TL; DR
- An open topic SBIR does not tell you what to build. Every DoW SBIR topic carries three pieces: a problem, a customer who owns it, and a solution. A standard topic gives you all three, an open topic gives you none, and a focused open topic gives you the first two and leaves the solution to you.
- The focused open topic is where most of the current activity sits. The agency brings a named problem and an identified customer, and you propose how to solve it. That removes the hardest part of a fully open submission, which is finding the customer yourself.
- AFWERX broke the standard DoW mold in 2018 with a fully open SBIR program, though you still had to find the customer yourself to advance. In 2026 it corrected toward the middle path. Now AFWERX brings the customer and the problem, and you pitch how to solve it. That keeps the innovation coming while giving you a clearer path to delivering for an Air Force customer.
- A 2022 law spread the model across the whole department. Section 7 of the SBIR and STTR Extension Act requires at least one open topic announcement at each DoD component every fiscal year, though it never defines how open an open topic has to be.
- The Army runs all three types at once, through xTech prize competitions, SBIR topics, and standing BAAs. xTech|Search is the Army’s only open topic prize competition, inviting proposals addressing “any valid Army need or capability gap.” The rest of the xTech lineup names a problem area.
- The Navy came late and went all in on focused open topics, publishing out of NAVSEA, NAVAIR, NAVWAR, and ONR separately. No fully open topic yet.
- Three of the widest doors are not SBIR topics at all. DARPA’s ERIS takes a seven minute video with monthly reviews, DIU publishes Areas of Interest that close in about 14 days, and SOCOM works through SOFWERX events months before any solicitation exists.
- NSF and NIH have run open topics for decades. NSF states that its programs “do not solicit specific technologies.” NIH describes its parent announcements as researcher initiated. Neither becomes your customer, which is the tradeoff.
- Per firm proposal caps are now in effect. Every agency had to set a limit on Phase I submissions before the fiscal year began. Published numbers are scarce, so check each agency before planning a year of submissions.
- Winning takes strategy, intelligence, and volume. The approach we take with our clients at Grant Engine is to submit across agencies and topic types without losing focus, do the research and talk to the technical points of contact, build every submission around the end user, and start early enough that a 14 day window is workable rather than impossible.
Open-topic SBIR funding lets startups propose their own technology within an agency mission instead of building only to a narrowly prescribed specification. The opportunities vary: some invite a company-defined application, while focused open topics name a problem and leave room for different solutions. This guide maps those pathways across defense and civilian agencies, explains how to assess customer requirements, and shows how to build a funding strategy for 2027.
Three Types of SBIR Topics: Standard, Open, and Focused Open
Every SBIR topic ever written points in one of three directions:
- Develop this specific solution for this specific challenge. The agency has defined the problem and largely decided the answer. Your job is to propose the best way to execute it, better than anyone else competing for the same topic.
- Develop anything within our mandate. Every agency has one. The Air Force’s is all things Air Force. DARPA’s is futuristic defense technology. NASA’s is space, NSF’s is science, NIH’s is health. Bring them something credible inside that boundary and they’ll consider it.
- Develop something for this specific challenge, or within this industry, and you decide how to solve it. The agency names the problem with real precision. You propose the solution. This is the focused open topic.
Today, for a dual use or deep tech company chasing defense funding, the focused open topic is often your best opportunity. Before we jump in, let’s first talk about how we got here.
The Department of War, previously the Department of Defense, historically asked almost exclusively for the first. Requirements flowed top down. A program office identified a gap, wrote a topic describing the technology it wanted, and industry competed to build it. If your company was doing something genuinely novel that nobody in the Pentagon had thought to ask for, there was often no door to knock on.
AFWERX broke that mold in 2018. Its open topic opened a channel that hadn’t existed before, letting companies outside the traditional defense industrial base bring the Air Force technology it hadn’t thought to ask for. The reasoning was practical. Adversary nations such as China, Iran, Russia and others are moving quickly on commercial technology, and an agency that funds only the ideas already inside its own head gives up innovation ground rapidly.
