What is LLM optimisation?
LLM optimisation is the practice of structuring content so that AI systems – ChatGPT, Claude, Gemini, Perplexity – can discover it, understand it, and cite it in their answers. Where SEO earns a ranking on a results page, LLM optimisation earns a citation inside the answer itself.
Why it exists
A growing share of searches never reach a results page: the user asks an assistant, the assistant answers, and the only sites that receive attention are the ones the answer cites. One of our ecommerce clients now receives over 22,000 AI-referred sessions a month – traffic that didn't exist as a category three years ago. When the answer is the destination, being citable becomes the ranking.
What it involves in practice
- Definitive statements. Clear, quotable answers placed at the top of a page – the sentences an LLM can lift and attribute.
- Verifiable data. Models preferentially cite content with concrete, sourced numbers over vague claims.
- Structured markup. Schema.org data (FAQ, Article, Organization) that makes meaning machine-legible.
- Question-shaped content. Pages that answer the questions people actually ask, not just pages that target keywords.
- Crawler access. Letting AI crawlers reach your content, and publishing a curated llms.txt.
How it relates to SEO
It's an extension, not a replacement. The fundamentals overlap almost entirely – content quality, authority, technical accessibility – which is why the same asset can rank in Google and be cited by an assistant. The discipline also travels under other names: see generative engine optimisation (GEO).