The Multi-Agency SBIR Strategy: How Serial Winners Build a Portfolio Across DOD, NIH, and NSF

Ninety-five companies — less than one percent of all SBIR awardees between 2009 and 2019 — captured more than 21 percent of every award the program handed out. That figure, drawn from a State Science & Technology Institute analysis, sparked a congressional fight that froze $4 billion in annual R&D funding for five months, rewrote eligibility rules, and forced every small business with multi-agency ambitions to rethink its approach. The question facing founders now is whether a deliberate portfolio strategy — winning SBIR awards across the Department of Defense, National Institutes of Health, National Science Foundation, and beyond — remains viable under the new regime, or whether the era of the multi-agency SBIR company is ending.

The answer, for firms that understand the distinction between a genuine technology portfolio and a grant-writing operation, is that the opportunity has never been larger.

The Arithmetic of Eleven Agencies and $4 Billion

The SBIR program channels more than $4 billion annually through eleven federal agencies, each of which sets aside between 1.5 and 3.2 percent of its extramural research budget. The Department of Defense dominates, accounting for roughly 48 percent of all SBIR funding — approximately $1.5 billion in fiscal year 2019 — followed by the Department of Health and Human Services at 31 percent (nearly $1 billion, mostly through NIH). The Department of Energy, NSF, and NASA together represent another 19 percent.

Those numbers create a structural advantage for companies whose technology serves more than one federal mission. A sensor platform relevant to both battlefield surveillance and clinical diagnostics can tap DOD's $1.5 billion pool and NIH's $1 billion pool without competing against itself. A materials science breakthrough with applications in spacecraft thermal protection and clean energy infrastructure can simultaneously pursue NASA, DOE, and NSF funding. The math is straightforward: the more agency missions a technology credibly serves, the more noncompeting funding pools become available.

CFD Research Corporation, based in Huntsville, Alabama, demonstrates what this looks like in practice. Founded on computational fluid dynamics modeling, the company has won SBIR awards from the Air Force, Army, DARPA, NASA, DOE, NSF, NIH, and the Department of Homeland Security. Its simulation software, originally developed under defense SBIR awards, was adapted for biomedical modeling, energy systems analysis, and materials science applications. The company has won the SBA's Tibbetts Award twice and secured a $10 million Phase III IDIQ contract from the Air Force Research Laboratory — the kind of commercial transition that separates a legitimate multi-agency strategy from the pattern Congress targeted.

What Each Agency Actually Evaluates

The single most common mistake companies make when submitting to a second or third agency is treating SBIR as a monolithic program. Each agency applies distinct evaluation criteria, uses a different review structure, and prioritizes different aspects of commercialization.

DOD evaluates proposals against specific acquisition-oriented topics. Applicants do not propose their own research directions — they respond to defined needs published in solicitations by agencies like the Army, Navy, Air Force, DARPA, and the Defense Health Agency. Reviewers assess technical merit and innovation of the approach, qualifications of key investigators, and the potential for commercial application in both government and private-sector markets. Reviews are conducted internally. DOD Phase I awards run up to $200,000, with Phase II awards reaching approximately $1.16 million.

NIH operates through a dual-review system. External peer review panels evaluate scientific and technical merit, followed by internal advisory council review. NIH reviewers ask whether an application challenges current research paradigms, whether the investigators are well-suited to the project, and whether the proposed approach is significant. NIH Phase I awards reach approximately $307,000, and Phase II awards can exceed $2 million — the most generous among all agencies.

NSF frames its criteria around Intellectual Merit and Broader Impacts. Rather than responding to defined topics, NSF uses a unique two-step process: companies first submit a Project Pitch, and only those invited may submit full proposals. NSF reviewers assess the potential to advance knowledge, team qualifications, and societal benefit. Phase I awards reach $305,000, with Phase II at $1.25 million.

DOE evaluates the strength of the scientific and technical approach, the ability of the PI and team to execute efficiently, and the likelihood that the work will produce a marketable product or attract Phase III funding. DOE Phase I awards are typically $200,000, with Phase II reaching $1.6 million.

