# Definition density How often a page defines its own terms — and whether an agent can lift those definitions verbatim into a "what is X" answer. By AgentSite · 3 min read · Updated 2026-05-23 Definition density measures how often a page defines its own terms inline — "X is Y" patterns near where each term first appears. Answer engines lift definitions verbatim into glossary-style answers and "what is X" responses. A page that uses domain vocabulary without defining it gets cited less for queries that depend on those terms. ## The pattern The working shape is `[Term] is [definition].` sentences placed inline, near where the term first appears in the body. Three examples from the surrounding corpus: - "Agent readability is the property that decides whether an AI engine can read your page well enough to quote from it." (Term plus single-sentence definition; standalone answer.) - "A bot wall is the same URL returning different status codes to different user-agents." (Same shape, opening of a section.) - "Statistics and citations is the AEO dimension that measures whether numerical claims are co-located with the source they came from." (Defines a multi-word concept the rest of the page builds on.) In each case the term is the anchor and the sentence is removable from the surrounding paragraph without losing meaning. That removability is the test. ## Why it works The Princeton GEO paper measured up to 40% visibility lift in generative-engine responses from a stack of tactics including inline source citation and statistics addition ([Aggarwal et al., KDD 2024](https://arxiv.org/abs/2311.09735); [project page](https://generative-engines.com/GEO/)). Definition density isn't itself called out as a separately-measured tactic in that paper, but it's the structural backbone of the citation-bait sentence. Without a defined term, the engine cannot quote a page as the canonical source for what that term means; it has to either drop the page or paraphrase, both of which lose attribution. [Schema.org](http://Schema.org)'s [`DefinedTerm`](https://schema.org/DefinedTerm) type formalizes the same idea for explicit glossary surfaces — "A word, name, acronym, phrase, etc. with a formal definition." A page that emits `DefinedTerm` JSON-LD alongside its in-prose definitions adds a second extraction surface for the same content. ## Three properties of a working definition 1. **Term named explicitly.** "AEO is the new SEO" beats "this stuff is the new SEO." The term itself is the anchor for downstream agent reference. 2. **Definition self-contained.** No "as we discussed above" or "as the docs explain." An agent should be able to lift the sentence and have a complete answer. 3. **Placed where the term first appears.** Don't wait for a glossary section at the bottom of the page. Agents weight content from the top of the body disproportionately. ## Where this fits Definition density is a Layer 4 dimension — content quality on the individual page. It compounds with [statistics and citations](/statistics-citations) (the fact backbone inside each definition) and the [direct answer](/direct-answer) (the page-level extractable lede that's often the definition of the page topic). The longer thesis is in [agent readability](/agent-readability).