Evidence Building
How to Use BLS OEWS Salary Survey Data to Document the High Salary Criterion in 2026
The BLS Occupational Employment and Wage Statistics survey is the primary tool for documenting the O-1A high salary criterion, but only when the correct SOC code, geographic market, and percentile comparison are used. Here is how to build a salary exhibit that holds up under USCIS review.
The high salary criterion and why data sourcing matters
The high salary criterion is one of the eight O-1A evidentiary categories listed at 8 C.F.R. § 214.2(o)(3)(iii)(B). It requires the petitioner to show that the beneficiary has commanded or will command a high salary or other significantly high remuneration relative to others in the field. In practice, the criterion is satisfied through documentary evidence comparing the beneficiary's actual or offered compensation to wage survey data for the relevant occupation in the relevant geographic market. The Bureau of Labor Statistics Occupational Employment and Wage Statistics survey — commonly called BLS OEWS — is the most widely used publicly available wage data source in O-1A petitions, and understanding how to use it correctly is essential to building a high salary exhibit that USCIS will find persuasive.
BLS OEWS data is collected through a semiannual survey of employers across all industry sectors and published annually, typically in the spring. The survey reports the 10th, 25th, 50th, 75th, and 90th percentile wages for hundreds of occupational categories, broken down by geographic area at the national, state, and metropolitan area level. For O-1A purposes, the relevant comparison is typically the 90th percentile for the petitioner's occupation in the petitioner's metropolitan area, because USCIS has generally accepted 90th percentile compensation as strong evidence of high salary. Petitions that compare the beneficiary's compensation to the national median, rather than the 90th percentile for the relevant occupation and market, consistently draw RFEs requesting additional wage comparison data.
The OEWS data is organized by Standard Occupational Classification codes. Selecting the correct SOC code — one that accurately reflects the beneficiary's occupation and not a broader or narrower classification that distorts the wage comparison — is a key technical step in building the high salary exhibit. A software engineer classified under SOC 15-1252 (Software Quality Assurance Analysts and Testers) will see substantially different wage benchmarks than one classified under SOC 15-1251 (Computer Programmers) or 15-1299 (Software Developers and Software Quality Assurance Analysts, All Other). Using an inaccurate SOC code, even inadvertently, produces a wage comparison that USCIS may reject as inapplicable to the petitioner's actual occupation. Attorneys building high salary exhibits should verify the applicable SOC code against the BLS occupational definition before citing the wage data.
What the regulation requires
The regulatory text at 8 C.F.R. § 214.2(o)(3)(iii)(B)(8) requires a showing that the beneficiary "commands or has commanded a high salary or other significantly high remuneration for services, evidenced by contracts or other reliable evidence." This language establishes two distinct components: the evidentiary vehicle — a contract or other reliable evidence of actual or committed compensation — and the comparative standard — high salary in relation to others in the field. USCIS interprets "others in the field" as peers in the same or comparable occupation who are working in the same or comparable geographic market. A compensation figure that appears high in absolute terms but falls below the 75th percentile for a high-wage metropolitan area like San Francisco or New York may not satisfy the high salary standard for purposes of the criterion.
The AAO has addressed the high salary criterion in several precedent and non-precedent decisions and has consistently held that the comparison must be to "others in the field" — meaning the beneficiary's actual occupation and industry, not a generalized comparison to all workers. An O-1A petitioner who is a research scientist at a biotechnology company earns compensation that should be compared to research scientists in the biotech industry, using BLS OEWS data for the relevant SOC code in the relevant metropolitan area, rather than to all occupations or to a national average that obscures geographic variation in compensation. The petition should make this comparison explicit and show the math: beneficiary's total annual compensation versus the 90th percentile figure from the relevant OEWS table.
"Other significantly high remuneration" in the regulatory text is intended to capture compensation structures that are not purely salary — equity, royalties, contingent compensation, or profit-sharing arrangements that can be more valuable than a fixed annual salary but that do not appear in a standard employment contract in a way that is immediately legible to an adjudicator. USCIS has accepted evidence of high equity compensation, royalty income, and performance-based bonuses as part of the remuneration comparison when the petition documents the actual or projected value of those components and compares the total to wage survey data for salaried workers in the same occupation. The key is documenting that the non-salary components have real, quantifiable economic value rather than speculative future worth.