The scale of what followed is worth seeing. AFWERX reports more than 9,400 SBIR and STTR contracts totaling $7.47 billion since 2020, and you can browse the range of what got funded in its portfolio dashboard.
Now AFWERX is reorienting toward the golden mean. Not wide open, and not hyper specific. We have a problem, how can you solve it. Even though AFWERX is moving toward that middle path, that doesn’t tell the whole story. AFWERX sent ripple effects across the Department of War, challenging every sub-agency to rethink how it engages industry and how it stays competitive.
The Army has made major changes, including xTech competitions, focused open topics of its own, and other ways to engage industry such as CATALYST. The Navy has moved significantly, especially since the SBIR reauthorization in 2026, with a heavy dose of focused open topics running out of separate systems commands. DARPA takes seven minute video submissions every month against no fixed deadline. SOCOM runs its front door as an event calendar rather than a solicitation. And outside the Department of War, NSF, NIH, and DOE have been funding the second and third directions all along.
What Is an Open Topic SBIR, and How Is It Different From a Standard SBIR Topic?
To understand how an open topic differs from a standard topic, start with what the Department of War SBIR program is actually for. It funds your development so that if you succeed, you can solve a challenge the DoW needs addressed. The purpose of Phase I and Phase II is Phase III, a sole source contract with the agency that funded you. Your non-dilutive patron intends to become your customer. That transition is never guaranteed, and plenty of solid Phase II efforts end at Phase II. But it is the outcome the program is built around, and it shapes how every topic gets written.
With that in mind, the open topic has a challenge built into it.
Every DoW SBIR topic involves three pieces: a problem, a customer who owns it, and a solution. What separates the three kinds of topic is how many of those the government hands you before you start writing. An open topic SBIR is the one that hands you none of them, which is why it asks the most of you even though it appears to ask the least.
A standard SBIR topic gives you all three. The problem is defined, the end user is behind it, and the solution is largely specified. A program office wrote that topic because it has a requirement and a budget line, or at the very least a high level of interest. Your job is to propose the best way to execute it, better than anyone else bidding on the same thing.
An open topic gives you none of them. No stated problem, no named customer, no required solution. You’re making three cases at once: that the problem is real, that a specific end user inside the department owns it, and that your technology solves it. You have to go find that customer yourself. A proposal that skips that work reads as a technology looking for a problem. It isn’t a lower bar. It’s a different one, and usually a heavier one.
A focused open topic gives you two of the three. The problem is named, often in real detail down to performance thresholds and cost targets, and the customer is baked in. Somebody in a program office wrote that problem down because they have it. What’s left open is the solution, or at minimum the path you take to reach the outcome they’ve described. That’s your interpretation to make, and it’s where a differentiated company can separate itself.
Why Does Every Department of War Sub-Agency Run an Open Topic SBIR?
Because a 2022 law told them to. Section 7 of the SBIR and STTR Extension Act of 2022, titled Program on Innovation Open Topics, required the Department of Defense to conduct “not less than 1 open topic announcement at each component of the agency per fiscal year.”
So the mandate spread the model AFWERX introduced across the entire department. But sub-agencies such as the Army and the Navy likely had their own reasons to want it. GAO found that 43 percent of the 1,001 companies receiving Air Force open topic awards between FY2018 and FY2020 had no prior federal contracts at all, against 14 percent of the companies winning conventional topics. Open topic proposals to the Air Force grew from 575 in FY2018 to more than 4,700 in FY2020. More companies applying, and more of them new to the government, means more cutting edge technology reaching the warfighter. That is an argument a program office makes to itself without needing a statute.
What the law did not do is define how open an open topic has to be. One announcement per sub-agency per fiscal year is a floor, not a format, and they have used that latitude very differently. Some built genuinely wide aperture programs. Others adopted a focused open topic approach. Others met the requirement with something closer to a conventional topic wearing an open topic label.
That’s what the rest of this article maps, sub-agency by sub-agency, starting with the Army.
