Schema.org
Schema.org is a shared structured data vocabulary that site owners use to describe a website's content in code for search engines and other programs. The project was launched on 2 June 2011 by Google, Microsoft and Yahoo!, and Yandex joined in November 2011. The vocabulary defines object types, such as Product, Organization or Event, and their properties. Markup is most often written in the JSON-LD format, which Google recommends. Correct structured data makes a page eligible for rich results in search.
What is Schema.org?
Schema.org is a vocabulary, that is a set of agreed names of types and properties for describing things, people, places and actions. It is neither a data format nor a tool. According to the project's website, the vocabulary can be used with various encodings, including RDFa, Microdata and JSON-LD.
| Feature | Description |
|---|---|
| Type | a structured data vocabulary |
| Founders | Google, Microsoft, Yahoo! and Yandex |
| Project launch | 2 June 2011 |
| Governance | a steering group and the Schema.org Community Group at W3C |
| Version | 30.1 (November 2026) |
| Vocabulary size | 826 types and 1,540 properties (November 2026) |
| Formats | JSON-LD, Microdata, RDFa |
| Validation | Schema Markup Validator, Rich Results Test |
According to the Schema.org website, in 2024 over 45 million domains used markup from the vocabulary. It described over 450 billion objects. The vocabulary is used by, among others, Google, Microsoft, Pinterest and Yandex.
The terms “Schema.org” and “structured data” are not the same. Structured data is the general name for an organised description of a page. Schema.org is the most popular vocabulary in which it is written. The correctness of structured data is usually checked during an SEO audit.
How did Schema.org come about?
Schema.org was announced on 2 June 2011 by Google, Bing and Yahoo! as a shared collection of schemas for markup on web pages. Before that, each search engine expected data in a slightly different form. Google wrote in the announcement that Schema.org was meant to be a single place with information for webmasters.
| Year | Event |
|---|---|
| 2011 | project launch (2 June), Microdata is the first format |
| 2011 | Yandex joins (November) |
| 2012 | Martin Hepp's GoodRelations e-commerce vocabulary is integrated |
| 2015 | the Schema.org Community Group at W3C becomes the main project forum (April) |
| 2020 | Google moves its structured data testing tool to the validator.schema.org domain |
The project was initiated by R.V. Guha of Google, who chairs the steering group. Day-to-day work is led by Dan Brickley. The steering group approves new releases, while the Community Group prepares changes and discusses them publicly on GitHub.
Schema.org is used not only by search engines. The project's website describes its mission as creating schemas for structured data on web pages, in email messages and beyond.

How does Schema.org markup work?
Schema.org markup assigns a type and properties from the vocabulary to a part of a page. The type says what the described object is, e.g. a product, a company or an event. The properties give its features, e.g. the name, price, address or date.
Types form a hierarchy in which a more general type includes more specific ones. For example, LocalBusiness is a kind of Organization. A specific type inherits the properties of the more general type.
The same information can be written in three formats. Google accepts all of them if the markup is correct.
| Format | How the data is written | Notes |
|---|---|---|
| JSON-LD | a separate <script type="application/ld+json"> block in the page code | the format Google recommends as the easiest to implement and maintain |
| Microdata | itemscope, itemtype and itemprop attributes added to HTML tags | the first format Schema.org launched with in 2011 |
| RDFa | vocab, typeof and property attributes added to HTML tags | an HTML extension also used outside Schema.org |
An example of a simple company description in JSON-LD:
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Organization",
"name": "Example Company",
"url": "https://www.example.com",
"logo": "https://www.example.com/logo.png"
}
</script>JSON-LD does not mix with the HTML content, which makes it easier to add and change, e.g. via GTM (Google Tag Manager) or a plugin in a CMS.
How does Google use Schema.org?
Google uses structured data to understand a page's content better and to show it in search results in a richer form. Such results are rich results, e.g. with ratings, a price or an event date. Google supports only some of the types in the vocabulary, as described in the Search Central documentation.
Google's documentation cites results of structured data implementations at publishers.
| Site | Result according to Google |
|---|---|
| Rotten Tomatoes | 25% higher CTR on 100,000 pages with structured data |
| Food Network | 35% more traffic after enabling search features on 80% of pages |
| Nestlé | 82% higher CTR on pages shown as rich results |
The range of supported types changes. Google has limited rich results several times, although the types themselves still exist in the Schema.org vocabulary.
| Date | Change at Google |
|---|---|
| 8 August 2023 | FAQ results only for well-known government and health sites, HowTo results only on desktop |
| September 2023 | HowTo results stop showing, Google removes their documentation |
| 12 June 2025 | announced phase-out of, among others, book actions, course info, estimated salary, ClaimReview, learning video, special announcement and vehicle listing |
| 7 May 2026 | FAQ results stop showing in Google Search |
According to Google, unused structured data does not harm a page, but it has no visible effect either. The status of markup supported by Google is shown in the enhancement reports in Google Search Console.
Does Schema.org help in AI search engines?
Google and Microsoft say different things about this. According to Google, structured data is not required for generative AI features in search. Microsoft states that Schema.org markup helps its language models understand content.
In its guide to optimising for generative AI, Google writes that there is no special Schema.org markup for AI Overviews and AI Mode. It still recommends using structured data for SEO, because it helps pages qualify for rich results. In its guidelines for AI features, Google stresses that structured data must match the visible text of the page.
Fabrice Canel of Microsoft said in March 2025 at the SMX Munich conference that Schema.org markup helps Microsoft's language models understand content. This applies, among others, to Copilot answers, which draw on the Bing search index.
| Search engine | Position on Schema.org in AI |
|---|---|
| Google (AI Overviews, AI Mode) | structured data not required, no special markup for AI |
| Microsoft (Bing, Copilot) | markup helps language models understand content |
Structured data does not replace readable content. The structure of the text itself also matters, e.g. chunking content.
Sources
- The history, organisation and size of the vocabulary are described on the Schema.org pages: the home page, “About” and “Schemas”.
- The project launch is given in Google's announcement of 2 June 2011.
- Formats, implementation examples and tools are described in the Google Search Central documentation.
- Changes to rich results are reported on the Search Central blog and in the documentation changelog.
- Microsoft's position is cited by Search Engine Roundtable and Search Engine Land from March 2025.
Frequently asked questions
Q.Is Schema.org markup needed for a site to appear in Google?
No. A page can be indexed and visible in results without structured data. According to Google, it is not required for AI Overviews and AI Mode either. Google still recommends using it for SEO. Schema.org markup helps a page qualify for rich results, e.g. with a price or a rating.
Q.Which Schema.org format should you choose: JSON-LD, Microdata or RDFa?
In most cases JSON-LD. Google recommends the format that is easiest to implement and maintain, and for most sites that is JSON-LD. The data then goes into a separate script block and does not mix with the HTML code. All three formats are equally good for Google if the markup is correct.
Q.Do you need to remove FAQPage markup after Google's changes?
No. FAQ results stopped showing in Google on 7 May 2026, but the FAQPage type still exists in the Schema.org vocabulary. Google wrote back in 2023 that unused structured data does not cause problems in search. It has no visible effect in Google results, though.
Q.How do you check whether Schema.org markup is correct?
General correctness is checked by the Schema Markup Validator at validator.schema.org, without Google-specific rules. The Rich Results Test shows whether a page qualifies for specific Google features. After implementation, the status of markup across the whole site is visible in the enhancement reports in Google Search Console.
Let's talk growth
Send a brief or drop your contact. We'll reply within 24h.
Get a quote