Structured data is code added to a web page, in a standard format, that labels what the page contains: a recipe's ingredients and cooking time, a product's price, an event's date. It gives search engines explicit clues about a page's meaning instead of leaving them to infer it, and Google says most of the structured data its Search uses is written with the schema.org vocabulary.
How structured data works
Google supports three formats: JSON-LD, a script block in the page's HTML; Microdata; and RDFa. It recommends JSON-LD where a site's setup allows, because it is the easiest to implement and maintain at scale.
The labels come from schema.org, a shared vocabulary founded by Google, Microsoft, Yahoo and Yandex and developed through a W3C community group since April 2015. It currently lists 826 types and 1,540 properties, and schema.org says that as of 2024 over 45 million web domains used it.
Google uses the markup it finds to understand a page and to gather facts about the people, books or companies it describes. Markup can also make a page eligible for rich results, the enhanced listings Google builds from it, such as a recipe result. A page needs all of a feature's required properties to be eligible, and eligibility is all it buys: Google says it does not guarantee that structured data will show up in results, even when the markup is correct.
Google's guidelines add two rules. The markup must describe content that readers can see on the page, and spammy markup can lead to a manual action that removes a page's eligibility for rich results.
An example: a recipe page
Google's documentation shows a recipe page carrying markup like this (trimmed here; Google's version has more properties):
<script type="application/ld+json">
{
"@context": "https://schema.org/",
"@type": "Recipe",
"name": "Non-Alcoholic Piña Colada",
"author": { "@type": "Person", "name": "Mary Stone" },
"totalTime": "PT3M",
"recipeIngredient": ["400ml of pineapple juice", "100ml cream of coconut", "ice"]
}
</script>
Because each element is labeled, Google says people can find the recipe by ingredient, calorie count or cook time. Google also publishes case studies. Rotten Tomatoes added structured data to 100,000 pages and measured a 25% higher click-through rate on those pages than on pages without it.
Why structured data matters for AI answers
Google is direct about this. Structured data is not required to appear in its generative AI features, and there is no special schema.org markup to add for Google AI Overviews or AI Mode. Google still recommends keeping markup as part of normal SEO, and its checklist for AI features includes making sure structured data matches the visible text on the page.
There is an indirect route as well. In 2020 Google said its Knowledge Graph draws partly on markup that site owners add, linking to its event markup documentation, and in 2025 it said AI Mode taps the Knowledge Graph. Neither statement says how much that path matters for a given page. Our guide to schema markup for AI search covers what is worth marking up.
Common misconceptions
- "Correct markup guarantees a rich result." Google says it does not.
- "FAQ markup earns FAQ dropdowns." Google limited FAQ rich results to well-known government and health sites in August 2023 and stopped showing them in Search from 7 May 2026. How-to rich results were deprecated in September 2023.
- "AI answers need special schema." Google says its AI features need no new machine-readable files, AI text files or special markup. That includes llms.txt, which Google Search ignores.
- "Markup can describe what readers cannot see." Google's guidelines say not to mark up content that is not visible on the page.