Organization schema for web extraction

An organization schema captures the facts that describe a company or institution: its legal or common name, industry, founding year, headquarters location, website, employee count, and a short description. Teams extract this to enrich a CRM, build company databases, run lead research, or map a market. Instead of reading an about page and copying details by hand, you describe the fields you want and get back a structured record you can store and join against your own data. Company information lives in many shapes, from Wikipedia infoboxes to corporate about pages to press footers, which is why describing fields beats writing per-site selectors. You define what an industry or a founding year is, and the model reads it from the rendered page. Anything the page does not state, like an employee count a private company keeps quiet, returns null rather than a guess. For firmographic data that honesty matters, because an invented headcount or founding year quietly corrupts the analysis you build on top of it.

Fields in a Organization schema

The fields teams most often extract for a organization, what type each one is, and whether it is usually present.

Field	Type	Required	Notes
name	string	yes	The organization's name as presented on the page. This is the anchor field for company records and for matching against existing data, so it is required rather than optional.
industry	string	optional	The sector the organization operates in, such as Software, Logistics, or Healthcare. Useful for segmenting a company database and filtering a market map by line of business.
foundingYear	number	optional	The year the organization was founded, parsed as a number so you can sort by age and compute company maturity. Returns null when the page does not state a founding year.
headquarters	string	optional	Where the organization is based, typically a city and country. Returned as written on the page so you keep the exact location detail for mapping and regional segmentation.
website	string	optional	The organization's official website URL. Acts as a stable key for deduplication and a starting point for deeper enrichment from the company's own pages.
employeeCount	number	optional	The reported number of employees, parsed as a number where a single figure is given. Returns null when only a vague range or nothing at all is published, rather than a guessed value.
foundedBy	string[]	optional	The founders credited on the page, returned as a list of names. Helpful for people-to-company mapping and for research that tracks founders across multiple ventures over time.
description	string	optional	A short summary of what the organization does, taken from the page. Gives you a human-readable blurb for cards and profiles without you having to write one per company.

The schema

Copy this and send it with any URL.

{
  "name": "string",
  "industry": "string",
  "foundingYear": "number",
  "headquarters": "string",
  "website": "string",
  "employeeCount": "number",
  "foundedBy": "string[]",
  "description": "string"
}

Example output

Validated JSON back, matching the schema.

{
  "name": "Brightgrove Analytics",
  "industry": "Software",
  "foundingYear": 2012,
  "headquarters": "Toronto, Canada",
  "website": "https://example.com",
  "employeeCount": 480,
  "foundedBy": [
    "Mara Ellison",
    "Theo Park"
  ],
  "description": "Brightgrove Analytics builds dashboards and data tooling for mid-market retailers."
}

Try this organization schema against a real page in the live demo, or read the docs to use it via the API.

Organization schema: FAQ

What if the page does not list an employee count or founding year?

Those fields return null rather than a guess. Because the output is validated against your schema before it is returned, any number you receive was actually present on the page. For firmographic data this is important: an invented headcount or founding year would skew company analysis, so a null clearly signals a gap you can choose to fill from another source.

Are Wikipedia company pages a good source?

Yes. Wikipedia is server-rendered, cooperative, and rich in structured facts in its infoboxes, which makes it a strong fit for organization extraction. Extracto renders the page in headless Chrome and reads those details against your schema. As always, respect Wikipedia's terms and attribution expectations when you reuse the content you collect.

Can I enrich a list of companies by running this schema over their about pages?

Yes. The same schema runs across many sites because it describes meaning rather than markup, so one definition handles varied about-page layouts. Every request runs through a real headless browser with a managed anti-bot bypass layer and proxies handled for you, so JavaScript-rendered and anti-bot-protected about pages render normally, with login-gated content the only exception.

How is the foundedBy list returned when there are several founders?

Defining foundedBy as a list of strings tells the model to return every founder it finds as a separate entry, rather than cramming them into one string. That keeps the data clean for people-to-company mapping. When a page credits no founders, the field returns an empty result or null, so you never get fabricated names attached to a company.

How can I trial this on a few companies first?

The free plan gives you 100 pages with no credit card, which is plenty to run this organization schema over a handful of Wikipedia or about pages and check that the firmographic fields land correctly. Paid plans start at 35 dollars per month once you scale up. Validating the output shape before paying is the recommended first step.