How USCIS judges a green card EB1A case

A founder does not qualify for EB-1A because she started a company, raised money, or uses an impressive title. She qualifies only if the evidence shows sustained national or international acclaim, recognized achievements, and a level of expertise placing her among the small percentage at the top of a defined field. USCIS judges the person, not the startup pitch.

That distinction controls the whole case. Company facts matter when they prove the founder's individual work and standing. A funding announcement, accelerator badge, revenue chart, or press profile can help, but none carries a fixed number of points. The petition must connect each fact to a regulatory criterion and then connect the whole record to extraordinary ability. This is general educational information, not legal advice; immigration strategy depends on the record, status history, visa availability, and current law.

How the green card EB1A route actually works

EB-1A is an immigrant classification for extraordinary ability in the sciences, arts, education, business, or athletics. A founder usually claims extraordinary ability in business, or in a technical field if her acclaim and proposed US work genuinely belong there. The classification can support permanent residence, but approval of Form I-140 does not by itself issue a green card. A person still completes adjustment of status in the United States or immigrant visa processing abroad, and an immigrant visa must be available.

The practical advantages are substantial. The regulation permits self-petitioning, requires neither a permanent job offer nor a Department of Labor certification, and does not require a particular degree. Those features make EB-1A attractive to founders whose work does not fit an ordinary employer-sponsored case. They do not lower the evidentiary standard.

The governing rule, 8 CFR 204.5(h), offers two routes through the initial evidence stage. A person may show a one-time achievement that is a major internationally recognized award, or meet at least three of ten listed criteria. Most founders use the second route. Comparable evidence may be available if a listed criterion does not readily apply to the occupation, but it is not a free substitute for evidence the founder simply lacks. The petitioner must explain why a criterion fits the occupation poorly and why the proposed substitute has comparable significance.

The case must also show clear evidence that the founder intends to continue work in the area of expertise in the United States. Contracts, commitments, a detailed plan, company records, and letters about prospective work can do that. The proposed work need not use the identical job title, but the connection to the field that produced the acclaim must make sense. A celebrated researcher who now offers only a vague plan to run an unrelated retail company has a field problem, not a formatting problem.

Define the field before selecting evidence

A workable field is narrow enough to make comparison meaningful and broad enough to cover the founder's real record. Calling the field simply business often leaves USCIS with no sensible peer group. Defining it as enterprise fraud prevention, medical-device commercialization, or cross-border logistics technology may let the evidence answer who the peers are, what excellence looks like, and why the founder's results matter. A label invented solely for the petition can fail just as quickly if no real professional community recognizes it.

Write one sentence that states the field, the founder's distinctive work, and the intended US work. Then test every exhibit against it. If an award honors general student entrepreneurship, an article covers the company's consumer product, and the proposed work concerns health-care analytics, the petition may contain three good facts that do not describe one sustained career.

Founders often blur the company, the product, and the individual. USCIS does not. Company revenue may show commercial adoption; it does not automatically show who caused it. A patent may identify an inventor; it does not automatically establish major significance. A CEO title proves a leading role more readily than a critical one, but the criterion also requires a distinguished organization. The petition must supply the missing link in each instance.

Field definition also governs comparisons. Salary evidence needs a peer group with comparable geography, seniority, company stage, and compensation form. Citation evidence needs a meaningful scholarly field and publication age. Funding evidence needs context for the company's sector and stage. Broad comparisons create dramatic percentiles that an officer can dismiss because the denominator has little to do with the founder.

Do this field exercise before asking for recommendation letters. Otherwise each writer will describe a different version of the candidate, and the cover letter will have to repair contradictions that should never have entered the record.

The ten criteria do not fit founders equally

The ten criteria in 8 CFR 204.5(h)(3) are exact legal categories, not a menu of flattering accomplishments. The following founder translations identify what each category asks and what usually supplies the missing proof.

