Last updated: April 29, 2026
The new SBIR mill proposal caps in 2026 are being framed as a compliance story. They are not. They are a slot redistribution from a small cohort of multi-award firms to first-time applicants, and the agencies that publish their caps first in May, June, and July 2026 will redirect roughly $200M+ toward newcomers.
If you have been waiting out the reauthorization lapse, your timing window is the next 90 days.
TLDR (60 seconds)
S.3971 (the Small Business Innovation and Economic Security Act of 2026) was signed April 13, 2026. It introduces the first SBIR proposal limits 2026 has seen at the per-firm level: caps that each agency must publish by July 3, 2026, a 50-Phase-I commercialization benchmark that requires firms with 50+ Phase I awards to show 25% revenue from prior SBIR work, and faster commercialization auditing. First-time applicants do not trigger any of the new gates. The structural advantage flows to them.
That is the answer. The rest of this piece is the math, the calendar, and what to do about it.
The Math: 14 Firms, $200M+ in Concentration
Before you understand the upside of the new caps, you need to see the problem they were written to fix.
GAO-24-105470 documented the pattern that congressional staffers eventually called the "SBIR mill" problem. A small cohort of firms had built repeat-winning into a business model. Some had won 100+ Phase I awards each. Several had crossed 200+. The 14 most-concentrated firms together absorbed somewhere in the 5% to 8% range of annual SBIR Phase I proposal slots across the 11 issuing agencies.
The math compounds in ugly ways:
- Repeat winners had institutional reviewer familiarity ("I have seen Company X's submissions for ten years")
- They had dedicated proposal teams (sometimes 10--15 people on payroll just for SBIR writing)
- They had perfected templates that clear the technical bar with predictable margins
- They had less to prove and more political capital with program officers
None of that is illegal. But it created a structural asymmetry: a first-time applicant was competing against a back-catalog of trust signals they could not match.
How concentrated were SBIR awards before 2026?
A handful of firms with 100+ Phase I awards each absorbed roughly 5% to 8% of annual Phase I proposal slots across the 11 SBIR-issuing agencies. The concentration was highest in DoD components and DOE, lower in NIH and NSF, and was sustained through reviewer familiarity, dedicated proposal teams, and perfected templates rather than any single rule violation.
What S.3971 Actually Changed
Three levers. All are live as of April 13, 2026.
Lever 1: Per-Firm Proposal Caps
Each SBIR-issuing agency must publish a per-firm proposal cap by July 3, 2026 (90 days after signing). The caps are agency-specific, not government-wide. NIH may publish 12 per fiscal year while DOE publishes 6 per fiscal year, and they will not match.
The bill does not specify a number. It specifies that the cap must be published, justified, and tied to historical proposal volume per agency. Agencies have discretion. That discretion is where most of the action is.
Lever 2: 50-Phase-I Commercialization Benchmark
If a firm has won 50 or more Phase I awards historically, it must demonstrate that at least 25% of the firm's revenue comes from prior SBIR-funded products. If it cannot, it loses Phase I eligibility until the benchmark is met.
This is the real teeth. Most multi-award firms have built their revenue around grant capture, not product commercialization. The 25% revenue test is the single biggest disqualification mechanism in the bill.
Lever 3: Faster Commercialization Auditing
The Inspector General now has a standing audit mandate, and agencies must self-report commercialization data on a 12-month cycle rather than the previous 36-month cycle. This shortens the lag between non-commercialization and disqualification.
Honest uncertainty
I should note: agency-by-agency cap numbers are not published yet. The first cap notices are expected in May 2026. The 50-Phase-I threshold may face legal challenge from affected firms. And "per firm" definitions could expand to include affiliated entities, which would dilute the effect for some multi-cap-firm structures. None of this changes the directional pressure, but it changes the timing and magnitude.
