β¦
", "description": "An exciting record label who are looking for a commercially minded Data Analyst to join their London team.\n\n This is a great opportunity for someone with 3-4 yeβ¦", "seniorityLevel": "Mid-Senior level", "employmentType": "Full-time", "industries": "Accounting", "salaryText": null, "salaryMin": null, "salaryMax": null, "salaryPeriod": null, "salaryCurrency": null, "applicantsText": "45 applicants", "jobPosterProfileUrl": "https://uk.linkedin.com/in/example-recruiter", "jobPosterName": "Alex Example", "jobPosterTitle": "Recruitment Director", "jobPosterPhoto": "https://media.licdn.com/dms/image/β¦/profile-displayphoto-scale_400_400/β¦", "applicantsCount": 45, "isDirectApply": true, "companyDescription": "Handle Recruitment has been supporting the creative industries since 1978, building our reputation in music before expanding across film, TV, entertainment, fasβ¦", "companyWebsite": "http://www.handle.co.uk", "companyIndustry": "Staffing and Recruiting", "companySize": "51-200 employees", "companyHeadquarters": "London", "companyType": "Privately Held", "companyFounded": "1978", "companySpecialties": ["Digital Recruitment", "Sales and Marketing Recruitment", "Office Support Recruitment", "Legal Recruitment", "β¦"], "companyId": "1234567", "companyEmployeesCount": 56, "companySlogan": "We've been supporting the UK's Creative Industries since 1978.", "companyFollowers": 295258, "companyLocations": [ "Ink Rooms, 25-37 Easton Street London, WC1X 0DS, GB" ], "scrapedAt": "2026-09-13T21:45:29.322Z" } ``` `postedAt` is a date, and `postedText` keeps LinkedIn's "6 days ago" wording beside it. The salary columns are filled on jobs with details where LinkedIn shows pay, and are null elsewhere. `closingAt` is always null: LinkedIn doesn't publish an expiry date on its public postings. ### How many jobs can you get, and how fast? | Run | Result | |---|---| | 100 jobs with details | 100 jobs in 18-26 seconds | | 6 searches, 1,100 jobs with details | 1,100 jobs in 2 minutes 14 seconds (about 490 a minute) | | 100 jobs with **Read the full job page** on | 100 jobs in 28-32 seconds | | 6 searches, 1,142 jobs with **Read the full job page** on | 1,142 jobs in 3 minutes 26 seconds (about 330 a minute) | These are whole runs on Apify, measured on 5 October 2026. At that pace an hour covers roughly 25,000 jobs with details, or about 19,000 with the full job page. LinkedIn serves at most about 1,000 results for any one search. A large run is therefore several keywords or several locations. Splitting a country into its cities is the usual way to go deeper. ### How much does it cost to scrape LinkedIn jobs? | Event | Free plan | Starter | Scale | Business | |---|---|---|---|---| | `actor-start`, per run | $0.002 | $0.002 | $0.002 | $0.002 | | `job-scanned`, per listing card fetched, before dedup or filters | $0.00005 | $0.00005 | $0.00005 | $0.00005 | | `job-detail`, per job returned with details | $0.0006 | $0.00054 | $0.00048 | $0.00034 | | `job-detail-enriched`, per job returned with details and company data, replacing `job-detail` for that job | $0.0006 | $0.00054 | $0.00048 | $0.00034 | | `job-listed`, per job returned without details | $0.0002 | $0.00018 | $0.00016 | $0.00014 | Enterprise plans pay the Business price. **$0.60 per 1,000 jobs** on the Free plan and $0.34 on Business, with company details included. Nothing is added on top: proxy usage and platform usage are in that price. Your bill follows the number of postings opened. A job removed by deduplication, by the only-new store or by a title or company filter costs the scan rate and no more. A posting that was opened is charged as a detail even if a later filter drops it. A posting that couldn't be read is returned listing-only and charged as `job-listed`. - Two keywords at 200 results each, 60 dropped by title filters, 340 jobs returned with details and company data: $0.002 + 400 Γ $0.00005 + 340 Γ $0.0006 = **$0.226** ($0.138 on Business). - A six-hourly monitor that scans 150 cards and finds 12 new jobs: $0.002 + 150 Γ $0.00005 + 12 Γ $0.0006 = **$0.017** ($0.014 on Business). A run of 100 jobs with details costs about $0.067. On Apify's Free plan, which comes with $5 of credits a month, that's about 74 runs, or roughly 7,400 jobs, before you pay anything. ### Limits: what LinkedIn doesn't show - **About 1,000 results per search.** Use "Locations (multiple)" to split a country into cities. - **No apply URL and no expiry date.** Neither is published on LinkedIn's public postings. - **No job function.** LinkedIn stopped showing it publicly (0 of 1,136 jobs in a check on 2 October 2026). Industries, seniority and employment type are returned. - **No workplace-type column.