1. Lesser nationally or internationally recognized prizes or awards. The award must recognize the founder's excellence in the claimed field, and the record must establish the award's reach, selection rules, judges, competitiveness, and prior recipients. A startup's investment round is ordinarily an equity transaction, not an award to the founder. USCIS administrative decisions have rejected attempts to relabel ordinary venture funding as a prize when the company received capital in exchange for equity.
2. Membership in associations requiring outstanding achievements. Paid membership, alumni status, and communities that accept applicants based on experience do not satisfy the wording. Supply the association's governing rules, admission criteria, evidence that recognized experts judge candidates, and proof that the founder received the qualifying membership level.
3. Published material about the person and her work. The material must be about the founder, relate to her work in the field, and appear in professional, major trade, or other major media. A passing quote, a founder-written guest post, a company press release, or a product roundup may not qualify. Include the title, date, author, full article, circulation or audience evidence, and a certified translation where needed.
4. Judging the work of others. Reviewing accelerator applications, serving on a pitch jury, refereeing papers, or judging an industry competition can fit when the founder actually performs the judging. An invitation alone is weak. Keep the invitation, selection basis, rubric, completed scores, organizer confirmation, event information, and proof showing the level of entrants.
5. Original contributions of major significance. Originality and significance are separate facts. A founder may prove that she created a method, product, business model, or technical advance, then show consequential use through customer adoption, licenses, standards, independent implementation, measurable industry effects, or detailed testimony backed by records. Patents, revenue, and praise can support this criterion, but none automatically proves field-wide significance.

The remaining five criteria often expose petitions built by matching words instead of reading the rule.

6. Authorship of scholarly articles. The article must be scholarly and appear in a professional or major trade publication or other major media. Most marketing posts and founder essays are not scholarly. Research method, citations, editorial standards, expert audience, and publication context matter more than the label thought leadership.
7. Display at artistic exhibitions or showcases. This criterion concerns artistic work. A software demo at a trade show does not become an artistic exhibition because the booth displayed a product. Founders in design, film, architecture, or another artistic field may use it when the venue and displayed work truly fit the wording.
8. A leading or critical role for a distinguished organization. Founders can often prove leadership with corporate records, titles, authority, and actual duties. They must separately prove that the company or relevant division has a distinguished reputation through credible coverage, respected awards, meaningful customers, funding in context, rankings, scale, or other external recognition. To prove a critical role, show what depended on the founder and what changed because of her work.
9. High salary or significantly high remuneration. The comparison concerns pay for services, not the company's valuation or bank balance. Tax records, payroll, contracts, equity compensation evidence, and reliable compensation surveys may work when the comparison accounts for location, role, industry, and company stage. A founder who deliberately takes a low salary should not force this criterion merely because her shares might become valuable.
10. Commercial success in the performing arts. The regulation points to box-office receipts and sales of recordings or video. It rarely applies to a business founder unless her claimed field and achievements are genuinely in the performing arts. Product sales belong elsewhere in the analysis.

A petition needs three criteria, not three exhibits. Twenty articles may support one published-material criterion. One judging engagement and one completed peer review still support the same judging criterion. Counting documents instead of legal categories produces false confidence.

Three criteria open the door but do not settle the case

USCIS policy describes a two-part review. At the first part, the officer asks whether the evidence meets the plain requirements of a major international award or at least three criteria. If it does, the officer considers the record as a whole and asks whether it proves sustained acclaim, recognized achievements, and placement among the small percentage at the top of the field. Practitioners call that second review the final merits determination.

Kazarian v. USCIS is often cited as the source of this framework. The Ninth Circuit rejected extra requirements that USCIS had inserted while deciding individual criteria, such as demanding proof of research-community reaction within the scholarly-authorship criterion. The court also said the agency could consider the quality of the evidence in a final merits analysis. That distinction matters: an officer should not rewrite a criterion at step one, but can examine what the qualifying evidence says about overall acclaim.

The legal status of the second step is now disputed. In January 2026, a federal district court in Nebraska, in Mukherji v. Miller, held that USCIS's final-merits procedure violated rulemaking requirements and ordered relief for that plaintiff. In June 2026, a federal district court in New York, addressing the issue in Taneja v. Miller, reached the opposite conclusion and expressly disagreed with Mukherji. A district court ruling does not automatically erase USCIS policy nationwide, and current USCIS materials continue to describe the two-part approach. Anyone filing or responding to a denial should have counsel check the law in the relevant jurisdiction rather than rely on a headline announcing that final merits is dead.

For petition planning, the safe approach remains straightforward: do not build a case that survives only if the officer stops counting after the third criterion. Ask what an unfamiliar reviewer will learn about scale, independent recognition, selectivity, duration, and the founder's personal responsibility after reading all exhibits together.

Sustained acclaim does not require a prize every year, and a career can have quieter periods. It does require a coherent time line. A burst of petition-oriented articles and judging invitations in the months before filing may meet literal boxes yet carry little weight beside a multi-year record of third-party recognition. Older achievements remain relevant when later evidence shows the standing continued.

