A JSON-based format for encoding Linked Data, enabling semantic annotations in web pages and APIs.
JSON-LD (JavaScript Object Notation for Linked Data) is a W3C standard for encoding Linked Data using the familiar JSON format. It allows developers to add semantic meaning to JSON data by embedding a @context that maps JSON keys to URIs in shared vocabularies. JSON-LD is the most widely deployed Semantic Web standard, used by Google, Microsoft, and millions of websites to provide machine-readable structured data.
JSON-LD is the primary mechanism by which websites communicate structured data to search engines and AI crawlers. Google uses JSON-LD to power rich search results — knowledge panels, FAQ boxes, recipe cards, and event listings. As LLM-powered search engines like Perplexity and ChatGPT increasingly crawl the web, JSON-LD structured data helps AI systems accurately understand and cite web content.
A JSON-LD document includes a @context (mapping terms to URIs), a @type (the class of the described resource), and properties with values. The context can reference standard vocabularies like Schema.org, Dublin Core, or custom ontologies. JSON-LD processors can convert JSON-LD to RDF triples for storage in triple stores, or expand it to fully qualified URIs for interoperability. The @graph keyword allows multiple resources to be described in a single document.
A recipe website embeds JSON-LD in each recipe page: {"@context": "https://schema.org", "@type": "Recipe", "name": "Chocolate Cake", "cookTime": "PT1H", "recipeYield": "12 servings"}. Google reads this JSON-LD and displays a rich recipe card in search results with cooking time, ratings, and ingredients — without needing to parse the human-readable HTML.