SBIR and STTR Explained

SBIR and STTR are federal programs that fund research and development at small businesses through competitive, non-dilutive awards. Participating agencies set aside a share of their extramural research budgets. STTR requires a nonprofit research institution partner; SBIR does not. Both are periodically reauthorized.

Current figures — verified 2026-08-11

Item	Value	Source
Authorization end date	September 30, 2031	15 U.S.C. § 638(m)
SBIR set-aside	Not less than 3.2 percent of extramural R&D, agencies above $100 million	15 U.S.C. § 638(f)
STTR set-aside	0.45 percent of extramural R&D, agencies above $1 billion	15 U.S.C. § 638(n)
Phase I award guideline	Up to $323,090, including modifications	SBIR.gov, About
Phase II award guideline	Up to $2,153,927, including modifications	SBIR.gov, About
Employee ceiling	Not more than 500 employees, including affiliates	SBIR.gov, Eligibility FAQ
Ownership and control	More than 50 percent directly owned and controlled by US citizens or permanent residents, or by other qualifying small business concerns	SBIR.gov, Eligibility FAQ
Investor-ownership exception (SBIR only)	Multiple venture capital, hedge fund, or private equity firms may hold the majority if no single firm owns or controls more than 50 percent; electing agencies may direct 25 or 15 percent of SBIR funds	SBA, Guide to SBIR/STTR Program Eligibility
SBIR subcontracting limits	Up to 33 percent of Phase I; up to 50 percent of Phase II	NIH SEED
STTR minimum work split	Small business at least 40 percent; research institution at least 30 percent	NIH SEED
Participating agencies	11 run SBIR; 6 of those also run STTR	SBIR.gov, Participating Agencies
Agency-published award ranges	Phase I roughly $50,000–$300,000; Phase II roughly $400,000–$2 million	SBIR.gov, Participating Agencies
Registration lead time	“It can take 6 weeks or more to complete the registration process”	NIH SEED, Register Your Company
Federal obligations	FY2022: $4.4 billion SBIR, $662.3 million STTR	CRS IF12874
Award concentration	22 firms received 50 or more Phase II awards FY2011–FY2020: under 1 percent of awardees, 10 percent of Phase II funding	GAO, via CRS IF12874

These figures change, and the authorization has a defined end date. Verify against SBIR.gov and the statute before relying on them. Report an outdated figure

Key takeaways

• SBIR and STTR are set-asides inside agency research budgets, not appropriations.
• The authorization expires on a fixed date and must be renewed.
• STTR requires a research institution partner and a minimum work split.
• Topic-driven and investigator-initiated agencies require different proposals.
• Phase III carries sole-source rights but no SBIR money.

What are SBIR and STTR?

SBIR — Small Business Innovation Research — and STTR — Small Business Technology Transfer — are competitive federal programs that fund research and development performed by small businesses. Both are established in Section 9 of the Small Business Act, codified at 15 U.S.C. § 638, and both operate as mandatory set-asides: an agency above a stated extramural research budget must spend a fixed percentage of that budget with small businesses (see current figures above).

The set-aside mechanic is what makes the programs exist. There is no separate SBIR appropriation. Each participating agency carves the required percentage out of research money it already has, writes its own solicitations, picks its own winners, and pays its own awards. Within the wider set of funding tracks by funder type, SBIR and STTR are the largest source of federal research capital available to a for-profit company.

The Small Business Administration does not fund SBIR or STTR. SBA issues the SBIR/STTR Policy Directive governing program structure, eligibility, solicitations, data rights, and reporting across agencies (SBIR.gov, Policies), operates SBIR.gov and the Company Registry, and reports to Congress. Money and topic selection are entirely agency-controlled, and misunderstanding that fact sends more first-time applicants to the wrong place than any other error.

Awards take no equity, no board seat, and no liquidation preference, which is why SBIR and STTR anchor most discussions of non-dilutive capital for startups. Awards are made as grants or as contracts depending on the agency, and that choice carries substantially different compliance obligations.

Are SBIR and STTR permanent programs?

