Public funding · United States
SBIR and STTR
Federal agencies pay small US companies to do research they need, without taking equity. The programs are large, slow and rule-bound, and eligibility depends on who owns your company.
The Small Business Innovation Research (SBIR) and Small Business Technology Transfer (STTR) programs are the largest source of non-dilutive early-stage funding in the United States. Federal agencies pay small companies to do research and development the agency cares about. The government takes no equity, and the company generally keeps the rights to its inventions. In exchange, founders accept a slow, rule-heavy process and a narrow definition of who is eligible.
This page explains how the programs are structured, who qualifies, how venture ownership changes the picture, and the mistakes that most often sink an application. It applies to US companies only.
How the programs work
Congress requires every federal agency with an extramural research budget above $100 million to set aside a percentage of it for SBIR, and every agency above $1 billion to do the same for STTR. As of today, SBIR.gov lists eleven agencies running SBIR programs, including the Department of Defense (also styled the Department of War), the Department of Health and Human Services (mainly the National Institutes of Health), NASA, the Department of Energy and the National Science Foundation, and six running STTR. SBIR.gov markets the two programs together as "America's Seed Fund".
The Small Business Administration (SBA) coordinates the programs. It issues the Policy Directive the agencies must follow, keeps the company registry and the award database, and reports to Congress. The SBA does not make SBIR or STTR awards itself. Each agency writes its own solicitations, reviews its own proposals and signs its own funding agreements, which may be grants (common at NIH and NSF) or contracts (common at the Department of Defense). The difference matters: a contract has deliverables the agency wants, while a grant funds research the applicant proposed within a topic.
The three phases
Both programs use the same three-phase structure. Phase I tests feasibility, Phase II funds the main research and development, and Phase III is commercialization paid for by someone else.
| Phase | Purpose | Typical length | Size limit before an SBA waiver is needed |
|---|---|---|---|
| Phase I | Establish technical merit, feasibility and commercial potential | About 6 months (SBIR) or 1 year (STTR); agencies may allow longer | $335,937 |
| Phase II | Continue the R&D begun in Phase I; generally open only to Phase I awardees | About 2 years | $2,239,578 |
| Phase III | Commercialize, or sell to the government, using non-SBIR/STTR money | Open-ended | No SBIR/STTR funds |
Size limits: SBIR.gov states that as of 16 October 2026 agencies may award up to these amounts (including modifications) without SBA approval. Most agencies publish their own, lower, typical award sizes in each solicitation; the solicitation figure is the one that applies to you.
Several agencies may make a Phase II award to a company that did not receive a Phase I award, if it can show feasibility another way ("Direct to Phase II"); the 2026 law extended this flexibility through 2031 and widened the list of agencies that may use it. Phase III has a quieter advantage: agencies can award follow-on contracts that build on SBIR/STTR work without a new competition, which is why defense-focused companies value the status.
Who is eligible
Eligibility is judged at the time of award, not when you apply, and it is checked against SBA regulations at 13 CFR 121.702. The main tests:
- The company is a for-profit business located in the United States. Nonprofits cannot receive awards directly.
- It is more than 50% directly owned and controlled by US citizens or permanent residents, by other small businesses that are themselves majority-owned that way, or by certain tribal and native entities, or a combination of these. The venture-capital exception is described below.
- Together with its affiliates, it has no more than 500 employees.
- For both phases, the principal investigator's primary employment (more than half their time) is with the company, or with the research institution for STTR. A professor with a full-time university job cannot be the SBIR principal investigator without a written deviation.
- The research is performed in the United States, except in rare approved cases.
- For SBIR Phase I, the company performs at least two-thirds of the research; for Phase II, at least half.
- The company has a Unique Entity ID from SAM.gov and has registered in the SBA Company Registry on SBIR.gov.
Ownership is measured on a fully diluted basis. The SBA counts options, warrants, the unallocated option pool and convertible securities as if they had converted. A SAFE or note held by a foreign investor or a fund can therefore change the ownership arithmetic before it converts. Run the numbers on your cap table before every award, not just the first.
