How to Write an SBIR Proposal (2026 Guide)

This guide is for: Startup founders, small business R&D leads, university spinout teams, and consultants writing SBIR or STTR Phase I proposals for NSF, NIH, DOD, DOE, NASA, USDA, or other federal agencies.

Full answer: A winning SBIR proposal needs five things: a clear innovation statement, a feasible research plan with measurable milestones, a commercialization plan that names real customers, a budget that matches the work scope, and strict compliance with agency formatting rules. About 17% in FY2025 of Phase I proposals get funded. For NSF in particular, the first gate is not the full proposal: NSF says Phase I applicants are required to submit a Project Pitch covering objectives, technical innovation, technical risks, market opportunity, and team before being invited to submit.

How to Write an SBIR Proposal: Structure, Scoring, and What Reviewers Actually Look For

Data note: Award amounts, success rates, and application rules were checked again on June 17, 2026 against SBIR.gov tutorials, NSF America's Seed Fund, NIH SBIR/STTR application guidance, DOD SBIR/STTR submission resources, DOE applicant resources, and USASpending.gov. Program parameters may change with annual appropriations and agency solicitations.

Writing an SBIR proposal is different from writing an academic paper, a business plan, or even a typical federal grant application. SBIR reviewers evaluate a unique combination: Is the science sound? Can this team do the work? Will anyone buy this?

If you're a startup founder applying to SBIR for the first time, a small business R&D lead preparing a Phase I proposal, or a consultant helping technology companies pursue federal funding, this guide covers the process from solicitation to submission. For the full scope of non-dilutive funding paths available to small businesses, including state programs and USDA grants beyond SBIR, see our small business grants complete guide. For a comparison of how SBIR and STTR differ, see our SBIR vs STTR guide. For agency-by-agency differences in funding priorities and success rates, see our SBIR agencies comparison.

Most guides tell you to "follow the solicitation instructions." That's necessary but not sufficient. What separates fundable proposals from the rest is often specificity. Specific technical milestones instead of vague objectives. Named potential customers instead of "the healthcare industry." Quantified success criteria instead of "we'll evaluate performance."

Building the first page around one technical risk, a concrete validation plan, and named industrial buyers lets NSF see the experiment, the risk, and the customer in one pass.

This guide covers the exact structure your proposal needs, the scoring criteria reviewers use at each major agency, formatting requirements that vary by agency, and the mistakes that sink otherwise promising proposals. Updated June 17, 2026.

If you are still choosing an agency, compare live small-business opportunities in the Grantsights grant database before drafting. The right agency fit is often worth more than another week of rewriting.

Last updated: June 17, 2026

---

According to NSF America's Seed Fund, the Project Pitch covers technology innovation, technical objectives and challenges, market opportunity, and company and team before a full proposal invitation. According to NSF America's Seed Fund, Phase I can provide up to $305,000 in non-dilutive R&D funding and Phase II can provide up to $1,250,000 over 24 months. According to SBIR.gov, SBIR and STTR proposal preparation depends on the agency solicitation, so the same company may need different structures for NIH, NSF, DOD, DOE, and NASA. According to DOE SBIR materials, Phase I work usually runs 6 to 12 months and Phase II can run up to 2 years, which changes how much technical risk belongs in each proposal. Our analysis is that the strongest SBIR proposals prove one fundable technical risk and one fundable market path, instead of trying to describe the whole company.

June 17, 2026 Source Check: SBIR Proposal Rules Before Drafting

This update keeps the proposal advice tied to agency instructions rather than a single generic template.

June 17 SBIR Proposal Path Table

Do this route check before drafting. It prevents the most common failure: writing one generic SBIR proposal and then forcing it into several agency formats.

For exact program context, compare the SBIR Phase I hub, the NIH and CDC parent SBIR record, and the NSF SBIR program record before writing.

For adjacent route checks, use SBIR grants complete guide, SBIR versus STTR, SBIR agency comparison, SBIR commercialization plan, and the stable SBIR Phase I grant index.

The SBIR Proposal Is Really 4 Different Proposals

Official SBIR.gov guidance says every applicant must respond to an agency solicitation, and the solicitation contains proposal preparation, submission, and evaluation instructions. NSF adds a separate Project Pitch gate before a full Phase I proposal. NIH, DOD, DOE, NASA, USDA, and other agencies then score different evidence. Grantsights' proprietary reading is that "how to write an SBIR proposal" should not be one template. It should be an agency-fit triage before drafting.

