The full data catalog feeding the Global Labor Graph. Every input is provenance-grade — we list provider, what we use, coverage, and refresh cadence. No marketing badges.
| Source | What we use | Coverage | Refresh |
|---|---|---|---|
Lightcast | Job postings · Profiles · Skills · Firmographics | Enterprises firmographic | Daily |
O*NET | Detailed Work Activities · Tasks · KSA ontology | 19,265 DWAs · 1,016 occupations | Per release |
BLS · OEWS | US workforce employment + p10–p90 wage distribution | 396 metros · 894 occupations | Per release |
ILOSTAT | Cross-country labour-market employment | 172 countries with employment data | Per release |
OECD Economic Outlook | Forward-looking employment projections | 38 OECD countries | Biannual |
Eloundou et al. 2023 (OpenAI) | GPT exposure α / β / γ per occupation | 100% of O*NET occupations | Static (research) |
Felten AIOE | Occupational AI exposure scores | 873 occupations | Static (research) |
Frey-Osborne · Oxford 2013 | Probability of full-task automation | 702 occupations | Static (research) |
WEF Future of Jobs 2025 | Employer-reported skills trajectory | 1,000+ CHROs surveyed across 22 sectors | Annual |
TAG (proprietary depth) | HR client telemetry · candidate interactions · placements | 100K enterprise clients · 104M placements/yr · 300M candidate interactions/yr | Continuous |
Every score in the Workbench cites which of these sources fed which dimension. Where a dimension can’t be sourced, we show a reason chip instead of guessing.