Structured Data | Definition

Publisher note

All ten sites in this network are published by Bay Area AEO. They cover answer engine optimization in the Bay Area from ten angles and link to one another so readers can move between related topics.

The general sense. Any data arranged in labelled fields, as opposed to running text.
The web sense. Markup added to a page, normally in the schema.org vocabulary, that states what the page is about in a form software can parse.
The AEO sense. The fourth of the five AEO signals: schema markup that tells AI the services, service area and hours of a business.

What it looks like

A page about a dental practice may say in prose that it is open until six on weekdays and treats children. Structured data says the same thing in fields: a type (Dentist), a name, an address, an opening-hours specification, an area served. The commonest format is JSON-LD, a block of code inside a script tag that sits in the page's source and is invisible to readers. Google prefers JSON-LD, and the schema.org vocabulary it uses is shared across the major search engines.

Schema.org has specific types for many local businesses, including Dentist and Plumber, Restaurant and Attorney, and RealEstateAgent, all under the general LocalBusiness type. Choosing the narrowest type that fits tells a machine more than the general one.

Why AI systems use it

Prose has to be interpreted. Fields do not. When an assistant fans a customer's question out into narrower searches, some of those searches are factual lookups: is it open on Sunday, does it serve Concord, does it offer heat pump installation. Structured data answers lookups like these without guesswork. It also helps entity resolution, because the same organization identifier can tie a business's website to its profiles elsewhere.

Restaurants show the point clearly. Bay Area AEO's restaurant page describes turning PDF and image menus into crawlable pages with dish names and prices alongside dietary labels, plus Menu and Restaurant schema. A menu locked in an image is unreadable to most crawlers; the same menu in text with markup lets an assistant answer “vegan restaurant near UC Berkeley” from the source itself.

Table. Facts a local business commonly marks up, and the question each one answers.
Field	Customer question it serves
Business type and services	“Ductless mini-split installer in Oakland”
Area served	“Plumber in San Bruno for a water heater”
Opening hours	“Emergency dentist open Saturday in Hayward”
Menu and price range	“Japanese restaurant in downtown San Mateo”
FAQ entries	Any question the page already answers in prose

The matching rule

Structured data must describe what the page visibly says. Google's guidelines require it, and AI systems reward it for the same reason they reward consistent listings: agreement between sources is what makes a fact trustworthy. Hours in the markup that differ from hours on the page, or services listed in schema that the site never mentions, create exactly the kind of conflict that entity consistency work exists to remove. The correct order is to write the page clearly first, then mark up what it says.

Structured data in an AEO engagement

Bay Area AEO lists structured data among its service types and builds it into its Answer-Ready Content service, where service pages and FAQs come with structured data and answer real customer questions in plain language. A new website is usually unnecessary for this. The agency's stated approach is to work with a business's existing site and add what AI needs, which for most sites means a block of JSON-LD per page type and FAQ markup where the page already answers questions.

This dictionary practises what it defines. Every entry carries JSON-LD for the publisher, its founder and the page, and each headword, including this one, is also marked up as a DefinedTerm so a machine can read the definition as a definition.

Tying the business together

Structured data can also state who a business is, beyond what a single page offers. Organization markup gives a business one stable identifier and lists the profiles that belong to it through a property called sameAs, which points at the business's listings and social pages. That helps a machine see the website, the Google profile and the Yelp page as one entity. It is the markup counterpart to entity consistency: listings make the facts agree, and the identifier tells machines that the agreeing records describe the same business. This network uses the same device. All ten of its properties declare one organization identifier for Bay Area AEO, so the publisher resolves to one thing wherever a page is read.

Usage

“Structured data” and “schema markup” are used interchangeably in practice, and the network's glossary treats the second as a pointer to the first. Strictly, structured data is the general idea and schema markup is the vocabulary most often used to express it. JSON-LD names the format and says nothing about the content; a sentence such as “add JSON-LD” is incomplete until it says which facts.