Why This Is Structural Advantage, Not Just Compliance
Here is the contrarian read: the new rules are not "everyone follows new compliance steps." They are a re-allocation of slots from a small cohort to a much larger pool.
The math:
- The 14 most-concentrated firms previously absorbed roughly 5% to 8% of annual Phase I proposal slots
- Capping each of them at 20--50 proposals per cycle (the most likely agency cap range) cuts their submission volume by 60% to 80%
- That redirects somewhere in the $150M to $250M range annually toward the rest of the applicant pool
First-time applicants benefit specifically because:
- They do not trigger the 50-Phase-I commercialization benchmark
- They are not at any per-firm cap
- They compete in a less-crowded field at the margin
- Reviewers see fewer "yet another from Company X" submissions and more variety, which shifts cohort psychology
You are also competing in a market where the loudest narrative ("the system is rigged against newcomers") is structurally less true than it was 6 months ago. That perception lag is an arbitrage window -- most founders still believe the old story, so fewer are moving now than should be.
Are first-time SBIR applicants advantaged by the 2026 reauthorization?
Yes. First-time SBIR applicants are the structural beneficiaries of S.3971 because they do not trigger the 50-Phase-I commercialization benchmark, do not sit at any per-firm cap, and compete in a less-crowded pool. The expected redirection of $150M to $250M annually toward the broader applicant pool flows disproportionately to first-time and lightly-experienced applicants.
The Agency Cap Calendar (May--July 2026)
The bill was signed April 13, 2026. Notices are required by July 3, 2026. That is a 90-day window. Based on how agencies have historically rolled out Phase I rule changes, here is the expected order:
| Agency | Expected Publication Window | What to Watch |
|---|---|---|
| NIH | May 15 -- June 15, 2026 | Cap likely 8--12 per fiscal year. Watch for resubmission treatment. |
| DOE | May 15 -- June 15, 2026 | Cap likely 4--8 per fiscal year. Watch for ARPA-E vs. Office of Science split. |
| NSF | June 1 -- July 1, 2026 | Cap likely 6--10 per fiscal year. Watch for whether STTR is bundled. |
| DoD components | June 15 -- July 3, 2026 | Each component may publish separately (Army, Navy, AFWERX, SOCOM, DLA). Watch for affiliated-entity definitions. |
| NASA | June 15 -- July 3, 2026 | Cap likely 4--6 per fiscal year. |
| USDA, EPA, ED, DHS, DOC, DOT | Late June -- July 3, 2026 | Lowest historical Phase I volume; may publish higher caps that bind less. |
These windows are predictions based on agency response patterns to prior SBIR amendments (1992, 2011, 2016 reauthorizations). Treat them as a planning calendar, not a guarantee.
When will SBIR agencies publish proposal caps under S.3971?
All 11 SBIR-issuing agencies must publish per-firm proposal caps by July 3, 2026 (90 days after the April 13, 2026 signing of S.3971). Based on historical agency response patterns, NIH and DOE are expected to publish first (mid-May to mid-June), with NSF, DoD components, and smaller agencies following. Agencies have discretion on cap number, fiscal-year reset rules, and resubmission treatment.
What First-Time Applicants Should Actually Do Right Now
Five tactics. None require waiting for the cap notices to publish.
1. Pick agencies likely to publish first
NIH and DOE will publish first. Build your application calendar around their next two cycles. The first cycle after a cap publishes will be the least-crowded reviewer pool you will see for years.
2. Watch for cap-announcement-week filings
The week an agency publishes its cap, multi-award firms will be re-allocating which proposals they submit and which they pull. There will be a brief window of reviewer-bandwidth abundance. Submit during that window if your timeline allows.
3. Avoid the highest-concentration topics
If a SBIR topic has historically been won by the same 2--3 firms cycle after cycle, avoid it for the next 12 months. Those topics will face the steepest cohort-rebalancing pressure, but they will also attract the most attention from displaced multi-award firms looking to redeploy.
