The FY2027 H-1B lottery took 211,600 registrations and handed out roughly 78,000 slots. Do the maths: about a 37% chance, and that is before you learn the new rule changed the odds based on what your employer pays you, not on who you are. A founder building something genuinely new can lose that draw three years running and never set foot in a US office. The visa never even looked at the work.
The O-1A does the opposite. It looks only at the work. No annual cap. No lottery. No coin toss. And in recent USCIS data the approval rate sits near 94%. The catch is real and we will be honest about it: the evidentiary bar is high, and this is not a backdoor for everyone. But here is the uncomfortable part most Indian founders never hear. If you have raised institutional money, been written about, judged a hackathon or accelerator cohort, or pull a salary well above your sector's median, you may already meet the standard and not know it.
We tell most serious founders to stop praying for the H-1B and start documenting for the O-1A. Here is the full head-to-head, with current 2026 rules.
Why the H-1B is a trap for founders specifically
The H-1B was built for one situation: a US company hires a foreign worker into a defined role and supervises them. That logic breaks the moment you are the company.
Three structural problems hit founders hardest.
It is a lottery, and the odds just got worse for early-stage work. On 29 December 2025, DHS implemented a wage-weighted selection system for the FY2026-2027 cap. Instead of one entry per registration, each beneficiary now gets entries scaled to their Occupational Employment and Wage Statistics (OEWS) wage level. Level IV gets four entries, Level III gets three, Level II two, Level I one. The projected selection probabilities tell the story: roughly 15% for Level I, 31% for Level II, 46% for Level III, and 61% for Level IV. An early founder paying themselves a modest salary sits at the bottom of that ladder by design.
It is employer-dependent. The H-1B requires a petitioning employer with genuine control over your work, the power to hire and fire. When you own and run the company, USCIS scrutinises whether a real employer-employee relationship exists at all. Founders can sometimes structure around this with an independent board, but it is fragile, expensive to defend, and a Request for Evidence is common.
It is cap-subject and slow. The cap is fixed: 65,000 regular plus 20,000 for US master's holders. Registration runs in March, the work cannot start until 1 October even if selected, and if you miss the draw you wait a full year to try again. For a founder with a US customer, a US co-founder, or a US fundraise in motion, that calendar is unworkable.
One note for the record, because the headlines were loud. The widely reported $100,000 H-1B fee aimed at consular cases was struck down by a federal judge on 8 June 2026 as unlawful. That removes one cost, but it does nothing about the lottery, the cap, or the employer dependency. The structural problems remain.
What the O-1A actually asks of you
The O-1A is the visa for a person of "extraordinary ability" in science, education, business, or athletics. The phrase scares people off. It should not. "Extraordinary" in regulation does not mean Nobel laureate. It means sustained acclaim and evidence that you sit near the top of your field.
You qualify one of two ways. Either you have a single major internationally recognised award, or you satisfy at least three of eight criteria. For founders, it is almost always the three-of-eight route. The eight:
- A leading or critical role for a distinguished organisation
- Commanding a high salary or other high remuneration relative to your field
- Membership in associations that demand outstanding achievement to join
- Published material about you in professional or major trade media
- Receipt of national or internationally recognised awards or prizes
- Judging the work of others in your field
- Original scientific, scholarly, or business-related contributions of major significance
- Authorship of articles in professional journals or major media
Read those again as a founder, not as an academic. A "critical role for a distinguished organisation" is your own venture-backed company, or an advisory seat. "High remuneration" can be your founder salary plus equity, benchmarked against sector data. "Judging" is being a hackathon judge, a YC or accelerator mentor, a startup-competition panellist. "Original contributions of major significance" is your product, your patent, your traction, evidenced by investor letters and press. "Published material about you" is the YourStory or Inc42 profile, the TechCrunch mention, the podcast.
USCIS evaluates this in two steps: first whether each piece of evidence meets a criterion, then a final-merits look at whether the whole record shows extraordinary ability. The 2026 policy manual gives extra examples aimed squarely at STEM and business founders. The bar is real, and a thin record fails. But a founder three or four years into a credible, funded company is often closer to it than they assume.
The petitioner question, where founders get this wrong
Here is the myth: "I can't get an O-1 because I don't have a US employer to sponsor me." Half true, and the half that is false matters enormously.
You cannot self-petition an O-1 the way you can an EB-1A green card. A petitioner must file Form I-129 for you. But that petitioner can be your own company. USCIS has clarified that a founder may be the beneficiary of a petition filed by their own US entity, and may even be the majority shareholder, provided a bona fide employer-employee relationship exists. In practice that means real governance: a board, other owners, or a structure with the authority to oversee, and yes to hire and fire, the beneficiary.
