USDA Will Fund 38 SBIR Phase II Projects and Let No Company Win Twice — $24.75 Million, November 17 Deadline

USDA's National Institute of Food and Agriculture posted its SBIR/STTR Phase II notice on October 1, 2026, and it closes November 17, 2026 at 5:00 P.M. Eastern. The funding opportunity number is USDA-NIFA-SBIR-012221, Assistance Listing 10.212.

The headline numbers are modest by federal standards: $19,500,000 for SBIR Phase II and $5,250,000 for STTR Phase II in FY2026, for $24.75 million total. The average individual award range is $600,000 to $650,000, and the maximum is $650,000 — that is a $600,000 project plus up to $50,000 in Technical and Business Assistance.

But this notice is one of the most tightly constrained competitions in the federal small-business portfolio, and three of its rules will determine outcomes more than the science will. They are all in the document. Most applicants will not read far enough to find them.

The Closed Universe

Start with who can apply, because it eliminates almost everyone: "This solicitation is only for SBIR/STTR Phase II applications, and only prior Phase I grant recipients are eligible to apply."

Not prior SBIR Phase I recipients generally — prior USDA SBIR or STTR Phase I recipients. A company with a strong NSF or NIH Phase I and an agricultural application has no entry point here. The entire competition is contested among firms that already cleared a USDA Phase I.

That makes this a rare federal competition where you can estimate the odds. NIFA publishes anticipated award counts per topic. Summed across the ten topics, the plan is 28 SBIR awards and 10 STTR awards — 38 total.

The topics and their anticipated SBIR/STTR splits:

Grant duration is 24 to 36 months across every topic, and every topic carries the same $650,000 ceiling.

Two observations fall straight out of that table.

First, the money concentrates. Four topics — natural resources management, food science and nutrition, biofuels and biobased products, and plant production engineering — carry double the SBIR slots of the other six. If your Phase I sits at a boundary between two topics, the topic with four anticipated awards is worth a careful second look. NIFA encourages Phase II applicants to respond to the topic corresponding to their funded Phase I work, so this is a framing decision rather than a free choice, but boundary cases are real.

Second, the STTR arithmetic does not close. Ten anticipated STTR awards at the $650,000 ceiling would cost $6.5 million against $5.25 million available. Either STTR awards land meaningfully below the ceiling — around $525,000 on average — or NIFA makes fewer than ten. Either way, an STTR applicant budgeting to the maximum is budgeting against a pot that cannot pay ten maximums. The SBIR side has slack: 28 awards at $650,000 is $18.2 million against $19.5 million.

You May Submit Two. You Will Win One.

This is the rule that reshapes strategy, and it is stated three ways in the eligibility section.

Applicants are restricted to one Phase II submission per individual Phase I award. A company holding multiple Phase I awards may submit a maximum of two distinctly different Phase II applications per solicitation. And then: "USDA NIFA limits issuance to one (1) SBIR/STTR Phase II award per SBC."

So a two-time Phase I winner can file two applications and, even if peer review recommends both, can receive only one award. The notice spells out what happens next: "In instances where multiple applications from a single SBC are recommended for funding via peer review, USDA staff will require the applicant to provide a written determination specifying which single application shall proceed."

You make the cut, not USDA. The non-selected application stays eligible for resubmission in a future Phase II cycle, subject to then-current review procedures.

The implication for a multi-Phase-I company is not "file both and hope." It is that filing two applications costs you two full proposals and buys you one award with two shots at it — a real hedge against a bad panel draw, but a hedge that doubles your writing cost for zero additional dollars. If one program is clearly stronger, concentrating on it is defensible. If two are genuinely close and either would be fundable, the second application is cheap insurance against reviewer variance. What is never rational is filing a weak second application to "cover" a topic, because a weak application cannot convert into money that a strong one would not have won.

The Published Tiebreaker Is a Letter

Most Phase II competitions leave tie-breaking to panel discretion. NIFA writes theirs down.

"Where Phase II applications are evaluated as having approximately equal technical merit, NIFA will consider optional letters of support or contingent commitment letters." Applicants with strong letters for non-SBIR/STTR follow-on funding "may receive special consideration as a point of merit in the review and evaluation process."

Then the calibration: "Greater value (in Phase II evaluation) will be given for an executed formal agreement with reasonable terms and funding equal to or in excess of the Federal investment requested in the Phase II application."

Read that as a specification. An executed agreement beats a letter of intent. Funding at or above your federal ask beats a token commitment. On a $600,000 request, a $600,000 follow-on commitment is the target the notice itself names.

The acceptable forms are broad: venture capital, a package including venture capital, contract R&D, a joint venture, an R&D limited partnership, or another agreement with a non-SBIR/STTR funding source. Two constraints matter. The terms cannot be contingent on an issued patent, because USDA acknowledges patents take too long. And no amortization, repayment, or repurchase of commitment funds may occur during the Phase II period of performance — so a convertible note that starts paying back inside 24 months does not qualify.

USDA also accepts that commitments will be contingent on technical objectives. Those objectives must be clearly defined, measurable, specified in the agreement at a threshold level that would justify the investment, and accomplishable within the scope of the proposed work. They need not be identical to the Phase II objectives, but they must be closely related.

If you have a commercial partner and five weeks, this is the highest-leverage document in your application — and it is the one that cannot be written the night before.

