Lightcast Job Postings
Lightcast (formerly Emsi Burning Glass)
Billions of scraped, deduplicated job ads, each tagged with skills: the data behind much skills research.
Posting-level job advertisement data collected from thousands of online sources, deduplicated, and coded for occupation, industry, employer, location, education, experience, salary, and specific skills. Each row is a unique posting. Formerly Burning Glass Technologies data, it is widely used in research on skill demand, upskilling, and AI-related hiring.
Access
- Cost
- PaidPriced by quote; no published prices. Typically institutional or enterprise licenses; some universities hold research licenses.
- License
- ProprietaryCommercial license
- Sign-up
- Purchase required
- How you get it
- API and Bulk downloadAPI, bulk files
What's in it
- One row is
- One row per deduplicated job posting
- Covers
- United States and many other countries (Lightcast data spans 165 countries overall)
- Years
- U.S. postings from 2010 (with some earlier years) to present
- Updated
- Daily
- Size
- About 3 billion job postings (provider's figure)
Quality
The most-used posting data in academic labor research, with rich skill coding, though postings over-represent white-collar jobs.
- Postings skew toward professional jobs
- Coverage shifts as source websites change
- Not every posting is a real vacancy
Connects to
- O*NET Database: Map posting occupations and skills to O*NET descriptors
- Job Openings and Labor Turnover Survey (JOLTS): Official openings to benchmark posting volumes
- Indeed Hiring Lab Job Postings Index: Free alternative index of postings
Questions it could answer
- How fast are AI-related skills spreading into non-tech job postings?
- Did employers drop college degree requirements after 2020?
- Which metros post the highest advertised salaries for nurses?