Schema.org Gives Search Engines Better Facts, Not Better Rankings

Schema.org is a shared vocabulary for describing things on a web page: an article, product, event, organisation, person, recipe, and much more. Search engines can use that structured data to understand facts that are less obvious in prose.

Adding it does not guarantee a ranking increase or a rich result. It makes the page easier for machines to interpret. Whether a search feature appears is still the search engine’s decision.

JSON-LD keeps the markup manageable

Schema.org can be written with JSON-LD, Microdata, or RDFa. For most sites, I prefer JSON-LD because it keeps the structured data in one script rather than scattering attributes through the HTML.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "Schema.org Gives Search Engines Better Facts",
  "datePublished": "2024-03-06",
  "author": {
    "@type": "Person",
    "name": "Ronalds Vilciņš"
  }
}
</script>

The structured values should match what a visitor can see on the page. Marking up a five-star review that does not exist in the content is not optimisation; it is misleading data with braces around it.

Choose the type from the page

Common starting points include:

• Article for editorial content;
• Product with Offer for an item and its price or availability;
• LocalBusiness or Organization for a business entity; and
• Event for something at a defined time and place.

Use the most specific suitable type in the Schema.org vocabulary, then add properties the page can support accurately. More markup is not automatically better markup.

For a product, price and availability need to stay in sync with the visible offer. For a local business, address and opening hours need maintenance. For an article, headline, author, image, and publication dates should come from the same content data used to render the page.

Schema.org and Google are different layers

Schema.org defines a broad vocabulary. Search engines document smaller sets of properties required or recommended for their particular rich-result features.

A block can therefore be valid Schema.org data without being eligible for a Google rich result. Conversely, passing a rich-results test does not promise that the result will appear.

Validate both syntax and meaning

Google’s Rich Results Test checks markup for supported Google search features. The Schema Markup Validator checks the wider Schema.org structure. Use both when the search feature matters.

Validation catches missing commas, invalid types, and absent required fields. It cannot know that your price is old or the author name is wrong. Compare the extracted values with the visible page as part of the test.

Generator tools and Google’s Structured Data Markup Helper can produce a starting block, but I would move the final markup into the site’s templates. Hand-maintained copies drift quickly when a title, URL, image, or price changes.

Structured data works best as another output of reliable content data, not a separate SEO paragraph hidden in the source. Describe what is really on the page, validate it, and leave the decision to decorate the search result to the search engine.