Structured data is machine-readable annotation embedded in web pages that describes the page's content semantically — what is a product, who is the author, when was it published, what is it about.
Three formats are supported by search engines:
- JSON-LD (the recommended format): a JSON block in a
<script type="application/ld+json">tag. - Microdata: HTML attributes (
itemtype,itemprop) directly on display elements. - RDFa: similar to Microdata, used in academic and government contexts.
The vocabulary is primarily defined by schema.org. Search engines (Google, Bing, Yandex) and AI engines (Perplexity, ChatGPT, Claude) parse structured data to understand page content beyond plain text. Rich snippets, knowledge panels, AI Overview citations, and merchant feed-style information all depend on structured data being correctly emitted and validated.
For ecommerce, the structured data inventory typically includes: Organization, Product (with 2026 fields), BreadcrumbList, FAQPage, HowTo (for guides), Article (for blog posts), CollectionPage (for PLPs), Review and AggregateRating, Person (for team members), DefinedTerm (for glossaries).