Salary numbers you can actually trust
Most pay sites online are blended self-reported data with no real source. Glassdoor, Payscale, tons of niche tools — the sample sizes are hidden, the methodology shifts, and the numbers drift over time.
Our pay data starts with the U.S. Bureau of Labor Statistics Occupational Employment and Wage Statistics (OEWS), May 2024. BLS surveys 1.2 million US workplaces over a three-year cycle. It's the biggest, most trusted wage dataset in the country.
How we build the table
- Job mapping. 49 real jobs mapped to their government codes (Warehouse Associate → 53-7065, Medical Assistant → 31-9092, HVAC Technician → 49-9021).
- National baseline. For each job, BLS gives us low (p25), median, high (p75), and top (p90) pay.
- Metro adjustment. BLS's area wage factors let us split it into 15 metros — so Atlanta pay isn't showing you San Francisco numbers.
What we publish and what we don't
We publish per-job, per-metro pay at four points: low, median, high, and top. We don't publish overtime, tips, shift pay, or sign-on bonuses — those vary way too much by employer to be honest. We do call them out when they move the total (nursing night shifts, delivery tips, trades overtime).
What we won't do
- Publish a number we can't source. Every pay page shows where the figure comes from and when it's from.
- Train a black-box estimator on scraped LinkedIn data. We use BLS.
- Use "up to" or "starting at" pay tricks. You see low, median, and high together every time.
The full table lives at /salary/. 735 rows, 49 jobs, 15 metros.
See something wrong? Email hello@jobeezy.com. We log corrections in the changelog.