Where open topics live across the federal government.
Where Are the Army’s Open Topics? xTech, SBIR, and Standing BAAs
The Army runs open topics primarily through three separate vehicles: xTech prize competitions, SBIR topics published directly, and standing Broad Agency Announcements. Each one takes a different kind of submission, and a company that only watches SBIR topic releases is seeing a fraction of what the Army offers.
Most of this sits under the Army Pathway for Innovation and Technology, or PIT, which houses the FUZE portfolio and its four pathways: xTech, SBIR|STTR, TMI, and ManTech. All four touch dual use technology. For a small business looking for a way in, xTech and SBIR|STTR are usually the two front doors.
xTech is a family of prize competitions, not a single program. Several run at any given time, each with its own aperture. You submit a white paper, selected companies advance to pitch, and winners take cash prizes. Since 2018 the program has awarded more than $30 million in non-dilutive cash prizes across more than 50 competitions, against $285 million in follow-on contracts.
That ratio tells you what xTech is actually for. The prizes are modest by design, typically running from $5,000 at the semifinalist stage up to a couple hundred thousand dollars for a top finisher, and roughly ten dollars of follow-on contract value has flowed for every prize dollar awarded. The money is not really the point. The value is exposure to Army stakeholders, which amounts to free marketing in rooms you would otherwise struggle to get into, plus a direct look at the problem sets they are worried about so you can tailor both what you build and how you pitch it.
The structural advantage comes after. Winners are typically invited to submit a Phase I proposal under a topic only they can apply to. You are no longer competing against an open field, which can meaningfully improve your odds of converting a win into a contract and starting a real relationship with the Army.
The cadence matters as much as the aperture. More than 50 competitions since 2018 works out to several new ones every year, with multiple open at any given time. xTech|Search alone has awarded over $14 million in cash prizes to more than 820 semifinalists, finalists, and winners across its first nine iterations. If one competition does not fit what you are building, another will likely open within a few months.
xTech|Search is the Army’s annual open topic. It is, in the Army’s own words, its only open topic prize competition. The most recent round, xTech|Search 10, invited submissions addressing “any valid Army need or capability gap.” While it does single out five priority areas, including all arms maneuver, command and control, layered protection, adaptive sustainment, and cross domain fires, it is genuinely an open topic and is looking for any technology that can help improve the Army’s capabilities. Up to 50 semifinalists receive $5,000, up to 20 finalists receive $20,000, and the top three take $200,000, $100,000, and $50,000. Finalists can then submit for up to $300,000 in Phase I funding under topic ARM26BX05-NP011.
The rest of the lineup spans the spectrum, and not every competition sits in the same place. Some are focused open topics in the fullest sense, naming a technology area and leaving the approach to you. xTech|Disrupt covers soldier power generation and storage alongside air and missile threat defeat. xTech|Phantum went after quantum and photonics. xTech|Kinetic Reach targets agile sustainment.
Others read much closer to a standard topic. xTech|Edge Strike: Ground asks for autonomous ground systems that can “detect, identify, track, and neutralize threats, deliver precision fires, emplace obstacles and defensive positions, carry and distribute supplies, or evacuate casualties,” against a $2 million prize pool. xTech|Disrupt Fires seeks low cost interceptors for cruise missiles and unmanned systems. xTech|Inversion is different again, built for companies willing to commercialize Army owned intellectual property.
The SBIR program publishes focused open topics directly. These name a problem area and leave the technology to you, and the range is wider than most companies expect. The AI/ML Focused Open Topic, A244-P037, covered six AI sub-fields from adversarial data validation to synthetic data generation, at up to $250,000 for Phase I or $2 million for Direct to Phase II. The Li-ion 6T Battery Focused Open Topic, A254-P050, named a materiel problem, batteries meeting MIL-PRF-32565 with better weight, energy density, and cycle life, and left the chemistry to you. The Canine Readiness and Performance Focused Open Topic, A254-P001, asked for technologies supporting the readiness and operational effectiveness of military working dogs, explicitly excluding purely veterinary use cases, at up to $250,000 over six months.
