Structured Data & Schema

Lesson 4: Structured Data & Schema for Local SEO

Learning objectives

By the end of this lesson you'll understand why structured data matters for local SEO and AI citation specifically, be able to implement basic LocalBusiness JSON-LD, add FAQPage schema correctly, and validate an implementation before considering it finished.

This lesson covers schema from a local-business angle — what a location page needs and why, building on the on-page optimisation lesson earlier in this track. For the deeper technical detail on how schema interacts specifically with AI Overviews and LLM citation, see schema markup for AI search, which this lesson deliberately doesn't repeat.

What structured data actually does

Structured data is code embedded in a page that states, explicitly and in a machine-readable format, exactly what a business is, where it is, and what it does — rather than leaving Google or an AI system to infer that from reading the page's prose. Without it, extraction is slower, less precise, and more prone to error; a system reading unstructured text has to guess at which sentence is the address and which is marketing filler. With it, the address, hours, and category are labelled fields a system reads directly, with no inference step at all.

The three schema types that matter most for a local business

LocalBusiness and its subtypes declare the business entity itself — name, address, phone, hours, category, services. FAQPage declares structured question-and-answer content that search engines can extract for "People also ask" results and AI Overview answers directly. BreadcrumbList declares the page hierarchy, which helps Google understand site structure and often produces a breadcrumb trail in the search results display itself.

Two others worth knowing exist, even if they're used less often on a typical location page: HowTo schema for genuinely step-by-step content, and Review or AggregateRating properties nested inside LocalBusiness when reviews are displayed directly on the page.

Implementing LocalBusiness schema

Add a JSON-LD block like this to the page for each location, not just the homepage:

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "MedicalClinic",
  "name": "Sharma Skin Clinic",
  "url": "https://sharmaskin.in/locations/bengaluru/koramangala/",
  "telephone": "+91-98XXX-XXXXX",
  "description": "Dermatology clinic in Koramangala, Bengaluru. Acne, laser & cosmetic treatments.",
  "address": {
    "@type": "PostalAddress",
    "streetAddress": "42, 5th Cross, 5th Block",
    "addressLocality": "Koramangala",
    "addressRegion": "Bengaluru",
    "addressCountry": "IN",
    "postalCode": "560095"
  },
  "geo": {
    "@type": "GeoCoordinates",
    "latitude": "12.9352",
    "longitude": "77.6245"
  },
  "openingHoursSpecification": [
    {
      "@type": "OpeningHoursSpecification",
      "dayOfWeek": ["Monday","Tuesday","Wednesday","Thursday","Friday"],
      "opens": "09:00",
      "closes": "20:00"
    }
  ],
  "sameAs": [
    "https://www.practo.com/bengaluru/clinic/sharma-skin-clinic"
  ]
}
</script>

Choose the most specific @type schema.org actually defines rather than defaulting to the generic LocalBusiness: MedicalClinic or Physician for a clinic, Restaurant for a restaurant, LodgingBusiness for a hotel, EducationalOrganization or School for a coaching institute, BankOrCreditUnion for a bank, LegalService for a law firm. The more specific type carries more implicit meaning to anything reading it, at no extra implementation cost.

Implementing FAQPage schema

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [
    {
      "@type": "Question",
      "name": "How long does it take to see local SEO results?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "GBP optimisation and citation corrections typically show initial movement within 4-8 weeks. Meaningful rank change in competitive categories usually takes 3-6 months of consistent execution."
      }
    }
  ]
}
</script>

The one rule that trips people up most often: the questions and answers marked up in FAQPage schema must match the visible content on the page exactly. Marking up a question in schema that doesn't appear anywhere in the visible page content is not a minor technical shortcut — it's exactly the kind of mismatch Google's structured data guidelines exist to catch.

A worked example: why the specific type matters

A physiotherapy clinic in Pune initially marked up its location pages using the generic LocalBusiness type, since it was the first result in a quick search for "schema for a clinic." Switching to the more specific MedicalClinic type didn't change a single visible thing on the page — no new rich result appeared, nothing looked different to a human visitor. What it did change was the specificity of what a crawler or AI system could confidently state about the business: with the generic type, all a system could extract was "this is some kind of local business at this address." With the specific type, it could extract "this is specifically a medical clinic." That distinction matters most exactly in the AI-citation context this lesson's companion piece, schema markup for AI search, covers in depth — a system deciding whether to cite a business for "physiotherapy near me" is more confident doing so when the schema itself confirms the category, rather than inferring it from a page description that might be ambiguous.

Validating an implementation

Run the page through Google's Rich Results Test first — enter the URL, check whether schema is detected at all, and read any errors it surfaces. Follow with a general schema.org validator, which catches syntax issues and flags missing recommended fields that the Rich Results Test sometimes doesn't surface on its own. After Google actually crawls the page — usually within one to two weeks of implementation — check Google Search Console's Enhancements section for anything the earlier two tools missed once real crawl data is available.

What practitioners get wrong

The most common mistake is copying a schema template from one location page to every other location without updating the address, geo coordinates, or sameAs links — a schema block that technically validates but describes the wrong location is worse than no schema at all, because it actively misinforms rather than simply lacking information. The second is marking up FAQ content in schema that doesn't match what's visibly on the page, discussed above — this is a Google guideline violation, not a stylistic choice. The third is adding schema once at launch and never revisiting it after a business detail changes — new hours, a new service, a relocated address — leaving the schema silently out of sync with reality while the visible page content gets updated correctly, the same drift problem covered in local content and landing pages.

FAQ

Do I need FAQPage schema on every page, or just a dedicated FAQ page? Any page with genuine, visible question-and-answer content can carry FAQPage schema — it doesn't need to be a dedicated FAQ page specifically. The requirement is that the marked-up questions and answers are actually visible on that page, wherever it lives on the site.

What happens if my schema has an error — does it break the page? No, a schema error doesn't break the page itself or its normal rendering; it just means the structured data either doesn't get read correctly or gets ignored entirely by systems trying to extract it. The page still works fine for human visitors regardless of a schema error.

Can I use more than one schema type on the same page? Yes — a location page commonly carries LocalBusiness (or a subtype) alongside FAQPage and BreadcrumbList simultaneously, each describing a different aspect of the same page. They don't conflict as long as each block is correctly formed on its own.

How does this relate to the LocalBusiness schema glossary entry versus the broader structured data glossary entry? The structured data entry covers the general concept and JSON-LD format; the LocalBusiness entry covers the specific schema type this lesson focuses on implementing. Both are worth reading alongside JSON-LD itself if the underlying format, not just the local-business application of it, is unfamiliar.

Next lesson: Local SEO measurement and attribution →