Schema Markup

Schema markup is structured data added to a page using the schema.org vocabulary, usually as a JSON-LD block in a script tag, that states in machine terms what the page is and what it contains: this is an organisation with this name and these profiles, this is a product with this identifier, this is an article with this author and this publication date. The everyday reason to add it is eligibility for rich results — the review stars, breadcrumbs, prices, event dates and expandable questions that appear alongside a listing. Eligibility is the operative word. Valid markup makes a rich result possible and never guarantees one; the engine decides per query and per page, and it periodically retires whole result types, so anything you gain this way should be treated as revocable rather than earned. There is one rule that decides whether markup helps or hurts. It has to describe what is actually on the page and visible to the person reading it. Marking up a review score the page does not display, an offer that has expired, or a question and answer that exists only in the JSON is the misuse that engines act against, and the penalty for it is losing rich-result eligibility across the site rather than on the one page. In practice the durable value is broader than search snippets. A structured statement of what a page asserts — the entity it is about, its author, when it was last updated — is legible to anything that parses the web mechanically, which increasingly includes retrieval systems collecting sources for an answer. It costs little on top of markup you were adding anyway. Two implementation notes save most of the maintenance. Generate the markup from the same data that renders the page, so it cannot drift into describing a version of the page that no longer exists, and validate it with each engine's own testing tool rather than a generic linter, because the vocabulary is permissive and the engines are the ones with the required-field opinions.