How to choose a LinkedIn scraper
Most people who look for a LinkedIn scraper want rows: profiles for a list of people, a company’s posts, everyone who commented on a post, the jobs that match a search. This guide covers how to collect each one and what to compare.
By PrimaryAPIReviewed
What people collect from LinkedIn
The same collection jobs come up again and again, and each one maps to a single API.
- Profiles for a list of people: headline, location, current position, experience, education, skills and certifications.
- Company pages: description, industry, headcount, followers, headquarters, specialities and similar companies.
- Posts from a company page or a member, with text, date, reactions, comments and media.
- The people who commented on or reacted to a post, with name, headline and profile URL.
- Job listings that match a title and location, with filters such as workplace type, experience level and date posted.
Your own scraper or an API call
A LinkedIn scraper you run yourself usually means a headless browser, often a signed-in account, and code that reads each page. All of it is yours to host, schedule and fix when LinkedIn changes a page.
An API call is one request and one JSON response, with nothing on your side to maintain. You pay per call instead of paying for servers and upkeep.
What LinkedIn’s own APIs read
LinkedIn’s official APIs read through members who sign in to your app: their own profile basics, and the company Pages and ad accounts they manage. Comments and reactions are read on Pages the signed-in member administers.
Most programs need LinkedIn’s approval, and the Marketing APIs limit how long member data can be kept. The dated details are in the official API section.
Collecting long lists without overspending
Comments, reactions, posts and job searches come back a page at a time with a nextCursor. Pass it back as the cursor to read the next page, and stop when it is null.
Use the date filters on posts, comments and job searches so a daily run reads only what is new, and give each recurring job its own key with a credit limit.
Checks before you collect
Run every candidate through the same five questions.
- Does it return every record I need, such as profiles, posts, comments and jobs, and not only one of them?
- What will I have to run, log in to or update myself?
- How often are the fields I rely on filled for my own sample of URLs?
- Can I page through long lists and filter them by date?
- What happens, and what do I pay, when a profile is not public or a post was deleted?
Your first LinkedIn pull with PrimaryAPI
Sign up, copy your API key and call the API for each job: a profile URL for a profile, a company page URL for its details or posts, a post URL for its comments or reactions, a title and location for jobs. Each API’s page shows what to send, the fields you get back and an example.
Results come back as JSON, public data only, and you only pay when data comes back. If an AI agent will do the collecting, point it at the MCP server or SKILL.md; it uses the same key.