What are the U.S. talent visa criteria? Ten for EB-1A and eight for O-1, in plain language

A head of engineering listed his grounds for EB-1A in half a minute on a consultation: a “best manager of the year” award from his employer, membership in a developers’ association, a salary above the Moscow average. Three of ten criteria — time to file. Twenty minutes later it was clear the officer would count none of them: the award was internal, the association admits people for a fee, and “above the Moscow average” is not a comparison with highly paid specialists in the field. He did have four real grounds — he just did not know they were called criteria. We will walk through the ten criteria the way an officer reads them, then the eight for O-1, then ten questions so you can count yourself honestly.

In brief

• EB-1A needs three of ten criteria, or one major international award; O-1A — three of eight; O-1B in the arts — three of six.
• Three counted criteria are the first step. The second is an assessment of the record as a whole: do the proofs show sustained national or international acclaim. Petitions assembled to tick boxes fail there.
• Three criteria are most often closed only on paper: awards, membership and high remuneration. The reason is one — recognition inside an organisation instead of recognition outside it.
• If a criterion does not apply to the occupation, it can be replaced with comparable evidence.

Why is “three of ten” not what it looks like?

Criteria are not a checklist. They are a way to show the officer acclaim. The rule does ask for three of ten, but since 2010, after Kazarian, the assessment follows the Kazarian two-step analysis. First the officer checks whether the evidence matches the wording of the criterion — a mechanical first step. Then the officer looks at all the evidence together and answers the main question: does it show that you sit among the small number of people at the top of the field, and that the acclaim is sustained, not a one-off. That second step is the final merits determination, and it is where petitions with three criteria closed on minimal evidence are denied.

The practical point: what matters is not the number of criteria, but the quality of each and whether they add up to one story. Four strong criteria beat six weak ones. Now the criteria themselves, in the order they appear in the rule, and with two questions on each: what counts, and what does not.

The ten EB-1A criteria: what does USCIS count, and what does it not?

1. Awards. National or international prizes for achievement in the field. What counts: trade awards with a known jury and open criteria, state honours, international contests with selection, team awards — if your role on the team is proven. What does not: “employee of the year” and other employer awards, participant diplomas, paid rankings with no selection, awards of local reach — regional contests read weaker than national ones. The officer asks: who gives it, who selects, who else has received it.

2. Membership in associations. Membership in associations that require outstanding achievements and where recognised experts decide on admission. What counts: Senior Member and Fellow levels of major professional societies, academies with election, guilds with portfolio selection. What does not: membership for a fee, automatic membership by diploma, clubs with no charter setting criteria. The proof is the charter or the admission rules, not the certificate: the officer reads the criteria for candidates.

3. Published material about you. Material about you and your work in professional, major trade or major general media. What counts: interviews, profiles, write-ups of your project with your name in the headline, in outlets with an editorial staff and a documented audience. What does not: press releases, articles you wrote yourself, name-checks in a list, pieces tagged “advertorial”, publications on sites with no audience. Attach outlet data to each piece: circulation, traffic, editorial staff. More on this in “How does an expert or founder get into the media?”.

4. Judging. Taking part in judging the work of others — alone or on a jury. What counts: peer review for journals and conferences, juries of contests and awards, grant expert panels, editorial boards, judging hackathons with a documentary trail. What does not: an invitation with no work performed, evaluating subordinates as a manager, “judging” inside your own company. The proof is the invitation plus confirmation of work done: a review, a letter of thanks from the editors, minutes.

5. Original contribution. Scientific, applied, artistic, sporting or business achievements of major significance to the field. The most demanding criterion: what is needed is not the result but its influence — adoption by others, citation, use of the method in the field, licensing of a patent, a change in practice. What does not: a patent with no use, a product with no external proof of significance, letters that the contribution is “important” with no facts on who uses it and how.

6. Scholarly articles. Authorship of articles in professional, trade or major media. What counts: articles in peer-reviewed journals and conference proceedings, expert pieces in trade outlets with an editorial staff; citations strengthen the record. What does not: blog and social-media posts, articles on platforms with no editorial staff, texts inside your own company. A degree is not needed for this criterion — an editorial staff on the other side is.

7. Exhibitions. Display of your work at artistic exhibitions and showcases. A criterion for art and design: galleries, museums, festivals, biennales, international exhibitions with selection. For engineers and scientists it usually does not apply — and then the comparable-evidence rule works: conferences and demonstrations of work, with an explanation of why the standard criterion does not fit.