** LinkedIn doesn't expose on-site, remote or hybrid as a field. - **Experience level, job type and workplace type no longer filter the search on LinkedIn.** Here they're folded into the keywords. They steer the results strongly without being exact. Pair them with "Keep only these seniority levels" or "Keep only these employment types", which use LinkedIn's own labels on the posting. - **Pay only where the employer states it.** On 5 October 2026, 193 of 1,142 London jobs across six searches had a range (12 of 100 for "software engineer"). The rest have empty pay columns. - **Seniority is often "Not Applicable" on the short advert**: 580 of 1,100 jobs in the same check. With "Read the full job page" on, 125 of 1,142 read "Not Applicable". All other columns are identical either way. - **A filter never guesses.** "Minimum annual salary" keeps a job whose pay can't be compared with a yearly figure (an hourly rate, or no pay shown). The seniority, employment-type and description filters keep a job whose page doesn't show that field. - **Sorting orders what the run collected.** LinkedIn no longer sorts its public job search, so "Sort" can't change which jobs LinkedIn selects. ### FAQ #### Do I need proxies or a LinkedIn account? Neither. The actor reads LinkedIn's public job search only, and you never sign in. Proxy usage is included in the price and there's nothing for you to configure. #### How do I get only the new jobs on a schedule? Give "Only new jobs" a store name, then schedule the run in Apify. From the second run on you get the postings that weren't there before, and the ones already seen cost $0.00005 each to skip. Set "Posted within the last N hours" a little wider than your schedule interval. #### How do I get more than 1,000 jobs for one search? Split it. Each keyword and each location is its own search with its own ceiling, so "nurse" across twelve cities reaches far further than "nurse" in "United Kingdom". #### How accurate is the posting date? It's the date LinkedIn holds for the posting, not a guess worked back from "1 day ago". The label is kept in `postedText` if you want it. #### Why is there no apply URL or closing date? LinkedIn stopped publishing both on its public job pages. `closingAt` stays in the row as null so the columns line up with our other job scrapers, and `isDirectApply` tells you whether the job uses Easy Apply. #### How do I make sure every row really is part-time (or entry level)? Use the pair. Set "Job type" to `part-time` to steer the search, and "Keep only these employment types" to `Part-time` to check each posting. With both set, every returned row is part-time. The same pairing works for experience level and seniority. #### What does the company data include? Description, website, industry, size, employee count, headquarters (with street, postcode and country code in `companyAddressParts`), type, specialties, follower count and office addresses. Each company is fetched once and cached for 30 days by default. #### Is scraping LinkedIn jobs legal? The actor reads only the job postings LinkedIn serves to logged-out visitors. Those are business content, not personal profiles. Web scraping of publicly available data is generally legal, but you're responsible for complying with LinkedIn's terms and applicable law in how you use the data. ### Run it on a schedule or from code The setup we'd recommend is a monitor: an Apify Schedule every few hours with an only-new store named. A webhook can fire at the end of each run, and the new postings can go to Slack, Google Sheets, Zapier, Make or Airbyte through Apify's integrations. From code, `POST https://api.apify.com/v2/acts/boncross~linkedin-jobs-scraper/run-sync-get-dataset-items` with the input JSON as the body returns the jobs in one call for short runs. Use the standard run endpoints for longer ones. Clients exist for Node.js (`apify-client` on NPM) and Python (`apify-client` on PyPI). For an AI agent, add the Apify MCP server (`mcp.apify.com`) to your MCP client and the actor appears as a tool. You could ask it for today's new postings at three named companies and get a summary back. ### More from BonCross - [LinkedIn Company Scraper](https://apify.com/boncross/linkedin-company-details-scraper): LinkedIn company pages by URL, slug or name, with no login. - [LinkedIn Company Posts Scraper](https://apify.com/boncross/linkedin-company-posts-scraper): The latest posts from any LinkedIn company page, with no login. - [LinkedIn Post Scraper](https://apify.com/boncross/linkedin-post-details-scraper): Any public LinkedIn post by URL, with no login. - [NHS Jobs Scraper](https://apify.com/boncross/nhs-jobs): NHS Jobs vacancies by keyword and location. - [Reed.co.uk Jobs Scraper](https://apify.com/boncross/reed-jobs-scraper): Reed.co.uk jobs by keyword and location with Reed's own filters: salary, contract,β¦ - [UK Government & NHS Jobs Scraper](https://apify.com/boncross/uk-public-sector-jobs): NHS Jobs, Civil Service Jobs (every department) and Ministry of Justice vacancies inβ¦ ### Your feedback Spotted a LinkedIn job that came back wrong, or a field you expected and didn't get? Tell us on the Issues tab in Apify Console. We read each report and reply there, and fixes go out as new builds. # Changelog This Actor's version history is a separate document: https://apify.com/boncross/linkedin-jobs-scraper/changelog.md # Actor input Schema ## `keywords` (type: `array`): One search per entry, for example "paralegal" or "compliance analyst". Results are deduplicated across searches. Leave this empty if you give start URLs instead. ## `location` (type: `string`): The default location, written as you would type it into LinkedIn, such as "London, United Kingdom". Ignored when the locations list is set. ## `locations` (type: `array`): Optional list. Every keyword runs against each of these and the results are deduplicated. Splitting a country into its cities is how you get past LinkedIn's ceiling of about 1,000 results per search. ## `distance` (type: `integer`): Radius around the location in miles, as on LinkedIn. 0 uses LinkedIn's default. ## `startUrls` (type: `array`): Optional. Paste full LinkedIn job-search URLs. Date-posted, company, Easy Apply and under-10-applicants filters are kept. The experience, job-type and workplace filters LinkedIn dropped are turned into keyword terms. ## `maxResultsPerSearch` (type: `integer`): Ceiling per keyword or URL. The search stops earlier when LinkedIn runs out of matches for your window. LinkedIn serves at most about 1,000 per search. ## `maxItems` (type: `integer`): A run-wide ceiling across every target. The per-target limit above applies to each one separately, so ten targets at 50 each returns 500; set this to cap the total. Zero means no run-wide limit. Rows that report an input the actor could not resolve are not counted here and are never charged for. ## `datePosted` (type: `string`): Only return jobs posted within this window. It is sent to LinkedIn as a real filter, so you do not pay for stale results. ## `postedWithinHours` (type: `integer`): A finer window than the presets: 6 for a run every six hours, 1 for hourly. Overrides date posted when set. LinkedIn honours any number of hours. ## `sortBy` (type: `string`): Most recent first, or the order LinkedIn returned. LinkedIn no longer sorts the public job search, so "Most recent" is applied to the results after they are collected: it orders what the run fetched, it cannot change which jobs LinkedIn selects. ## `experienceLevels` (type: `array`): Experience level. Since August 2026 LinkedIn no longer filters on this, so it is added to your search terms instead. That steers results strongly but is not exact, and it narrows the pool. For exact results pair it with "Seniority levels" below. ## `jobTypes` (type: `array`): Job type. Since August 2026 LinkedIn no longer filters on this, so it is added to your search terms instead. That steers results strongly but is not exact, and it narrows the pool. For exact results pair it with "Employment types" below. ## `workplaceTypes` (type: `array`): Workplace type. Since August 2026 LinkedIn no longer filters on this, so it is added to your search terms instead. In testing this raised remote roles from about 1 in 9 to about 11 in 15, but it is not exact and it narrows the pool. ## `under10Applicants` (type: `boolean`): LinkedIn's own "under 10 applicants" filter, applied as a real URL filter. ## `easyApply` (type: `boolean`): LinkedIn's own Easy Apply filter, applied as a real URL filter. ## `companyIds` (type: `array`): Optional numeric LinkedIn company ids to restrict the search to. Applied as a real URL filter. If you do not know the ids, use Company names above instead. ## `companyNames` (type: `array`): Company names to restrict the search to, e.g. "Monzo". Each is resolved to the numeric LinkedIn id that the company filter needs, so you do not have to find ids yourself. The run log shows what every name resolved to, and lists the other candidates it matched, because names are not unique. Resolution costs one lookup per name and is paced, so keep the list short. ## `seniorityLevelsFilter` (type: `array`): A real filter on the seniority label printed on the job page (Internship, Entry level, Associate, Mid-Senior level, Director, Executive, Not Applicable). It runs after the detail fetch, so those jobs are still charged as details. ## `employmentTypesFilter` (type: `array`): Filter on the employment-type label from the job page (Full-time, Part-time, Contract, Temporary, Internship, Volunteer, Other). Runs after the detail fetch. ## `maxApplicants` (type: `integer`): Drop jobs with more applicants than this, for example 25. Finer than LinkedIn's under-10 toggle. Runs after the detail fetch. ## `minAnnualSalary` (type: `integer`): Keep only jobs whose declared yearly pay reaches this. Needs "Fetch job details" on. A job that states no pay, or only an hourly or daily rate, is kept rather than dropped. ## `titleInclude` (type: `array`): Optional, case-insensitive. Jobs whose title contains none of these are dropped before the detail page is fetched, so you are not charged a detail for them. ## `titleExclude` (type: `array`): Optional, case-insensitive. Handy for cutting agency noise, for example "recruitment consultant" or "sales". ## `companyExclude` (type: `array`): Case-insensitive substrings matched against the company name before the detail fetch, so these cost only the scan rate. ## `descriptionInclude` (type: `array`): Case-insensitive substrings the full description must contain. Runs after the detail fetch. ## `descriptionExclude` (type: `array`): Case-insensitive substrings that drop a job, for example "security clearance" or "driving licence". Runs after the detail fetch. ## `excludeStaffingAgencies` (type: `boolean`): Needs company details switched on. Drops jobs whose company industry is Staffing and Recruiting. ## `fetchDetails` (type: `boolean`): Fetch each job's page for the full description, seniority, employment type, industries and pay when the employer states it. Turn it off for a cheap listing-only run. ## `fetchFullJobPage` (type: `boolean`): Reads each job's full LinkedIn page instead of the short advert. The columns are the same; the difference is the seniority level, which names a real level on far more jobs (in a London check of 1,100 jobs, "Not Applicable" fell from 580 to 125). Runs are about 45% slower. Declared pay is included either way. Needs "Fetch job details" on. ## `scrapeCompany` (type: `boolean`): Also fetch each unique company's public LinkedIn page for description, website, industry, size, headquarters, type, founding year and specialities. One extra request per company rather than per job, cached within the run. ## `maxNewJobs` (type: `integer`): Stop after this many jobs have been returned. 0 means no cap. ## `dedupStoreName` (type: `string`): Name a key-value store, for example "linkedin-seen", and the actor remembers every job id it has returned. On later runs those jobs are skipped before the detail fetch, so a scheduled run gives you only what is new. ## `companyCacheDays` (type: `integer`): With a dedup store set, company details fetched within this many days are reused rather than fetched again. ## `geoId` (type: `string`): LinkedIn's numeric location id, such as 102257491 for London. More precise than the location text. Ignored when the locations list has more than one entry. ## `maxConcurrency` (type: `integer`): How many requests run at once across the whole run. The default is the fastest level measured clean; lower it only if a run reports refusals or errors. ## `delayMs` (type: `integer`): Extra pause before each request, jittered by about 30 percent either way. The run already keeps to a request rate measured safe for LinkedIn, so the default is 0; raise it only if a run reports refusals. ## `restartAttempt` (type: `integer`): Set by the actor when it restarts itself in a fresh container after being refused. Leave it alone. ## Actor input object example ```json { "keywords": [ "paralegal", "compliance analyst" ], "location": "London, United Kingdom", "locations": [], "distance": 0, "maxResultsPerSearch": 100, "maxItems": 0, "datePosted": "past24h", "postedWithinHours": 0, "sortBy": "date", "experienceLevels": [], "jobTypes": [], "workplaceTypes": [], "under10Applicants": false, "easyApply": false, "companyIds": [], "companyNames": [], "seniorityLevelsFilter": [], "employmentTypesFilter": [], "maxApplicants": 0, "minAnnualSalary": 0, "titleInclude": [], "titleExclude": [], "companyExclude": [], "descriptionInclude": [], "descriptionExclude": [], "excludeStaffingAgencies": false, "fetchDetails": true, "fetchFullJobPage": false, "scrapeCompany": false, "maxNewJobs": 0, "dedupStoreName": "", "companyCacheDays": 30, "geoId": "", "maxConcurrency": 8, "delayMs": 0, "restartAttempt": 0 } ``` # Actor output Schema ## `dataset` (type: `string`): LinkedIn jobs: every row the run returned. ## `stats` (type: `string`): Counts for the run: what was resolved, fetched, filtered and returned. # API You can run this Actor programmatically using our API. Below are code examples in JavaScript, Python, and CLI, as well as the OpenAPI specification and MCP server setup. ## JavaScript example ```javascript import { ApifyClient } from 'apify-client'; // Initialize the ApifyClient with your Apify API token // Replace the '