For example, a battery startup writing one Phase I narrative for NSF and DOE is likely under-scoring both. The NSF version should make the Project Pitch test easy: high-risk technical innovation, market opportunity, and team. The DOE version should map each task to the published topic language and show why the work advances DOE's mission. The science can be the same, but the score logic is different.

The Grantsights pattern from SBIR pages is that weak proposals usually fail before the technical idea gets a fair read: the solicitation fit is fuzzy, the customer is unnamed, or the milestone is not measurable. That is why this guide starts with agency fit and reviewer evidence, not a blank proposal template.

May 2026 Grantsights Reviewer Memo: The Proposal Must Pass Three Gates

The public SBIR instructions tell applicants what to submit. They do not tell founders which evidence should appear first so a reviewer can score the proposal quickly. Our May 2026 reviewer memo uses three gates before drafting: agency fit, risk proof, and commercial buyer proof.

According to SBIR.gov's proposal preparation tutorial, applicants should understand evaluation criteria, build a proposal preparation schedule, assemble the right team, and address Phase I commercialization. According to NSF America's Seed Fund, the Project Pitch must cover technology innovation, technical objectives and challenges, market opportunity, and company and team before a full proposal invitation. Those two sources point to the same practical rule: do not write the full proposal until the reviewer can see the risk, the experiment, and the buyer.

Grantsights' apply/skip rule: apply when the company can describe one fundable technical risk, one agency reason, and one customer path in plain language. Skip or delay when the project is ordinary engineering, services work, or general startup funding. The right next step may be NSF Project Pitch, NIH specific aims, DOD topic fit, DOE topic mapping, or a non-SBIR funding route.

NSF SBIR Project Pitch Requirements in 2026

For NSF SBIR/STTR, the Project Pitch is the first gate. According to NSF America's Seed Fund, the pitch explains whether the proposed technology fits NSF SBIR/STTR requirements before the company is invited to submit a full proposal.

NSF says Project Pitch responses typically take about 1 to 2 months. An invitation lets you submit a full proposal, but it does not mean NSF has committed to Phase I funding.

The most common pitch failures are easy to spot:

• The pitch reads like a service business or consulting offer, not R&D.
• The technical risk is vague or already solved.
• The market section names a giant sector but no buyer.
• The team section ignores missing engineering, regulatory, or customer discovery roles.
• The pitch describes product features but not the research needed to prove the core technology works.

Best Fit and Bad Fit Checklist Before Writing an SBIR Proposal

Use the Grantsights grant database to compare agency fit, then use the agency solicitation as the source of truth for page limits, forms, budget caps, and review criteria.

Before You Write: The 3 Things You Need to Know

You need to answer three questions before you write a single word of your proposal. If you can't answer all three clearly, you're not ready to submit.

1. What specific problem does your technology solve? Not a general category of problems. One specific, well-defined problem. "Antibiotic resistance" is too broad. "Rapid point-of-care detection of carbapenem-resistant Enterobacteriaceae in emergency departments" is specific enough.

2. What's new about your approach? Reviewers need to understand your innovation in the first paragraph. If your approach is an incremental improvement to existing technology, you'll score low on Innovation. If it's a fundamentally new method, material, algorithm, or architecture, say so directly and explain why it hasn't been tried before.

3. Who will pay for this when SBIR funding ends? The commercialization question. You need at least a preliminary answer: specific customer segments, estimated market size, and a realistic path from Phase I feasibility to Phase III revenue.

Once you can answer these three questions in two sentences each, you're ready to start writing.

Proposal Structure by Agency

Every agency has its own format, but the underlying logic is the same: technical merit, team qualifications, work plan, budget, and commercial potential. Here's what each major agency requires.

NIH SBIR Proposal Structure (R43)

NIH uses the SF424 (R&R) application package submitted through eRA Commons. The key sections are:

NIH scores on five criteria: Significance, Investigator(s), Innovation, Approach, and Environment. Each criterion gets a score from 1 (Exceptional) to 9 (Poor). Three assigned reviewers score independently, then the full panel discusses and assigns an overall impact score.

The Specific Aims page is the most important page in your proposal. Reviewers read it first, and many form their initial impression here. It should state the problem, your proposed solution, the specific objectives of Phase I, and the expected outcomes. One page, no more.

