Yes, if your startup has genuine R&D to do and your technology aligns with a federal agency's mission. Government grants -- particularly SBIR -- provide $150K to $2M+ in non-dilutive funding you never repay and that costs you zero equity. But they take 4-9 months, require serious proposal effort, and aren't right for every company.
We've written 500+ grant proposals across 30+ federal agencies. Some of those companies never should have applied. Others left millions on the table by not applying sooner. Here's how to tell which camp you're in.
What government grants actually exist for startups
Before you decide whether to apply, you need to know what's actually out there. The landscape is broader than most founders realize -- and narrower than most listicles suggest.
SBIR/STTR (Small Business Innovation Research) is the flagship. Eleven federal agencies set aside a combined $4B+ annually to fund R&D at small businesses. This is the big one for tech startups. Phase I awards range from $150K-$315K, Phase II from $650K-$2.1M, and Phase III is a sole-source government contract with no funding ceiling.
But SBIR isn't the only game:
| Funding Type | Typical Amount | Dilution | Repayment | Best For |
|---|---|---|---|---|
| SBIR/STTR grants | $150K-$2.1M per phase | None | None | Deep tech R&D aligned with a federal mission |
| State innovation programs | $25K-$250K | None | None | Early-stage companies in eligible states |
| Foundation grants | $50K-$500K | None | None | Social impact + technology (limited for-profit eligibility) |
| IRAP (Canada) | Up to $1M CAD | None | None | Canadian SMEs doing technology-driven innovation |
| R&D tax credit (US) | Up to $500K/year against payroll tax | None | N/A (credit) | Any startup with qualified R&D expenses |
| SR&ED tax credit (Canada) | Up to $2.1M/year refundable | None | N/A (credit) | Canadian companies with eligible R&D |
Most of this guide focuses on SBIR because it's the largest, most accessible, and highest-ROI program for US tech startups. If you're Canadian or cross-border, see our IRAP vs. SBIR comparison.
SBIR by the numbers
Numbers first, narrative second. Here's what the program actually looks like across major agencies.
Award amounts by agency
| Agency | Phase I | Phase II | What They Fund |
|---|---|---|---|
| NSF | $305,000 | $1,000,000 | Broad science and tech (AI, biotech, materials, climate) |
| NIH | Up to $314,363 | Up to $2,095,748 | Biomedical, health tech, diagnostics, therapeutics |
| DoD (AFWERX) | $50K-$250K | $750K-$1.7M | Air Force dual-use technology |
| DoD (Navy/Army) | Up to $250,000 | Up to $1,700,000 | Military-specific technology |
| DOE | $250,000 | $1,600,000 | Energy, nuclear, grid, climate tech |
| NASA | $150,000 | $750,000 | Space, aeronautics, earth science |
| USDA | $200,000 | $650,000 | Agriculture, food safety, rural tech |
The SBA sets maximum award guidelines ($314,363 for Phase I, $2,095,748 for Phase II as of 2025). Each agency interprets these differently. For the full breakdown, see our SBIR funding amounts by agency reference.
Success rates
| Agency | Phase I Success Rate | Notes |
|---|---|---|
| NSF | 20-25% (pitch-to-award) | Two-step process filters early |
| NIH | 13-24% (varies by Institute) | Preliminary data is critical |
| DoD | ~15% | Largest volume (~1,440 Phase I awards/year) |
| Overall (all agencies) | ~17% Phase I | Phase I-to-Phase II conversion varies by agency (30-65%) |
Those rates are for all applicants. First-time applicants do worse -- roughly 75-80% of first submissions are rejected. The learning curve is steep, and most proposals fail on framing, not on the underlying technology.
Timeline from submission to award
| Agency | Typical Timeline |
|---|---|
| NSF | 4-6 months from full proposal |
| NIH | 6-9 months (3 fixed receipt dates per year) |
| DoD (AFWERX) | 2-4 months (rolling deadlines) |
| DoD (Navy/Army) | 4-6 months (annual cycles) |
| DOE | 6-9 months |
If you haven't applied before, add 2-6 weeks for first-time federal registrations: SAM.gov (your company's federal contractor registration, the bottleneck), Grants.gov (where proposals are submitted), and login.gov (federal identity verification). Start SAM.gov first.
