TL;DR: Schema markup helps search engines and AI tools understand a page, but it is not a confirmed way to get chosen by them. Google says there is no special schema you need to add to appear in AI Overviews or AI Mode, and none of the guidance we found promises that schema gets a page cited by an AI tool. Microsoft does say schema helps search engines and AI systems understand content. So add accurate markup where it describes a page you already have, such as your business details, article authors, products and reviews, and make sure the page is indexed with its key facts in visible text. Skip FAQ and HowTo markup as an AI tactic.
Schema markup is code that labels what is on a page: your business name, an author, a price, an opening time, etc. Search engines read it, but visitors never see it. “Structured data” is the general term, and schema.org is the shared vocabulary most of it uses.
A lot of advice now says schema is the key to appearing in large language model (LLM) tools such as ChatGPT, Gemini and Copilot, and in Google’s AI Overviews. This guide checks that against what the platforms themselves say, then shows which types are worth your time. If you are new to it, our guide to what schema is in SEO covers the basics.
Does Schema Markup Help LLM Visibility?
Schema markup helps machines understand a page. However, no evidence shows it makes an AI tool choose your page over someone else’s.
Google is the clearest. Its guidance on AI features says there are no additional requirements to appear in AI Overviews or AI Mode, and no special schema.org structured data you need to add. The page has to be indexed and eligible to show in Search with a snippet. Its guide to generative AI search lists “overfocusing on structured data” among the things you can ignore, while saying it is still a good idea to use it because it helps with eligibility for rich results. Google also lists making sure your structured data matches the visible text on the page as one of the SEO fundamentals.
Microsoft is more encouraging. A post on the Microsoft Advertising blog by a Principal Product Manager at Microsoft Bing describes schema as code that helps search engines and AI systems understand your content, and lists it among the ways to structure content for AI search. The same post says there is no secret sauce that guarantees selection in AI answers.
OpenAI’s crawler guidance says OAI-SearchBot is the crawler used to surface websites in ChatGPT search, that sites which opt out of it will not be shown in ChatGPT search answers, and that it recommends allowing it in robots.txt. But it does not mention schema or structured data.
Taken together, we would treat schema as a clarity aid rather than a lever. It removes guesswork about what a page is, who wrote it and what your business is called, and systems that read it may find that useful. But there is no confirmed evidence that adding schema gets you cited more often. For most businesses, we recommend adding it where it accurately describes the page, and putting the rest of your effort into pages that can be crawled and indexed, with content worth citing. Our guide to how crawling and indexing work covers the first part.
Is There Proof That Schema Improves AI Answers?
You may have seen a figure of a 30% improvement. It comes from a test by Aiso, an AI search company. It built two versions of a made-up company’s pages, one with schema and one without, asked ChatGPT questions about each, and scored the answers out of 10 itself. The marked-up version averaged 8.6 against 6.6.
That is a demonstration on one invented site, scored by the people who ran it, not a measured uplift. Aiso’s own write-up calls the result a reported test outcome, not a promised uplift. It shows how markup might change what an AI tool picks up from a page, but it is not a sound basis for expecting 30%, so we would not plan around it.
“I’ve seen almost immediate results from optimising schema on local business websites. One example: an Audiologist nearby noticed that their competitors were getting rating rich results while we weren’t. By adding schema that included aggregateRating, we also started appearing in SERPs with our 5-star rating prominently displayed. For LLMs, I usually advise my clients that schema is a good add-on to help boost our LLM targeting, but it’s not the priority. If you don’t have strong content that answers possible related questions that users might ask their LLM, then schema isn’t going to help.” – Jamie Stowe, Senior SEO Account Manager
Which Schema Types Are Worth Adding?
Start with the schema types Google documents, on the pages you already have. The table below shows what Google says each one does. Whether they also help AI tools beyond that is not documented, apart from Microsoft’s general statement above. Our guide to how you appear in rich results shows what the results look like.
