Structured data for AI search: which schema matters for local businesses, which rich results no longer show, a JSON-LD example, and how to validate it.
Structured data for AI search means adding schema markup, usually in JSON-LD, that labels the facts about your business and pages so machines can read them without guessing: who you are, where you are, what you offer, and when you're open. It helps AI systems and search engines understand and describe you accurately, but it isn't a switch that guarantees citations, and anyone who tells you otherwise is overselling.
I want to be direct about that, because this topic is full of inflated claims. Google has said that its AI features rely on the same foundations as Search, and that you don't need special markup to appear in them. What structured data does is remove ambiguity. When your address, hours, services, and social profiles are labeled clearly and match what's visible on the page and in your Business Profile, you make it easier for any system to trust and repeat your information. That's worth doing, and it's cheap.
This guide covers what schema is, which types matter for local businesses, what has changed (several once-popular rich results no longer appear for most sites), a working example, how to implement and validate it, and the mistakes that cause problems. For the wider AI picture, see our guides to Google AI Overviews and local search and getting recommended by Gemini.
Structured data is code added to a page that labels information in a standard vocabulary, most often the one at schema.org. Instead of an AI inferring that "Open 9-5, Mon-Fri" means your business hours, structured data states your opening hours in a labeled, machine-readable format.
Google recommends JSON-LD, a block of code placed in the page that doesn't change how the page looks. For a local business it can communicate:
The honest summary is that structured data helps machines understand your content, and no one has shown it directly causes AI citations. Here's what has been said publicly, and what follows from it.
Google: Its guidance for AI features in Search says the same SEO fundamentals apply, that no special markup is required, and that structured data should match the visible content on the page. Structured data is used to understand pages and to power rich results where eligible.
Microsoft: Microsoft has publicly said that schema markup helps its search and Copilot systems understand content.
What that means for you: treat schema as a clarity and consistency layer. It's one input among many, alongside a complete Business Profile, clear page content, reviews, and independent mentions. It works best when it agrees with everything else.
📊 Flento Data: Local businesses with complete, accurate structured data were better positioned for accurate entity recognition than those relying on unmarked content, giving machines the labeled facts they prefer.
The Flento Schema Stack is a five-layer way to decide what to mark up, in priority order. Build from the top and stop when you've covered what your site genuinely contains.
| Schema type | What it labels | Notes |
|---|---|---|
| LocalBusiness (or a subtype such as Dentist, Plumber, Restaurant) | Name, address, phone, hours, area served, price range | The foundation. Use the most specific subtype available |
| Organization | Brand, logo, official social profiles (sameAs) | Good for the homepage and About page |
| Service | The specific services you offer | Helps machines understand your range |
| Product and Offer | Items and prices | For businesses that sell products |
| FAQPage | Questions and answers visible on the page | See the limits below |
| Person | Authors and practitioners | Supports credentials and expertise |
| BreadcrumbList | Site structure | Supports clean navigation signals |
| Review and AggregateRating | Ratings | Subject to strict self-serving limits, see below |
Several schema types no longer produce visible search features for typical local business sites, and guides that promise otherwise are out of date.
FAQ rich results: Google limited FAQ rich results to well-known authoritative government and health sites in 2023, so most businesses won't see expandable FAQs in results. Marking up an FAQ can still clarify the page, but don't expect a visual feature.
How-to rich results: Google removed these from Search results in 2023.
Review stars for your own business: Google doesn't show star ratings in search results for a business's own reviews marked up on its own website (self-serving reviews of LocalBusiness or Organization). The stars people see for local businesses come from the Business Profile. Our guide to embedding Google reviews on your website explains this in more detail.
Rules change, so check Google's current structured data documentation before you invest in a specific type.
Here's a simplified JSON-LD example for a local plumber. Replace every value with your own accurate details, and make sure each one matches what's visible on your site and in your Google Business Profile.