SBIR and STTR are not permanent. The statute contains an explicit termination date for the authority to carry out both programs, and Congress has extended that date repeatedly rather than making the programs permanent (see current figures above; 15 U.S.C. § 638).

The durable structure to plan around is this: SBIR and STTR are periodically reauthorized programs with a statutory sunset. Authorization has lapsed before, and during a lapse agencies cannot make new awards even though existing awards continue on already-obligated funds. Any company building a multi-year plan on SBIR revenue should check the authorization end date at SBIR.gov and in the statute before committing to a sequence of awards that extends past it.

Reauthorization is also where the program’s rules change. Award ceilings, phase mechanics, eligibility conditions, research-security requirements, and per-firm proposal limits have all been modified through reauthorization statutes rather than through routine rulemaking. A solicitation written under one authorization period may not describe the rules of the next.

The practical instruction is narrow and worth stating plainly: never rely on a secondary description of SBIR rules, including this one, for a live submission. Read the agency’s current solicitation and the current SBA Policy Directive, and treat every dollar figure and percentage as something to verify at the source.

What are the three phases of an SBIR or STTR award?

SBIR and STTR are structured in three phases: Phase I funds feasibility, Phase II funds the principal research and development effort, and Phase III is commercialization funded entirely from non-SBIR sources. Phase I and Phase II have published award guidelines and duration norms; Phase III has neither.

Phase I asks whether the technical approach could work. It is a feasibility study with a decision gate, scoped to answer a question rather than to build a product. Phase II funds the main development effort — prototyping, testing, and maturation — and is competitively awarded to Phase I performers, not granted automatically. Award guidelines for both phases are published by SBA and indexed for inflation (see current figures above).

Phase III is the part almost every first-time applicant misreads. Phase III receives no SBIR or STTR funds by definition; it is a status conferred on follow-on work, funded by private capital or by non-SBIR federal money. What it confers is legal position.

SBIR.gov states that “Section 4(c)(2) of the May 2, 2019 SBA SBIR Policy Directive states that a Phase III is by nature an SBIR, and must be accorded SBIR status, including SBIR Data Rights,” that “the right to receive sole-source funding agreements is a key Phase III right,” and that “Phase IIIs are also exempt from SBA’s size standards” (SBIR.gov, Data Rights Tutorial). The sole-source justification is that the new award must “derive from, extend, or complete” prior SBIR effort.

Several agencies also operate alternate entry and bridge paths where authorized: Fast-Track, which reviews Phase I and Phase II together; Direct-to-Phase II, an SBIR-only authority allowing an award to a firm that established feasibility with non-SBIR funds; sequential Phase II awards; and matched supplements requiring investor or customer money alongside federal dollars. The specific solicitation governs availability.

What is the difference between SBIR and STTR?

The core difference between SBIR and STTR is partnership. STTR requires the small business to partner with a nonprofit research institution and to divide the work according to statutory minimums; SBIR requires no partner and allows the small business to subcontract within stated limits. The table below compares the two programs on the dimensions that change how a proposal is built.

Dimension	SBIR	STTR
Research partner	Optional	Required nonprofit research institution
Work split	Subcontracting capped by phase	Statutory minimums for both parties
Principal investigator	Primarily employed by the small business	At most agencies, either organization
Set-aside size	Larger	Substantially smaller
Direct-to-Phase II	Available where authorized	Not available

The small business is always the applicant and always the prime recipient in both programs. A university cannot apply for an STTR award; it participates as the research institution partner. SBIR.gov states the rule directly: “The small business concern is the prime contractor or grantee” (SBIR.gov, Eligibility FAQ).

Work-split and subcontracting percentages, and the principal investigator employment rules, are set by statute and agency policy and are summarized by NIH (NIH SEED). Exact figures appear in the current figures callout above, and one agency applies the SBIR principal-investigator rule to its STTR awards, so the solicitation controls.

STTR is not the easier program. It is a different structure, chosen when the core intellectual property or the essential instrumentation sits in a university or federally funded laboratory. It also imports a second institution’s contracting office, intellectual property policy, and timeline into the proposal schedule.

Which agencies run SBIR and STTR, and how do they differ?