If you have not formed the entity yet, or your cap table makes the company a poor petitioner, a US agent can file on your behalf. Agent petitions are explicitly allowed for beneficiaries who are self-employed, working on short engagements, or operating through a foreign employer. This is the mechanism most founders miss entirely. It is the difference between "I have no sponsor" and "I have two valid routes to a sponsor."
O-1 vs H-1B, head to head
| Factor | H-1B | O-1A |
|---|---|---|
| Lottery | Yes. Wage-weighted from FY2027 | None |
| Annual cap | 85,000 (65k + 20k master's) | No cap |
| FY2027 odds | ~37% overall; ~15% at Level I | Not applicable; ~94% approval |
| Who can petition | A US employer with control over you | Your own US company, or a US agent |
| Founder / self-employment fit | Poor. Employer-employee relationship hard to prove | Strong. Built to accommodate it |
| Start timing | 1 October at earliest, if selected | Any time; no cap-season calendar |
| Initial duration | Up to 3 years | Up to 3 years |
| Extensions | Up to 6 years total (extendable with green card in process) | Unlimited 1-year increments to continue the work |
| Premium processing | Yes. 15 business days, $2,965 (from 1 Mar 2026) | Yes. 15 business days, $2,965 (from 1 Mar 2026) |
| Evidentiary bar | Specialty occupation + degree | High. 3 of 8 acclaim criteria |
| Natural green-card path | EB-2 / EB-3, brutal India backlog | EB-1A self-petition, far shorter for India |
The green card argument almost nobody makes out loud
For Indian founders this is the part that should change the decision. The H-1B usually flows into EB-2 or EB-3, where the India-born backlog runs into many years, sometimes a decade or more, because of per-country limits.
The O-1A flows naturally into the EB-1A extraordinary-ability green card. The evidence you assemble for the O-1A is roughly the same evidence the EB-1A demands, three of ten criteria plus sustained acclaim. And critically, EB-1A lets you self-petition with no employer at all, and it sits in the first preference category, which for India-born applicants moves far faster than EB-2 or EB-3. You are not just choosing a better work visa. You are choosing a faster runway to permanent residence.
This is why we frame the O-1A as a campaign, not a form. The O-1 is year one. The EB-1A, built on the same record, is the destination.
So who should not chase the O-1A
We are honest brokers, so the counter-case. If you are an early-stage founder with no funding, no press, no salary to speak of, and no external recognition, the O-1A is not your visa this year. The two-step adjudication will find the gaps. Forcing a thin case wastes money and burns a denial onto your record. For that profile, the right move is often to spend twelve to eighteen months deliberately generating the evidence, the funding round, the press, the judging invitation, the patent filing, and file when the record is real.
And if you genuinely fit a standard specialty-occupation role at a large employer who will sponsor and run the green card for you, the H-1B may still be the cleaner path despite the lottery. The O-1A is not universally better. It is dramatically better for founders, and that is a different claim.
How SaathiVisa thinks about this
We start every founder conversation by auditing the evidence you already have against the eight O-1A criteria, not by selling a filing. Most serious founders are one or two artefacts short, not five. If you are close, we build the petition and the petitioner structure, own company or US agent, and line up the EB-1A on the same evidence. If you are genuinely not close yet, we tell you that, and we tell you exactly what to go earn first.
FAQ
Can a startup founder really sponsor their own O-1 visa?
Not as a true self-petition, but close. You cannot file for yourself the way you can with an EB-1A. However, your own US company can be the petitioner, and USCIS allows you to be the majority shareholder, as long as a genuine employer-employee relationship exists through a board or other governance. Alternatively, a US agent can petition on your behalf, which is the standard route for self-employed or early-stage founders.
How much faster is the O-1A than the H-1B?
The O-1A has no lottery and no cap, so there is no March-to-October waiting game and no annual rejection by chance. Both visas offer premium processing, a 15-business-day decision for $2,965 as of 1 March 2026. The bigger speed gain is downstream: the O-1A feeds the EB-1A green card, which for India-born applicants moves years faster than the EB-2 or EB-3 path most H-1B holders are stuck in.
What if my O-1A case is borderline?
Then do not file it yet. A denial is recorded and can complicate future petitions. The smarter play is to spend the next year deliberately closing the gaps, secure a funding round, earn press coverage, accept a judging or advisory role, file a patent, then file from strength. A borderline case forced through is the most expensive mistake we see founders make.