SBIR & Research
How to Write an SBIR Proposal: A Step-by-Step Guide
Grant Writing Service Editorial Team
June 12, 2026 · 6 min read
Table of contents
Key takeaways
- SBIR and STTR were reauthorized on April 13, 2026 through September 30, 2031, after a lapse that began October 1, 2025; agencies are issuing solicitations again under some new rules.
- Each agency runs its own solicitation and format, so the agency you target decides the structure of the proposal.
- Registrations (SAM.gov, the SBA Company Registry, and the agency portal) can take weeks and should start first.
- Phase I proposals win on a narrow, testable feasibility question with numeric success criteria and a credible early view of the market.
To write a Small Business Innovation Research (SBIR) proposal, choose the agency and solicitation that fit your technology, confirm your company is eligible, start the required registrations, and then write to that agency's format: a tight statement of aims or objectives, a technical plan with measurable milestones, a budget that traces to the work, and the commercialization content the agency scores. The agency decides the structure. NIH, NSF, the Department of Defense (DoD), and the Department of Energy (DOE) each ask for different documents in a different order.
Where the program stands in 2026
SBIR and its sister program, Small Business Technology Transfer (STTR), lost their statutory authority on October 1, 2025, and agencies stopped issuing new solicitations and awards. The Small Business Innovation and Economic Security Act (S. 3971), signed on April 13, 2026, reauthorized both programs through September 30, 2031. Agencies have since reopened, but on their own schedules, and the law brought new rules with it:
- Application caps. The law requires agencies to limit how many proposals one company can submit. For HHS, which includes NIH, the limit is nine new and resubmission Phase I, Fast-Track, and Direct to Phase II applications and contract proposals per small business per fiscal year, for due dates on or after April 13, 2026. Other agencies set their own limits; NSF's current solicitation, for example, allows two Project Pitches per company in any 12 months.
- Foreign ties. The law expanded foreign-ownership disclosure and research security review, and NIH has issued notices implementing it. Expect more questions about investors, affiliations, and foreign relationships at application and before award.
- A larger Phase II tier. Agencies with large SBIR budgets can make Strategic Breakthrough awards. NSF's version allows up to $30,000,000, by invitation, for existing Phase II awardees.
Because implementation is still rolling out, treat the current solicitation as the only authority on dates, award sizes, and page limits.
Step 1: Choose the agency and read the solicitation
SBIR runs through eleven participating agencies, and they do not share a format. The differences that matter most:
| Agency | How topics work | What the core proposal looks like |
|---|---|---|
| NIH | Investigator-initiated through the parent announcement, plus targeted opportunities | One-page Specific Aims plus a Research Strategy (6 pages for Phase I and 12 for Phase II in the standard instructions); commercial potential argued inside Significance at Phase I; a separate Commercialization Plan of up to 12 pages for Phase II, Fast-Track, and Direct to Phase II |
| NSF | Open topics within broad technology areas | A required Project Pitch first; if invited, a full proposal whose project description includes a commercial opportunity discussion (SBIR.gov's tutorial notes a recommended two to four pages) |
| DoD | Narrow, published topics written by military components | A technical volume answering one topic, with a commercialization strategy of about a page at Phase I and a clear path to a military end user |
| DOE | Published topics | A technical narrative plus a mandatory two-page Phase I Commercialization Plan; DOE declines applications that omit it without review |
Read the full solicitation and, for topic-based agencies, the topic itself before writing anything. A strong NIH-style application can fail an NSF or DoD review on structure alone. DoD topics are released in a pre-release period during which you can contact the topic author directly, and that window closes when the solicitation opens. Use it.
Step 2: Confirm eligibility
SBIR funds for-profit U.S. small businesses with 500 or fewer employees, including affiliates, that are more than 50 percent owned and controlled by U.S. citizens or permanent residents (some agencies, including NIH, also accept companies majority-owned by multiple venture capital, hedge fund, or private equity firms, under specific rules). For SBIR, the principal investigator's primary employment must be with the company at the time of award and during the project. The company must perform at least two-thirds of the work in Phase I and at least half in Phase II.
STTR requires a formal partnership with a U.S. research institution. The small business performs at least 40 percent of the work and the institution at least 30 percent, and the principal investigator may be employed by either. If a university lab is central to your technology, compare the two programs before choosing.
Step 3: Register early
Registrations gate submission and routinely take longer than expected:
- SAM.gov, which issues your Unique Entity Identifier and must be renewed every year.
- The SBA Company Registry at SBIR.gov, which issues the registry identifier every SBIR and STTR application needs.
- The agency portal: eRA Commons and Grants.gov for NIH, Research.gov for NSF, the Defense SBIR/STTR Innovation Portal for DoD.
Start these the day you decide to apply. Our guide to SAM.gov and Grants.gov registration walks through the common delays.
Step 4: Write the aims or objectives first
Whatever the agency calls it, the proposal hangs on a short statement of what you will prove. For NIH this is the Specific Aims page; for NSF and DoD it is the technical objectives. A Phase I version should define one central feasibility question, two aims or objectives at most for most projects, and a numeric success criterion for each. Write it before anything else, because every later section exists to support it. For the method, see writing a Specific Aims page that frames the proposal, and for a complete SBIR Phase I aims page, see our annotated Specific Aims examples.
Step 5: Build the technical plan
Reviewers want feasibility, not certainty. Cover:
- Innovation: what is new and why current approaches fall short.
- Approach: methods, milestones, and go or no-go criteria for each aim.
- Risks and alternatives: the likeliest failure points and what you will do if they happen.
- Preliminary data: even informal bench results that show the idea is plausible.
- Team and facilities: who does each piece of work, including consultants and any research institution partner.
Tie every experiment to an aim. A reviewer should never have to guess which work answers which question.
Step 6: Prepare the budget
Award ceilings come from SBA guidelines that are adjusted for inflation, and agencies set their own levels within them (NSF's current Phase I limit is $305,000, for example; NIH and DoD publish their own). Take the figure from the solicitation, not from an older article. Every line should trace to an activity in the technical plan, and indirect costs should reflect your negotiated rate or the agency's rules for companies without one. Our guide to research and project budgets covers building the justification.
Step 7: Make the commercial case
Commercialization is scored at every agency, though its weight and format vary. At Phase I most agencies want the customer, the market, the competition, and how the company will make money, in a page or a few pages. At NIH Phase II it becomes a 12-page document of its own. Our guide to the SBIR commercialization plan, agency by agency covers what each one asks for.
Step 8: Check compliance, then edit
A noncompliant proposal is returned before anyone reads the science. Check page limits, fonts and margins, required attachments, registration identifiers, and any certifications against the solicitation. Then have someone outside the project read it cold. If they cannot restate your innovation in one sentence, a reviewer will not either.
Plan for more than one cycle. Many funded companies were declined first and won after responding to reviewer feedback; our look at the SBIR success rate by agency explains what the odds look like and why the new caps make each submission count.
When you want a specialist team to take the drafting off your plate, our SBIR and STTR grant writing service works across NIH, NSF, and DoD formats, and our DoD SBIR proposal service covers the topic-driven defense cycle.