{
"@context": "https://schema.org",
"@type": "Plumber",
"name": "Example Plumbing Co.",
"url": "https://www.example.com/",
"telephone": "+1-555-123-4567",
"image": "https://www.example.com/images/storefront.jpg",
"priceRange": "$$",
"address": {
"@type": "PostalAddress",
"streetAddress": "123 Main St, Suite 4",
"addressLocality": "Austin",
"addressRegion": "TX",
"postalCode": "78701",
"addressCountry": "US"
},
"geo": {
"@type": "GeoCoordinates",
"latitude": 30.2672,
"longitude": -97.7431
},
"openingHoursSpecification": [
{
"@type": "OpeningHoursSpecification",
"dayOfWeek": ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday"],
"opens": "08:00",
"closes": "18:00"
}
],
"areaServed": ["Austin", "Round Rock", "Cedar Park"],
"sameAs": [
"https://www.facebook.com/exampleplumbing",
"https://www.linkedin.com/company/exampleplumbing"
]
}
Notice what isn't in it: no AggregateRating. Only add rating markup if it follows Google's review snippet guidelines, and don't expect stars for self-published reviews.
Implementing local schema takes an afternoon for most sites. Do it in this order.
Multi-location businesses: give each location its own page with its own markup for that location's address, phone, and hours. Don't combine several locations into one block.
🛠️ Action Step: Run your homepage through Google's Rich Results Test today. Seeing what's already detected shows you the gap.
Most schema problems come from markup that disagrees with the page, and a few can create real risk.
Markup that doesn't match visible content. Marking up hours, prices, or FAQs that aren't on the page violates Google's guidelines. What you label must be visible to visitors.
Marking up fake or self-serving reviews. It can lead to a manual action against your site's structured data, and it won't produce stars.
Conflicting information. Schema that says one address while your Business Profile says another confuses rather than helps. Our guide to NAP consistency explains why consistency matters.
Over-marking. Adding every possible type to every page adds noise, not value.
Set and forget. Outdated hours in schema are worse than none.
Assuming it replaces content. Schema labels what's there. It can't make thin content useful.
Schema is one layer, and the bigger drivers of AI visibility are elsewhere. Systems that answer local questions lean on clear page content, a complete and consistent Business Profile, recent reviews, and independent mentions.
Our guide to optimizing for Perplexity covers one AI assistant in depth.
Small businesses need less schema than they think. A correct LocalBusiness or Organization block on the homepage, matching your Business Profile, delivers most of the benefit. Add Service markup to service pages and BreadcrumbList site-wide if your platform makes it easy.
B2B companies without a storefront should use Organization markup with an accurate description, logo, contact details, and official profiles in sameAs, and can add Service markup for what they sell. The goal is the same: an unambiguous, consistent identity.
✅ Done? See how Flento audits the profile your schema should mirror → Get started free →
What structured data helps AI search engines cite my pages? Accurate LocalBusiness or Organization markup, plus Service markup where relevant, that matches your visible content and Business Profile. It helps systems understand you, but it isn't a guarantee of citations.
Does schema markup help with AI citations? Google says no special markup is needed for its AI features, and structured data should match the visible content. Microsoft has said schema helps its systems understand content. Treat schema as a clarity layer, not a citation switch.
What is the best schema for local businesses? LocalBusiness or its most specific subtype, on your homepage or contact page, with accurate name, address, phone, hours, and area served.
Should I add FAQ schema? It can clarify a page, but most sites no longer get FAQ rich results in Google. Only mark up questions and answers that are visible on the page.
Will review schema give me star ratings? Not for your own business's reviews published on your own site. Google doesn't show stars for that kind of self-serving markup.
How do I test my structured data? Use Google's Rich Results Test and the Schema Markup Validator, and check Search Console for errors after launch.
Do I need structured data for a small business? A single accurate LocalBusiness or Organization block on your homepage delivers most of the benefit. Add more only where it reflects real content.
Structured data won't make an AI cite you by itself, but it removes ambiguity about who you are and what you offer, and it costs an afternoon. Label the basics accurately, match everything to your Business Profile, validate it, and put your effort into the bigger signals: clear content, reviews, and consistent information.