Home / Blog / Technical guide

Technical guide

Schema markup for AI visibility: what actually works

20 June 2026 · Zeb Choudhry

Schema markup is structured data you add to your pages in JSON-LD format. It communicates facts directly to search engines and AI systems in a language they can parse without inference. Of all the changes you can make for AI visibility, correctly implemented schema is among the highest-leverage — because it removes ambiguity about who you are and what your pages contain.

This guide covers which schema types matter most for AI citations, what to include in each, and the most common mistakes that make schema invisible to the systems you are trying to reach.

Why schema matters specifically for AI

Google's traditional ranking algorithm has processed billions of pages and learned to infer meaning from text even without explicit signals. AI language models are also capable of inference — but they are more likely to cite a page confidently when the facts are stated explicitly rather than implied.

When a user asks ChatGPT or Perplexity "what is the best [type of business] in [city]", the models pull from pages where they can clearly determine: the business name, its category, its location, and its reputation. Schema provides all four of those facts in a single, machine-readable block at the top of the page.

The five schema types that matter most for AI visibility

1. Organization All sites

The foundation. Every site should have Organization schema on the homepage (and ideally on every page via a site-wide template). It establishes who you are as an entity — the starting point for everything else.

Key fields for AI visibility:

{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "Your Business Name",
  "url": "https://yourdomain.com",
  "description": "One sentence: what you do, where, for whom.",
  "foundingDate": "2020",
  "contactPoint": {
    "@type": "ContactPoint",
    "telephone": "+44-...",
    "contactType": "customer service"
  },
  "sameAs": [
    "https://www.linkedin.com/company/...",
    "https://www.crunchbase.com/organization/...",
    "https://en.wikipedia.org/wiki/..."
  ]
}

The sameAs field is critical: it links your entity to external profiles where AI systems can cross-reference and validate your existence. LinkedIn, Crunchbase, and Wikidata are the three most impactful.

2. LocalBusiness (or subtype) Local businesses

For any business with a physical location or service area, LocalBusiness schema (or a more specific subtype like Restaurant, Dentist, or Plumber) is essential. It adds location, hours, and geo-coordinates that AI systems use to answer "near me" and city-specific queries.

{
  "@context": "https://schema.org",
  "@type": "LocalBusiness",
  "name": "Your Business Name",
  "address": {
    "@type": "PostalAddress",
    "streetAddress": "123 High Street",
    "addressLocality": "Manchester",
    "postalCode": "M1 1AA",
    "addressCountry": "GB"
  },
  "telephone": "+44-...",
  "openingHours": ["Mo-Fr 09:00-17:00"],
  "geo": {
    "@type": "GeoCoordinates",
    "latitude": 53.4808,
    "longitude": -2.2426
  }
}

3. SoftwareApplication SaaS / tools

For SaaS products and web tools, SoftwareApplication schema communicates the product category, platform, and pricing model — all facts AI systems use when answering "what is the best tool for X" queries.

{
  "@context": "https://schema.org",
  "@type": "SoftwareApplication",
  "name": "Your Product Name",
  "applicationCategory": "BusinessApplication",
  "operatingSystem": "Web",
  "offers": {
    "@type": "Offer",
    "price": "0",
    "priceCurrency": "GBP",
    "description": "Free audit included"
  },
  "description": "What the product does in one sentence."
}

4. FAQPage Use carefully

FAQPage schema is powerful for AI visibility specifically — even though Google no longer shows FAQ rich results in standard search for most sites. AI systems read FAQPage schema to extract clean Q&A pairs they can use directly in generated answers.

Include your most common customer questions with direct, factual answers. Avoid marketing language — AI systems extract these for factual queries, not promotional ones.

5. Article / BlogPosting Blog content

For blog posts and articles, Article or BlogPosting schema adds authorship, publication date, and topic signals that contribute to E-E-A-T assessment. Always include author, datePublished, and dateModified.

Common schema mistakes that hurt AI visibility

Missing sameAs on Organization

The single most common gap. Without sameAs, your entity exists only on your own domain. AI systems have no external reference point to validate you. Add LinkedIn, Crunchbase, or Wikidata links even if those profiles are minimal.

Schema that contradicts the page content

If your schema says you are located in London but your contact page says Manchester, AI systems detect the inconsistency and reduce confidence in your entity data. Schema and on-page content must agree.

Using Microdata instead of JSON-LD

Microdata is valid but harder to maintain and more prone to implementation errors. JSON-LD in a <script> tag is the recommended format — it sits separately from your HTML and can be updated without touching layout.

Schema on the wrong page

Organization schema belongs on the homepage and ideally in a sitewide template. LocalBusiness schema belongs on the homepage and contact page. SoftwareApplication schema belongs on the pricing or product page. Putting all schema on one page that is not the homepage confuses the entity graph.

Omitting description

The description field is often skipped because it is not required. For AI visibility it is one of the most important: it is the sentence AI systems use to describe your business when they have no other clean passage available.

How to validate your schema

Use Google's Schema.org Validator or the Rich Results Test to check for syntax errors. Neither tool validates AI-specific considerations (sameAs completeness, description quality) — for those, run a Visus audit.