8. Critical role. A leading or critical role in an organisation with a distinguished reputation. Two parts, and both are needed: prove the organisation’s reputation — awards, rankings, published material about it, scale — and prove that your role was critical: you were responsible for a result the organisation would not have reached without you. What does not: a title with no description of influence, a role in a company with no proven reputation, letters from colleagues about what a valuable employee you are.

9. High remuneration. A salary or other remuneration substantially above that of others in the field. The key word is “others”: the comparison is with highly paid specialists in your field on independent statistics, not with the average salary in the country or the city. What does not: a comparison with the average, company income instead of personal, a founder’s revenue with no split from costs. The evidence is income statements and official salary statistics for your role.

10. Commercial success in the performing arts. Box-office receipts, record sales, streaming, print runs — for musicians, actors, directors. For other occupations it does not apply and is replaced with comparable evidence: product sales, audience, project revenue — with an explanation of the analogy.

The ten criteria have been read. The officer does not line them up in this order: three of them the officer sees closed on paper more often than the rest.

Three criteria that are most often closed only on paper

Awards. Almost every petition we review after someone else’s denial has an employer award or a paid ranking. People bring them first because they look like awards. The officer cuts them with one question: who, besides your organisation or the organiser, knows about this award?

Membership. A membership certificate looks convincing until the officer opens the charter and sees that admission requires paying a fee. The difference between “member” and “Senior Member” is the difference between zero and a closed criterion, and it usually costs a year of work on the application.

High remuneration. The most common failure among strong professionals: the income is genuinely high, but the petition compared it with the national average. The officer expects a comparison with the upper end of the market in your occupation — and if on that scale you sit in the top ten percent, the criterion works; if you compared with the average, it does not count even at a large figure.

What the three share is this: recognition inside an organisation was presented as recognition outside it. That is why people with four genuine external criteria in the petition get through, and people with six internal ones do not. Now the same logic for O-1.

How do the eight O-1A criteria and the six O-1B criteria differ?

The O-1A criteria almost repeat EB-1A word for word — without exhibitions and commercial success; the bar is lower in practice because the visa is temporary and what is assessed is whether the grounds suffice for the specific work. An extra O-1 condition: the evidence must relate to the field of the forthcoming work. O-1B for the arts is built differently: not extraordinary ability, but distinction — a high degree of skill and recognition.

For those planning the sequence “O-1 now, EB-1A later,” the table is a working tool: evidence is gathered once against the shared rows, and the difference is made up in a year of work in the United States. How O-1 is structured and who files the petition — in “The O-1 visa to the United States”; how EB-1A differs from the other first-preference categories — in “What is the EB-1 visa”.

A self-check: ten questions in place of ten criteria

Answer “yes” only if the answer is backed by a document you can attach to the petition today.

• Do you have an award given neither by your employer nor by an organiser for a fee — with a known jury and open criteria?
• Do you belong to an association whose charter requires achievements for admission, and where experts decide?
• Has an interview or a profile about you personally run in an outlet with an editorial staff and a verifiable audience?
• Have you judged other people’s work — as a reviewer, a jury member, an expert — and is there a document of that?
• Do other organisations use your method, product or result — and can that be shown in words that are not yours?
• Have you published articles in outlets with an editorial staff — journals, conference proceedings, trade media?
• Have your works been shown at exhibitions or festivals with selection? If your field is not art — skip.
• Have you held a role without which a known organisation would not have reached a result — and can you prove both the organisation’s reputation and your role?
• Is your income higher than that of highly paid specialists in your field, on independent statistics?
• Do you have measurable commercial success in the performing arts? If not — skip.

How to read the result. Four or more “yes” answers with documents — there are grounds for EB-1A; the rest is the strength of the evidence and the story that holds it together. Exactly three — too early to file: at the second step of the assessment three minimal criteria are usually not enough; each must be strengthened or a fourth added. One or two “yes” answers, but a strong project for the United States — look toward EB-2 NIW. Many “yes” answers, but on questions where you hesitated at the word “document” — the most common case: the acclaim is there, the trail is not, and it takes several months to build. What is built, and through which tools — in the PR catalogue: every item there is tied to a criterion number from this article.

The engineering lead from the first paragraph answered “yes” to questions 4, 5, 6 and 8 after the review: he had reviewed papers for a trade conference, three outside companies used his tool, he had two articles in a professional journal, and a critical role in a product the press wrote about. He named none of those four on the consultation — they were not in his picture of “the criteria.” You can check your chances in two minutes in the quiz below; if it comes out mixed — we will go through the documents on a consultation, not the words.