Compare any of those to a standard topic from the same program, affordable Ka-band metamaterial based electronically scanned array radar, topic A254-049, where the technology itself is specified. Same program, opposite ends of the spectrum.
Standing BAAs are the quietest option. A Broad Agency Announcement carries no topic list and accepts white papers on a rolling basis. There are still deadlines, but they run in years rather than weeks, and an expiring BAA is usually renewed or replaced by an updated one.
Those vehicles are more common than most companies realize, and the three below are only examples. The DEVCOM Army Research Laboratory BAA, W911NF-23-S-0001, takes submissions across ARL’s research portfolio year round. The DEVCOM C5ISR Research and Technology Integration BAA, W56KGU-23-R-RTID, runs through November 2027 and covers soldier and platform mission equipment, spectrum dominance and intelligence, and advanced technology including AI, machine learning, sensors, and autonomous systems. The Army Applications Laboratory BAA for Disruptive Applications, W911NF-24-S-0008, stays continuously open through April 2029 across 16 technology areas, can award through Other Transactions as well as contracts and grants, and explicitly invites “the widest possible range of sources.” All three start with a white paper, and all three welcome small businesses alongside universities and nonprofits.
Most of these lean toward research rather than procurement of a specific system, which makes them a better fit for an earlier stage technology than a prototype ready to transition. They also draw nowhere near the proposal volume of an SBIR topic release.
Where Are the Navy’s SBIR Open Topics? NAVSEA, NAVAIR, NAVWAR, and ONR
The Navy embraced the open topic later than the Air Force or the Army. That changed after the SBIR program was reauthorized in the spring of 2026, following a lapse that began in late 2025 and that we covered in SBIR Reauthorization: What Just Happened and What It Means for You. The Navy appears to have used that pause to reorganize how it runs its SBIR program, and what came out the other side leans noticeably harder on pulling in a wider range of innovation. The Navy has not described it that way publicly, so take that as our read rather than official policy, but the before and after is visible in the topic lists.
There has not been a fully open topic. There have been quite a few focused open topics.
Part of what makes the Navy harder to track is how it is organized. Its acquisition work splits across systems commands, roughly along the lines of sea, air, and information warfare. NAVSEA covers ships and ship systems. NAVAIR covers aviation and the weapons that go with it. NAVWAR covers networks, communications, and positioning, navigation, and timing. Each publishes its own topics, so there is no single Navy open topic page and no single schedule.
The counter unmanned air systems topic, DON26BX03-NP002, is the clearest example of the pattern. NAVAIR and NAVSEA ran it jointly, with a technical question and answer period before proposals opened in the summer of 2026. The threat is specified. The required outcome is specified. How you detect, track, and defeat the threat is entirely your call.
NAVWAR has been the most active of the three, and its open topics are easy to spot because the Navy puts the words in the title. “Open Topic for Resilient Wideband RF Photonic Architectures for Assured Communications and PNT in Contested Electromagnetic Environments,” topic DON26BX01-NP001, named the environment and the problem, reducing receiver noise, holding signal integrity under jamming, capturing wideband signals, and fitting on ships and aircraft, then left the architecture to the proposer. A later NAVWAR topic, DON26BX05-NP004, asked for unified assured positioning, navigation, and timing decision support.
The pattern has carried into the new fiscal year. FY-27 Release 1 includes DON27BX01-NP001, an ONR Open Topic for low cost heads up displays for virtual and augmented reality, alongside five conventional topics. Same structure as everything above: a named need, an open solution.
For tracking any of this, use the Defense SBIR/STTR Innovation Portal. That holds for every DoW SBIR opportunity, and it matters more for the Navy than for most. AFWERX, Army SBIR, and xTech all run modern, well organized sites that are genuinely useful on their own. The Navy’s is dated by comparison, which makes DSIP the practical place to watch.
