LinkedIn Jobs Scraper with Salary & Company Enrichment

LinkedIn Hiring Radar: Jobs, Pay & Employer Profiles

Turn public LinkedIn job pages into a research-ready dataset: one row per posting with the full description, a parsed pay range and where that pay came from, applicant volume, Easy Apply status, hiring badges, recruiter details and the employer's own LinkedIn profile. Search by keywords, places, companies or exact job links. No LinkedIn account, cookies or proxy setup is needed.

What this actor does

• Pay that can be compared. The pay box and pay ranges written in the description are read, normalised to yearly figures and tagged with salarySource so you can tell employer-provided pay from description text.
• Employer intelligence, not just a name. For every distinct employer, the actor reads the public LinkedIn company page once and attaches industry, headcount band, headquarters, website, founding year, follower count, specialties, offices, tagline, description and numeric company ID.
• Broad discovery. Combine several keyword searches with several places, search entire companies, paste LinkedIn search links, or feed a list of job links and IDs. Results are merged and de-duplicated by job ID.
• Filters that narrow the dataset. Date posted from the past hour to three months, search radius, Easy Apply, fewer than ten applicants, companies to include or leave out, experience level, job type, industry, staffing agencies, salary and recruiter presence.
• Past the 1,000-result wall. When one search reaches LinkedIn's result ceiling, the actor can continue with narrower posting-date windows and merge the unique jobs.
• Clean rows. Empty fields are omitted at every depth; a field the source does not show for a posting is never filled with blanks, zeroes or guesses.

Output per job

Field	What it holds
id, title, jobUrl	LinkedIn job ID, title and canonical job link
companyName, companyUrl, companyId, companyNumericId, companyLogo	Employer identity and logo
location	Location exactly as LinkedIn displays it
postedDate, postedDateText, postedAt, expiresAt	Posting date, relative label, exact UTC time when available, and LinkedIn's expiry time when its structured data includes one
salary, salaryRange, salarySource	Raw pay text, parsed {min, max, currency, interval, minYearly, maxYearly} and employer or description
applicantCount, applicantCountNumeric, applicantCountKind, applicantInsights	Applicant text, parsed number, whether it is exact/more/fewer, and early-applicant notes
listingBadge, activelyHiring	Search-card badge such as "Actively Hiring" or "Be an early applicant"
easyApply, applyUrl	Whether the job can be applied to on LinkedIn, plus the LinkedIn application page
descriptionText, descriptionHtml	Full posting description in both formats
responsibilities, qualifications, preferredQualifications	Bulleted sections when the description uses a recognisable heading and list
seniorityLevel, inferredSeniorityLevel, employmentType, industries, industriesList	Criteria as LinkedIn publishes them, plus a title-based seniority fallback
experienceYearsMin, experienceYearsMax	Experience range stated in the description
visaSponsorshipMentioned, visaSponsorshipPhrase	Positive or negative sponsorship statement, with the matched phrase
remoteAllowed	Set only on explicit fully-remote evidence (title, wording or a TELECOMMUTE structured-data flag); never inferred from hybrid/on-site text
recruiter	{name, title, profileUrl, photoUrl} when LinkedIn shows the poster card
contactEmails	Email addresses written in the description
atsPlatform	Applicant-tracking platform named or linked in the posting, when detectable
companyIndustry, companySize, companyHeadquarters, companyType, companyWebsite, companyFounded, companySpecialties, companyLocations, companyFollowersCount, companySlogan, companyDescription	Employer profile from the company's public page
companyAffiliatedPages, companySimilarPages	Related employer pages as listed on the company profile
companyRecentPosts	Recent public company posts with reaction and comment counts, when shown
similarJobs, peopleAlsoViewed, similarSearches, industryIds, titleId, titleCategoryLabel, titleCategoryUrl	Related postings and LinkedIn taxonomy hints (full job-page depth only)
searchQuery, searchLocation, searchUrl	The search that surfaced this job
scrapedAt	Collection timestamp in UTC

The dataset includes three ready-made views: Jobs, Salaries and Employers.

Input

At least one of searchQueries, searchUrls, jobUrls or companies must be provided. The form is grouped as "What to search", "Where", "Filters", "Data to include" and "Limits".

Field	Type	Default	Meaning
searchQueries	string list	["Python Developer"] in the prefill	Job titles, skills or phrases; each is searched in every place
searchUrls	URL list	-	LinkedIn job-search links; their keywords, place and filters are kept
jobUrls	string list	-	Job links or numeric job IDs to read directly
locations	string list	["United States"] in the prefill	Places in English; unknown places are reported and skipped, not silently treated as the US
geoIds	string list	-	Numeric LinkedIn place IDs for an exact market
distance	select	LinkedIn default	Radius around a city: 0, 5, 10, 25, 50 or 100 miles
timePosted	select	Any time	Past hour, 24 hours, 3 days, week, 2 weeks, month or 3 months
companies	string list	-	Employer names, company/university links or numeric company IDs
excludeCompanies	string list	-	Employers to skip by name or LinkedIn page
easyApply	boolean	false	Keep only jobs applicable on LinkedIn
under10Applicants	boolean	false	Keep only jobs with fewer than ten applicants
experienceLevel	select	Any	Internship, Entry level, Associate, Mid-Senior level, Director, Executive
jobType	select	Any	Full-time, Part-time, Contract, Temporary, Internship, Volunteer, Other
industry	multi-select	[]	LinkedIn industry of the employer (all 434 categories)
companySizes	multi-select	[]	Employer headcount bands from 1-10 to 10,001+
excludeStaffingAgencies	boolean	false	Remove staffing and recruiting firms
requireSalary	boolean	false	Keep only postings that state pay
minYearlySalary	integer	-	Drop postings whose yearly pay ceiling is below this amount
salaryCurrency	select	Any	Keep only this pay currency; also the currency used by minYearlySalary
requireRecruiter	boolean	false	Keep only postings where LinkedIn names the poster
includeTitleKeywords	string list	-	Title must contain one of these words
excludeTitleKeywords	string list	-	Title must not contain any of these words
excludeJobIds	string list	-	Job IDs seen in an earlier run
scrapeJobDetails	boolean	true	Read each job page for description, pay, criteria, applicants and recruiter
includeCompanyDetails	boolean	true	Read each employer's public profile once
enrichmentDepth	select	standard	full adds related jobs, related searches, taxonomy IDs and extra posting-time retries
language	select	en	Language of LinkedIn's labels: English, Spanish, French, German or Portuguese
splitLargeSearches	boolean	true	Continue a capped search with narrower posting-date windows
maxItems	integer	100	Stop after this many saved jobs (1-10,000)

