PAYDOSSIER_

How we mark to market

Every number in a mark is fitted from the sources below — nothing else, and nothing from model memory. The full methodology document is being prepared for this page; the sources and their honest limitations are already final.

[1] Occupational Employment and Wage Statistics (OEWS)

U.S. Bureau of Labor Statistics · May 2025 · https://www.bls.gov/oes/

Annual wage percentiles (10th, 25th, median, 75th, 90th) for every occupation in every metropolitan area, from a survey of roughly 1.1 million establishments. This is the shape of the distribution, and it is the most authoritative wage data that exists for the United States.

What it misses: OEWS measures wages: base pay and, for some occupations, commission. It does not capture equity grants, and for most white-collar occupations it excludes annual bonuses. It is also a survey of people already in the job, not of what is being offered to new hires today, and it carries roughly a one-year lag.

[2] LCA Programs (H-1B, H-1B1, E-3) Disclosure Data

U.S. Department of Labor, Office of Foreign Labor Certification · Fiscal years 2024–2025 · https://www.dol.gov/agencies/eta/foreign-labor/performance

The base wage an employer legally attested it would pay, for a named job title at a named worksite, filed continuously rather than annually. Because employers write their own job titles here, it is the only public source that distinguishes a staff engineer from a junior one, and it is uncapped at the top end.

What it misses: It covers visa-sponsored roles only. That population skews toward large technology employers, toward engineering, and away from sales and commission work almost entirely. It records base wage alone — no bonus, no commission, no equity. Filings are also frequently made at the bottom of an intended range, so it reads low at the median.

[3] Employer Costs for Employee Compensation (ECEC), National Compensation Survey

U.S. Bureau of Labor Statistics · 2026 Q1 · https://www.bls.gov/ncs/ect/

The share of total compensation that employers spend on each component, including "nonproduction bonuses" — cash bonuses not tied to output, which is where year-end, performance and retention bonuses land. Dividing that share by the wages-and-salaries share gives cash bonus as a percentage of base pay, by industry.

What it misses: ECEC EXCLUDES stock options and stock grants from its definition of compensation entirely, so it says nothing whatever about equity. It is also an average across every worker in an industry, including those who receive no bonus at all and including non-professional staff, so it runs well below the target bonus a senior professional would be quoted.

[4] Core Based Statistical Area (CBSA) delineation files, July 2023

U.S. Census Bureau · July 2023 · https://www.census.gov/geographies/reference-files/time-series/demo/metro-micro/delineation-files.html

The official county-to-metropolitan-area definitions used to place a worksite in a metro. This is how a filing in Redmond becomes a Seattle data point rather than being discarded.

What it misses: Nothing about pay. It is a geography reference only, cited because it determines which metro a wage record lands in.