Beyond SBIR, the Office of Naval Research runs a Long Range Broad Agency Announcement for Navy and Marine Corps science and technology. It is reissued annually, accepts white papers throughout the year, spans basic research through advanced technology development, welcomes industry alongside academia with no set-asides, and can award through contracts, grants, cooperative agreements, or Other Transaction Agreements.
What Changed at AFWERX SBIR in 2026, and What It Means for Your Proposal
AFWERX has replaced the propose anything open topic with focused open topics across its 2026 releases. The aperture narrowed, and the award sizes went up.
The old model is worth seeing for comparison. AFWERX Open Topic 25.5/E Release 10 closed on August 5, 2025, built on the same fully open topic model AFWERX was founded on: bring us whatever you are building and tell us what problem it solves for the Air Force. A Phase I award was $75,000 for SBIR and $110,000 for STTR. Small checks, wide door. The catch was that the door only opened if you found the customer yourself and showed your technology solved a challenge they were actually facing.
AFWERX, then and now.
Take topic DAF26BX06-DP027 to see how the approach has changed. It is titled “OPEN, Proliferated Low-Altitude Detection for C-sUAS.” The word open is right there in the title. So is a requirement specifying Group 1 through 3 detection, a threshold range of 1 to 300 meters against an objective of 500 meters, 100 meter per node coverage building to 10 kilometers in aggregate, integration with Medusa command and control, and a full rate production cost under $100,000 per unit. It is a Direct to Phase II topic worth up to $2 million plus $50,000 in technical and business assistance over 24 months. The Air Force is telling you plainly that it has this need. It is still leaving it to you to say how to meet it.
SpaceWERX shows the same pattern from a different angle. Topic DAF26BX06-NV510 names the mission, next generation missile detection, warning, tracking, and reporting with Space Systems Command’s Space Sensing System Delta 84 and the OPIR Tap Lab, then states that “focused topic areas can include any part of the following” and lists four broad lanes, one of which invites new sources and methods for missile warning that sit outside the current operational paradigm. That last one is close to an open invitation inside a focused topic.
So the trade is real. You no longer get $75,000 to propose anything. Depending on the topic you can win a Phase I worth a couple hundred thousand dollars, or go straight to a Direct to Phase II worth $2 million, to solve something they have already defined. And in the SpaceWERX case you still get meaningful room to define how.
There is a second half to that trade, and it matters more than the award size. A focused open topic arrives with the customer already identified. Somebody in the Air Force wrote that requirement because they have the problem, which means the path from Phase I or Phase II to a Phase III contract is visible before you start writing. That is the work the old model left entirely to you.
So what does that mean going into 2027? AFWERX stated in an April 2026 post reauthorization webinar that the focused open topic model applied to every 2026 release, and that is what the 2026 releases delivered. We have found no announcement that the wide open topic is permanently discontinued, so it could come back. If it does, that is good news for a company with something genuinely novel and no obvious customer yet. But if 2026 is any indication, 2027 is more likely to bring a heavy dose of focused open topics. The early FY27 Air Force releases have so far carried conventional topics rather than open ones. We covered the full picture, including what this means for how you position, in AFWERX in 2026: What Changed, Why It Matters, and How to Win.
Where Can You Get DoW Funding Outside the SBIR Program? ERIS, DIU, and SOCOM
Some of the widest doors in the Department of War are not SBIR topics at all. Three in particular: DARPA’s ERIS program, the Defense Innovation Unit’s rapid prototyping OTAs, and SOCOM’s events run through SOFWERX. Both DARPA and SOCOM do run SBIR programs, which may include the occasional focused open topic. But the more broadly open programs are the ones below, and for a dual use company they are often faster and less crowded than a topic release.
DARPA runs a standing open call called ERIS. The Expedited Research Innovation System has no submission deadlines, just monthly review cycles and topic areas that change over time. The submission is a video: up to seven minutes, unclassified, plus a form and four supplemental slides. No proposal, no template. DARPA collects monthly and returns results within 30 days of each period closing.
