Schema types that still earn results

This section covers the types of schema markup that still do visible work in Google Search as of November 2026, and the ones AI answer engines lean on when they read a page. A schema type is a label from the schema.org vocabulary, such as Review, Product or LocalBusiness, that tells a machine what kind of thing a page describes. Each guide takes one type, lists the properties Google documents for it, and shows a JSON-LD example with example values.

Which types does Google still show?

Google keeps a public list of the features it supports in its structured data search gallery. Review snippets, product results, local business details, software apps and carousels are all still on it. Some once-common features have been narrowed or dropped: Google limited FAQ rich results and retired HowTo results in 2023, which our guide to FAQ and HowTo schema after the cuts covers.

Types that never had a star or a carousel still matter. Organization and Person markup describe who stands behind a page, and the handbook covers them in Organization schema and entity markup and author and Person schema. This section stays with the types that describe the thing a page is about.

Valid markup makes a page eligible for a feature; it never guarantees one. Google's general structured data guidelines say so plainly, and every guide here repeats the point where it matters.

Reviews and ratings

Start with review schema and review snippets if you publish reviews written by an editor or a single reviewer. It explains the difference between Review and AggregateRating, the item types Google shows stars for, and the 2019 rule against self-serving reviews.

If your ratings come from many visitors instead, read AggregateRating schema for user reviews. It deals with ratingValue and the two counts, the ban on importing ratings from other sites, and the 2024 FTC rule on fake reviews.

Business and products

Local business schema describes a physical place: its address, coordinates and opening hours, and the most specific subtype to use. It also explains why a business cannot mark up reviews of itself for stars, and how the markup relates to a Business Profile.

Product schema and AI shopping answers covers Offer, price and availability, the split between product snippets and merchant listings, and how product data reaches shopping features. SoftwareApplication schema is the smaller relative for apps, plugins and web tools: a name, an offer and a rating, plus category and operating system.

Lists and tables

ItemList schema for ranked lists and tables explains how ListItem and position describe an ordered list, why Google limits host carousels to courses, movies, recipes and restaurants, and why a plain HTML table usually matters more than the markup around it.

Which types do AI engines lean on?

Google's guidance on AI features says no special schema.org markup is needed to appear in AI Overviews or AI Mode, and asks that structured data match the visible text. The types that describe concrete things, such as products, places, prices and ratings, are the ones that carry facts an answer can reuse, as long as the page shows the same facts.

The AI answers section covers how engines fetch and read pages. The guides here note, type by type, what Google documents about AI features as of November 2026 and where the documentation is silent.

How do you pick the right type?

Pick the type that names what the page is mostly about, not the one with the most attractive result. A recipe page is a Recipe even if it has reviews; the reviews nest inside it. A page about one product uses Product; a page listing ten products is a list. When two types fit, use the more specific one, which is also what Google's guidelines recommend.

It also helps to decide what result you are hoping for before you write any markup. If the answer is stars, check the review guides first, because the item type and who wrote the reviews settle most cases. If the answer is shopping features, start with products.

Then check the required properties in the guide, write the visible content first, and add markup that repeats it. Test the result with the tools in testing structured data the right way. If the idea of markup is new to you, our overview of structured data explains what it is and why machines read it.