Job Postings Data
1B+ unique job postings since 2022, collected from Indeed, LinkedIn Jobs and 200,000+ employer career portals (Greenhouse, Lever, iCIMS, Workday and others). Each posting is enriched with 100+ fields: occupation (SOC) and industry (NAICS) codes, ~100K normalized titles, 43,000+ skills and credentials, salary, seniority, work arrangement, visa and clearance signals, and per-field confidence scores. US-primary plus 240+ other countries and territories, delivered at least daily.
One row per posting. The posting file is split into modules by subject, so you take only the columns you need; every module carries jobID to re-join. 157 posting columns in catalog 8.4.
Key Highlights
- Normalized title on 99.8% of postings and an SOC 2018 occupation code on 99.9% (73% top-1, 93% top-5 across 867 codes)
- NAICS 2022 industry code on 98.0% of postings, resolved per employer (89% top-1, 98% top-5 across 1,012 codes)
- Annual salary in USD on 90.5% of postings; 42.6% of those are employer-stated, and a salarySource column marks which rows are model estimates
- Skills, certifications or licenses extracted on 89.8% of postings, matched against a 43,000+ entry taxonomy
- Geocoded location on 97.8% of postings and a metro area on 89.1%
- 23.3% of postings linked to a public-company ticker, with the home listing carried point in time
- Seniority, employment type, work arrangement, worker classification (W-2, 1099, C2C), visa sponsorship and security clearance on every module where the posting states them
- A stable jobID across deliveries, and a flag marking one representative posting per duplicate group for cross-source deduplication
Use Cases
- Hiring-activity signals for investment research
- Labor market and occupation analytics
- HR tech product features: job search, matching and market pricing
- Competitive hiring intelligence
- Staffing demand sizing
- Economic research and training data for job and skills models
Sample FieldsView full schema
jobIDjobTitlenormalizedTitlesocCodesenioritynormalizedCompanyNamenaicsCodefinalCityfinalStatefinalCountryremoteStatusemploymentStatusannualSalaryAverageUSDsalarySourcetechnicalSkillstickerSymbolWhat You Receive
Delivered as separate files that join on one key, so you take only what you need. Catalog version 8.4.
| File | What it carries | Availability |
|---|---|---|
jobs_core | Identifiers, title, URL, source, posted and seen dates, the board's own function and industry labels, location and metro | Standard |
jobs_role | Normalized title, seniority (8 bands, IC or manager track) and SOC code, each with its confidence | Standard |
jobs_arrangement | Employment type, work arrangement, shifts, travel, advertiser type and worker classification | Standard |
jobs_eligibility | Visa sponsorship, security clearance, citizenship and other screening gates | Standard |
jobs_skills_extracted | Technical skills, soft skills, certifications, licenses, languages and minimum degree | Standard |
jobs_salary | Annual pay in USD (minimum, average, maximum), pay period, its source, benefits and equity mentions | Standard |
jobs_company | Employer firmographics on the posting, NAICS 2022 and the ticker with its home listing | Standard |
jobs_changes | Postings updated or removed between deliveries | Included with subscriptions |
jobs_descriptions | Full cleaned description text, one row per distinct text | Add-on |
jobs_signals | The evidence phrases behind each classification | Add-on |
jobs_open_duration | Posting status and how long each posting stayed listed | Opt-in |
Delivery
Cadence: At least daily (several times a day available)
Raw vs. Enriched
See how Canaria transforms a basic job posting into a fully enriched record with 100+ structured fields.
Raw Data
What scrapers give you
No SOC codes, no salary prediction, no skills extraction,
no seniority, no deduplication...
Building enrichment in-house costs $500K-$1M Year 1
Canaria Enriched
100+ fields per record
+ 62 more fields (location, dedup metadata, qualifications, clearance, travel...)
Relevant Solutions
Labor Market Data for Investment Professionals
Hiring velocity as an economic leading indicator with SOC-level granularity
Job Market Data for Competitive Intelligence
Competitor hiring patterns, skills trends, and geographic expansion signals
Job Market Data for HR Tech Platforms
Add salary benchmarking and skills intelligence to your platform without building ML
Job Market Training Data for AI & ML Teams
Pre-enriched, deduplicated job market training data. Skip 6 months of pipeline building.
Job Market Data for Workforce Planning
Enterprise-grade labor market intelligence with full data transparency, no vendor lock-in
Job Market Data for Academic Research
Longitudinal dataset for labor economics: wage dynamics, skill demand, remote work adoption
Job Market Data for Recruiting & Staffing
Know where hiring is heating up, what competitors pay, and which skills are pulling demand, before your clients ask.
Job Market Data for Consulting Firms
Project-ready labor market data with no annual contract required.
Job Market Data for Healthcare Workforce Planning
Track clinical hiring by facility type, patient population and credentialing timeline across 132M+ clinical job postings.
More Datasets
Job postings, company, Google Maps and public-company files share one company key; salary and skills join on occupation and job title.
Healthcare Job Postings Data
132M+ clinical postings with facility type, patient population, caseload, credentialing and compliance detail
Company Data
29M company profiles with firmographics, NAICS codes and monthly hiring activity
Public Company Hiring Data
Point-in-time job-posting counts for 7,900+ public-company security series since 2022
Google Maps Business Data
200M+ business locations in 243 countries with categories, ratings, hours and coordinates
Salary Benchmark Data
42K+ salary benchmarks by occupation, for every US state and nationally, with percentile ranges and 95% confidence intervals
Job Skills Data
Skill demand by job title and month: 36K skills across 70K job titles
Frequently Asked Questions
Common questions about the Job Postings Data dataset.
- How many job postings are in the dataset?
- 1B+ job postings since 2022, unique by URL and source job key, collected from Indeed, LinkedIn Jobs and 200,000+ employer career portals. US-primary plus 240+ other countries and territories.
- What enrichment is included?
- 157 delivered columns per posting: normalized title, SOC occupation and NAICS industry codes, seniority, work arrangement, worker classification, skills and credentials, salary, visa and clearance signals, employer firmographics and ticker linkage, with per-field confidence scores.
- How complete are the main fields?
- On the current release, a normalized title is populated on 99.8% of postings, an SOC code on 99.9%, a NAICS code on 98.0%, an annual salary on 90.5% and at least one skill or credential on 89.8%. The data dictionary lists the fill rate of every column.
- How are duplicate postings handled?
- The same job posted to several boards keeps one row per posting, and a flag marks one representative per duplicate group, found with embedding similarity, MinHash and graph resolution. Filter on it to count each job once.
- How is the data delivered?
- At least daily. Parquet by default, CSV on request, delivered to Amazon S3, Google Cloud Storage, Azure Blob Storage, Google Drive or SFTP. Every product carries its own catalog version, so a delivered schema never changes without notice. Request a sample to evaluate the files before you buy.