Submissions assessed as awardable go into a marketplace where Department of War activities can select, negotiate, and award them through Other Transaction authority, with negotiable terms and no stated price ceilings. Being assessed as awardable does not guarantee an award. It makes you visible to buyers, which is not the same as being bought.
The signal worth watching is the Shopping Notice. DARPA offices publish these to announce they are actively searching the marketplace for a specific capability, with a date range and a note that awards may come before the notice closes. One recent notice had the Strategic Technology Office looking for moderate power high energy lasers that can run at high duty cycle in size, weight, and power constrained platforms. Read it the way you would read a focused open topic: the need is named with precision, the solution is not. The difference is that you do not submit against the notice. DARPA shops among solutions already in the marketplace, which is an argument for getting your video in before a notice appears that fits you.
DARPA’s ERIS program is open to US based firms of any size, universities, nonprofits, and individuals, and DARPA states that the technology readiness level you report “is not used to determine whether solutions are assessed as awardable/not awardable.”
The Defense Innovation Unit runs the closest thing to a focused open topic outside SBIR. DIU publishes Areas of Interest, each roughly a page, each live on diu.mil for about 14 days. An AOI names a capability gap and asks what you have. You respond with a solution brief of no more than 5 pages or 15 slides, which DIU’s own guidance says must “stand entirely on its own merits, no cross references, no assumed context,” and tie explicitly to the gap the AOI defined.
From there it moves fast. Phase one is a pass or fail screen on relevance, innovation, and technical merit, evaluated within about 10 days. Phase two is a 60 minute live session with DIU and mission experts where you show commercial viability, a realistic schedule, and a rough order of magnitude price. Phase three is a request for prototype proposal, where scope, funding, and terms lock in over roughly 30 to 60 days. Awards come as Other Transaction Agreements under 10 U.S.C. 4022, outside the FAR entirely. The eligibility rules favor companies like yours: a prototype project generally needs participation by a nontraditional defense contractor, or all significant non government participants being small businesses, or a cost share of at least one third.
The thing to understand about DIU is the clock. Fourteen days is not enough time to work out your positioning, build a customer story, and write a brief from scratch. By the time an AOI appears that fits what you build, the preparation needs to already be done.
What you get for that compressed window is scale. GAO reported that DIU awarded 450 prototype agreements between FY2016 and FY2023 against roughly $1.7 billion in obligations, with individual FY2023 projects running to $51.3 million for Real-Time Information and Effects and $52.2 million for Blended Wing Body. DIU also reports transitioning 62 prototype agreements into production awards and contracts with a combined ceiling value above $5.5 billion. Two weeks of work, in the right case, buys a shot at something an order of magnitude larger than a Phase II.
An undervalued pathway is the SOFWERX event calendar. Special Operations Command works largely through SOFWERX, which hosts assessment and collaboration events tied to specific program offices. A recent stretch of the calendar included air launched effects work with PEO Fixed Wing, distributed micro sensors with PEO SOF Warrior, an unmanned surface vehicle event with PEO Maritime, and a small business networking event capped at two attendees per organization.
The benefit is access. These events put you in a room with SOCOM stakeholders, which gives them a chance to get to know you and what you have built, and gives you a chance to understand what they actually need. When a program executive office posts a collaboration event, it is publishing its problem, often months before any solicitation exists. That is slower than responding to a topic, but it puts you in the conversation while the requirement is still being shaped rather than after it has been written.
Elsewhere across the department, DTRA, DMEA, DLA, DHA, and the Office of Special Warfare all appear regularly in the FY26 SBIR releases, though generally with conventional topics rather than open ones.
Open Topic SBIR Outside Defense: NSF, NIH, and DOE
The open topic model the Department of War spent the last several years adopting is one NSF and NIH have been running for decades, and it fits what those agencies exist to do. Both are there to push cutting edge science forward. Neither has much reason to tell you what to build.
For NIH that means health related science, broadly defined: diagnostics, medical devices, therapeutics, digital health tools, biotech instrumentation. For NSF it means science across essentially every domain. DOE has been moving in the same direction more recently.
