Structured Data and JSON-LD

Concept

Structured data states facts about a page in a form machines read directly: this is an article, published on this date, by this author. A crawler can infer some of that from the markup. Structured data removes the guessing.

The vocabulary is Schema.org, shared across search engines. The recommended format is JSON-LD - a script block in the head, separate from the visible markup.

It does not improve rankings. What it does is make a page eligible for rich results: star ratings, event dates, recipe times, breadcrumb trails, FAQ accordions in the result itself. Those change how much of the page a searcher sees before clicking, which changes how many click.

Syntax

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "How to read a bus timetable",
  "datePublished": "2026-03-02",
  "dateModified": "2026-08-11",
  "author": {
    "@type": "Person",
    "name": "Meera Iyer"
  },
  "publisher": {
    "@type": "Organization",
    "name": "Riverside Journal",
    "logo": {
      "@type": "ImageObject",
      "url": "https://example.edu/logo.png"
    }
  },
  "image": "https://example.edu/images/bus-stop.jpg",
  "mainEntityOfPage": "https://example.edu/journal/bus-timetable"
}
</script>

The script has a specific type attribute, so browsers never execute or display it. Three keys are structural: @context names the vocabulary, @type names the thing, and the rest are its properties.

The types worth knowing

BreadcrumbList

Shows the site hierarchy in the result instead of a bare URL. Straightforward to add and applicable to most sites.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "BreadcrumbList",
  "itemListElement": [
    { "@type": "ListItem", "position": 1, "name": "Home", "item": "https://example.edu/" },
    { "@type": "ListItem", "position": 2, "name": "Courses", "item": "https://example.edu/courses" },
    { "@type": "ListItem", "position": 3, "name": "Design", "item": "https://example.edu/courses/design" }
  ]
}
</script>

FAQPage

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [
    {
      "@type": "Question",
      "name": "When do applications close?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "Applications close on 30 June at five in the afternoon."
      }
    },
    {
      "@type": "Question",
      "name": "What documents do I need?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "A photocopy of your identity card, two photographs and the original receipt."
      }
    }
  ]
}
</script>

Every question and answer must be visible on the page. Marking up content that is not there is a violation and the markup will be ignored or penalised.

Organization and LocalBusiness

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "CollegeOrUniversity",
  "name": "Riverside College",
  "url": "https://example.edu",
  "logo": "https://example.edu/logo.png",
  "telephone": "+91-20-2600-1234",
  "address": {
    "@type": "PostalAddress",
    "streetAddress": "14 Sadar Bazaar Road",
    "addressLocality": "Pune",
    "postalCode": "411001",
    "addressCountry": "IN"
  },
  "sameAs": [
    "https://www.linkedin.com/company/example",
    "https://en.wikipedia.org/wiki/Example"
  ]
}
</script>

sameAs links the organisation to its profiles elsewhere, which helps a search engine connect them as one entity.

Product

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Product",
  "name": "Two litre electric kettle",
  "image": "https://example.com/images/kettle.jpg",
  "description": "Boils in ninety seconds. Two year warranty.",
  "sku": "KET-2L-001",
  "offers": {
    "@type": "Offer",
    "url": "https://example.com/products/kettle",
    "priceCurrency": "INR",
    "price": "2499",
    "availability": "https://schema.org/InStock"
  },
  "aggregateRating": {
    "@type": "AggregateRating",
    "ratingValue": "4.4",
    "reviewCount": "87"
  }
}
</script>

Others in common use

Combining several on one page

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@graph": [
    { "@type": "BreadcrumbList", "itemListElement": [ ... ] },
    { "@type": "Article", "headline": "...", "author": { "@id": "#author" } },
    { "@type": "Person", "@id": "#author", "name": "Meera Iyer" }
  ]
}
</script>

@graph holds several objects in one block, and @id lets them reference each other rather than repeating the same author details three times.

The rules Google enforces

• Mark up only what is visible on the page. Invisible or contradictory structured data is a violation.
• Be accurate. Fabricated ratings or prices lead to a manual penalty.
• Include the required properties for each type; without them the page is not eligible for the rich result.
• Rich results are never guaranteed. Valid markup makes a page eligible, not entitled.
• Render it server side. Injecting it with JavaScript can work but delays or prevents recognition.

Testing

• The Rich Results Test - paste a URL or the markup and see which rich results the page qualifies for.
• The Schema Markup Validator - checks the syntax against Schema.org itself, beyond what Google supports.
• Search Console Enhancements - reports valid and invalid structured data across the whole site over time.

Test before shipping. A missing required property is silent - nothing breaks, the rich result simply never appears.

Microdata and RDFa

<!-- microdata: attributes woven into the markup -->
<article itemscope itemtype="https://schema.org/Article">
  <h1 itemprop="headline">How to read a bus timetable</h1>
  <span itemprop="author">Meera Iyer</span>
</article>

Both are still supported, and both entangle the data with the markup so that a design change can break it. JSON-LD sits in one block, is easy to generate from a template, and is what Google recommends. Use it.

Important rules

• Structured data affects appearance, not ranking.
• Only mark up visible content.
• Each type has required properties.
• JSON-LD in a application/ld+json script block is the recommended form.
• Server render it.
• Test before shipping.

Common mistakes

• Marking up content that is not on the page.
• Invented ratings and review counts.
• Missing required properties, so nothing appears.
• Invalid JSON - a trailing comma silently breaks the whole block.
• Relative URLs where absolute ones are required.
• Structured data contradicting the visible page.
• Adding it to every page indiscriminately rather than where a rich result exists.

Best practices

• Start with BreadcrumbList and Organization; both apply almost everywhere.
• Add the type that matches what the page actually is.
• Generate it from the same data that renders the page, so the two cannot diverge.
• Use @graph for several objects rather than several script blocks.
• Validate with the Rich Results Test on every template.
• Watch the Enhancements report for regressions.

Practice

1. Add BreadcrumbList to a deep page and test it with the Rich Results Test.
2. Mark up a real FAQ section where every question is visible on the page.
3. Add Article markup generated from the same data that renders the byline and date.
4. Remove one required property and observe what the test reports.