About Usage Statistics

What is this?

As the web grows, the way people use this language changes. To help everyone see these changes, Schema.org is publishing the Public Usage Statistics dataset via a collaboration with Google.

This dataset looks at a huge sample of the public web to show a big-picture view of how different Schema.org terms (like "Person", "Product", or "price") are used across millions of unique websites.

This data is designed to help:

• Website Owners, Developers and Toolmakers: See which terms are most popular so you can focus on using the ones that have the most impact, use official data to justify your work, improve website plugins (like WordPress SEO tools), and build better analysis tools.
• Community Groups and Researchers: Track how the Schema.org language is growing, see what's new, and decide when old terms are no longer needed.

This dataset is compiled from observations within Google's public web crawling infrastructure and contributed directly to the Schema.org community. By sharing this data openly on the canonical Schema.org GitHub repository, we want to make the semantic web more transparent and collaborative for everyone.

How the Data is Gathered and Cleaned

Google uses an internal pipeline to gather, clean, and share this data:

1. Gather: Schema.org term frequencies are measured within Google's public web crawling infrastructure. The data is aggregated at the domain level (like example.com), not by individual pages. This means if you use the same term on 100 pages of your site, it still only counts as one domain using it.
2. Group: Instead of showing exact, raw numbers (which change daily and can be noisy), websites are grouped into range "buckets" (like "10K - 100K" domains). This keeps the data more stable and protects website privacy.
3. Publish: An updated file is pushed to GitHub every month.

The Raw Data Files

The raw files are available on the GitHub site at Google Public Stats dataset on GitHub. They are available in JSON and CSV with the same data and in a JSON summary format with aggregated bucket distributions.

To make it easy for anyone (including independent developers and startups) to use this data, the files are very simple. The dataset has three main fields:

• Term Type: The type of term. This is either Type (like "Person" or "Event") or Property (like "price" or "telephone").
• URI: The official URI of the term (for example: http://schema.org/Person).
• Domain Count Bucket: The range of unique domains using the term (for example: 100K - 1M domains).

Frequently Asked Questions (FAQs)

Q: Why numerical ranges instead of exact numbers?

• Filtering Out Daily Noise: Web crawling isn't perfect. Temporary server issues, network slow-downs, or minor website updates can make exact counts jump up and down daily. Using ranges gives a much more stable, reliable "big picture" view.
• Protecting Website Privacy: Showing exact numbers could allow competitors or bad actors to track minor changes on specific websites or reverse-engineer how search engines crawl the web.

Q: Does the data distinguish between different formats like JSON-LD, Microdata, or RDFa?

A: No. They are all combined into a single stat. If a website uses JSON-LD on the homepage and Microdata on product pages, it still only counts as one website using those terms.

Q: Why is this updated monthly?

A: Web adoption trends change slowly, making monthly updates sufficient to track meaningful shifts. Furthermore, each release requires manual validation and approval to ensure data quality and filter out anomalies before publication.

Q: A term I want to use is in the "< 1K" bucket. Does that mean search engines ignore it?

A: Not necessarily. The "< 1K" bucket includes both brand-new terms and highly specialized terms (like medical or government terms). Because the total number of medical websites is small compared to the whole internet, these terms will naturally stay in the lower buckets. However, using them still builds deep authority for your specific niche.

Q: How representative are these statistics of the entire web?

A: No single crawl can capture the entire internet simultaneously, so all web-scale datasets inherit biases based on the crawler's scope and indexing methodology. These statistics reflect the web as indexed by Google. This also means that there is no data captured on robots.txt blocked websites, for instance.

Q: Why do you aggregate by domain and not URL or number of objects?

A: URL counts and number of objects are very heavily influenced by the index composition in search indices. There would need to be more sampling and complexity to present this in a more objective manner rather than be biased toward head websites. Domain count is slightly more universal in at least measuring wide-scale adoption.

Q: Can other search engines or web archives contribute their own statistics?

A: Yes. The data format used for these statistics is fully open and documented in Schema.org GitHub repository. Schema.org welcomes other search engines and web-scale crawlers to publish their own adoption statistics in this format so the community can compare datasets and build a multi-provider view.