LLM Optimization
Making content retrievable and quotable by language models — HTML-visible, self-contained, consistent.
LLM optimisation is making content usable by large language models: reachable without JavaScript, permitted by robots.txt for AI crawlers, and written so a single passage answers a question completely enough to be quoted. It is less about keywords than about whether an extracted paragraph still makes sense on its own.
Definition
LLM optimization (LLMO) is the practice of making content easy for large language models to retrieve, quote and attribute correctly. It overlaps with SEO but optimises for a different consumer: a model assembling an answer, not a ranking algorithm ordering links.
Three properties do most of the work. Availability in HTML, because most AI crawlers do not execute JavaScript. Self-contained passages, because retrieval works on chunks, not whole documents. Internal consistency, because a source that contradicts itself gets discounted.
Why It Matters
AI assistants are already a meaningful traffic and influence channel, and they reward properties that traditional SEO never scored: whether a claim stands alone without its surrounding page, and whether your own numbers agree with each other across the site.
That last point is easy to underestimate. A site that says "11 pillars" on one page and lists ten on another gives a retrieval system a reason to prefer a competitor that does not contradict itself.
Example
Practical moves: publish llms.txt describing what the site is and what is worth citing; keep every number consistent across pages; open sections with a direct answer; ship the content server-side; and use structured data so claims are typed rather than inferred.
Know the term.
Check your own site.
A free audit tells you whether this is currently costing you score points — and exactly what to change.
Run a free audit