One difference from the DoW is worth naming upfront. These agencies are not trying to become your customer. There is no Phase III sole source contract waiting at the end. What they offer instead is substantial non-dilutive funding and a serious validation signal, and both investors and future customers tend to take that signal seriously.
NSF does not tell you what to build, and says so. Its current solicitation states that “the NSF SBIR/STTR programs fund across nearly all technology areas and market sectors; the programs do not solicit specific technologies or procure goods and services from startups and small businesses.” The program publishes 31 technology topic areas, from advanced manufacturing and AI to quantum information technologies, semiconductors, and space, and you pick the one that fits. Those are organizing categories, not requirements.
NSF funds across 31 technology areas.
There is good money here. Phase I runs up to $305,000, Phase II up to $1.25 million, and Fast-Track up to $1,555,555, combining a $400,000 Phase I component with $1,155,000 for Phase II. NSF restarted the program in May 2026 with $250 million behind it, including a $40 million pilot emphasis on next generation scientific instrumentation. Full proposal deadlines run on a fixed annual cycle, with the next two on March 4 and July 7, 2027.
There is also a gate, and it works in your favor. Before you can submit a full Phase I or Fast-Track proposal, you have to submit a Project Pitch and be invited. The pitch is a short document covering your technical innovation, the technical objectives and challenges, the market opportunity, and your team. NSF comes back with either an invitation to submit or a decline. The point is to spare you the weeks of work a full NSF proposal takes when the answer was always going to be no.
NIH gives you two doors. You can apply under the parent announcements, which is the fully open route, currently PA-27-100 for SBIR and PA-27-102 for STTR. Or you can respond to a targeted announcement from one of the individual Institutes and Centers, many of which publish their own funding opportunities and Notices of Special Interest. Twenty three Institutes and Centers participate, plus the Office of Research Infrastructure Programs. NIH is explicit that applications under the parent announcement are not limited to the areas those targeted announcements describe, so the open door stays open either way.
NIH small business funding runs through 24 doors.
Standard due dates are September 5, January 5, and April 5. Budget guidelines follow SBA levels, most recently published at $314,363 for Phase I and $2,095,748 across the duration of Phase II, and they adjust annually. Individual Institutes can set higher caps, and several do. The National Cancer Institute lists $700,000 for Phase I and $2.5 million for Phase II. The National Institute of General Medical Sciences lists $700,000 and $3 million. Check the cap for the Institute you are targeting before you build a budget.
DOE and NASA have both moved toward the model. DOE’s FY26 Phase I Genesis Mission opportunity named four broad areas, scaling the biotechnology revolution, AI for quantum computing and networking, designing materials with predictable functionality, and achieving AI driven autonomous laboratories, then required a 700 word pitch before any full application. Roughly 40 awards of up to $250,000 against $10 million total. Those are focused open topics in structure, with the added benefit of sitting squarely inside DOE’s energy and science mandate. A broader Phase I opportunity has been expected since the fall, and DOE has roughly $147 million open for Phase II.
NASA’s SBIR program runs defined subtopics rather than open ones, but NASA keeps a genuine open door elsewhere. NASA Innovative Advanced Concepts, or NIAC, solicits “visionary, yet credible, advanced concepts” with no topic list at all. Any US organization can submit at the first step, and a subset are invited to submit at the second. It is aerospace focused and deliberately early stage, which makes it a poor fit for a prototype ready to transition and a strong one for a concept that needs credibility before anyone will fund the build.
How to Win an Open Topic SBIR in 2027: Build a Multi-Submission Strategy
Winning SBIR funding works the way raising from investors works, and the way winning customers works. More shots on goal is better. You would not pitch one investor and wait. You would line up every relevant one you could find. Same with customers. Federal funding is no different.
Some classic specific topics will be a perfect fit for what you already build. Sometimes the opportunity is a focused open topic where the problem is named and the solution is yours to define. Sometimes it is a fully open competition like xTech|Search. Our approach has always been that a multi-submission strategy beats a single bet, and you can absolutely win more than one program.