Eleven federal agencies run SBIR and a subset of those also run STTR (see current figures above; SBIR.gov, Participating Agencies). The difference that matters most is not which agency but which of two models it uses: investigator-initiated or topic-driven.

Investigator-initiated agencies — NIH and NSF most prominently — publish broad solicitations and let the applicant define the problem. Selection runs through scientific peer review. The government is not the eventual customer. NIH states this explicitly: “the NIH is generally not the final purchaser of technologies generated through the programs” (NIH SEED). Commercialization therefore means private markets, and there is effectively no procurement path at the end of the pipeline.

Topic-driven agencies — defense, energy, space, homeland security, and transportation — publish numbered topics that encode a specific mission requirement, sometimes down to platform integration constraints. Awards are frequently contracts rather than grants, which imports federal acquisition regulations, cost accounting, and contract data rights clauses; the distinction is covered in grants vs contracts vs cooperative agreements. Here the government is a potential customer, and Phase III sole-source procurement is a real path.

The strategic consequence is that the same technology needs two entirely different documents. An investigator-initiated proposal is argued to scientists and must survive a merit review panel, in the same review culture described in research grants at NIH and NSF. A topic-driven proposal is argued to a program manager who holds a requirement and a budget line, and fit against the stated topic dominates. A strong technology submitted against the wrong topic loses to an adequate technology submitted against the right one.

Who is eligible for an SBIR or STTR award?

Eligibility rests on the definition of a small business concern: organized for profit, with a place of business in the United States, majority owned and controlled by US citizens or permanent residents, and under an employee ceiling that counts affiliates (see current figures above; SBIR.gov, Eligibility FAQ).

Six mechanics decide most real eligibility questions:

• Employee counting includes affiliates. Full-time, part-time, and leased employees all count, with no full-time-equivalent conversion, and the count aggregates across affiliated concerns.
• Affiliation turns on control, including the power to control. A minority investor holding blocking rights over ordinary business decisions can create affiliation, which aggregates employee counts and can disqualify a firm at award.
• Investor ownership is an SBIR-only exception. Majority ownership by multiple venture capital, hedge fund, or private equity firms is permitted at participating agencies only where no single such firm holds control, and only a limited share of an agency’s SBIR funds may go to such firms (see current figures above; SBA Eligibility Guide). No such authority exists for STTR.
• Work must be performed in the United States. The state is irrelevant; the country is not.
• The principal investigator’s employment is a rule. Under SBIR the principal investigator must be primarily employed by the small business, and agencies apply specific thresholds to “primarily.”
• Certification happens at award. The SBA guide notes that certification occurs “at the time of award,” and that “any intentional or negligent misrepresentation within the certification may result in criminal, civil or administrative sanctions.”

Registrations sit on the critical path. SAM.gov and the unique entity identifier, the SBA Company Registry and its control number, Grants.gov, and agency-specific systems each take time to clear, and NIH warns applicants that the full sequence takes weeks rather than days (see current figures above; NIH SEED). Start them before a target solicitation exists, using the process in SAM.gov registration and the UEI. Governance terms negotiated in a seed round can silently terminate eligibility years later, so eligibility belongs on the term-sheet agenda rather than in proposal review.

What do SBIR and STTR reviewers score?

SBIR and STTR review reduces to four dimensions across agencies — technical merit and innovation, feasibility of the approach, the team, and commercial potential — with weighting that differs sharply by agency. Investigator-initiated agencies weight scientific merit most heavily; topic-driven agencies weight fit to the stated requirement and evidence of a transition path.

NSF names three explicit criteria for its SBIR and STTR proposals: Intellectual Merit, Broader Impacts, and Commercial Impact. Under Commercial Impact, NSF asks whether there is a significant market opportunity, whether there is a compelling potential business model, whether the team has the elements suggesting strong commercial outcomes, whether NSF support “will serve as a catalyst to improve substantially the technical and commercial impact of the underlying commercial endeavor,” and — the question that separates strong from weak proposals — “Does the company possess a significant and durable competitive advantage, based on scientific or technical innovation, that would be difficult for competitors to neutralize or replicate?” (NSF SBIR/STTR Merit Review).

