"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."
"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"."
Many in the community believe implementing structured data/schema on pages improves AI search visibility. Few empirical tests exist, and those tests indicate structured data does not yet improve AI search visibility. Large language models tokenize text into character-sequence tokens and learn from billions of token windows to predict subsequent tokens. Tokenization breaks schema markup into regular tokens, so schema elements like "@type": "Organization" become separate tokens for "type" and "Organization", making schema indistinguishable from ordinary words. Inclusion of schema in training data would likely only slightly affect token probabilities, with insignificant impact.
Read at Search Engine Roundtable
Unable to calculate read time
Collection
[
|
...
]