Reauthorization note (March 2026): SBIR's authorization lapsed in September 2025. Congress has since passed reauthorization legislation extending the program through 2031 -- including new "Strategic Breakthrough Awards" of up to $30M for Phase II alumni -- but the bill is awaiting presidential signature. Some agencies paused solicitations during the gap. Check sbir.gov for current status.
Pros of government grants for startups
Here's what makes grants genuinely powerful for the right company.
Zero dilution, zero repayment
This is the obvious one, but it's worth quantifying. A $305K NSF Phase I grant preserves the equity that a $305K seed check would cost you 8-15% of your company. At a $5M pre-money valuation, that's $400K-$750K in equity value preserved. And if you go on to Phase II ($1M), the math gets even more lopsided.
Third-party validation that investors actually respect
This isn't hand-waving. Sabrina Howell's 2017 study in the American Economic Review found that an early-stage SBIR Phase I award approximately doubles the probability of receiving subsequent venture capital. The mechanism isn't just signaling -- it's that the grant lets you build the prototype that makes the investment case tangible.
The alumni list makes the point. Genzyme's first grant was $62K from HHS -- seven SBIR awards later, Sanofi acquired the company for $20.1B. Illumina funded its early genome sequencing work with 14 NCI SBIR grants and now does $3.3B in annual revenue. 23andMe launched with a $250K NIH SBIR. Symantec was literally the first SBIR awardee in 1982 (back when it was still called Machine Intelligence Corporation).
You're funding R&D you'd do anyway
If your startup's next 12 months involve building a prototype, running experiments, or proving technical feasibility, you're doing SBIR-eligible work whether or not you apply. The grant just pays for it without dilution.
Phase III: the part nobody talks about
Phase I and Phase II are grants. Phase III is a sole-source government contract -- meaning the agency can buy your product without competitive bidding, with no funding ceiling. This is how SBIR startups go from research awards to real government revenue. For defense tech and govtech companies, Phase III is the entire point. But even outside defense, the pattern holds: NIH and NSF Phase II completers regularly land follow-on contracts, collaborative agreements, and commercial partnerships that trace directly back to the SBIR work.
Cons and risks of government grants for startups
Now the part most grant advocates skip.
The time math is brutal
A competitive SBIR Phase I proposal takes 100-300+ hours to write. If you're a 2-person team with 6 months of runway, spending 6-8 weeks on a proposal that has a 15-25% chance of winning is an enormous bet. That's time not spent on product development, customer discovery, or revenue.
For context: a well-prepared NSF Project Pitch takes about 4 hours and gets you a yes-or-no signal in 3 weeks. That's a reasonable time investment. A full NIH proposal that takes 160 hours? That's a different calculation entirely.
4-9 months is a long time in startup years
If you need capital in the next 60 days, grants won't save you. Even the fastest agencies (AFWERX, 2-4 months) are slower than most seed rounds. Grants work best as planned capital, not emergency funding.
Strict usage restrictions
Grant funds are restricted to the proposed R&D activities. You can't redirect SBIR money to marketing, sales hires, or general operations. If your biggest need is go-to-market capital, grants solve the wrong problem.
Reporting and compliance overhead
Quarterly technical reports, financial reporting, potential audits. It's not crushing, but it's not zero either. And SBIR income is taxable -- plan accordingly.
The mission drift risk
This one is underrated. Sometimes a solicitation topic is close-but-not-quite to what you're building, and the temptation is to bend your R&D plan to fit. That's how startups end up spending a year building something their market doesn't need because a government agency was willing to pay for it. If you'd have to meaningfully pivot your technical roadmap to match a solicitation, that's a red flag.
