Introduction to LLM-Optimized Pages
Google Search Advocate John Mueller has expressed his thoughts on creating separate Markdown or JSON pages for large language models (LLMs). He questioned the need for LLMs to have pages that no one else sees. This discussion started when Lily Ray asked about creating separate markdown or JSON pages for LLMs and serving those URLs to bots.
The Discussion on LLM-Optimized Pages
The question drew attention to a developing trend where publishers create "shadow" copies of important content in formats that are easier for AI systems to understand. Mueller replied that he isn’t aware of anything on Google’s side that would call for this kind of setup. He notes that LLMs have worked with regular web pages from the beginning and have no problems dealing with HTML. Mueller argued that if file formats made a meaningful difference, companies running those systems would be very vocal about that.
What Mueller Said About LLM-Only Pages
Mueller said that some pages may still work better for AI systems than others, but he doesn’t think that comes down to HTML versus Markdown. He also mentioned that excluding JavaScript, which still seems hard for many of these systems, is important. Taken together, Mueller’s comments suggest that, from Google’s point of view, you don’t need to create bot-only Markdown or JSON clones of existing pages just to be understood by LLMs.
How Structured Data Fits In
Other individuals in the thread drew a line between speculative "shadow" formats and cases where AI platforms have clearly defined feed requirements. For example, OpenAI’s eCommerce product feeds use JSON schemas to govern how ChatGPT ingests and displays product data. This example supports the idea that structured feeds and schemas are most important when a platform publishes a spec and asks you to use it.
Why This Matters
If you’re questioning whether to build "LLM-optimized" Markdown or JSON versions of your content, this exchange can help steer you back to the basics. Mueller’s comments reinforce that LLMs have long been able to read and parse standard HTML. For most sites, it’s more productive to keep improving speed, readability, and content structure on the pages you already have, and to implement schema where there’s clear platform guidance.
Looking Ahead
The conversation highlights how fast AI-driven search changes are turning into technical requests for SEO and dev teams, often before there is documentation to support them. Until LLM providers publish more concrete guidelines, this thread points you back to work you can justify today: keep your HTML clean, reduce unnecessary JavaScript where it makes content hard to parse, and use structured data where platforms have clearly documented schemas.
Conclusion
In conclusion, creating separate Markdown or JSON pages for LLMs is not necessary, according to Google’s John Mueller. LLMs have been able to read and parse standard HTML from the beginning, and there is no need to create bot-only clones of existing pages. Instead, focus on improving speed, readability, and content structure on your existing pages, and implement schema where there’s clear platform guidance. By doing so, you can ensure that your content is optimized for both users and AI systems, without needing to create separate pages for LLMs.