What drives AI citations is search results and model brand knowledge
What Drives AI Citations? Content, Retrieval and Model Knowledge in Generative Engine Optimization (GEO)
Information RetrievalComputation and LanguageMachine LearningSocial and Information Networks
Summary
When AI tools like ChatGPT choose web pages to cite, they mostly pick pages they found during their own searches rather than relying on obvious page features. Pages that use words matching what users ask are more likely cited because they fit the AI’s search queries. The AI also tends to mention brands it already knows, even if no linked page supports that mention. The authors analyzed a large set of answers and citations in different markets and showed how search and model memory explain citations more than webpage speed or structure.
What this means in practice
- •For digital marketing teams: Optimize website content by aligning titles and paragraphs with user queries to improve citations from AI-powered search engines.
- •For ai product managers: Adjust brand presentation in AI systems to influence how often products or companies are named in AI-generated answers.
Authors
Mahe Chen, Nick Koudas
Abstract
Generative search engines such as ChatGPT, Gemini and Perplexity now decide which Web pages users see as sources, and a growing industry of generative engine optimization advises publishers on how to earn these citations. We study what shapes citation in two 90-day collections of answers to commercial questions in contrasting markets: 158,099 answers with 468,037 citations. Comparing pages of the same website, content matters in one way: pages whose title and paragraphs match the questions people ask are cited more, whereas FAQ blocks, structured data, summary boxes and page speed appear effective only across websites. Page properties nevertheless explain little of the variation between pages of a website. Replaying the questions through the engines' APIs shows why. Every page ChatGPT cites is one its own search retrieved, and which pages that search retrieves, for sub-queries the model rewrites on every answer and that rarely coincide with Google's or Bing's results, explains far more of the variation than page properties (32% against 2.8%). Content acts mainly through this step, by matching the engine's queries in meaning. The model's own knowledge shapes it too: ChatGPT writes brands it knows into its searches, prefers websites it knows, and names brands without any supporting page, and controlled experiments show that brand names reorder its rankings of products with identical facts. Pages are thus cited when searched; brands are also named when remembered. The findings replicate in a market of small B2B vendors. We release data and code.