Types of Schema in SEO: The Schema.org Types Worth Using (and Which to Skip)

The types of schema in SEO are the entity types defined by the Schema.org vocabulary, such as Organization, Article, Product, BreadcrumbList and FAQPage, that you add to a page as structured data. Schema.org lists hundreds of them. Most sites need a handful, chosen by what the page actually is.

Last reviewed: October 16, 2026.

Two things get mixed up under this heading, so separate them first:

• Schema types are what you describe: an organization, an article, a product.
• Schema formats are how you write it: JSON-LD, Microdata, or RDFa.

Neither one raises rankings by itself. Structured data gives machines an explicit description of the page. Whether a search engine or AI system uses it, and how, is that system’s decision.

What are the main types of schema in SEO?

Group them by the job they do. This is easier to remember than an alphabetical list, and it tells you which types belong on which page.

Identity types: who is behind the site

Type	Describes	Typical page
Organization	A company, brand, or publisher: name, logo, URL, social profiles	Home or about page
WebSite	The site as a whole: name and URL	Home page
Person	A named individual, such as an author	Author or team page
LocalBusiness	A business with a physical location, hours, and address	Location or contact page

These are the entity types. They tell a machine that a name on your site refers to a specific company or person, which is the groundwork for entity optimization. Put Organization once, on the page that is canonical for the brand, not repeated with different details on every route.

Content types: what the page is

Type	Describes	Typical page
Article (and BlogPosting, NewsArticle)	An editorial piece: headline, author, dates, image	Blog post
FAQPage	A page of questions, each with one authoritative answer	FAQ or support page
HowTo	Step-by-step instructions	Tutorial
VideoObject	A video: title, description, thumbnail, upload date	Page with a primary video
ProfilePage	A page whose main subject is one person or organization	Author profile

FAQPage and HowTo are the cautionary examples. Google deprecated FAQ rich results in May 2026 and had already scaled back HowTo rich results years earlier. The types are still valid Schema.org vocabulary; only Google’s display feature changed. What is FAQ schema? walks through what the markup still does.

Commerce and software types: what you sell

Type	Describes	Typical page
Product	A physical or digital product: name, image, brand, SKU	Product page
Offer	A price and availability, nested inside a product	Product page
AggregateRating and Review	Ratings and reviews of a product or service	Product or review page
SoftwareApplication	An app or software tool: category, operating system, offers	Software product page
Service	A service provided by a business	Service page

The rule that matters most here: ratings and prices must match what visitors can see. Marking up reviews that are not on the page, or a price that is out of date, is the quickest way to make structured data a liability.

Navigation and event types: how the site is organized

Type	Describes	Typical page
BreadcrumbList	The page’s position in the site hierarchy	Any page below the top level
ItemList	An ordered list of items	Category or listicle page
Event	A dated event with a location	Event page
JobPosting	An open role	Job page
Recipe	A recipe with ingredients and steps	Recipe page

BreadcrumbList is the safest one to roll out site-wide because it can be generated from your URL structure. The rest only apply when the page really is that thing.

What are the three schema formats?

You can express any of these types in three syntaxes:

1. JSON-LD: a <script type="application/ld+json"> block separate from the visible HTML. This is the format Google recommends and the easiest to generate from code.
2. Microdata: attributes added inline to your HTML elements.
3. RDFa: another attribute-based inline syntax, less common on new sites.

For a developer-run site, JSON-LD is almost always the right choice, because it lives in one place, can be built from the same data as the page, and doesn’t tangle with your markup. A minimal Article looks like this:

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "Types of Schema in SEO",
  "author": { "@type": "Organization", "name": "SEOAgent" },
  "datePublished": "2026-10-16"
}
</script>

Which schema types has Google stopped showing?

Google has retired several structured data features over the years, and the list keeps changing, so check the current state rather than relying on an old guide. The ones to know:

• FAQ rich results: deprecated in May 2026.
• HowTo rich results: reduced in 2023 and no longer a dependable feature.
• Sitelinks search box: removed in 2024.

Each of these is a Google feature going away, not the Schema.org type becoming invalid. Google’s structured data documentation lists the types that currently qualify for a search appearance, and the Schema.org full hierarchy lists every type that exists. They are different lists, and the difference is where most schema confusion comes from.

How do you choose which schema types to add?

Start from the page, not from the list.

1. Name what the page is. A blog post is an Article. A pricing page for software is a SoftwareApplication with Offer data. A team page is Person entries.
2. Add the site-wide basics once. Organization and WebSite on the home page, BreadcrumbList on inner pages.
3. Add a content or commerce type only if the visible page supports it. Every property you mark up should be readable by a visitor.
4. Skip types that need content you don’t have. No reviews, no Review markup. No steps, no HowTo.
5. Prefer fewer, accurate types over many partial ones. One correct Article beats five half-filled blocks.

How do you check which types a site already uses?

Search the repository for application/ld+json, @type, and itemtype, then fetch the deployed pages and read the raw HTML, because markup injected only by client-side JavaScript may not be seen by every crawler. Validate with the Schema Markup Validator, and compare each block against the visible page. The validator catches syntax mistakes; it cannot tell you that a price is stale.

If you maintain many pages, don’t hand-edit each block. Derive the JSON-LD from the same data the page renders, as described in automating schema markup in your codebase. For the other tags that sit alongside schema in a page’s head, see what are SEO tags?.

Doing this with a coding agent

SEOAgent’s free Skill and seoagent CLI audit your live site from inside your coding agent, flag pages that are missing structured data or carry a type that doesn’t match the page, and write the proposed JSON-LD into your repository for you to review. It doesn’t publish to a CMS. In Claude Code you read the diff, check the markup against the visible page, and ship it in your normal pull request. The SEO optimization feature covers audits, metadata, and schema together.

The practical answer to “which types of schema should I use?” is short: the identity basics once, the type that matches each page, and nothing the page can’t back up.