USCIS applies the preponderance of the evidence standard, commonly described as more likely than not. That standard is lower than certainty, but it does not rescue unexplained documents. The petitioner must make each necessary fact more probable through relevant, credible, and probative evidence.

Founder achievements need an attribution chain

A founder record becomes persuasive when it connects an external result to the founder through documents created for business reasons, not immigration reasons. I use a four-part chain: event, context, attribution, effect. If one link is missing, an officer can accept that the event happened and still reject the claimed significance.

Take venture funding. The closing documents and bank record prove that an investment occurred. Independent data about comparable rounds explains its scale for that sector and stage. Investment committee material or a detailed investor letter can identify why the founder influenced the decision. Later records may show that her work funded a product adopted beyond the company. Funding can therefore support the company's distinguished reputation, the founder's critical role, remuneration, or the overall record. Calling the round a personal award usually overreaches.

Revenue works similarly. Audited or tax records are stronger than a pitch-deck chart for the amount. Customer contracts and renewal data may show adoption. Product ownership records, board minutes, dated plans, and testimony can tie the result to the founder. Market comparisons can show whether the result is exceptional. Revenue without attribution proves the company sold something; attribution without reliable revenue records leaves the effect unverified.

Accelerator admission may support selectivity and reputation, but the label accelerator proves little on its own. Document the cohort size, applicant pool if the organizer publishes it, selection criteria, named decision-makers, program standing, and whether the founder or company received the honor. Do not recycle one admission as an award, membership, and judging evidence unless each criterion's wording is independently satisfied.

Press evidence needs provenance. Earned reporting by an independent journalist carries a different evidentiary story from a contributed article arranged by a publicist. Both may be legitimate publications, but only the first reliably shows that an outside editor considered the founder newsworthy. Preserve the complete article and publication data when it appears. Screenshots taken after pages disappear are harder to authenticate and often omit the author or date required by the regulation.

Patents prove issuance and named inventorship. To prove major significance, add licenses, citations in later patents where relevant, product integration, independent adoption, standards use, technical evaluations, or market consequences. A recommendation letter that merely calls a patent groundbreaking repeats the conclusion USCIS must decide.

Build an evidence ledger before writing the petition

An evidence ledger forces the team to identify what each document proves, who created it, and what corroboration is missing. It also prevents a glossy 900-page filing in which the strongest proof disappears behind duplicate screenshots. One row per factual proposition works better than one row per exhibit.

Use these columns: claimed fact; relevant criterion; exhibit; source; date; independent or interested source; field connection; comparison baseline; attribution to founder; weakness; needed corroboration. A useful row might read: Founder designed fraud model adopted by three unaffiliated banks; original contribution; contracts plus implementation records; customers; 2023-2025; independent; enterprise fraud prevention; adoption by comparable institutions; product records name founder; contracts redact usage detail; obtain customer declarations with permitted metrics.

Then run four tests.

1. Plain-language test: Does the document meet every noun, verb, and modifier in the criterion?
2. Attribution test: Does it prove the founder's work rather than the company's general success?
3. Baseline test: Does it show why the result is unusual among relevant peers?
4. Consistency test: Do dates, titles, amounts, ownership, and field descriptions agree across exhibits?

The ledger also reveals when one strong fact should appear in more than one part of the legal argument. That is acceptable when the fact genuinely answers different questions, but duplicate copies add bulk rather than weight. Cite the same exhibit and explain the distinct inference.

Do not hide contrary facts. If an acquisition ended a product, a company later failed, or a founder shared credit with a team, state the accurate scope. Extraordinary ability does not require a flawless company history. An unexplained contradiction damages credibility far more than a candid limit on one achievement.

Create a second view that groups evidence by time rather than criterion. Put the founder's earliest recognition, major work, outside adoption, later judging, press, and current US plan on one line. This catches a common final-merits weakness: every exhibit may be genuine, yet the record shows one isolated success followed by silence. It also catches date conflicts between media profiles, corporate records, and letters before an officer does.

Source quality belongs in the ledger as a fact, not a feeling. For media, record whether editorial staff or the subject placed the article, how the publisher measures audience, and which source supports that figure. For an award, attach the rules from the awarding body rather than a nominee's description. For compensation, keep the underlying survey methodology and matching job definition. For customer results, separate numbers the customer verified from estimates the company created.

Redaction needs discipline too. Commercial records often contain confidential pricing, customer names, or product details, but blacking out every identifying fact can make an exhibit impossible to connect to the claim. Retain dates, document type, signatories where permitted, relevant obligations, and enough context to authenticate the record. A customer declaration can explain a necessary redaction and confirm the unredacted fact. Counsel should decide whether to submit sensitive material, request appropriate handling, or use another form of proof.

