KhaqanWhat is JSON-LD structured data?
JSON-LD structured data is a block of JSON, placed in a script tag with type application/ld+json, that describes a page's content using schema.org vocabulary (organisation, article, product, FAQ) so search engines and AI systems can read the facts without parsing the visible HTML.
JSON-LD stands for JavaScript Object Notation for Linked Data. It is a W3C standard, and it is the format Google recommends for structured data because it sits in one block, usually in the head, rather than being woven through the HTML as Microdata or RDFa attributes. The vocabulary almost everyone uses is schema.org, a shared set of types such as Organization, LocalBusiness, Article, Product, FAQPage and BreadcrumbList, each with named properties.
It matters because it removes guesswork. A crawler can work out from the HTML that a page is probably about a company that installs CCTV in a particular emirate, but the JSON-LD says so directly, with the legal name, address, phone number, opening hours and links to the same entity on other sites through the sameAs property. Search engines use it for rich results, and AI systems use it as a source of clean facts about who you are. For a business that wants to be described correctly by an assistant, entity schema is the cheapest signal available.
A minimal example is an Organization block with name, url, logo, telephone, address and sameAs, placed on every page and expanded on the home page. A blog post adds Article with headline, author, datePublished and dateModified. Several types can be grouped under a single @graph array so the page carries one script block rather than many.
The common mistake is invalid JSON, typically a trailing comma or an unescaped quote, which fails silently: the browser ignores it and so does every crawler. The second is describing content that is not on the page, which Google treats as a violation. Validate with the Schema Markup Validator after every template change; ai-visibility-audit parses the blocks, including nested @graph, and reports any that fail.
Related terms
- FAQPage schema: FAQPage schema is a schema.org structured data type that marks up a page's questions and answers as a list of Question items, each with an acceptedAnswer, so search engines and AI assistants can identify the exact answer text to a specific question.
- speakable schema: Speakable schema is a schema.org property (speakable, with a SpeakableSpecification) that points, using CSS selectors or XPath, to the sections of a page best suited to being read aloud by a voice assistant, typically the headline and a short summary.
- AI Overviews (Google): AI Overviews are the AI-generated summaries Google shows at the top of some search results, built by its Gemini models from pages in Google's index, with links to the sources it used; they launched in 2024 and now appear for many queries.
- AEO (answer engine optimisation): Answer engine optimisation (AEO) is the practice of structuring web content so that answer engines, meaning AI assistants, voice assistants and featured snippets, can extract a direct, accurate answer to a question and attribute it to your site.