The Research Strategy is where proposals live or die. Six pages sounds short, but it's enough if you're concise. Structure it as:

• Significance (1-1.5 pages): Why does this problem matter? What gap does your innovation fill?
• Innovation (0.5-1 page): What's new? How does it differ from existing approaches?
• Approach (3-4 pages): What will you do? How? What are your milestones and success criteria?

Don't waste space on extensive literature reviews. Cite key references, but focus on your approach.

NSF SBIR Proposal Structure

NSF uses Research.gov for submission, but the Project Pitch comes first. According to NSF America's Seed Fund, Phase I SBIR/STTR applicants must submit a Project Pitch and receive an official invitation before a full proposal will be reviewed. NSF says the pitch covers the technology innovation, technical objectives and challenges, market opportunity, and company and team.

The key full-proposal sections are:

NSF evaluates on three criteria: Intellectual Merit, Broader Impacts, and Commercial Impact. Unlike NIH, NSF's SBIR program is open-topic: you don't respond to a specific research area. You propose your own innovation.

NSF gives you more space (10-15 pages) but expects a different emphasis. Commercial Impact carries significant weight. Your Project Description should dedicate at least 3 to 4 pages to market analysis, customer discovery, competitive positioning, and commercialization timeline.

The Project Summary is your elevator pitch. One page, three mandatory sections: Overview, Intellectual Merit, and Broader Impacts. If any of these three sections is missing, your proposal will be returned without review. This is a hard administrative requirement.

Project Pitch example: A strong NSF pitch does not say "AI platform for hospital operations." It says: "We will test a scheduling model that reduces emergency department imaging wait time by 25% in a 12-week pilot using de-identified historical queue data from two hospital partners. The technical risk is whether the model can predict scanner bottlenecks at least 90 minutes in advance with 80% precision. The first paying customer is a 300-bed regional hospital that currently spends $420,000 per year on overtime tied to imaging delays." That tells NSF what is new, what is risky, who cares, and why a company can be built.

DOD SBIR Proposal Structure

DOD uses the Defense SBIR/STTR Innovation Portal (DSIP) for submission. The proposal consists of multiple volumes:

DOD evaluates on: Technical Merit, Qualifications of Key Personnel, and Potential for Commercialization. Each DOD component (Army, Navy, Air Force, DARPA, MDA, etc.) may weight these criteria differently, so check the specific solicitation.

DOD proposals must directly address a specific topic. Unlike NSF, you don't propose your own idea. You respond to a published topic that describes a military or defense need. Your proposal must demonstrate that you understand the operational problem and that your technology addresses it specifically.

The Technical Volume limit is solicitation-specific and may include figures, tables, and references. Use space wisely. Do not include a 2-page org chart when a half-page table would suffice.

DOE SBIR Proposal Structure

DOE uses the DOE SBIR/STTR portal. Format requirements are similar to NIH but with DOE-specific sections:

DOE provides detailed topic descriptions with specific technical requirements. Your proposal should map directly to the topic's objectives. DOE reviewers are typically subject matter experts who will evaluate your technical depth rigorously.

Writing the Technical Approach: What Gets High Scores

The technical approach is the core of your proposal. It needs to accomplish three things: show that you understand the problem deeply, demonstrate that your solution is feasible, and lay out a work plan that a reviewer can believe you'll actually execute.

Start with Your Innovation, Not the Background

Many applicants spend the first two pages reviewing the literature and describing the problem space. Reviewers already know the field. They're reading your proposal to learn about your solution.

Lead with your innovation. By the end of paragraph one, the reviewer should know exactly what's new about your approach. Then provide just enough background to contextualize why your innovation matters.

Bad opening: "Cancer is the second leading cause of death in the United States, affecting millions of patients annually. Current diagnostic methods include..."

Good opening: "We've developed a microfluidic chip that detects circulating tumor DNA from a standard blood draw in under 30 minutes, compared to the 7 to 14 days required by current liquid biopsy services. In Phase I, we'll validate detection sensitivity and specificity using banked clinical samples across three cancer types."

Define Measurable Milestones

Every Phase I proposal should include 3 to 5 specific, measurable milestones. These are the benchmarks reviewers use to evaluate whether your Phase I will actually generate useful data.

Weak milestone: "Develop initial prototype of the sensor platform."

Strong milestone: "Fabricate sensor prototypes achieving a limit of detection below 10 ppb for lead in water, validated with 20 blind samples and compared against ICP-MS reference measurements."

