JSON-LD and Common Schema Types

JSON-LD is your website's semantic brain—it tells machines what your pages mean, not just what they say. This memo covers what JSON-LD does for Model Context Marketing and the essential Schema.org types you should implement.

What JSON-LD Does for Model Context Marketing

JSON-LD performs three critical functions that make your content understandable to large language models:

1. Tells LLMs Exactly What Your Entity Means

Without structured data, LLMs must infer what your content represents. Is this page about a company? A product? A concept? JSON-LD removes ambiguity by explicitly declaring entity types. When you mark up your page as an Organization, LLMs know they're looking at company information, not just text that mentions a company.

2. Gives Them Structured Understanding

LLMs process structured data more reliably than unstructured text. JSON-LD provides clear fields like name, description, datePublished, and author that AI systems can extract with confidence. Instead of guessing which text represents the article title versus the author name, the schema explicitly defines these relationships.

3. Creates Semantic Relationships

JSON-LD doesn't just describe individual entities—it shows how they relate to each other. An Article has an author who is a Person, published by an Organization, about a specific topic. These connections help LLMs understand context and build knowledge graphs.

Common Schema Types for Marketers

Here are the essential Schema.org types you should implement, with practical use cases for each:

Organization

What it is: Your company or brand entity.

Use case: Homepage, About page, company information.

Key fields: name, url, logo, description, founder, foundingDate, contactPoint

{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "Model Context Marketing",
  "url": "https://modelcontextmarketing.com",
  "description": "Educational resource for creating LLM-friendly content",
  "founder": {
    "@type": "Person",
    "name": "Stephen Newman"
  }
}

Product / SoftwareApplication

What it is: Products, SaaS applications, tools, or software.

Use case: Product pages, feature descriptions, pricing pages.

Key fields: name, description, offers, applicationCategory, operatingSystem

Example use cases:

• SaaS application product pages
• Online courses or educational products
• Software tools and utilities
• Digital products and services

FAQPage

What it is: Frequently asked questions with answers.

Use case: FAQ pages, Q&A sections, knowledge base articles.

Why it works: LLMs handle FAQs exceptionally well—they mimic conversational training data.

{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [{
    "@type": "Question",
    "name": "What is JSON-LD?",
    "acceptedAnswer": {
      "@type": "Answer",
      "text": "JSON-LD is JavaScript Object Notation for Linked Data..."
    }
  }]
}

HowTo

What it is: Step-by-step instructions or tutorials.

Use case: Guides, tutorials, implementation instructions, setup processes.

Key fields: name, description, step (with text, image, url)

Example use cases:

• "How to Implement JSON-LD on Your Website"
• "Complete Guide to Setting Up Schema Markup"
• "Tutorial: Building a Knowledge Graph"

Article

What it is: Articles, blog posts, and written content.

Use case: Blog posts, memos, long-form articles, news content.

Key fields: headline, description, author, datePublished, dateModified

Important: Always include both datePublished and dateModified to signal content freshness.

Review

What it is: Customer reviews, testimonials, case studies.

Use case: Product reviews, service testimonials, case study results.

Key fields: reviewRating, author, reviewBody, itemReviewed

Challenge: Getting authentic, real reviews is difficult. Don't fake these—LLMs can detect patterns of inauthentic reviews. Use actual customer feedback or skip this schema type.

Event

What it is: Events, webinars, launches, demos.

Use case: Webinar announcements, product launches, live demos, conferences.

Key fields: name, startDate, endDate, location (physical or virtual), organizer

Example use cases:

• Webinar registration pages
• Product launch announcements
• Demo session scheduling
• Conference or meetup listings

Key Insight: Meaning vs. Saying

This distinction is crucial. Consider these two examples:

Without Schema (What It Says)

❓ LLMs must infer: Is this a company? A product? Who is the founder? When was it created?

With Schema (What It Means)

{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "Model Context Marketing",
  "foundingDate": "2025",
  "founder": {
    "@type": "Person",
    "name": "Stephen Newman"
  },
  "description": "Helps marketers optimize content for LLMs"
}

✅ LLMs know exactly: This is an Organization, founded in 2025, founder is Stephen Newman, purpose is stated.

Implementation Priority

If you're starting from scratch, implement schemas in this order:

1. Organization — Establish your brand entity first (homepage, about page)
2. Article — Mark up all blog posts, memos, and written content
3. FAQPage — Add to any Q&A sections (high LLM value)
4. Product/SoftwareApplication — If you have products or SaaS
5. HowTo — For tutorials and guides
6. Event — When you have webinars or launches
7. Review — Only when you have authentic testimonials

Multiple Schemas on One Page

You can (and should) use multiple schema types on a single page when appropriate. For example:

• An Article about your company can also include Organization schema
• A HowTo guide can include FAQPage schema for common questions
• A product page with SoftwareApplication can include Review schemas

Just include multiple <script type="application/ld+json"> blocks with different schema objects.

Validation

Always validate your JSON-LD before publishing:

• Google Rich Results Test — Shows how Google sees your markup
• Schema.org Validator — Validates against Schema.org standards
• Google Search Console — Monitor for schema errors after publishing

Why This Matters

Every schema type you implement is another way for LLMs to understand your content with precision. The more structured data you provide, the more confidently AI systems can extract, cite, and recommend your information.

Without JSON-LD, you're forcing LLMs to guess. With it, you're giving them certainty. That certainty translates to citations, recommendations, and authority.