SBIR and STTR are sister programs that fund the same thing -- early-stage R&D at small businesses -- with the same award amounts, the same review criteria, and the same three-phase structure. The difference is structural: SBIR is designed for companies that do the work in-house, while STTR is designed for companies that need a research institution partner.
Most startups should default to SBIR. STTR exists for a specific situation -- when your R&D genuinely requires university resources, equipment, or faculty expertise that you can't replicate in your own shop. If that's you, STTR is the right path. If it's not, STTR adds partnership overhead with no funding upside.
This guide explains the real differences, when STTR is worth the complexity, and how to decide.
The core difference: who does the work
The fundamental distinction between SBIR and STTR is how R&D work gets allocated between your company and outside partners.
| SBIR | STTR | |
|---|---|---|
| Research institution partner | Optional | Required |
| Phase I: minimum work by your company | 67% (two-thirds) | 40% |
| Phase I: minimum work by research institution | N/A | 30% |
| Phase II: minimum work by your company | 50% | 40% |
| Phase II: minimum work by research institution | N/A | 30% |
| Remaining work allocation | Flexible (subs, consultants) | 30% flexible |
In plain terms: SBIR says "you do most of the work." STTR says "you and a university do the work together, with a defined split."
This has real implications for your budget. On an STTR, at least 30% of your Phase I funding goes to the research institution through a subcontract. On a $305,000 NSF Phase I, that's roughly $92K that flows to the university. But here's what most guides don't mention: universities charge facilities and administrative (F&A) rates on subawards, typically 50-60% of modified total direct costs. On that $92K subaward, $30-35K may go to university overhead before a dollar reaches the lab. Your effective research budget from the university partnership is closer to $58-62K in direct research costs. If you don't genuinely need that university work, that's a significant chunk of your grant paying for a partnership you didn't need.
PI employment: the rule that matters most
For most startup founders, the PI (Principal Investigator) requirement is where the SBIR vs. STTR decision actually gets made.
SBIR rule: The PI must be primarily employed by the small business -- meaning more than 50% of their total working time. A faculty member with a full-time university appointment cannot serve as PI on an SBIR.
STTR rule: At most agencies (DOD, NIH, DOE, NASA), the PI can be primarily employed by either the small business or the research institution. This means a university professor can lead the project while staying on faculty.
The exception: NSF requires the PI to be primarily employed by the small business even on STTR proposals. If you're targeting NSF and your PI is university faculty, this is a problem.
The adjunct/part-time edge case: "Primarily employed" means more than 50% of total working time. If your technical lead holds a part-time or adjunct faculty appointment but spends the majority of their time at the startup, they likely satisfy SBIR's PI requirement already. In that case, you may not need STTR at all -- you can apply SBIR and subcontract the university work separately. Check with the specific agency's program office if the split is close to 50/50.
| Agency | STTR PI can be at university? |
|---|---|
| DOD (AFWERX, Navy, Army, DARPA) | Yes |
| NIH | Yes |
| DOE | Yes |
| NASA | Yes |
| USDA | Yes |
| NSF | No -- same as SBIR rules |
The practical test: If your technical lead is a university professor who isn't ready to leave their appointment, STTR at a non-NSF agency is probably your only path. If your PI is already full-time at the startup, SBIR is simpler and gives you more control over the work.
Funding: identical amounts, different pool sizes
SBIR and STTR award the same dollar amounts at every agency. The SBA Policy Directive sets caps that are adjusted for inflation annually:
- Phase I: up to $314,363 (as of FY2025; SBA adjusts annually)
- Phase II: up to $2,095,748 (as of FY2025; SBA adjusts annually)
Each agency sets its own limits within those caps, and those limits are the same for both programs. NSF pays $305,000 for Phase I regardless of whether it's SBIR or STTR. NIH pays up to $314,363 for both.
The difference is in how much total money is available:
| SBIR | STTR | |
|---|---|---|
| Set-aside | 3.2% of extramural R&D | 0.45% of extramural R&D |
| Annual funding | ~$4.4 billion | ~$660 million |
| Share of total | ~87% | ~13% |
| Participating agencies | 11 | 6 |
STTR is a much smaller pool. But it's also a smaller applicant pool -- roughly 13-15% of total SBIR/STTR applications go to STTR (per SBA annual report data). The competition isn't necessarily easier, but you're competing against a different set of applicants (typically university-affiliated teams rather than pure startups).
Which agencies offer STTR?
Only 6 of the 11 SBIR agencies participate in STTR. The threshold is a $1 billion extramural R&D budget:
| Agency | SBIR | STTR |
|---|---|---|
| DOD (AFWERX, Navy, Army, DARPA, etc.) | Yes | Yes |
| HHS / NIH | Yes | Yes |
| DOE | Yes | Yes |
| NSF | Yes | Yes |
| NASA | Yes | Yes |
| USDA | Yes | Yes |
| EPA | Yes | No |
| DOC (NIST, NOAA) | Yes | No |
| DHS | Yes | No |
| DOT | Yes | No |
| ED (Education) | Yes | No |
If your best-fit agency is EPA, NIST, DHS, DOT, or Education, STTR isn't an option. You'll apply through SBIR regardless.
The STTR partnership: what's actually required
STTR requires two things before you can receive an award:
1. A subcontract with the research institution
Your company is the prime awardee. The research institution is a subrecipient. You'll need a formal subcontract that specifies the research institution's scope of work, budget, deliverables, and timeline. The research institution must perform at least 30% of the R&D.