Finally, label the inference as precisely as the document. An investment agreement proves investment terms. It does not, without more, prove acclaim. A customer contract proves a commercial relationship. It does not, without performance records, prove successful adoption. This sentence-level discipline keeps the petition credible when the same exhibit supports several parts of the case.

Recommendation letters should explain mechanisms

A strong letter supplies informed facts that ordinary records cannot explain, then identifies how the writer knows them. A weak letter strings together adjectives and regulatory phrases. Ten weak letters do not become independent evidence merely because ten people signed them.

Choose writers for knowledge, not title alone. A customer can explain why adoption changed an operating result. An investor can describe the diligence and the founder's role in the investment decision. A recognized technical peer can compare a contribution with prior methods. A former executive can document why a role was critical, provided company records corroborate the account.

Each letter should establish the writer's qualifications, relationship to the founder, specific observations, comparison basis, and concrete consequence. If the writer says the founder's system cut review time, the letter should state the starting process, the founder's work, the observed change, and the records behind the number. Avoid drafting every letter in the same voice. Identical structure and recycled praise make independent witnesses look coordinated.

Interested witnesses are not disqualified. Co-founders, employees, and investors often know the work best. Their letters gain weight when independent documents confirm them. A contract, version history, board record, publication, customer metric, or contemporaneous email can turn opinion into a verifiable account.

Give writers the final field definition and an accurate time line, but do not ask them to decide legal eligibility. The useful sentence is not that the founder is among the top percentage in business. It is that the writer compared her method with a defined set, selected it after a stated review, observed a specific result, and saw others adopt it.

Most founder petitions fail at the missing link

Many denials accept the headline facts and reject the conclusion drawn from them. USCIS may accept that a company raised capital but find no proof that the round was nationally recognized, exceptional for its stage, or attributable to the founder. It may accept that the founder led a startup but find that the startup lacks a distinguished reputation. It may accept a patent but find no major significance in the field.

The usual failure points are predictable.

• The petition treats company success as personal acclaim without proving attribution.
• It repeats a small set of achievements under several criteria without meeting each criterion's exact wording.
• It uses self-published rankings, paid awards, sponsored profiles, or unauthenticated audience claims without explaining provenance.
• It compares the founder with the wrong peer group or offers no comparison at all.
• It proves an impressive past in one field and proposes US work in another.

A request for evidence often identifies a repairable documentary gap, but it cannot change whether the founder qualified on the filing date. New letters may explain facts that already existed. A post-filing award, new contract, or later judging assignment generally cannot retroactively create eligibility that was missing when USCIS received the petition. Keep contemporaneous records long before filing.

Volume creates its own failure mode. Officers must review the record, but the petitioner still benefits from precise exhibit labels, complete source information, reliable translations, and short explanations of what each document proves. A table of contents should let a skeptical reader move from claim to primary proof without guessing.

The recommendation to claim every plausible criterion is popular because more boxes appear safer. It can be wrong. A strained claim gives the officer easy adverse findings, distracts from stronger evidence, and may expose inconsistent descriptions. Claim the criteria the record can prove cleanly, then use other achievements where they belong in the overall analysis.

File when the record tells one finished story

A filing date should follow evidence maturity, immigration timing, and counsel's review, not anxiety. Build backward from the proposed field and US work. Confirm that at least three criteria are supported under their exact wording, then test the whole record for sustained acclaim and top-of-field standing. Check visa availability and the separate requirements for adjustment or consular processing before assuming an I-140 approval solves the entire immigration plan.

The intended US work deserves its own evidence. A founder may submit a detailed statement covering the company or projects, role, market, planned activities, expected collaborators, and how that work continues the established field. Contracts and commitments help, but the regulation permits a detailed plan. Generic language about contributing to the American economy is not a plan.

Before filing, have one reader who knows the field challenge the comparisons and one immigration lawyer challenge the legal mapping. Peer communities can help expose where a founder's story relies on insider knowledge; Sisters gives women building companies a place to ask practical US visa questions and get perspective from women who have dealt with the system. That conversation does not replace legal advice, but it can reveal which claims an outsider will not understand.

The final packet should let a reviewer answer five questions without inference: What is her field? What did she personally do? Who outside her company recognized it? How unusual was it among relevant peers? Why will her US work continue the same record? If one answer depends only on a recommendation letter's adjective, the case is not ready.