NASA requires demonstrated understanding of the current state of the art and clear plans for developing and verifying the innovation. Like DOD, NASA channels proposals into specific topic areas aligned with agency missions. NASA uniquely weights commercialization experience, co-funding commitments, and Phase III strategy. Reviewers include both internal staff and at least two independent external experts.

A company submitting the same boilerplate language to three of these agencies will lose at all three. The technology may overlap, but the framing, the language, and the commercialization narrative must be rebuilt for each mission context.

Building a Staggered Pipeline

The agencies do not operate on the same calendar, and that asynchrony is a strategic asset. DOD typically opens solicitations on the first Wednesday of a month and closes on the last Wednesday of the following month, cycling through multiple releases per year. NIH historically uses standard receipt dates — September 5, January 5, and April 5 — though those dates were disrupted by the authorization lapse. NSF accepts Project Pitches on a rolling basis, providing responses within about 30 days before inviting full proposals.

Smart multi-agency companies build a staggered submission pipeline that treats the calendar as a resource. A firm might spend Q1 responding to a DOD Open Topic solicitation, shift in Q2 to an NIH Phase I application targeting a September receipt date, then use Q3 to pitch NSF on a related technology angle while preparing a DOE submission for the fall cycle. This cadence prevents the bottleneck that sinks many multi-agency efforts: trying to write four proposals simultaneously with the same three-person team.

The staggered approach also creates a compounding advantage. A Phase I award from NIH generates preliminary data that strengthens a subsequent DOD proposal. A DOD Phase II deliverable provides the kind of validated prototype that NSF reviewers want to see when evaluating commercialization potential. Each award feeds the next.

The Line Between Portfolio Strategy and SBIR Mill

Congress drew a hard line in the Small Business Innovation and Economic Security Act, which the House passed 345 to 41 on March 17, 2026, and which now awaits the president's signature. The bill does not ban multi-agency strategies. What it targets is the pattern that prompted the legislation: companies that treat SBIR as a permanent revenue source without meaningfully commercializing their technology.

The data behind the crackdown is stark. DefenseScoop reported that Physical Sciences Inc. accumulated 1,728 SBIR awards worth $650 million since 1983 — more SBIR/STTR funding than all small businesses combined in 26 states. Triton Systems collected 906 awards totaling $365 million. GAO analysis found that these multi-award winners produced patents at a rate of 3.4 percent, compared to 4 to 5 percent for companies with far fewer awards, and generated just $0.89 in commercial revenue per Phase II dollar received.

The reauthorization bill addresses this by requiring each agency to establish caps on the number of proposals any single company can submit per Phase I or Phase II solicitation. Agencies retain discretion to set limits by fiscal year, by solicitation, or by individual topic. Waivers require written justification, senior approval, and congressional notification, and are capped at 5 percent of topics annually.

For companies building a legitimate multi-agency portfolio, the distinction is clear. A genuine portfolio strategy starts with a core technology platform, adapts it to serve distinct agency missions, and converts SBIR awards into Phase III contracts, commercial products, or venture investment. An SBIR mill submits variations of the same proposal to every open topic, optimizes for Phase I win rates, and never crosses the Valley of Death into production.

The New Strategic Breakthrough Awards

The most consequential provision in the 2026 reauthorization creates Strategic Breakthrough Awards — a post-Phase II funding mechanism offering up to $30 million over 48 months. Only agencies with $100 million or more in annual SBIR obligations qualify to issue them, which limits the pool to DOD, HHS, DOE, NASA, and NSF. The total allocation is capped at 0.50 percent of each agency's extramural R&D budget.

The catch is the matching requirement: applicants must demonstrate 100 percent matching funds from private investment, non-SBIR government contracts, or revenue. A company seeking a $10 million Strategic Breakthrough Award needs $10 million in matching capital. This provision explicitly rewards companies that have translated earlier SBIR work into commercial traction — precisely the behavior a multi-agency portfolio strategy is designed to produce.

Companies that have built relationships with defense primes through DOD Phase II work, attracted NIH follow-on funding, or secured licensing deals from NSF-funded technology are positioned to meet the match. Companies that have collected hundreds of Phase I awards without commercializing anything are not.