Eligible research institutions include:
- Universities and colleges
- Nonprofit research organizations (501(c)(3) with research missions)
- Federally funded research and development centers (FFRDCs)
2. An intellectual property allocation agreement
This is the statutory requirement that catches people off guard. Before receiving an STTR award, both parties must sign an IP agreement that spells out:
- Who owns what IP generated during the project
- Rights to carry out follow-on research
- Rights to commercialize the results
University tech transfer offices move slowly. If you start negotiating this agreement the week before the proposal deadline, you're going to miss it. Start the IP conversation at least 6-8 weeks before submission.
Why this matters more than you think: Under SBIR, the small business retains IP rights by statute (via the Bayh-Dole Act). Under STTR, those rights are subject to the IP allocation agreement you negotiate with the university. If the university's tech transfer office negotiates aggressively, you could end up with weaker commercialization rights than you'd have under SBIR -- including complications for Phase III sole-source contracts, where clear IP ownership is critical.
Common IP negotiation sticking points:
- Universities often want to retain rights to publish research findings (standard, generally manageable)
- Some university IP policies claim ownership of inventions made using university resources -- negotiate this explicitly
- Exclusive commercialization rights for the startup need to be spelled out clearly
- Background IP (what each party brings in) vs. foreground IP (what gets created during the project) should be defined
The decision framework: SBIR or STTR?
For most startup founders, the decision comes down to three questions:
1. Does your PI work at a university?
If your technical lead is a faculty member who won't (or can't) leave their university position, STTR is likely your path. SBIR's primary employment requirement would disqualify them as PI.
If your PI is already full-time at the startup, use SBIR. Don't add partnership complexity you don't need.
2. Does your R&D genuinely require university resources?
Some projects need things that only a research institution can provide:
- Specialized lab equipment (biosafety level 3 labs, particle accelerators, clean rooms)
- Patient cohorts or clinical trial infrastructure
- Unique datasets or biological repositories
- Faculty expertise in a narrow scientific domain
If your R&D requires these and you can't build or lease the capability yourself, STTR formalizes that partnership in a way reviewers understand.
If you could do the work in-house (or with a simple consulting arrangement), SBIR keeps things simpler. You can still subcontract university work on an SBIR -- you just can't exceed the work allocation limits (one-third in Phase I, one-half in Phase II).
3. Is the partnership real or manufactured?
Reviewers can tell the difference between a genuine research collaboration and a university name stapled to a proposal for credibility. If the research institution's contribution is vague or could easily be done by your team, you'll get dinged for it.
The best STTR proposals show clear complementarity: the startup brings commercial expertise, market knowledge, and development capability. The research institution brings scientific infrastructure, domain expertise, or access to something the startup can't replicate. The 40/30 split should feel natural, not forced.
Common scenarios
University spinout, founder still on faculty: STTR. This is the textbook case. Apply to a non-NSF agency so the faculty member can serve as PI. Use Phase I to build feasibility data while transitioning the founder to the startup.
Startup needs access to university lab equipment: STTR if the lab work is 30%+ of the project. SBIR with a university subcontract if it's less than 30% and your PI is already at the startup.
Startup wants a university name for credibility: Don't use STTR for this. Reviewers see through it. If you want academic validation, get a letter of support or advisory board member instead.
Startup applying to EPA, NIST, DHS, DOT, or Education: SBIR only. These agencies don't have STTR programs.
Phase I feasibility needs university work, Phase II is all in-house: Start with STTR Phase I, then switch to SBIR Phase II. This is common and perfectly fine -- each application is evaluated independently.
Faculty co-founder has fully transitioned to the startup: SBIR. If the PI is now primarily employed by the company, STTR's PI flexibility doesn't help you, and you're giving up 30% of your budget to a research institution you may not need.
What most startups should do
Default to SBIR. It's simpler, gives you more control over your budget, and doesn't require an IP agreement or formal research institution partnership. The vast majority of startup founders applying for grants -- 85-87% of all SBIR/STTR applications -- go through SBIR.
Switch to STTR when:
- Your PI is university faculty and can't meet SBIR's employment requirement
- Your R&D genuinely requires research institution resources (not just "nice to have")
- You're spinning out of a university lab and the partnership is natural
- The IP agreement terms are workable (check before committing)
Don't use STTR to:
- Add academic credibility you haven't earned
- Access a "less competitive" pool (it's not meaningfully less competitive)
- Force a partnership that doesn't serve the science
The right program is the one that matches how your R&D actually works. If you're doing the work in-house, that's SBIR. If your R&D is genuinely collaborative with a research institution, that's STTR. Don't overthink it.
Quick eligibility check
Both SBIR and STTR share the same basic eligibility requirements. Before you spend time choosing between programs, confirm you qualify:
- For-profit US small business (under 500 employees)
- At least 51% owned by US citizens or permanent residents (or other qualifying small businesses)
- PI primarily employed by the small business (SBIR) or by the SBC or research institution (STTR, at most agencies)
- Registered in SAM.gov and the SBA Company Registry
If you meet these, you're eligible for both programs. The SBIR vs. STTR choice is about structure, not eligibility. For a deeper dive, see our SBIR eligibility requirements guide.
Now that you've decided
Once you know which program fits, the next step is finding the right agency and solicitation for your technology:
- New to SBIR? Start with our complete SBIR guide for an overview of how the program works across all 11 agencies
- Know your agency? Jump to our agency-specific guides: NSF SBIR Project Pitch, AFWERX SBIR, or DARPA BAAs
- Not sure if grants are right for you? Read Should Your Startup Apply for Government Grants? for an honest framework
- Want a personalized assessment? Book a strategy review and we'll map your technology to the best-fit programs