How to export LinkedIn company pages to JSON

LinkedIn company pages are one of the richest sources of B2B company data: company size, industry, specialties, headquarters, founding year, follower count, and recent posts. Extracting this into structured JSON lets you build prospect lists, enrich a CRM, or track competitor activity programmatically.

What's on a LinkedIn company page

• Company name and LinkedIn URL
• Industry and company type (public, private, non-profit, educational)
• Headcount — LinkedIn's employee count, which is typically inflated vs. actual
• Headquarters — city and country
• Founding year
• Specialties — self-reported keyword tags
• Tagline and 'About' description
• Follower count
• Website URL
• Recent posts and engagement (on public pages)

Method 1: LinkedIn's official API

LinkedIn's Marketing Developer Platform gives access to company profile data via the Organizations API. The official path requires applying for API access (which LinkedIn reviews and often rejects for non-marketing use cases), OAuth 2.0 authentication with a company page admin, and then GET /v2/organizations/{id}.

Practically speaking: LinkedIn's API program has extremely restricted access. Unless you're building a LinkedIn-approved marketing tool, the application will be rejected or left pending indefinitely. Most developers who need company page data don't use the official API.

Method 2: The unofficial API (for controlled use)

Several projects reverse-engineered LinkedIn's internal API. linkedin-api (Python, MIT license) wraps the unofficial endpoints and can fetch company data by LinkedIn URL or company ID. This violates LinkedIn's ToS and may result in account banning — use only on accounts you're willing to risk, and with significant rate limiting.

• pip install linkedin-api
• from linkedin_api import Linkedin
• api = Linkedin('your@email.com', 'password')
• company = api.get_company('openai') # slug from the URL
• print(company['name'], company['staffCount'], company['specialities'])

linkedin-api is a liability for production use — LinkedIn rate-limits aggressively and accounts used this way get restricted. Acceptable for research or one-time data pulls; not suitable for an automated pipeline.

Method 3: Scraping public pages

LinkedIn blocks unauthenticated requests with a login redirect after a few page loads. Tools like Playwright or Selenium with a logged-in session can access public company pages, but LinkedIn detects and blocks automation rapidly. Proxies, delays, and browser fingerprint spoofing are an arms race. Consider this path only if you have a strong need and the resources to maintain it.

Method 4: Paste the page text or PDF export

For a small number of company pages — a prospect list of 20-50 companies — the most practical approach is manual: save each page as PDF (browser Print → Save as PDF), or copy the visible text from the page, and run AI extraction to structure the fields. This sidesteps authentication and rate limits entirely.

What JSON fields to expect

A well-extracted LinkedIn company page produces:

• company_name, linkedin_url, website
• industry (e.g. 'Software Development'), company_type ('Privately Held')
• employee_count or headcount_range ('201-500 employees')
• headquarters (city, country)
• founded (year)
• specialties (array of string tags)
• about (the company description text)
• follower_count (if visible)

Enriching a prospect list

The common workflow: start with a list of company names or domains, find their LinkedIn URLs (by searching LinkedIn or using a prospecting tool), extract each company page into JSON, and merge the structured fields back into your CRM or outreach tool. AI extraction is the right tool for the extraction step — it handles the field variation across company types and sizes without per-company configuration.