4. Lead with your clean commercialization record
In your research aims and commercialization plan sections, explicitly state your zero-commercialization-debt position. Multi-award firms competing in your topic must now spend pages defending their 25% revenue ratio. You do not. That is two pages of substantive content you have that they do not.
5. Build the agency cap tracker into your application calendar
Put each agency's cap publication date on your calendar. The day after publication, your agency-specific application strategy may shift materially. Most founders will not be watching this. Watching gives you a 30--60 day information edge.
The Honest Uncertainties
This depends on how agencies interpret the bill. Watch for:
- "Per firm" definitions that include affiliated entities -- some multi-cap-firm structures will lobby for this. If granted, the cap effect dilutes.
- Cap numbers set high enough to not bind -- politically possible if agencies are captured by prior winners.
- Court challenges to the 50-Phase-I threshold -- expected from at least 2--3 of the most-affected firms. Likely 12--18 months to resolve.
- Agency self-reporting that under-counts commercialization gaps -- the IG audit cycle will surface this eventually but with a 12+ month lag.
None of this changes the directional pressure. All of it changes the magnitude.
Track the Cap Calendar. Get an Honest Eligibility Read.
If you want to track agency cap publications in real time, Cada is publishing a weekly-updated agency cap tracker through the July 2026 deadline. It includes each agency's publication status, cap number once published, and a flag for changes that affect first-time applicants disproportionately.
Get the SBIR Agency Cap Tracker -- updated weekly through July 3, 2026.
If you want a straight answer on which agencies are likely to be most winnable for your specific company in this window, we do a 15-minute roadmap eligibility check. No pitch. Just an honest read on where you are competitive -- and where you are not.
Book a 15-minute roadmap check
FAQ
Does the SBIR cap apply to STTR too?
S.3971 applies to both SBIR and STTR. Per-firm caps are likely to be set jointly by each agency, though agencies have discretion to publish separately. NIH historically separates SBIR and STTR rules; NSF historically bundles them. Watch each agency's notice for the specific scope.
What counts toward the 50 Phase I threshold?
Historical Phase I awards across all SBIR-issuing agencies, summed at the firm level (not per agency). Affiliated entity treatment is unclear and is one of the most-watched interpretive questions. If you are unsure, assume the conservative interpretation and track your firm's cumulative Phase I count now.
If my agency has not published its cap, can I still apply?
Yes. Pre-publication, the standard SBIR rules apply. Once an agency publishes its cap, applications submitted in the same fiscal year count toward it. The cap does not retroactively invalidate prior submissions in that fiscal year.
Are foreign-owned firms affected by S.3971?
The existing FOCI and >50% US-individual ownership rules remain unchanged. S.3971 layers per-firm caps and the commercialization benchmark on top of existing eligibility, not in place of it. If you were eligible under the old rules, you are eligible under the new ones.
How does this interact with the FY2026 budget?
The bill is reauthorization, not appropriation. It changes who can win, not how much money is available. FY2026 SBIR appropriations are largely set; the new caps redistribute who within the eligible pool gets the existing money.
About Cada
Cada has written 100+ grant proposals across 30+ agencies, with a strategic focus on first-time applicants -- the population most advantaged by the new SBIR rules. We do honest filtering before paid engagements: if a program is not a fit, we tell you on the 15-minute call, not after you have committed cash.
If you want to know whether you are competitive for SBIR in this window, that is the question to answer first -- before investing 40--80 hours in an application.
Sources
- S.3971 (Small Business Innovation and Economic Security Act of 2026), Public Law signed April 13, 2026
- GAO-24-105470, "SBIR/STTR: Agencies Could Improve Their Reporting on Fraud, Waste, and Abuse" (concentration data)
- SBA SBIR/STTR Public Award Database (firm-level cumulative Phase I counts)
- Cada internal: 100+ grant proposals across 30+ agencies, strategic-priority focus on first-time applicants