical context. Cross-brand citations, chudi.dev naming citability.dev in a sentence about AI visibility, and vice versa. Industry write-ups that use your entity label instead of your category label.
§4, The canonical-pair timeline
Watch the timeline. First co-mention in a transcript. First partial citation in AI Overviews. First Perplexity citation that pairs the two entities. First ChatGPT citation with a direct quote. The pattern is weeks, not months. Engines that ingest the web continuously pick up new co-occurrence patterns fast.
§5, Measurement is a graph problem
Spreadsheets cannot hold this data well. The unit of measurement is an entity-pair plus a co-mention count plus a source-diversity score. The right instrument is a graph database or a purpose-built tracker that ingests citation logs from each answer engine.
Bridge, measurement
citability.dev tracks the entity pair chudi.dev + AI Visibility Engineering across Perplexity, ChatGPT, and AI Overviews. Run it against your own canonical pair to see what engines are picking up.
· Sources & further reading
Sources & Further Reading
Further reading
- Entity Optimization for Brands in AI Search /blog/entity-optimization-brands-ai-search Rank is a single-page game. Entity coherence is the compounding game. How sub-DR-20 brands engineer a Person + Organization graph that AI search engines actually cite.
- Schema.org for Answer Engines, the 40 Properties That Matter /blog/schema-org-answer-engines-guide A tactical guide to the Schema.org properties answer engines actually read. Which fields move citation decisions, which are noise, and how sub-DR-20 operators compress a full JSON-LD graph into the forty that matter.
- Originality Signals and Citation Patterns /blog/originality-signals-ai-citation-patterns AI engines deprioritize pages that look like everything else. The originality signals that move a post from the summary layer into the quote layer, and why recap content is the new thin content.
- Perplexity vs ChatGPT: Different Citation Rules /blog/perplexity-vs-chatgpt-citation-rules Perplexity quotes liberally. ChatGPT quotes selectively. The engine-level differences in citation behavior that change what a sub-DR-20 brand should optimize for, engine by engine.
- How AI Answer Engines Decide Which Sources to Cite: 6 Factors /blog/aeo-answer-engine-optimization-explained How answer engines like ChatGPT and Perplexity decide which sources to cite: the 6 AEO factors, plus citation-integrity and attribution guidance.
Reading Path
Continue the AI Visibility Engineering track
Contextual next reads
How AI Answer Engines Decide Which Sources to Cite: 6 Factors
How answer engines like ChatGPT and Perplexity decide which sources to cite: the 6 AEO factors, plus citation-integrity and attribution guidance.
Entity Optimization for Brands in AI Search
Rank is a single-page game. Entity coherence is the compounding game. How sub-DR-20 brands engineer a Person + Organization graph that AI search engines actually cite.
Schema.org for Answer Engines, the 40 Properties That Matter
A tactical guide to the Schema.org properties answer engines actually read. Which fields move citation decisions, which are noise, and how sub-DR-20 operators compress a full JSON-LD graph into the forty that matter.
What do you think?
I post about this stuff on LinkedIn every day and the conversations there are great. If this post sparked a thought, I'd love to hear it.
Discuss on LinkedIn