Structured Data

Structured data is a standardized way to describe a page to Google and classify its content. Google works hard to understand pages and asks you to help with explicit clues. Two conclusions follow: markup can enable search results that are more engaging (rich results), but Google does not guarantee they will show up, even if the markup passes the Rich Results Test. Also, some types once treated as results "activators" have stopped appearing: first of all FAQ.

The basis is the introduction to structured data, the general guidelines and the search gallery.

What markup is for

Google uses structured data to understand the content of the page and to gather information about the world, such as the people, books or companies in the markup. The documentation's example: on a recipe page the markup describes the title, author, time and calories, and Google can show a rich result. Structured data is coded using in-page markup on the page that the information applies to, and describes the content of that page.

Formats

Three formats are supported: JSON-LD, Microdata and RDFa. JSON-LD is recommended. Microdata and RDFa use HTML attributes and typically go in the body, but can be in the head. JSON-LD can also be injected by script; more in the lesson on JavaScript and SEO. If you work in a CMS, direct access to the HTML may be closed: then look for a settings page or a markup plugin, as in the lesson on choosing a CMS.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "Structured data",
  "datePublished": "2026-10-03",
  "author": [{ "@type": "Person", "name": "Author Name",
    "url": "https://example.com/authors/name" }]
}
</script>

The rules for the author are in the lesson on E-E-A-T and on the Article page.

What changed: FAQ and HowTo

FAQPage was long recommended as a way to get expandable questions in results, with cases such as "50 articles, CTR from 2.1% to 4.7%". That no longer works. The documentation changelog says the FAQ feature was first limited to well-known, authoritative government and health sites, then deprecated, and from 7 May 2026 the FAQ rich result is no longer shown in Google Search results. Its documentation has been removed. For HowTo it is similar: the rich result stopped being shown, and its documentation was removed too. If you already have such markup you may keep it for other services, but it does not affect visibility or rankings in Google. The current list of features is in the search gallery; before implementing any type, check it is still there.

Google's requirements

• Markup describes what is visible. Don't mark up content that is not visible to readers of the page: if the JSON-LD describes a performer, the page body must describe that same performer. Do not create empty pages just to hold markup;
• Relevance. The markup must be a true representation of the page. Examples of violations: a woodworking site labeling instructions as recipes;
• Completeness. Items missing required properties are not eligible for rich results; the more recommended properties, the higher the quality of the result;
• Access. Don't block pages with markup from Googlebot using robots.txt, noindex or similar — see the lesson on robots.txt and meta robots;
• Images. Image URLs in markup must be crawlable and indexable;
• Duplicates. If you have duplicate pages for the same content, Google recommends the same markup on all duplicates, not only on the canonical page.

A violation: what it costs

A markup issue can result in a manual action. But an important clarification: a structured data manual action means a page loses eligibility for appearance as a rich result; it doesn't affect how the page ranks in web search. Violating quality guidelines can prevent display or, in the extreme, be marked as spam.

Why markup does not guarantee display

The documentation lists reasons: markup enables a feature to be present, it does not guarantee that it will be; the algorithm tailors results to the user (history, location, device) and may decide a text result or another feature is better; the markup is not representative of the main content or is misleading; an error the Rich Results Test could not catch; content hidden from the user. So the claim "markup raises CTR" cannot be treated as a law.

Measuring the effect yourself

Google itself cites cases: for example Rotten Tomatoes measured a 25% higher click-through rate for pages with markup than for pages without, and Nestlé measured an 82% higher click-through rate for pages shown as rich results. Those are specific sites' data, not a promise. For your own site Google suggests a before and after test on a few pages: take pages without markup that have several months of data in Search Console, are not seasonal and don't change much; add the markup; confirm it is valid and that Google found it with URL inspection; then record performance for a few months in the Performance report, filtering by URL.

How to check

1. Open the page in the page analyzer: it shows whether JSON-LD is present, which types were found and whether there are JSON parse errors.
2. Check eligibility in the Rich Results Test and whether the markup was found in Search Console URL inspection.
3. Make sure the marked-up values are visible on the page.
4. Crawl the site with the crawler and find templates with no markup or with duplicated markup.
5. For a bulk check of statuses, the URL audit; for speed, Core Web Vitals.

What is not in the documentation

• A promise that markup raises positions or guarantees a rich result;
• Current relevance of FAQ markup for results: the feature is no longer shown;
• A claim that Article with an author directly "helps E-E-A-T evaluation": markup helps understand who the author is;
• A universal figure for CTR growth: there are only specific sites' cases and the advice to measure yourself.

Checklist

1. The type is chosen from the current features gallery, not from an old article.
2. The format is JSON-LD, on the page it describes.
3. Marked-up content is visible to the reader and matches the page.
4. Required and as many recommended properties as possible are given.
5. Pages are not blocked from Googlebot.
6. The effect is measured before and after on a few pages.

Mistakes

• Implementing FAQPage for blocks in results.
• Marking up what is not on the page, or someone else's reviews.
• Expecting a guaranteed rich result.
• Treating a valid test result as a guarantee of display.
• Blocking marked-up pages from crawling.

What comes next

Next is Core Web Vitals: how to measure page speed and stability.