Each milestone should include:

• What you'll measure
• What threshold constitutes success
• How many samples, tests, or experiments you'll run
• What method you'll use to evaluate results

Address Risks Directly

Reviewers give higher scores to proposals that acknowledge technical risks and describe contingency plans. A proposal that claims everything will work perfectly signals either naivety or dishonesty.

For each major technical risk, state:

• What could go wrong
• How likely it is
• What you'll do if it happens
• Whether it would be fatal to the project or just require a course adjustment

"If the polymer membrane shows insufficient selectivity at room temperature, we'll test performance at higher temperatures (40-60C) and consider alternative polymer formulations identified in our preliminary screening (see Table 2)" is much more convincing than ignoring the risk entirely.

Include Preliminary Data When Possible

Preliminary data isn't formally required for most Phase I applications, but proposals that include it score higher. Even early-stage results, computational predictions, or literature-based feasibility arguments strengthen your case.

If you've done any work on this technology, show it. A graph, a table, a photograph of a bench prototype. Anything that demonstrates you've started and have some basis for believing your approach will work.

If you don't have preliminary data, you can still strengthen your proposal by citing published work from your team (even if in different contexts), referencing validated methods you plan to adapt, or presenting computational models or simulations that support feasibility.

Writing the Commercialization Plan: Where Most Proposals Fail

The commercialization plan is the section reviewers consistently identify as the weakest part of SBIR proposals. Most applicants are scientists or engineers. They write strong technical sections and then phone in the business case.

Don't make this mistake. At NIH, the commercialization plan is an explicitly scored criterion. At NSF, Commercial Impact is one of three evaluation criteria. At DOD, your company's commercialization track record (via the Company Commercialization Report) directly factors into scoring.

What a Strong Commercialization Plan Includes

Target market with numbers. Not "the global diagnostics market is $80 billion in FY2025." Instead: "There are approximately 4,500 clinical labs in the U.S. that perform toxicology screening. The average lab spends $150,000 per year on screening reagents. Our addressable market in year 3 is 200 labs at $50,000 per lab, representing $10 million in annual revenue."

Named potential customers. "We've had preliminary conversations with Quest Diagnostics, LabCorp, and three academic medical centers about beta testing our platform. Letters of support from two of these organizations are included in the appendix."

Competitive analysis. Identify 3 to 5 competitors or alternative approaches. For each, explain what they offer, what they charge, and why your solution is better. "No competitors" is never true and will hurt your score.

Revenue model. How will you charge? Per unit? Subscription? Licensing? What's your target price point? What are your estimated gross margins?

Phase III funding path. How will you fund the transition from Phase II to market? Will you seek venture capital? Strategic partnerships? Additional federal contracts? Revenue from early adopters? Be specific about your funding strategy.

Regulatory pathway (if applicable). If your product requires FDA clearance, EPA registration, or other regulatory approval, outline the pathway. Which regulatory category? Predicate devices? Estimated timeline and cost?

Letters of Support Make a Difference

Letters of support from potential customers, partners, or end users carry real weight with reviewers. They demonstrate that you've done customer discovery and that real organizations are interested in your technology.

A strong letter of support:

• Comes from a decision-maker (VP, Director, CEO), not a random contact
• Describes a specific problem the organization faces
• Explains why the proposed technology could address that problem
• Expresses willingness to test, evaluate, or purchase the technology
• Mentions specific quantities, budgets, or timelines when possible

A generic letter that says "we support this new technology" adds nothing. A letter that says "We currently spend $2M annually on manual inspection processes that your automated system could reduce by 40%, and we'd be willing to conduct a 6-month pilot at our facility" is extremely valuable.

Budget Preparation: Getting It Right

Your budget must be reasonable, justified, and within the agency's cap. Reviewers evaluate whether your budget matches your work plan. A $250,000 budget for a project that could clearly be done for $150,000 raises concerns. A $100,000 budget for a project that requires $300,000 worth of work raises different concerns.

Key Budget Categories

Personnel (typically 60-70% of total). List each team member by name, their role, their percent effort, and their salary basis. The PI must be budgeted at 51% effort or more to meet the SBIR employment requirement.

Consultants. Rate must be reasonable, typically $100-$300/hour depending on expertise. Justify why you need external expertise and what they'll contribute.

Equipment. Items over $10,000 require justification. Most Phase I budgets don't include major equipment because the feasibility stage rarely needs it.

Supplies and materials. List categories and estimated costs. Reviewers want to see that your estimates are based on real prices.