STTR and the research-institution partner
STTR exists to move research out of universities and labs. Every STTR project is a formal collaboration between the small business and a single partnering research institution: a US nonprofit college or university, a domestic nonprofit research organization, or a federally funded research and development center (FFRDC).
- The small business must perform at least 40% of the work and the research institution at least 30%. These percentages are fixed in statute and cannot be waived.
- The two parties must sign an agreement allocating intellectual property rights and the right to carry out follow-on work.
- The principal investigator may be employed by either party, which is why STTR suits a faculty founder who has not yet left the university.
Negotiating the IP agreement with a university's technology transfer office can take longer than writing the proposal, so start early.
How venture funding interacts with eligibility
A typical seed round leaves founders in majority control and creates no problem. Trouble starts when funds collectively own more than half the company, which can happen after a Series A or earlier on a fully diluted basis.
- SBIR only. A company that is more than 50% owned by several venture capital operating companies, hedge funds or private equity firms (in any combination) can still win SBIR awards, but only from agencies that have chosen to use this authority. The rules cap such awards at 25% of an agency's SBIR funds at NIH, DOE and NSF and 15% at other agencies. No single fund may own more than 50% unless that fund itself qualifies as a US-owned small business.
- Such a company must register as majority-owned by multiple funds in the SBA Company Registry before it submits the proposal, and include a specific certification with it.
- STTR has no venture-owned path. An STTR awardee must meet the individual-ownership test.
- Shared investors do not by themselves make you an affiliate of a fund's other portfolio companies, so a fund's portfolio usually does not count toward your 500-employee limit. Affiliation can still arise from control, for example where one investor holds more than 50% of the voting equity (the SBA may also find control at 40% or more), or where the same investor controls your board and another company's.
The 2026 law also created "strategic breakthrough" awards. Agencies with large SBIR budgets may use up to 0.50% of their extramural R&D budget for Phase II awards of up to $30 million, over no more than 48 months, to companies that already hold a prior Phase II award and can show matching funds of at least 100% from new private capital or other new government funding. These are for later-stage companies, but they make prior SBIR history worth more to investors.
Solicitations, topics and timelines
SBIR and STTR do not fund unsolicited ideas. Each agency publishes solicitations (also called funding opportunity announcements, broad agency announcements or requests for proposals) listing topics, and a proposal must answer a specific topic in an open solicitation. Agencies differ widely:
- Defense components publish narrowly specified topics describing a capability they want, often several times a year.
- NIH and NSF publish broad research areas, and their reviewers judge scientific merit and commercial potential.
- Some agencies run one solicitation a year; others run several. SBIR.gov keeps a topic search and a funding planner, but its copies may lag the agency's official solicitation, which always governs.
From fiscal year 2027, the 2026 law requires each agency's program director to cap the number of Phase I and Phase II proposals any one company may submit, by fiscal year, solicitation or topic. Repeat applicants should check the cap before planning a batch of proposals. Companies with many past awards must also meet SBA's commercialization and Phase I-to-Phase II transition benchmarks to remain eligible for new Phase I awards.
Plan for a long cycle. SBIR.gov says review times vary by agency; between a solicitation opening and money arriving, several months is normal, and Phase II follows only after Phase I results. A startup that needs cash in the next quarter should not count on an SBIR award to provide it.
What the award data shows
The figures below summarize every SBIR and STTR award in SBIR.gov's public award file, downloaded on 2026-10-31. Years are federal fiscal years (October to September) as recorded by agencies; agencies keep adding awards after a year ends, so the most recent year can still grow.
| Fiscal year | Awards | Phase I | Phase II | Total awarded | Firms |
|---|---|---|---|---|---|
| FY2021 | 6,881 | 4,495 | 2,386 | $3.84 billion | 3,973 |
| FY2022 | 6,639 | 3,888 | 2,751 | $4.49 billion | 3,738 |
| FY2023 | 6,325 | 3,790 | 2,535 | $4.72 billion | 3,670 |
| FY2024 | 6,412 | 3,811 | 2,601 | $4.91 billion | 3,776 |
| FY2025 | 5,340 | 3,014 | 2,326 | $4.11 billion | 3,283 |
Across FY2021 to FY2025, the database records 31,597 awards worth $22.06 billion to 10,167 distinct firms. In FY2025 the median Phase I award was $198,986 and the median Phase II award $1.25 million.