Evidence that routinely satisfies the criterion
The most reliable evidence for the high salary criterion is a compensation-related document from the current employer: an executed employment agreement specifying base salary, bonus structure, equity terms, and benefits; or an offer letter from the prospective employer signed by both parties. Where the petitioner is already employed, the most recent W-2 or pay stubs covering the most recent complete year provide actual compensation records rather than offer-stage commitments. W-2 Box 1 wages combined with employer-paid benefits documented through the petitioner's benefits enrollment records give the adjudicator a total compensation figure grounded in actual tax records. This approach is generally more persuasive than relying solely on the offer letter because it reflects actual earned compensation rather than a commitment not yet fulfilled.
The BLS OEWS data is published at data.bls.gov and accessed through the occupational employment statistics tools. When using the data in a petition, the attorney should include a printed copy or screenshot of the relevant OEWS table for the specific SOC code and metropolitan area, annotated to show: the SOC code used, the metropolitan area selected, the year of the data, and the 10th through 90th percentile wages. The annotation should highlight the 90th percentile figure and include a statement showing that the beneficiary's total annual compensation equals or exceeds that figure. For petitions where the beneficiary is offered a salary that falls between the 75th and 90th percentile, the annotation should explain the comparison and include any additional wage data sources — professional association salary surveys — that support the argument.
Industry salary surveys from recognized professional associations can supplement or provide context for BLS OEWS data, particularly in technical fields where compensation varies substantially by specialization within a single broad BLS category. The IEEE annual compensation survey, the Computing Research Association's Taulbee Survey for computer science faculty, the American Chemical Society Salary Survey for chemists, and the Association of University Technology Managers compensation surveys are examples of field-specific salary surveys that USCIS has accepted as supporting evidence in high salary exhibits. These surveys often break out compensation by career stage, specialization, and sector in ways that BLS OEWS data does not, and can show that the beneficiary's compensation is high relative to a more narrowly defined peer group in their specific discipline.
Evidence USCIS regularly discounts
BLS OEWS data selected from the wrong geographic market is one of the most common technical errors in high salary exhibits. The OEWS survey reports wages for individual metropolitan statistical areas, and compensation levels vary dramatically across markets. A software engineer earning $220,000 per year in San Francisco may be at the 75th percentile for their market, while the same salary would represent the 95th percentile or above in a smaller metropolitan area. A petition using national-average OEWS data when the beneficiary works in a high-wage metropolitan area significantly overstates the relative standing of the beneficiary's compensation, producing a comparison that a careful USCIS adjudicator will recognize as potentially misleading. Using metropolitan area data for the relevant market is the technically correct approach.
National median comparisons — rather than the 90th percentile for the specific occupation and market — are consistently less persuasive and regularly draw RFEs. A petition asserting that the beneficiary's compensation is "above the national median for software engineers" does not satisfy the high salary criterion because being above the median means only that the beneficiary earns more than half of workers in the occupation — a standard that does not represent extraordinary compensation. USCIS's operationalization of "high salary" in practice corresponds to compensation at or near the 90th percentile for the relevant SOC code and market. Petitions using median comparisons as the primary benchmark should be supplemented or corrected before filing.
Compensation from foreign employers — for petitioners who earned their high salary outside the United States — requires a different data source and more explanatory work. BLS OEWS data reflects U.S. wages and is not a valid comparison benchmark for foreign compensation. Using BLS OEWS to compare compensation earned abroad gives the adjudicator an inaccurate picture of the beneficiary's relative compensation standing. For foreign compensation, the petition must use wage survey data from the relevant country's statistical agency or a recognized international compensation survey — and must include a citation to the data source, the year, the occupation category used, and a clear statement of the comparison result.
Presenting borderline salary evidence
When a beneficiary's compensation falls between the 75th and 90th percentile for the relevant occupation and market — above most peers but not at the very top of the distribution — the petition should present the comparison at multiple levels and from multiple data sources rather than relying on a single OEWS table. Presenting both the 75th and 90th percentile figures, noting where the beneficiary's compensation falls between them, and supplementing with an industry survey showing that the compensation exceeds the survey's reported upper quartile for the relevant specialization provides a richer evidentiary picture. The totality-of-evidence standard means that a near-90th-percentile salary, combined with four other satisfied criteria, is substantially more likely to result in approval than when high salary is the only strong criterion in the petition.
