Valid structured data

What we check

We read all structured data the home page delivers: JSON-LD blocks, Microdata and RDFa. Then we check every item, whatever its type:

• Readability: Is every JSON-LD block valid JSON? A forgotten comma or one quotation mark too many makes the whole block useless.
• Vocabulary: Do the types and properties used exist in Schema.org, and do the properties fit the type? No machine understands Organisation instead of Organization or telephon instead of telephone.
• Required details: For the enhanced display in search results (stars, prices, dates, breadcrumbs) Google requires certain details per type. We check them for LocalBusiness, Product, Offer, Review, AggregateRating, Event, Recipe, BreadcrumbList and FAQPage.
• Values: empty details, web addresses without "https://" and dates that are not a date in the format 2026-10-27.

Unreadable JSON and missing required details are warnings, the rest notices. If the home page delivers no structured data, there is nothing to check; the check "Structured company data" reports that.

Why it matters

Structured data is written for machines, and machines forgive little. A person sees that "www.muster.ch" is a web address. A program reads it as a path on the same site. A block with a syntax error is not read in part, it is discarded as a whole, including name, logo and contact details.

Google only shows enhanced search results when the required details are complete. AI assistants use the same data to place the website and its content. Faulty items are therefore often worse than none: they need upkeep and bring nothing.

How to fix it

The evidence of every finding names the item, the path to the property (for example $.@graph[2].offers.price) and the faulty value. Correct the detail where it is created:

• Plugin or website builder: in the settings of the SEO or schema plugin (Yoast SEO, Rank Math, Schema & Structured Data for WP) fill in missing fields such as address or logo and enter web addresses with "https://".
• Custom JSON-LD: check the block in a JSON validator, correct typos in type and property names (upper and lower case count) and write dates in the format 2026-10-27 or 2026-10-27T19:30+01:00.

A complete product with a price looks like this:

{
  "@context": "https://schema.org",
  "@type": "Product",
  "name": "Espresso machine",
  "offers": {
    "@type": "Offer",
    "price": "290.00",
    "priceCurrency": "EUR"
  }
}

How to check it yourself

The Schema Markup Validator (validator.schema.org) checks the vocabulary, Google's "Rich Results Test" the required details. Both show every occurrence.