Structured Data Does Not Help With Visibility In AI Search

Sep 15, 2025 - 7:21 am 12 by

Code Data

There are a lot of folks in the community saying that implementing structured data / schema on your pages will help you with AI Search visibility. But few have really tested it until now. And those few tests show that adding structured data / schema does not help with your visibility in AI search, at least not yet.

The first to test this was Mark Williams-Cook who posted on LinkedIn an experiment he conducted where he posted a "visual explanation of why your favourite LLM does not use schema in their core training data." He explained how when the LLMs process the page, it actually "destroys" the schema markup and thus does not use it.

He wrote:

LLMs work by "tokenising" content. That means taking common sequences of characters found in text and minting a unique "token" for that set. The LLM then takes billions of sample "windows" of sets of these tokens to build a prediction on what comes next.

The image below is some example schema that has a colour change applied which represents that set of characters is a unique token as made by the GPT-4o model.

What you will notice is that the schema gets "destroyed". For instance, the schema "@type": "Organization", gets broken down so there are separate tokens for "type" and "Organization", which means that in terms of tokenisation the regular words "type" and "Organization" are not distinguishable from schema.

If schema was included in this training data, all it would do in reality is say there is a slightly (likely insignificant) probability of tokens such as "@ appearing before the word "content".

Here is his screenshot:

Token Llm Schema Destroyed

If that is not good enough for you, Julio C. Guevara also tested it and wrote about his test on LinkedIn as well. He said "We set up two product pages of the same made-up product that both Gemini and ChatGPT had never seen before. One page had all content visible in the HTML as text + structured data, the other page had only structured data and else nothing visible as text (visually empty)."

The result show no benefit. He wrote, "We tried different extraction prompts, hundreds of times, to see if the LLMs could give back information like price, colors, SKU numbers. Surprise, surprise: this only worked on the page with information visible as text."

His test show the LLMs couldn't even see the text within the structured data.

Of course, this can all change in the future but here is some early testing done on this.

Forum discussion at LinkedIn.

 

Popular Categories

The Pulse of the search community

Google Search Volatility

More Details

Search Video Recaps

 
Video Details More Videos Subscribe to Videos

Most Recent Articles

Search Forum Recap

Daily Search Forum Recap: June 18, 2026

Jun 18, 2026 - 10:00 am
Bing SEO

Bing Webmaster Tools May Support More Country Reporting

Jun 18, 2026 - 7:51 am
Google Maps

Google Business Profiles Dashboard Activities Button

Jun 18, 2026 - 7:41 am
Google

Google Tests 3D Model From Attribution In Product Results

Jun 18, 2026 - 7:31 am
Google

Google Discover Feed With More Google News Showcase Results

Jun 18, 2026 - 7:21 am
Google Search Engine Optimization

Google: For Site Moves, Add All Domain Variants To Change Of Address Tool

Jun 18, 2026 - 7:11 am
 
Previous Story: FTC Investigating Google Over Ad Pricing & Terms On Websites