Every figure in every report traces back to an official federal source — BLS, O*NET, IPEDS — published by agencies whose data any reviewer can access independently. No proprietary estimates, no black boxes, no vendor you have to trust on faith. Here's exactly what we use, how we use it, and how you can verify every number yourself.
Six public datasets. All maintained by federal agencies or the federally funded state projection consortium. All independently accessible.
Regional wages, employment levels, and location quotients for 393 metro areas across 800+ occupations. Published by the Bureau of Labor Statistics.
National 10-year growth rates, annual openings, and typical education requirements for 831 occupations. BLS now publishes a new 10-year projection every year.
A relative AI exposure category (Low, Moderate, High, Very high) for each of the 831 projected occupations, introduced by BLS alongside the 2025–2035 projections. Built by combining three theoretical exposure measures with two observed AI-usage measures and ranking occupations against one another. National only. Per BLS, exposure does not imply job loss, automation probability, or wage effects. Used on the Workforce Pell State Eligibility Tracker; every occupation's category appears in the expanded card and the CSV export.
State-level long-term (10-year) and short-term (2-year) occupational employment projections — annual openings and projected growth for each state. PMP is the U.S. Department of Labor / ETA-funded consortium through which state labor market information offices produce the official projections their states report to DOL and use for WIOA in-demand determinations.
Canonical task statements, technology profiles, and certification data for 1,000+ occupations. Developed by the DOL.
Institution data, program completions, and student outcomes (employment rate, median earnings, median debt) for 5,000+ institutions. IPEDS from NCES; outcomes from College Scorecard.
Four steps between the federal source and the number in your compliance document. Nothing proprietary happens to the data along the way.
Sourced job postings are semantically compared against O*NET occupation task statements. Each posting sentence is matched to the canonical task list for your SOC code, identifying which tasks employers are actively requesting.
Wages, employment, growth projections, and location quotient are pulled directly from BLS datasets using your specific SOC code and metro area code. No transformation is applied — the values are read as-is.
SOC codes, MSA codes, and area titles are checked for consistency across all data sources. If an occupation–region combination has suppressed data in BLS, the report flags it rather than estimating.
Data is organized into the report format required by your framework — Workforce Pell, Perkins V, WIOA, accreditation, or planning — with the source cited next to every number.
Pick a data point and follow it step-by-step from the federal dataset to your report — then verify it yourself on the source website.
Each May, BLS surveys 1.1 million employers and publishes wage data for every occupation in every metro area. The raw dataset covers 393 metros × 800+ SOC codes.
When you select a metro area and SOC code, we query the OEWS dataset for that specific combination — for example, area_code: 12060 (Atlanta) × soc_code: 31-9092 (Medical Assistants). The value returned is the annual_median field with no transformation.
The wage is displayed alongside the MSA name and area code so reviewers can see exactly which region's data they're looking at. Reports that evaluate wage thresholds (Workforce Pell, WIOA) compare this value against the BLS national all-occupations median ($50,981, BLS OEWS May 2025).
Go to bls.gov/oes → One occupation for one area. Select your metro area and enter the SOC code shown in the report header. The Annual median wage column will match the figure in your report.
Each year, BLS projects employment levels 10 years out for every occupation. The latest edition, 2025–2035, was released August 27, 2026; reports currently draw on the 2024–2034 edition and will move to 2025–2035 in the next pipeline update, with the vintage stated in the report. This is national data — BLS does not publish regional projections.
The change_percent and annual_openings fields are looked up by SOC code from BLS Table 1.7. Annual openings combines growth openings with replacement needs (retirements, career changes). These are national figures clearly labeled as such. Occupation-level median wages in the projections dataset are joined with BLS OEWS May 2025 national data so all wage figures reflect the most current release.
The Accreditation report displays these in a dedicated "National Employment Projections" section, and the Perkins report uses them in demand validation. The Workforce Pell sweep instead uses state-level projections from the Projections Managing Partnership for its in-demand determination, since a state approval decision turns on demand within that state. Reports always note "national" scope when displaying BLS Employment Projections data.
Go to bls.gov/emp → Table 1.7. Search for the SOC code in the report header. The percent change column will match. Median wages reflect BLS OEWS May 2025 national figures — verify at bls.gov/oes.
For each occupation, O*NET publishes 15–25 task statements describing what workers actually do. These are developed by occupational analysts and validated by worker surveys.
Each sentence in the sourced job postings is embedded and compared to the O*NET task statements using semantic similarity. We report which tasks were confirmed by employer postings and which were not — even when the wording differs.
Reports show the task coverage percentage (e.g., "17 of 22 O*NET tasks confirmed across 15 postings"), with each task listed as core (70%+), common (30–70%), or occasional (<30%) based on confirmation rate.
Go to onetonline.org. Search for the SOC code in the report header. Under "Tasks," you'll see the same task statements listed in the report.
Location quotient measures how concentrated an occupation is in your metro area compared to the national average. An LQ of 1.2 means the occupation is 20% more concentrated locally. BLS computes this directly from OEWS survey data — it is not an estimate.
The location_quotient field comes straight from the OEWS dataset for your metro × SOC combination. No computation or adjustment. We display it with standard interpretation: ≥1.2 above-average, 0.8–1.19 near-average, below 0.8 below-average.
All five regional reports display LQ with the same thresholds and color coding, ensuring consistent interpretation regardless of which compliance framework you're using.
Go to bls.gov/oes → One occupation for one area. Select your metro area and SOC code. The "Location quotient" column will show the same value.
Every report draws from the same underlying data. This shows exactly which data points appear in each report type.
| Data Point | Source | Workforce Pell | WIOA ETPL | Perkins V | Accreditation | Curriculum | Employer |
|---|---|---|---|---|---|---|---|
| Median Wage (regional) | BLS OEWS | ||||||
| Wage Distribution (10th/90th) | BLS OEWS | ||||||
| Regional Employment | BLS OEWS | ||||||
| Location Quotient | BLS OEWS | ||||||
| Growth Rate (national) | BLS Projections | ||||||
| Annual Openings (national) | BLS Projections | ||||||
| Task Alignment / Coverage | O*NET + Postings | ||||||
| Certifications | O*NET + Postings | ||||||
| Technologies | O*NET + Postings | ||||||
| Beyond O*NET | Postings (derived) |
Every data source follows a public release schedule. Here's when to expect new data and when this platform will reflect it.
May survey, released following April.
National 10-year outlook, published annually.
Quarterly updates; major release in Q3. Task layer anchored to v30.2 — see note below.
Annual institutional data release.