This resource is intended to provide insights into website schema and what Forge and Smith does (and doesn’t) do with it as part of our regular project process.

What is schema?

In relation to websites, schema is a standardized snippet of code. Its purpose is to provide more information about the site, the page, and the content to search engines and AI tools.

Schema (also called ‘structured data’ or ‘markup’) doesn’t add content to a page. It labels content that’s already there, in a clear format. Think of it like a shipping label on a box. The label doesn’t change what’s inside, it just means the courier knows exactly what’s in it and where it needs to go without opening it.

You can learn about all the possible types of schema here—there are a lot.

Does my site need schema to rank and have AI visibility?

No. Approximately 88% of all websites don’t use schema, and an Ahrefs study found that only 53% of sites cited by AI tools had schema.

Then why would I need schema?

Search engines and AI tools can read the text on your site’s pages. If you do a great job of using clear labeling, proper heading hierarchy, and well-written page content, search engines will have no problem understanding what your site is about and how to rank it. They also get a lot of this information from crawling your XML sitemap.

But there are some areas where search engines and AI tools can get confused or make mistakes. For example:

  • If your copy is poorly written, search tools will have as much trouble understanding it as humans
  • Ambiguous words can be misinterpreted
  • People and place names can also be ambiguous
  • Dates and times can be misread (especially variations of day-month-year vs. month-day-year)
  • Prices and stock statuses can be hard for tools to parse from complex pages

Schema is also useful to build rich snippets, which are special features like star ratings, recipe cooking times, event details, and, as mentioned above, product prices. These appear in Google and other search engine results.

Is schema included in my project scope?

Yes, some schema configuration is always included in our project scopes.

We install and configure a WordPress plugin called Yoast SEO on every site that we build. Yoast has many SEO purposes, one of which is automating schema.

Yoast detects specific details from your content and injects schema into the code. For each page and post on your site, the Yoast box has drop-down options for Page Type and Article Type – the default is what Yoast has detected, but you can change them.

Yoast also covers other types of schema, such as Organization, infrastructure types (WebSite, WebPage, etc.), and Person (for author bios).

That’s the outside layer of schema, and it’s absolutely sufficient for the majority of websites.

Additional schema beyond Yoast automation is custom work that requires additional scope.

I heard I need FAQ schema, is that true?

Google turned FAQ rich results off entirely as of May 7, 2026, FAQ rich results reporting was removed from Search Console in June, and API support for this type of schema ended in August 2026. Existing FAQPage markup is still technically valid and it will not hurt your site, but it’s no longer producing anything extra on Google.

As far as AI systems are concerned, they do not read schema the way search engines do. They ingest the page as ordinary text and pull from what is visibly written.

In independent testing across ChatGPT, Claude, Perplexity, Gemini, and Google’s AI Mode, content that existed only in the structured data and not in the visible page was almost never extracted. Claude got zero percent. Gemini was the best performer at fifty percent, and its successes came from the visible HTML rather than the markup.

That’s why we always recommend putting FAQ sections on the pages where your audience is most likely to have questions. Answer those questions in plain language. That’s the kind of content that gets cited by AI tools and search features like ‘People Also Ask’.