What is structured data?
Structured data is a block of typed facts about a page, in the schema.org vocabulary, so machines can read it without parsing the prose. Google recommends JSON-LD, a script tag holding the facts as JSON. It must match the visible page. Search engines use it for rich results and to classify the page; other crawlers can lift the same facts.
Part of Glossary: the terms behind MCP, coding agents and shared pages, defined plainly.
Pick the schema.org type that fits: Article for a post, FAQPage for questions and answers, DefinedTerm for a glossary entry, BreadcrumbList for its place.
Write the object in JSON-LD with the type's required properties, using the same facts the visible page shows. Nothing hidden, nothing extra.
{"@context":"https://schema.org","@type":"DefinedTerm","name":"Structured data"}Put it in a script tag of type application/ld+json, in the head or the body. Google reads it either way, including when a script injects it.
Test it. Google's Rich Results Test validates the markup; an item missing a required property is not eligible for a rich result.
Crawlers parse the block alongside the text. Google may use it for a rich result or to understand the page; it has to stay consistent with what readers see.
Source: Google Search Central: introduction to structured data
Example: a glossary page's JSON-LD
Every glossary page on intraface.cc carries three JSON-LD objects. A BreadcrumbList places the page: Glossary, then the term, with each ListItem carrying a position starting at 1, because schema.org says the order of elements in the markup is not enough to indicate order. A FAQPage lists the same two or three questions that appear at the bottom of the page, each a Question with an acceptedAnswer. And a DefinedTerm holds the term, with the 40 to 60 word answer as its description, pointing at the glossary hub through inDefinedTermSet. The hub itself is a DefinedTermSet, which schema.org defines as a set of defined terms such as a glossary, dictionary or classification scheme.
{
"@context": "https://schema.org",
"@type": "DefinedTerm",
"name": "Structured data",
"description": "Structured data is a block of typed facts about a page...",
"url": "https://intraface.cc/glossary/structured-data/",
"inDefinedTermSet": "https://intraface.cc/glossary/"
}
None of these earns a Google rich result today. They are there because the facts on the page become typed fields for anything that reads the HTML, and because generating them from the same content file as the visible text keeps the two from drifting apart.
Why it matters now
The rich-result reasons for structured data have shrunk. Google removed the HowTo rich result in September 2023. In the same month it restricted FAQ rich results to well-known, authoritative government and health websites, and on 7 May 2026 it retired the FAQ rich result altogether, removing the documentation the following month. Breadcrumbs, articles, products and a handful of other types still produce visible results.
What is left is the original purpose: telling a machine what the page is about in a vocabulary it already knows. That matters at least as much now that answer engines read pages to write summaries. Google is explicit that no special markup is needed for AI Overviews. But a page whose type, name, description and questions are stated in a standard form is cheaper for any crawler to classify, and the discipline of keeping markup and text identical is the same discipline that makes a page quotable.
Common mistakes
- Marking up what is not on the page. Google's guidelines say not to mark up content invisible to readers, even if it is accurate. FAQ markup with questions the page does not show is the classic case.
- Expecting a rich result because the markup validates. Valid is necessary, not sufficient. Google says structured data might enable rich results, and a missing required property makes the item ineligible outright.
- Treating FAQ schema as a lever. It was one for most sites until 2023. Keep it for the description it gives other engines, not for Google's results page.
- Letting markup drift from the template. Google warns that pages can break after release because of templating or serving problems. Generate the JSON-LD from the same source as the text and re-test after each deploy.
FAQ
- Is FAQ schema still worth adding?
- Not for Google rich results. Google limited the FAQ rich result to well-known government and health sites in 2023 and retired it entirely on 7 May 2026. FAQPage markup is still valid schema.org and still tells any crawler which questions a page answers, so sites keep it for other engines and AI assistants. Treat it as description, not as a way to win space in Google.
- Should I use JSON-LD or microdata for structured data?
- Google treats JSON-LD, Microdata and RDFa as equally fine when valid, and recommends JSON-LD because it is the easiest to implement and maintain at scale: one block, separate from the HTML, easy to generate from a template. Microdata and RDFa are attributes woven through the markup, which is harder to keep in sync with the text.
- Does structured data help a page appear in AI Overviews?
- Google says no special structured data is needed for AI Overviews or AI Mode, and that there are no requirements beyond being indexed and eligible for a snippet. It does ask that any structured data you have stays consistent with the visible page. So it does not get you in, but markup that contradicts the page can count against you.
Stop sending HTMLover Slack.
Create an account in a minute, connect your coding agent, and make your first board.