Government grants vs. venture capital vs. revenue vs. debt
These aren't competing. But it helps to see them side by side.
| Government Grants | Venture Capital | Revenue | Venture Debt | |
|---|---|---|---|---|
| Dilution | 0% | 15-30% per round | 0% | 0-2% (warrants) |
| Repayment | None | None (exit-driven) | N/A | Yes, with interest |
| Typical amount | $150K-$2M | $500K-$10M+ | Varies | $1M-$5M |
| Speed to funding | 4-9 months | 2-6 months | Immediate | 2-4 weeks |
| Control impact | Low (reporting) | High (board seats, governance) | None | Medium (covenants) |
| Best for | R&D, deep tech, pre-revenue | Scaling, go-to-market | Product-market fit | Extending runway between rounds |
| Main cost | Time and opportunity cost | Equity and control | Slow growth | Repayment pressure |
The smartest founders combine these. Use grants to de-risk R&D and extend runway, then raise VC from a position of strength (validated technology, preserved equity, longer runway). Research backs this up: SBIR recipients raise more VC, not less, because the grant removes technical risk that investors don't want to fund.
The decision framework
Here's the part you actually came for. Be honest with yourself on each criterion.
Apply if:
- You have genuine R&D to do (not just product development or feature engineering)
- Your technology has 6-18 months of technical work ahead
- You can identify at least one federal agency whose mission aligns with what you're building
- You have (or can get) preliminary data, a prototype, or published research
- You can absorb 4-9 months of waiting without a cash crisis
- Equity is precious to you right now (pre-seed, seed, or bootstrapped)
Skip if:
- You need cash in the next 60 days
- Your product is a pure consumer app, marketplace, or SaaS tool with no novel technology
- You can't identify a federal use case or societal benefit for your technology
- You have no R&D component (just integrating existing tools)
- Your company has more than 500 employees or is majority foreign-owned
- The grant amount is immaterial relative to your burn rate (you've raised $20M+ and $305K doesn't move the needle)
The gray zone
"We're building AI, but is our model architecture novel enough?" or "We have a medtech device, but no clinical data yet." These come up constantly.
The best test: search sbir.gov for recent awards in your technology area. If companies similar to yours are winning, that's a strong signal. If the awards all go to universities or large defense primes, the program might not be structured for companies at your stage.
Another test: can you articulate a genuine technical uncertainty you're trying to resolve? Not "will customers buy this" (that's market risk, not R&D). But "can we achieve X accuracy with Y architecture under Z constraints" -- that's fundable.
If you can pass both tests -- similar companies winning awards, and a real technical uncertainty you can articulate -- you're a viable candidate regardless of stage.
How to get started (if you decide yes)
Start with NSF's Project Pitch
If you've never applied for a grant, the NSF Project Pitch is the lowest-friction entry point in the entire federal system. It's roughly 2 pages, takes about 4 hours to write, and NSF reviews it in ~3 weeks. They'll tell you whether to submit a full proposal. If they say no, you've lost a day of work, not two months.
Register early
You'll need to register at three federal portals before you can submit anything: SAM.gov (your company's federal contractor registration, takes 2-4 weeks to process), Grants.gov (where proposals are submitted), and login.gov (federal identity verification). Start SAM.gov first -- it's the bottleneck. The others take under an hour.
Match your technology to an agency
Different agencies fund different missions. NSF covers the broadest range of science and engineering. NIH is biomedical and health. DoD branches fund defense-relevant technology. DOE covers energy and climate. See our SBIR guide for the full agency-by-agency breakdown.
Decide on DIY vs. consultant
First-time applicants face that 75-80% rejection rate. A qualified consultant compresses the learning curve and helps you avoid the framing mistakes that kill otherwise-strong proposals. But even with a consultant, you need to bring the core ingredients: your technology, your data, your market insight. Nobody can write a credible proposal without deep input from the founding team.
For more on this decision, see our post on choosing a grant writer payment model.
The bottom line
Government grants are not free money. They're R&D capital with real costs (time, compliance, opportunity) and real benefits (zero dilution, validation, Phase III contracts). The founders who benefit most are the ones who treat grants as a strategic tool in their capital stack, not as a lottery ticket or a distraction.
If your technology is genuine, your R&D roadmap is real, and you can find an agency match, the expected value of applying is strongly positive. Just go in with clear eyes about what it costs and how long it takes.
Want to know if grants are realistic for your company?
We've helped secure $1.6B+ in non-dilutive funding across 500+ proposals. If you want a straight answer on whether SBIR (or another program) fits your startup before you invest weeks in an application, that's what our Strategy Review is for. Fifteen minutes, no obligation, specific to your technology.