It's a familiar frustration: a competitor with seemingly less detailed content gets cited by AI tools, while a more thorough page on the same topic gets passed over. The difference often isn't about writing quality at all — it's about recognisability.

Recognisable entities get cited more easily

When an AI system is assembling an answer and deciding which sources to draw on or cite, sources tied to a recognisable entity — one that has a presence in structured knowledge bases, consistent identity signals, and corroborating context — are easier for the system to "place." It already has a framework for understanding what that source is and why it might be relevant.

A source with excellent content but no recognisable entity identity is, in a sense, harder for these systems to contextualise. It might still get used — but it's competing without the scaffolding that makes a source easy to trust and attribute quickly.

Consistency across the web matters

Another factor is consistency: does the name, description, and core facts about an entity match across the places where it appears — its own site, structured knowledge sources, and any descriptive content about it? Inconsistencies (different names, conflicting details, no clear "canonical" version) can make a source feel less authoritative, even if each individual piece of content is accurate.

A useful question: if an AI system encountered your business for the first time through three different sources — your website, a knowledge base entry, and an article about you — would all three agree on who and what you are?

This is solvable — deliberately

None of this is about gaming citation algorithms. It's about making sure that, when these systems do the work of figuring out "what is this, and can I trust it," there's a clear, consistent, well-corroborated answer waiting for them — across all three layers of the entity infrastructure, not just on your own site.