The Work-Split Rules Decide Which Program You Apply Under

Applicants choose SBIR only, STTR only, or both in a single submission. The percentages are not interchangeable.

SBIR Phase II: The Project Director/PI's primary employment must be with the small business — more than half their time, with the company, for the entire award period. Primary employment with the SBC precludes full-time employment at another organization or academic institution. By budget expenditures, at least 50 percent of the research or analytical work must be performed by the proposing company. Research-institution participation stays under 50 percent. No formal cooperative agreement on intellectual property is required.

STTR Phase II: The PI may be employed by either the small business or the partnering research institution, at more than half time. A single partnering research institution must perform at least 30 percent of the work, and the company at least 40 percent. A formal cooperative agreement outlining IP rights is required, and research-institution participation may run 30 to 60 percent.

Both: PI primary employment with the company, a single research institution at 30 percent or more, and the company at 50 percent or more. Research-institution participation runs 30 to 50 percent.

Eligible STTR partners must be U.S.-based and fall into one of three categories: a nonprofit college or university, a domestic nonprofit research organization, or a federally funded R&D center.

Size eligibility is the standard 500-employee cap including affiliates — with a trap worth naming. A subsidiary must document that its parent company or companies are also small business entities, and "if the parent company or one of the parent companies is a nonprofit organization, then the subsidiary is not eligible." Employee stock ownership plan trustees and plan members count as owners; trust trustees and beneficiaries do too.

All Phase II R&D must be performed in the United States, with exceptions available only in rare circumstances where a supply, material, or project requirement is unavailable domestically — and requested after award, not before.

Money Mechanics Worth Knowing

No match. "The SBIR/STTR programs have no matching requirement. NIFA will not factor matching resources into the review process as an evaluation criterion." Cost share is encouraged but explicitly unscored — which, paired with the follow-on-funding tiebreaker, tells you where to put your partner's money: into a commitment letter for Phase III, not into a match line that earns no points.

TABA is additive, restricted, and use-it-or-lose-it. Up to $50,000 on top of the $600,000 project for Technical and Business Assistance. You identify and select your own provider, and the costs go into the application as a formal TABA Plan. TABA funds may not be used for indirect costs, profit, or fee calculations, and cannot be included in the base for computing indirect cost rates. Any unspent TABA funds may not be rebudgeted to other project costs and must be returned to USDA. Allowable uses include minimizing technical risk, solving technical problems, intellectual property protection, and screening for potential foreign involvement in commercialization.

Indirect costs. Use your negotiated rate from your cognizant federal agency if you have one; you will have to produce the rate agreement to recover indirect costs. If you have no negotiated or provisional rate, you may elect the 15 percent de minimis rate on modified total direct costs under 2 CFR 200.414(f), usable indefinitely. For a first-time Phase II company without a rate agreement, electing de minimis at application is the clean path.

What Will Get an Application Rejected or Returned

• Pure product development. Projects seeking funds solely to develop a product from previously completed research, with no research involved, "will not be accepted."
• Literature surveys as scope. They must be completed before the application and cannot be proposed as part of the R&D effort.
• Large-scale demonstration or classified research. Not permitted in Phase II.
• Prototype misuse. A laboratory prototype is allowed where necessary, but must comply with 7 CFR 3403 for disposition at the end of the grant, cannot be used for Phase III activities, and must serve only as a model for manufacturing commercial versions.
• Missing foreign-relationship disclosure. Every applicant must complete the Required Disclosures of Foreign Affiliations or Relationships to Foreign Countries form — including to certify that no such relationships exist. Disclosure does not mean disqualification, but omission is a compliance failure.
• Dangerous gain-of-function research, as defined in Section 8 of Executive Order 14292.

Project narratives are capped at 20 pages of written text. The response to previous review does not count against that limit, and applicants must include their Phase I results.

Two SBA benchmarks also gate eligibility for experienced firms. Companies with more than 20 Phase I awards in the past five fiscal years must maintain a 25 percent Phase I-to-Phase II conversion rate; above 50 Phase I awards, 50 percent. On commercialization, companies with more than 15 Phase II awards over ten years must average at least $100,000 in sales or investment per Phase II award, or hold patents equal to at least 15 percent of those awards, with higher thresholds at 50 and 100 Phase II awards.

The Cost of Missing November 17

This NOFO covers two fiscal years. The FY2027 deadline is listed as September 21, 2027 (tentative), with the same anticipated funding — $19.5 million SBIR, $5.25 million STTR.

So the penalty for missing this window is roughly ten months, not two. For a company whose Phase I funding has ended and whose runway is measured in quarters, that gap is the real deadline pressure — not the November date itself.

Projects should also align with Secretary's Memorandum 1078-020, which sets five departmental R&D priorities: increasing profitability of farmers and ranchers, expanding markets and new uses of U.S. agricultural products, protecting American agriculture from invasive species, promoting soil health, and improving human health through precision nutrition and food quality. NIFA separately encourages applicants to engage farmers, ranchers, and foresters directly when applying.

Five weeks is enough time to write a good Phase II application. It is barely enough time to negotiate an executed follow-on funding agreement — which is the one element USDA has told you, in writing, it will use to break a tie. Start there.

For the wider agency calendar, see our SBIR and STTR deadlines calendar for 2026, and for how commercialization plans changed after reauthorization, our analysis of the SBIR commercialization plan post-reauthorization.