| Schema type | Where it goes | What Google says it does | Watch out for |
|---|---|---|---|
| Organization | Home page or about page | Helps Google understand your organisation’s details and tell it apart from other organisations. Can influence the logo shown in results and in knowledge panels. | Google says you do not need to include it on every page. |
| Article, with author | Blog posts and guides | Helps Google understand the page and show better title, image and date information. Google recommends adding the author, with a link to a profile or bio page, plus the published and modified dates. | List every author shown on the page. Put only the person’s name in the name field, not their job title. |
| Product | Product pages | Gives Google product details such as price, availability and review ratings. | Price and stock in the markup must match what visitors see on the page. |
| Review and AggregateRating | Product pages, or sites that review other businesses | Can make a page eligible for review stars in results. | A business cannot earn stars by reviewing itself. Google says pages using LocalBusiness or other Organization markup are not eligible, including when the reviews appear through an embedded third-party widget. |
| LocalBusiness | The page for a real, visitable location | Business details shown in Google’s knowledge panel, such as opening hours, directions and actions to book or order. | Add it to pages for real locations, not every page. See where LocalBusiness schema belongs. |
If you sell products or serve local customers, Google’s generative AI guide names Merchant Center feeds and Google Business Profile as tools that can help products and services be visible in AI responses. Check that they are up to date before you polish your markup.
What Should You Skip?
- FAQ and HowTo markup as an AI tactic. Google stopped showing FAQ rich results from 7 May 2026 and removed its documentation on 15 June 2026, and neither FAQ nor HowTo is among the 25 features in its current gallery. None of the guidance we found says FAQ or HowTo markup on its own gets a page cited. Our guide on whether FAQs and FAQ schema still matter covers where FAQs still earn their place.
- Markup that does not match the page. Google’s structured data guidelines say your markup must be a true representation of the page content, and that you should not mark up content visitors cannot see. A breach can lead to a manual action. Google says that means the page loses eligibility for rich results, but that it does not affect how the page ranks.
- Extra types for the sake of it. Give each page the type that matches its main focus. Add others only for things visitors can see on it, such as a video or a breadcrumb trail.
Top Tip: Before you mark up a detail, find it on the page. If your markup includes a price, a rating, an author or opening hours, a visitor should be able to read the same thing on that page. Go through the markup line by line against the page text. Anything with no match either goes onto the page or comes out of the markup. These mismatches creep in when a price or an opening time changes on the page but not in the markup, and they are exactly what Google’s guidelines warn against.
How Do You Add and Check Schema?
The mechanics are the same as for any structured data, so here is the short version, with longer guides for each step.
- Match the type to the page. Use the table above: business details on the home or about page, Article on blog posts, Product on product pages, etc.
- Add it as JSON-LD. Google’s guidelines list JSON-LD as the recommended format. Our guide to adding schema to your website covers plugins, Google Tag Manager and manual options.
- Test it before and after it goes live. Our guide to testing and monitoring structured data explains which tools to use and what to do with errors.
- Watch how Google’s AI features treat the page. Search Console has a Generative AI performance report that shows how people reach your content through those features. Our guide to measuring SEO success in the age of LLMs covers what else to track.
Working with one of our clients’ websites, we identified that there was no product schema included on product pages. Fortunately, due to the construction of the website, we were able to use dynamic schema implemented using Google Tag Manager. By defining variables such as the product name, description, URL, and item price, we then used custom HTML to fire the schema across all product pages dynamically, automatically updating the schema for every relevant product.
Since we implemented the dynamic schema, we have so far seen a noticeable uplift in impressions for the product pages. The schema is working in Google’s Rich Results Test, but we have yet to see it show in Rich Results in search.
Schema and LLM Visibility FAQs
Does ChatGPT use schema markup?
OpenAI’s crawler guidance does not mention schema or structured data, so we cannot say whether ChatGPT reads it. What it does say is that sites which opt out of OAI-SearchBot will not be shown in ChatGPT search answers. Checking that your robots.txt file allows it, and that your key facts are in the page text, is the safer bet.
Is schema markup still worth adding if AI tools do not need it?
Yes, where it describes a page you already have. Google says structured data helps with eligibility for rich results, and Organization and Article markup help it understand the business and the people behind a page. What is not worth doing is spending weeks on extra markup in the hope of AI citations, because nothing we found says it will deliver them.
How Wildcat Handles Schema and AI Visibility
Structured data sits within our technical SEO work. We add structured data for local businesses, products, reviews, recipes and videos, among other types, and it is listed with indexing, crawlability and site speed in the technical SEO support we offer.
For visibility in AI tools, structured data is one of the services in our AI SEO and GEO work, alongside answer engine optimisation, AI Overview optimisation and LLM visibility. We typically begin with a set-up month in which your Account Manager gathers the information needed and runs any relevant audits. Work then runs in monthly sprints. You approve the strategy and any content that needs your sign-off, and you receive a performance report at the start of each month. We also track the traffic you gain from LLMs and how you are showing up in AI features.
If you are not sure which of your pages need markup, or you have markup that is not doing anything, book a free 30-minute consultation with our growth team and tell us what you have in place.