| Agency, FY2021–FY2025 | Awards | Total awarded | of which STTR |
|---|---|---|---|
| Department of Defense | 16,385 | $12.14 billion | $1.65 billion |
| Department of Health and Human Services | 6,471 | $6 billion | $773.01 million |
| Department of Energy | 2,736 | $1.45 billion | $167.53 million |
| National Science Foundation | 2,110 | $1.03 billion | $138.24 million |
| National Aeronautics and Space Administration | 2,367 | $941.37 million | $122.07 million |
| Department of Agriculture | 582 | $161.69 million | $11.07 million |
| Department of Commerce | 300 | $87.43 million | $0 |
| Department of Homeland Security | 182 | $79.55 million | $0 |
| Department of Transportation | 175 | $73.24 million | $0 |
| Department of Education | 123 | $66.25 million | $0 |
| Environmental Protection Agency | 166 | $28.64 million | $0 |
Source: SBIR.gov award data, U.S. Small Business Administration, file dated 2026-10-21, downloaded 2026-10-31. Public domain (U.S. federal government work). Dollar amounts are as reported by agencies and are not adjusted for inflation; Phase II includes Direct to Phase II awards; firms are counted by Unique Entity ID.
Common mistakes
- Registering late. SAM.gov, the SBA Company Registry and the agency's own portal each need registering, and SAM.gov can take weeks. Do them before the solicitation opens.
- Proposing what you want to build rather than what the topic asks for. Reviewers score against the topic. A strong technology that does not answer the topic loses.
- Ignoring the cap table. A round that tips ownership past 50% to funds, or a foreign investor's convertible, can make the company ineligible at the moment of award. Check fully diluted ownership before each award, and keep your data room ready to show it.
- A principal investigator who does not meet the primary-employment rule, usually a founder with another full-time job.
- Submitting essentially equivalent work to two agencies and accepting both. Duplicate funding for the same work is fraud, not double luck. Disclose overlapping proposals.
- Treating the award as unrestricted cash. Awards come with cost principles, reporting and audit rights, and contracts can require an accounting system the agency will review. Budget for compliance.
- Underestimating research-security review. Agencies now screen applicants for foreign ties, and the 2026 law added specific restricted-entity lists. Expect detailed disclosure questions about foreign investors, contracts and personnel.
SBIR money rarely replaces a venture round, but it can extend runway without dilution and give a technical team validated results before it raises. For the other SBA programs, see SBA programs for startups.
Sources
- About SBIR and STTR, U.S. Small Business Administration (SBIR.gov). Public domain (U.S. federal government work).
- Frequently Asked Questions, U.S. Small Business Administration (SBIR.gov). Public domain (U.S. federal government work).
- How to Apply, U.S. Small Business Administration (SBIR.gov). Public domain (U.S. federal government work).
- SBIR/STTR Policy Directive, U.S. Small Business Administration (SBIR.gov). Public domain (U.S. federal government work).
- 13 CFR 121.702: size and eligibility standards for the SBIR and STTR programs, Electronic Code of Federal Regulations (eCFR). Public domain (U.S. federal government work).
- S. 3971, Small Business Innovation and Economic Security Act (enrolled bill), U.S. Government Publishing Office (GovInfo). Public domain (U.S. federal government work).
- Administrator Loeffler Applauds SBIR-STTR Reauthorization (13 April 2026), U.S. Small Business Administration. Public domain (U.S. federal government work).
- Funding Opportunities (topics and solicitations), U.S. Small Business Administration (SBIR.gov). Public domain (U.S. federal government work).
- SBIR.gov award data and data dictionary, U.S. Small Business Administration (SBIR.gov). Public domain (U.S. federal government work).