Example: pay research across markets

{
    "searchQueries": ["Python Developer", "Data Engineer"],
    "locations": ["United States", "Germany", "United Kingdom"],
    "timePosted": "r604800",
    "requireSalary": true,
    "minYearlySalary": 80000,
    "salaryCurrency": "USD",
    "maxItems": 300
}

Example: every current role at two employers

{
    "companies": ["Microsoft", "https://www.linkedin.com/company/affirm"],
    "includeCompanyDetails": true,
    "maxItems": 200
}

Example: audit a list of postings

{
    "jobUrls": [
        "https://www.linkedin.com/jobs/view/4479196690",
        "4465604021"
    ]
}

Choosing the right depth

scrapeJobDetails off returns search-card data only: title, employer, place, posting date, card salary when shown and the job link. It is the cheapest and fastest mode.

scrapeJobDetails on reads the public job page for every posting and adds the full description, pay parsing, criteria, applicants, recruiter and apply signals. includeCompanyDetails then adds the employer profile once per company.

enrichmentDepth: full uses the richer job-page layout and also adds related jobs, related searches and LinkedIn taxonomy IDs. It costs more time per posting and is unnecessary for most salary or hiring research.

How many jobs can I collect?

LinkedIn returns at most about 1,000 jobs for one keyword/place/filter combination. The actor exposes three ways past that wall:

• locations or geoIds spread one keyword across several markets.
• searchQueries multiply the number of distinct searches.
• splitLargeSearches continues a capped search with past-month, past-week, past-3-days and past-24-hours windows, removing overlaps by job ID.

The actor stops on LinkedIn's own end-of-results signals, not on a timer: an empty page, repeated pages with no new job IDs, or the 1,000-result ceiling with no narrower window left.

Use cases

• Recruiting and sourcing - compare public postings, pay ranges and employer profiles for a role
• Compensation research - normalise salary data across markets and seniority levels
• Job-market monitoring - track hiring volume and employer demand over time
• Sales and account research - identify companies actively hiring in a target segment
• Competitive intelligence - compare how employers present roles, pay and benefits

FAQ

Do I need a LinkedIn account, cookies or a proxy? No. Every request reads a public page as a signed-out visitor. The actor chooses its own connection automatically and moves to a fresh session when LinkedIn throttles it; no user proxy configuration is required.

Why is salary missing on some rows? Many employers do not publish pay. The actor reads LinkedIn's pay box and explicit pay ranges in the description and never estimates. salarySource tells you which source produced the range. Coverage varies by market and employer.

Can I filter for remote, hybrid or on-site jobs? LinkedIn does not show the workplace label to signed-out visitors, and the public search ignores its workplace parameter. remoteAllowed is therefore only set when the posting itself is explicit about being fully remote (title/wording, or a TELECOMMUTE structured-data flag). Hybrid and on-site are never guessed. Adding "remote" to your keywords is the practical workaround. Older saved inputs that still contain the old workplace filter keep running; that setting is ignored and the status message says so.

Can I get the employer's own apply link? LinkedIn exposes the external apply destination only to signed-in members. applyUrl is the LinkedIn application page and easyApply tells you whether the application can be completed on LinkedIn. atsPlatform is filled only when the posting text itself names or links an applicant-tracking system.

Why does postedAt sometimes differ from postedDate? postedDate is the date shown on the search card. postedAt is the exact UTC timestamp from the job page's structured data or posting-time metadata, and it is omitted when LinkedIn serves a page variant without it. postedDate is always present.

Do all links open? Job links and company logos were checked from a clean client: job pages return 200 and logos return 200 with an image content type. LinkedIn company pages and some employer websites block automated HEAD/script requests with 403 even though they open normally in a browser; those URLs are the real public pages, not broken builders. An employer's own site can also be down or geo-restricted independently of this actor.

Why do I see more jobs than maxItems? You should not: the actor reserves each output slot atomically and never pushes more than maxItems. If a search has fewer matching jobs, the run ends with fewer and explains why in the status message.

How do I get only new postings on a schedule? Use "Past 24 hours", or save the id values from one run and send them through excludeJobIds on the next.

What is intentionally not available? Remote/hybrid/on-site labels for non-remote jobs, the external apply URL, repost history, view counts, job state and the job-function criterion are gated to signed-in LinkedIn members or have been removed from public pages. Skills and benefits are not exposed as structured lists for signed-out visitors; their raw text remains in descriptionText. The actor reports these limits rather than inventing values.

Is this affiliated with LinkedIn? No. It is an independent tool that reads LinkedIn's public job pages.

Data source

Public LinkedIn job-search result cards, public job pages (linkedin.com/jobs/view/...), public company profiles (linkedin.com/company/...) and LinkedIn's public location/company suggestion endpoints. All requests are cookieless.