Both framings carry drafting consequences. The catalyst question means a proposal reading “we would do this anyway” fails on its face. The durability question is about defensibility rather than novelty — patents, trade secrets, data advantage, regulatory position — so a proposal that only asserts novelty answers the wrong question.

Commercialization plans are read as testable claims, not marketing copy. Reviewers look for named customer segments and named early adopters rather than analyst-report market sizing, an explanation of why the advantage persists, a description of who funds the next stage and at what milestone, and evidence that the team includes someone who has sold something. Success rates vary widely by agency and topic; NIH publishes phase-level success-rate files through NIH RePORT, and SBA publishes program-level data in its annual reports.

What does an SBIR or STTR award actually cost a company?

An SBIR or STTR award costs no equity, but it is not free. The price is paid in cycle time, compliance obligation, and constraints on how the company can be owned, staffed, and located — and those costs are predictable enough to compare against a priced round before committing.

The equity comparison is the honest argument for the programs. A Phase II award at the published guideline (see current figures above) costs zero percent of the company, no board seat, no liquidation preference, and no protective provisions. Raising the same amount at an early-stage valuation typically costs a double-digit percentage, which dilutes further through every later round while the preference stack sits ahead of common stock in every downside outcome.

The costs sit on the other side of the ledger. Proposal preparation consumes weeks of senior technical and executive attention. Submission to award commonly spans several months to the better part of a year, so award money does not solve a near-term cash problem. Ownership, employee-count, and place-of-performance rules constrain the cap table and hiring, and contract-type awards import acquisition and cost accounting requirements a five-person company has never had to satisfy.

Award concentration data frames a real failure mode: a small number of firms hold a disproportionate share of Phase II awards (see current figures above), and a company can optimize for winning awards while never building a business. Award count is not the diagnostic; whether non-SBIR revenue and outside capital are growing year over year is.

Pursue an SBIR or STTR award when the agency’s mission genuinely wants the technology, when the proposed work is work the company would do anyway, and when the prototype the money buys measurably improves the terms of the next raise. Decline when the project is reverse-engineered to fit a topic, when the award would consume the only technical staff with no commercial workstream, or when cash is needed within two quarters — the same logic as any other go/no-go decision.

What goes wrong in SBIR and STTR applications?

SBIR and STTR applications fail in patterns that are mostly structural rather than stylistic. Seven failure modes account for most declines and most withdrawn awards.

• Eligibility defects discovered at award. Affiliation, investor control, principal investigator employment, or foreign ownership problems surface at certification and are not curable inside the cycle.
• Wrong agency or wrong topic. At a topic-driven agency, a mismatch against the published requirement is unrecoverable regardless of technical quality.
• Phase I scoped like Phase II. Phase I answers a feasibility question. A proposal that promises a finished product in a Phase I budget reads as infeasible to reviewers who know the phase.
• No articulated technical risk. A proposal claiming no risk describes work that does not need federal research funding, which fails the catalyst test directly.
• Work-percentage violations. A budget that breaches SBIR subcontracting caps or STTR minimum work splits is administratively non-compliant before review.
• Undisclosed essentially equivalent work. Submitting substantially the same research for funding to more than one agency, or twice to the same agency, without disclosure is a compliance violation defined in the Policy Directive.
• Expired or missing registrations. A lapsed SAM registration or a stale Company Registry record blocks submission on the day it is due.

The most expensive of these is the first, because it is invisible until the award is at stake. Reviewing governance terms, affiliation exposure, and principal investigator employment against the current eligibility rules before the proposal is written costs a few hours and protects the entire effort.

This article is general information about SBIR and STTR structure, not legal, tax, or contracting advice. Eligibility, ownership, and data rights questions for a specific company should be reviewed with counsel and against the current solicitation.

Frequently asked questions

Do you have to win Phase I before applying for Phase II?

Usually, but not always. Phase II is normally competed among Phase I performers. Some agencies operate a Direct-to-Phase II authority, available for SBIR only where authorized, allowing an award to a firm that established Phase I-equivalent feasibility using non-SBIR funds. Availability is set in each solicitation.

Can a company get Phase I from one agency and Phase II from another?

