LinkedIn job scraping: how to collect job postings you can rely on
LinkedIn job scraping done right: set clear criteria, read job postings at their source, keep open roles only and put a link and date on every row.
Free app. Runs on your Mac with the Claude or Codex plan you already have.
What LinkedIn job scraping is
LinkedIn job scraping is the work of collecting job postings for a role and a place into a table: title, employer, location, contract type, date, salary when shown and a link to the posting. Sales teams use it to find companies that are hiring, since a new team often means new purchases. Recruiters and founders use it to see who competes for the same people and what pay is advertised.
The method below is how a market researcher does it by hand. Each step says what to do and why, then how 10folds carries it out. The 10folds job reads employers' own public job boards and pages you supply, and it does not log in to LinkedIn, so its limits are stated in each step.
What it does for the business
- Find companies at the moment they growA company hiring several people in one function is building something and will need tools, services and suppliers. Job postings point sales teams to those accounts early.
- Benchmark the pay you offerSalary ranges stated in current postings show what other employers advertise for the same role in the same city, which helps set an offer that gets accepted.
- See where competitors investThe roles a competitor opens, and where, show which products, markets or regions it is betting on, often months before any announcement.
- Track a labor market over timeRepeating the same collection each month shows whether demand for a role is rising or falling in your area, as long as the scope stays the same.
LinkedIn job scraping, step by step
Write the criteria and the question
A job title is a family of titles. Search for the family.
What to do
- Name the role and its common variants, for example data engineer, analytics engineer and data platform engineer.
- Name the place, and decide whether remote roles count.
- Set the period, such as posted in the last seven days, and an as-of date.
- List any employers you want to follow by name.
- Write the question: who is hiring, what is paid, or how demand changes.
Why
Employers title the same job in different ways, so a single exact title misses much of the market. The question decides the fields: finding hiring companies needs employer and date, a salary benchmark needs stated pay and contract type. An as-of date makes the result reproducible and stops old postings from being read as current.
Choose sources that allow it
The best copy of a job posting is the one the employer wrote.
What to do
- Read LinkedIn's terms of service. They prohibit automated collection without its permission.
- Go to the source: most postings come from an employer's applicant tracking system, such as Greenhouse or Lever, which often publishes a public job board.
- Use official public feeds and JobPosting structured data where the career page provides them.
- Check robots.txt, keep to a gentle rate limit and stop at any sign-in page.
Why
An applicant tracking system is the software employers use to publish roles and manage applicants. Its public board is the original copy of the posting, kept current by the employer and often offered as a feed for exactly this use. LinkedIn and other job sites mostly repeat these postings, sometimes late, sometimes after the role has closed. Any LinkedIn jobs scraper that logs in or rotates accounts to get past limits breaks the site's rules and puts accounts at risk.
Record each posting with its link and date
A job posting without a date is a rumor about a vacancy.
What to do
- Record title, employer, location, contract type, date posted, closing date and status.
- Record salary only when the posting states it, with currency, period and whether it is gross.
- Keep the posting's address and the date you saw it on every row.
- Leave out recruiters' names and other personal details unless you have a clear reason.
Why
Postings change and disappear, so the observation date is what makes a row checkable. Salary figures mean nothing without their currency and period, and an annual range cannot be compared with an hourly rate. Job postings rarely contain personal data, but recruiter names and contact details do count as personal data under GDPR.
Keep open roles and remove duplicates
The same job on four boards is one job.
What to do
- Merge the same posting found on several sites, using the employer's own posting as the reference.
- Drop postings marked closed or past their closing date.
- Keep postings with no closing date, marked as unknown status.
- Filter to the criteria you wrote, and record what was removed.
Why
Job sites repost the same role, recruiters advertise it again, and old postings linger long after the role is filled. Counting every copy inflates demand. Treating a missing closing date as open inflates it again, so the honest label is unknown.
Read the signals and hand over the table
Count what you saw, and say what you could not see.
What to do
- Count postings per employer to find the companies hiring most.
- Compare salaries only within the same currency, period and contract type.
- State the scope and the observation date at the top.
- Save the snapshot so the next run can show what is new.
- Name the next action, such as researching the top hiring companies.
Why
A table of postings becomes useful once it answers the question you wrote: who is hiring, or what is paid. Coverage is never complete, so a company that does not appear may simply use a board you did not read. Change over time is only meaningful between two runs with the same scope, sources and criteria.
Common mistakes
- Searching one exact titleEmployers name the same role many ways.
- Counting reposts as new demandDeduplicate across boards before counting.
- Treating every listed job as openCheck the status and closing date, and mark the rest unknown.
- Averaging salaries of different kindsHourly, monthly and annual pay, or different currencies, cannot be mixed.
- Logging in to scrape moreIt breaks the site's terms and risks the account.
What a good result looks like
- A one-line scope: role family, place, period and as-of date.
- One row per open posting, with employer, location, contract type and closing date.
- Salary only where the posting states it, with currency and period.
- A source link and observation date on every row.
- A list of sources that could not be read.
And then what? Using the result
- Research the companies with the most open roles and add them to your prospect list.
- Use the stated salary ranges to check your own offer.
- Share competitors' new roles with product and sales leaders.
- Run the same scope next month and compare with the snapshot.
When to run it
- Building a sales prospect listCompanies hiring in the function you sell to are growing and likely to buy.
- Setting pay for a new roleStated ranges in current postings show what the local market advertises.
- Watching competitorsTheir open roles show which teams and regions they are building.
- Tracking demand for a skillMonthly runs on the same scope show how demand moves in your area.
Questions people ask
Is it legal to scrape LinkedIn jobs?
LinkedIn's terms of service prohibit automated collection without its permission, and court decisions on scraping differ by country and by facts. Job postings rarely contain personal data, but the terms still apply. Reading employers' own public job boards is the safer route, and it is the route the 10folds job takes.
Can you scrape LinkedIn without logging in?
Some LinkedIn job pages can be viewed without an account, but the site limits and blocks automated visitors, and its terms still apply. The 10folds job never logs in. It reads employers' public boards on Greenhouse and Lever and up to 25 public pages you supply.
What data can you get from a job posting?
Usually the title, employer, location, contract type, date posted, closing date and description, and the salary when the employer states it. Many career pages publish these as JobPosting structured data, which is designed to be read by machines.
How do you find companies that are hiring?
Collect current postings for the roles that signal a need for what you sell, deduplicate them and count open roles per employer. The companies with several new roles in one function are the strongest leads.
Does the LinkedIn job scraping job read LinkedIn itself?
No. It does not search LinkedIn. It reads employers' public job boards and the public pages you supply, and its standard request caps a run at 100 records. Figures in the example screens on this site are illustrative.
Does it apply to jobs or contact recruiters?
No. It writes a table and a report in your project folder. It does not apply, message or write to any site.
More jobs in Scraping
Scrape LinkedIn job posts, on your own Mac.Free app. Runs on your Mac with the Claude or Codex plan you already have. Your files stay on your computer.
Download for Mac