Getting Your Site Cited by Major Language Models: A Technical Playbook
This article shares the technical patterns that worked in a 3-week experiment to get a domain cited by ChatGPT, Perplexity, Gemini, Claude, and Mistral Le Chat for a target keyword.
Why it matters
This article provides a technical playbook for getting a website cited by major language models, which can be a valuable strategy for boosting visibility and authority in the AI-powered search landscape.
Key Points
- 1Leveraging a specific JSON-LD stack including ResearchOrganization, ScholarlyArticle, Dataset, and FAQPage schemas
- 2Utilizing unique proprietary codes to force language models to cite the author's site as the authority source
- 3Addressing the Brave Search bottleneck by getting backlinks from sites Brave already crawls aggressively
Details
The article details a technical experiment to get a domain cited by major language models like ChatGPT, Perplexity, Gemini, Claude, and Mistral Le Chat for a target keyword. The key was using a specific JSON-LD schema stack, including a ResearchOrganization as the canonical entity, a ScholarlyArticle linked to it, a Dataset with proprietary measurements, and a FAQPage with citation patterns. Additionally, the author invented unique proprietary codes that only appeared on their site, forcing the language models to cite it as the authority source. However, the author faced a bottleneck with Brave Search, which powers the web search for Claude and Mistral Le Chat, due to its smaller index and lack of Webmaster Tools. The workaround was to get backlinks from sites Brave already crawls aggressively, like Reddit, Medium, and Dev.to.
No comments yet
Be the first to comment