Subcontracts. Remember the limits: 33% maximum for SBIR Phase I. Each subcontract needs a statement of work and budget.

Travel. One to two trips to the agency for project meetings is standard. Excessive travel budgets draw scrutiny.

Indirect costs. Use your negotiated rate if you have one. If you don't, most agencies accept a de minimis rate of 15% of modified total direct costs. Our budget categories guide explains this in detail. For NIH-specific SBIR guidance, see our NIH SBIR/STTR guide. For NSF-specific programs, see our NSF SBIR guide.

Common Budget Mistakes

• Personnel effort doesn't match the work plan. If your PI is budgeted for 51% effort but the work plan shows them leading every task, reviewers will question whether 51% is enough time.
• No justification for subcontract costs. Every subcontract needs a scope of work, budget breakdown, and explanation of why this work can't be done in-house.
• Forgetting indirect costs. If you don't include them, you're leaving money on the table. If you overstate them, reviewers will flag it.
• Budget exceeds the cap. This sounds obvious, but 2% to 3% of proposals are returned because the total budget exceeds the agency's Phase I maximum.

Formatting Requirements: The Details That Get Proposals Rejected

Between 5% and 10% of SBIR proposals are rejected before peer review for administrative non-compliance. Formatting violations are the most common cause. Here are the agency-specific rules you must follow.

NIH Formatting

• Font: Arial, Georgia, Helvetica, or Palatino Linotype at 11 points or larger. Times New Roman is NOT acceptable at NIH.
• Margins: 0.5 inches minimum on all sides
• Line spacing: No more than 6 lines per vertical inch (approximately single-spaced)
• Page limits: Specific Aims = 1 page; Research Strategy = 6 pages (Phase I)
• File format: PDF only

NSF Formatting

• Font: Times New Roman, Arial, Courier New, or Palatino at 10 points or larger
• Margins: 1 inch on all sides
• Line spacing: Single or double (check solicitation)
• Page limits: Project Summary = 1 page; Project Description = 10-15 pages; Budget Justification = 5 pages
• File format: PDF uploaded through Research.gov

DOD Formatting

• Font: Times New Roman or Arial at 11 or 12 points
• Margins: 1 inch on all sides
• Page limits: Technical Volume = typically 20 pages (varies by component)
• Must include header/footer with company name, topic number, and proposal number on each page
• File format: PDF uploaded through DSIP

Don't test the limits. If the solicitation says 11-point font minimum, don't use 10.5. If it says 1-inch margins, don't use 0.9 inches. Reviewers and administrative staff check these requirements, and violations result in automatic rejection at most agencies.

Reviewer Psychology: How Reviewers Read Proposals

Understanding how reviewers read proposals can help you write a better one. Here's what actually happens during review.

Reviewers Are Tired

Each reviewer reads 8 to 15 proposals per review cycle, often in a compressed timeframe. By proposal number 10, they're skimming. Make your proposal easy to read. Use clear headings, short paragraphs, bullet points for lists, and bold text for key findings.

The First Two Pages Set the Tone

Reviewers form an initial impression within the first two pages. If your Specific Aims (NIH) or executive summary (DOD) is poorly written, vague, or hard to follow, the reviewer starts with a negative bias that's difficult to overcome.

Front-load your strongest material. Lead with the innovation. State the commercial opportunity early. Don't build up to the punchline on page 8.

Reviewers Score Against Criteria, Not Against Other Proposals

This is a common misconception. Reviewers don't rank proposals against each other. They score each proposal independently against the published evaluation criteria. A reviewer can give multiple proposals the same score. This means your job isn't to be "better than the competition." Your job is to score well against each criterion.

Reviewers Look for Reasons to Fund, Not Reject

Most reviewers genuinely want to fund good science. They're looking for proposals that excite them. If your innovation is compelling and your plan is feasible, a few minor weaknesses won't sink you. But if your proposal is mediocre across the board, there's nothing for the reviewer to champion.

The Discussion Matters

After individual scoring, the panel discusses each proposal. Strong proposals get brief, positive discussions. Weak proposals get brief, dismissive discussions. Borderline proposals get lengthy discussions where one reviewer's enthusiasm (or concerns) can shift the group score. This is why it pays to have at least one section of your proposal that's genuinely impressive: it gives your advocates something to argue for.

Common Rejection Reasons and How to Avoid Them

Here are the most frequent reasons SBIR proposals don't get funded:

1. Innovation isn't clear. Reviewers can't identify what's new. Fix: State your innovation in the first paragraph. Contrast it explicitly with the current state of the art.