Managing the Compliance Load

Running simultaneous awards across multiple agencies is an operational challenge that destroys underprepared companies. Each agency imposes its own reporting cadence, financial management standards, and intellectual property provisions. DOD awards often require DCAA-compliant accounting systems, timekeeping that captures labor by project and reconciles payroll to timesheets, and indirect cost allocation using written, consistently applied rates. NIH requires progress reports that track milestones and commercialization plan updates. NSF mandates annual and final project reports through Research.gov.

Three operational practices separate companies that scale multi-agency portfolios from those that collapse under the weight:

First, install a project-level cost accounting system before winning the second award. The system must identify and accumulate direct costs by contract, allocate indirect costs using a documented methodology, and produce audit-ready financial reports on demand. Retrofitting accounting after an audit finding is orders of magnitude more expensive than building the system upfront.

Second, designate a compliance lead — even if that person is a fractional hire or outsourced specialist — whose sole job is tracking reporting deadlines, maintaining award files, and ensuring each agency's terms and conditions are met. A missed annual report can trigger termination of the award and future ineligibility.

Third, establish clear intellectual property boundaries between awards. Different agencies maintain different IP provisions. A DOD contract may grant the government unlimited rights to certain deliverables, while an NIH grant preserves the small business's patent rights under Bayh-Dole. When the same underlying technology platform generates work products for multiple agencies, sloppy IP management creates conflicts that can jeopardize every award in the portfolio.

Adapting Technology Without Double-Dipping

Federal rules prohibit funding the same work twice, but they explicitly permit funding different applications of the same core technology. The distinction is scope, not subject matter.

Consider a company that develops an advanced biosensor platform. A DOD SBIR might fund the development of that sensor for detecting chemical threats on the battlefield — a specific application with specific performance requirements tied to a military topic. An NIH SBIR might fund adaptation of the same sensor architecture for point-of-care sepsis detection in emergency rooms — different performance requirements, different regulatory pathway, different end users. A DOE SBIR might fund the sensor's application in monitoring microbial contamination in biofuel production systems.

Each proposal addresses a different agency mission, solves a different technical problem, serves a different end market, and generates different deliverables. The underlying sensor platform benefits from cross-pollination — manufacturing improvements developed under the DOD award reduce costs for the NIH application, and data from the NIH clinical validation strengthens the DOE proposal's credibility. But no two awards fund identical work.

The key is documenting distinct scopes of work, maintaining separate project files, and being transparent in each proposal about related SBIR awards. Every agency requires disclosure of current and pending support. Reviewers expect to see related awards; what triggers rejection is overlap in the proposed work, not overlap in the technology domain.

Where the Opportunity Sits Now

The five-month SBIR shutdown created a backlog of unfunded research priorities across every participating agency. DOD and NIH are expected to publish the first post-reauthorization solicitations in spring 2026, with NSF, DOE, and NASA following through early summer. Companies that used the shutdown period to refine their technology positioning across agencies — mapping their capabilities to specific topics in multiple agency portfolios — will have a significant head start when solicitations reopen.

The new proposal caps will thin the field of serial submitters who previously flooded agencies with dozens of Phase I applications per cycle. For companies with genuine multi-agency technology platforms, reduced competition from volume submitters means higher odds on each individual proposal.

Phase I success rates have historically hovered around 15 percent across most agencies, with NIH averaging higher at roughly 24 percent. Phase II selection rates run substantially higher — around 59 percent for SBIR Phase II applicants government-wide. A company that wins Phase I awards from two or three agencies and converts even half to Phase II is building a portfolio worth several million dollars, with Strategic Breakthrough Awards offering a pathway to tens of millions more.

The companies that will thrive under the new rules are the ones that always should have: firms with real technology, real commercialization plans, and the operational maturity to deliver across multiple federal missions simultaneously. The 2026 reauthorization did not kill the multi-agency strategy. It killed the excuse for not having one.

Granted tracks SBIR solicitations across all eleven participating agencies and matches them to your technology profile, so you can build a cross-agency pipeline without manually scanning thousands of topics.