For petitioners whose compensation includes significant equity or bonus components, the petition should document both the target and actual values of those components separately from base salary. A $150,000 base salary supplemented by a performance bonus targeted at 30 percent of base salary and an equity grant currently worth $200,000 — calculated at fair market value based on a recent 409A appraisal or secondary market transaction — presents a total compensation argument substantially different from the base salary alone. The petition should include the 409A appraisal or other equity valuation document, an explanation of the vesting schedule that clarifies what portion of the equity is currently vested and therefore realized, and a calculation of total annualized compensation including the vested equity portion.
Occupation reclassification can sometimes improve the high salary analysis legitimately — not by misclassifying the beneficiary to find a favorable benchmark, but by accurately identifying a more specific BLS SOC code that better reflects the petitioner's actual work and that has a higher wage distribution reflecting a specialization premium. A physician researcher classified under SOC 29-1228 (Physicians, All Other) may have a different wage distribution than one classified under SOC 19-1042 (Medical Scientists, Except Epidemiologists), depending on whether their work is primarily clinical or research-oriented. The correct classification is determined by the primary duties of the beneficiary's position, not by whichever SOC code produces the most favorable percentile comparison. Misclassification to chase a favorable benchmark is ethically problematic and vulnerable to challenge in any RFE.
Building and auditing the salary exhibit
A well-constructed high salary exhibit has the following components: the executed employment agreement or offer letter showing the full compensation package; the relevant OEWS table for the correct SOC code and metropolitan area, with the applicable year noted; a calculation showing total annual compensation compared to the 90th percentile figure; any supplementary professional association salary survey data where available; and a brief attorney narrative explaining the comparison and why it satisfies the high salary standard. These components, assembled in a logical order with clear annotations, give the adjudicator everything needed to evaluate the criterion without requiring independent research into occupational wage norms.
Attorneys reviewing high salary exhibits before filing should audit three things: the SOC code accuracy, the geographic market accuracy, and the total compensation calculation. The SOC code should be verified against the BLS occupational descriptions at bls.gov/soc, not selected based on what the beneficiary's employer calls the job title. The geographic market should correspond to where the beneficiary will actually perform services, using the relevant Metropolitan Statistical Area. The total compensation calculation should include all elements of remuneration that are actual and quantifiable — base salary, vested equity, realized bonuses, employer-paid benefits with quantified monetary values — but should exclude unvested equity and speculative future bonuses. An exhibit that passes these three checks is substantially less likely to draw an RFE on the high salary criterion.
As of the BLS May 2025 OEWS release — the most current data available in 2026 — practitioners should confirm that the wage tables they cite are from this release or the upcoming 2026 release when available, rather than from older surveys. Wage data from 2022 or 2023 significantly understates current compensation levels in high-wage occupations due to post-pandemic labor market changes, and an adjudicator who independently verifies the cited OEWS figures against the current release will find a discrepancy. Using outdated wage data also understates the beneficiary's relative compensation standing — a salary that appeared to be at the 90th percentile in 2022 data may be at the 80th percentile in 2025 data as market wages have increased. Always use the most recently published OEWS release.
What we typically gather for this kind of case
| Document | Where to source | Why it matters |
|---|---|---|
| Expert letters | 5–8 independent recognized experts | Quality and independence beat volume |
| Certified translations | ATA-certified translator | Required for any non-English source document |
| Exhibit cover sheets | Drafted by counsel, one per exhibit | Tells the adjudicator what each piece shows |
| Bibliometric reports | Web of Science / Scopus | Quantifies impact for original-contributions criterion |
What we see go wrong, again and again
- 01Sending exhibits without a one-paragraph framing memo explaining what each shows and why it matters.
- 02Relying on volume over specificity — five well-targeted expert letters beat fifteen generic recommendations.
- 03Skipping certified translations or using AI translation for foreign-language source documents.