2. Work plan is vague. "We'll optimize the formulation" tells reviewers nothing. Fix: Specify what you'll vary, how many conditions you'll test, and what metrics you'll use to define "optimal."

3. Milestones aren't measurable. "Demonstrate feasibility" isn't a milestone. Fix: Define success criteria with numbers: accuracy targets, throughput requirements, cost thresholds.

4. Commercialization is an afterthought. Generic market statements without customer validation. Fix: Talk to potential customers before you write the proposal. Include their feedback and letters of support.

5. Team can't do the work. The PI lacks relevant experience, or the team is too small for the proposed scope. Fix: Add consultants, subcontractors, or co-investigators who fill gaps. Include their biosketches.

6. Budget doesn't match the work plan. Too much money for too little work, or not enough money for an ambitious plan. Fix: Walk through each task in your work plan and estimate the personnel time, supplies, and costs required. The budget should map directly to the activities.

7. Didn't follow instructions. Wrong font, exceeded page limits, missing required sections. Fix: Create a compliance checklist from the solicitation and verify every item before submission.

Phase I to Phase II: Planning Ahead

While you're writing your Phase I proposal, think about Phase II. Reviewers want to see that you have a vision beyond the feasibility study.

Your Phase I proposal should briefly describe:

• What a successful Phase I would demonstrate
• What questions Phase II would answer
• What the Phase II prototype or product would look like
• How Phase II results would position you for Phase III commercialization

You don't need a detailed Phase II plan. But a paragraph or two showing that Phase I connects to a larger development trajectory demonstrates strategic thinking that reviewers value.

At NIH, Phase II success rates for applicants with strong Phase I results are significantly higher than Phase I success rates. A well-executed Phase I sets up a strong Phase II application. Think of Phase I as the first chapter of a multi-chapter story, not a standalone project.

The 10-Week Writing Timeline

Here's a realistic schedule for writing an SBIR Phase I proposal, starting 10 weeks before the submission deadline. Adjust based on your experience level and the complexity of your project.

Don't compress this timeline. Rushed proposals read like rushed proposals, and reviewers can tell. If you're 6 weeks out from a deadline and haven't started, consider waiting for the next solicitation cycle.

If you're new to federal proposals in general, our guide to writing federal grant proposals covers foundational skills that apply across SBIR and traditional grant programs. For a deeper look at SBIR funding, award data, and agency-by-agency strategy, see our complete SBIR grants guide.

What SBA's SBIR Portal Doesn't Tell You About SBIR Proposals

Each agency's SBIR program has different evaluation criteria weightings, and SBA's website doesn't show this clearly. DOD weighs technical merit and military relevance roughly equally. NIH weighs significance and investigator qualifications more heavily. NASA's SBIR program has a strong emphasis on commercialization potential from Phase I. Read each agency's solicitation, not just the SBA overview.

The commercialization section of an SBIR proposal is where most first-time applicants lose points. Reviewers want to see a specific customer, a realistic price point, and evidence of market demand. 'We will sell to government agencies' is not a commercialization plan; name the agency, the program office, and the contract vehicle.

Phase I awards are typically $150,000 to $275,000 depending on the agency, but the real value is the Phase II option. Companies that don't plan their Phase I with Phase II milestones in mind often struggle to win the follow-on. Build your Phase I work plan so its deliverables directly feed Phase II objectives.

Related Guides

• SBIR Commercialization Plan

---

Sources

• SBIR.gov: Preparing a Proposal - Official proposal preparation guidance
• SBIR.gov: Proposal Evaluation Criteria - Official evaluation-criteria tutorial
• SBIR.gov: Agency Solicitation Overview - Why each agency solicitation controls proposal structure
• NIH SBIR/STTR Application Guide - NIH-specific formatting and submission requirements
• NIH Page Limits - Current NIH application page-limit table
• NSF SBIR: Project Pitch - NSF's required invitation gate before full Phase I proposal review
• NSF SBIR: How to Submit - NSF proposal submission overview
• NSF SBIR: Full Proposal - NSF proposal structure and evaluation criteria
• DOD SBIR/STTR: Proposal Templates - DOD submission portal and volume requirements
• NIH SBIR/STTR Guide for Reviewers - What reviewers are instructed to score and how
• DOE SBIR Applicant Resources - DOE proposal preparation guidance