Yes. Cross-agency progression is permitted, and an STTR Phase I awardee may receive an SBIR Phase II and the reverse. Each receiving agency applies its own review process and its own solicitation rules, so the transition is competitive rather than administrative.

Does an SBIR award mean the government will buy the product?

No. At investigator-initiated agencies the government is generally not the customer at all. At topic-driven agencies, Phase III sole-source authority creates a legal path to procurement, but a path is not a purchase order, and follow-on funding frequently takes the form of more research money rather than acquisition.

Can a venture-backed startup win an SBIR award?

Sometimes. The baseline rule requires majority ownership and control by US individuals or other qualifying small businesses. An SBIR-only exception allows majority ownership by multiple investment firms where no single firm holds control, but only at agencies that have elected the authority, and never for STTR.

Who owns the intellectual property from an SBIR project?

The small business retains title to subject inventions under the Bayh-Dole framework, subject to reporting deadlines and government license rights. Separately, SBIR data rights restrict the government’s ability to disclose or use technical data developed under the award for a defined protection period.

How long does it take from submission to award?

Several months to the better part of a year, varying by agency and mechanism. Registrations add lead time before submission is even possible (see current figures above). Treat SBIR and STTR as a planning-horizon capital source rather than a solution to a near-term cash shortfall.

Related topics

• Funding Tracks by Funder Type — the hub covering federal, state, foundation, corporate, and innovation funding
• Non-Dilutive Capital for Startups
• Research Grants at NIH and NSF
• Grants vs Contracts vs Cooperative Agreements
• Federal Grants Explained

Sources

1. Office of the Law Revision Counsel, 15 U.S.C. § 638, Research and development (SBIR and STTR set-asides, program termination date). https://uscode.house.gov/view.xhtml?req=granuleid:USC-prelim-title15-section638&num=0&edition=prelim (accessed 2026-08-11)
2. U.S. Small Business Administration, SBIR.gov, “About the SBIR and STTR Programs” (award guidelines). https://www.sbir.gov/about (accessed 2026-08-11)
3. U.S. Small Business Administration, SBIR.gov, “Eligibility Requirements” FAQ. https://www.sbir.gov/faq/eligibility-requirements (accessed 2026-08-11)
4. U.S. Small Business Administration, SBIR.gov, “Participating Agencies.” https://www.sbir.gov/participating-agencies (accessed 2026-08-11)
5. U.S. Small Business Administration, SBIR.gov, “Policies” (SBIR/STTR Policy Directive). https://www.sbir.gov/about/policies (accessed 2026-08-11)
6. U.S. Small Business Administration, SBIR.gov, Data Rights Tutorial 4 (Phase III rights and Policy Directive § 4(c)(2)). https://www.sbir.gov/tutorials/data-rights/tutorial-4 (accessed 2026-08-11)
7. U.S. Small Business Administration, Guide to SBIR/STTR Program Eligibility. https://www.sbir.gov/sites/default/files/elig_size_compliance_guide.pdf (accessed 2026-08-11)
8. National Institutes of Health, SEED, “Understanding SBIR and STTR.” https://seed.nih.gov/small-business-funding/small-business-program-basics/understanding-sbir-sttr (accessed 2026-08-11)
9. National Institutes of Health, SEED, “Register Your Company.” https://seed.nih.gov/small-business-funding/how-to-apply/before-you-apply/register-company (accessed 2026-08-11)
10. National Institutes of Health, RePORT, “Success Rates” data files (includes SBIR and STTR by phase). https://report.nih.gov/funding/nih-budget-and-spending-data-past-fiscal-years/success-rates (accessed 2026-08-11)
11. U.S. National Science Foundation, America’s Seed Fund, “Merit Review” criteria. https://seedfund.nsf.gov/resources/review/merit-review/ (accessed 2026-08-11)
12. Congressional Research Service, Small Business Innovation Research (SBIR) and Small Business Technology Transfer (STTR) Programs, IF12874, December 15, 2025. https://www.everycrsreport.com/files/2025-12-15_IF12874_da2141a7cd905bd8c7e9925380c4de66989fcb88.html (accessed 2026-08-11)