Schema markup is code that labels the parts of a page, such as an article, a product or a question, in a standard vocabulary. Google uses it to understand a page and sometimes to show rich results. Google also says AI Overviews and AI Mode need no special schema, so add markup that matches your visible text and nothing more.
What is schema markup?
Schema markup is structured data written in the shared vocabulary from schema.org. Google defines structured data as a standardized format for providing information about a page and classifying the page content. For a recipe page, that could mean marking the ingredients and cook time. For a blog post, the author and publish date.
The people and the code read the same page. The markup just labels it so a machine does not have to guess.
Which format should you use?
Google supports three: JSON-LD, Microdata and RDFa. It states that it recommends JSON-LD if your setup allows it. JSON-LD sits in a single script block, separate from your visible HTML, which makes it easier to add and to maintain than markup woven through the page.
A minimal JSON-LD example for an article looks like this in concept: a script block of type application/ld+json that names the type (Article), the headline, the author and the date. Many CMS platforms and SEO plugins generate it for you.
What does schema markup do for search?
Google says structured data can enable more engaging results, which are called rich results. It does not guarantee them. Eligibility depends on including all required properties and following the general guidelines and any feature-specific rules, and valid markup can still produce a plain result.
Two further points from Google's structured data policies:
- A manual action for misleading markup removes rich result eligibility. It does not change how the page ranks in web search.
- Markup must be a true representation of the page, and you should not mark up content that readers cannot see.
So schema is mostly about how a result looks and how clearly Google reads the page. It is not a ranking lever by itself.
Does schema markup help you appear in AI answers?
For Google's AI features, Google is direct about it. In its guidance on AI features, it says you do not need to create new machine readable files, AI text files or markup to appear, and that there is no special schema.org structured data you need to add. It still lists making sure your structured data matches the visible text on the page as a best practice.
For other assistants, such as ChatGPT or Perplexity, we are not aware of an official statement that schema is required, so treat any claim that it is as unproven. What those tools reliably need is covered in how to get cited by ChatGPT: be crawlable, and publish a clear, quotable answer on the page.
Which types are worth adding?
Add types that describe what the page really is and that you can keep accurate.
- Article or BlogPosting for posts: headline, author, dates.
- Organization on the home page: name, logo, links to official profiles.
- Product, Recipe, Event, LocalBusiness where those describe the page.
- BreadcrumbList to show the page's place in your site.
FAQ markup is a special case, because its visibility in Google results has changed. We cover that in whether FAQ schema is still worth adding.
How do you add and test it?
- Pick the types that fit the page. Check the required properties in Google's documentation for the feature you want.
- Add JSON-LD through your CMS or plugin, or paste the script block into the page.
- Validate with the Rich Results Test while building.
- After launch, watch the rich result status reports in Search Console, since templating or serving issues can break markup later.
- Use the URL Inspection tool to confirm Google has found the structured data on a given page.
Also make sure the page can be crawled. Google says not to block pages carrying structured data with robots.txt or noindex. For how those two differ, read robots.txt vs noindex.
How much effort is it worth?
For a small site, a sensible scope is one sitewide Organization block, Article markup on posts, and BreadcrumbList if your theme does not add it. That takes an afternoon with a plugin. Beyond that, add a type only when a specific Google feature needs it and you can keep the data true.
Common mistakes
- Marking up content that is not on the page.
- Fake or inflated review markup, which can trigger a manual action.
- Copying one markup block across many pages without changing the details.
- Letting markup go stale after a price, date or author changes.
Frequently Asked Questions
Is schema markup required for SEO?
No. Pages rank without it. It helps Google understand a page and can make you eligible for rich results, but it is optional.
Is schema markup the same as structured data?
Nearly. Structured data is the general idea of labelling page information in a standard format. Schema.org is the shared vocabulary most markup uses.
Does schema help me show up in ChatGPT?
There is no official guarantee that it does. Clear page content and crawler access matter more, so treat schema as supporting detail.
Should I use JSON-LD or Microdata?
Use JSON-LD if you can. Google recommends it, and it keeps the markup separate from the visible HTML.
Can wrong schema hurt my site?
Misleading markup can earn a manual action that removes rich result eligibility. Google says this does not change ranking in web search, but it is still worth avoiding.