There is a real constraint on that, though. To accept multiple awards, your R&D roadmap has to be genuinely distinct for each one. Two awards covering essentially equivalent work is not allowed, and if that is what you have, you will be choosing one and giving up the other. So the submissions need to be planned as a portfolio from the start, not assembled opportunistically and reconciled later.
None of this means spray and pray. In the era of AI, a lot of companies are submitting a lot of low quality proposals, and those proposals do not get it done. Our approach starts with the end user. Build the pitch with that person’s problem in mind, in a way that funds the roadmap you actually want to pursue rather than a distraction that pulls your team sideways for eighteen months. That is a fine art, and it demands a balanced, thoughtful, and smart approach. It is how we work with clients at Grant Engine, and we have found it to be highly effective.
This has become more nuanced now that the law has added a constraint. Per firm Phase I proposal caps are in effect this fiscal year. Every participating agency was required to set its limit at least 90 days before the fiscal year began, and agencies choose the structure themselves, per company, per solicitation, or per topic. Waivers are capped at 5 percent of an agency’s topics in any year. Published numbers are still thin on the ground, so check the cap at each agency you are targeting before you plan a year of submissions, rather than assuming last year’s volume still works.
The good news is that a disciplined process is also a higher win rate process.
That is where our approach comes in. Four things carry most of the weight.
A multi-submission, multi-agency strategy that still holds focus. Spread across agencies and topic types, but not so wide that the portfolio pulls your roadmap in four directions at once.
Do the research, including the conversations. Talk to the technical points of contact. Understand the need behind the topic and the problem your solution solves for the people who wrote it. That is the difference between a proposal that reads like it was written for them and one that reads like it was written for anyone.
The end user first. The topic is the artifact. The person with the problem is the target. Build every submission around them rather than around which topic sits closest to your capability statement.
Start early. The windows are short and the research cannot happen inside them. Being ready in advance is what makes a 14 day deadline workable instead of impossible.
The open topic did not disappear in 2026. It changed shape and it spread, which means there is more ground to cover than there used to be and less room to cover it carelessly.
For a dual use small business, non-dilutive funding at this scale can change what you are able to build and how fast. So be strategic about how you approach it. If you think you could benefit from how we work, schedule a call. It costs nothing to find out what the real scope of opportunity looks like for your technology, and if things line up, maybe we work together and win some of it.
Open Topic SBIR: Frequently Asked Questions
Can I submit my technology if it does not match any published topic?
Inside the Department of War SBIR program, increasingly no. The wide open door that let you propose anything has narrowed, and most current topics expect you to map to a named problem. Outside of it, several doors still take what you have. xTech|Search invites submissions addressing “any valid Army need or capability gap.” DARPA’s ERIS takes a seven minute video with no topic to match. Standing Broad Agency Announcements from Army Research Laboratory, the Army Applications Laboratory, and the Office of Naval Research accept white papers year round with no topic list at all. And NSF and NIH do not specify topics in the first place.
Do I need a government customer before I apply?
Less than you used to. Under the old AFWERX open topic model you had to bring your own, formally, through a signed Customer Memorandum. The focused open topic changes that. The sponsoring office is already behind the topic, which is the entire point of the model, so the customer arrives with the opportunity rather than being something you go find first.
That does not make the customer irrelevant. In a fully open competition like xTech|Search, or in a standing Broad Agency Announcement, you are still responsible for showing whose problem you solve and why they care. NSF and NIH are different again, requiring no customer at all, since they fund the science rather than buy the result. And where a topic does name its sponsor, talking to the technical point of contact before you write is usually what separates a winning proposal from a merely competent one.
One to watch: if AFWERX brings the fully open topic back in 2027, the Customer Memorandum would likely come back with it, since that model requires a named Air Force customer to move you into Phase II.
What is the difference between an open topic and a focused open topic?
Every DoW SBIR topic involves three pieces: a problem, a customer who owns it, and a solution. A standard topic gives you all three. An open topic gives you none, so you have to establish the problem, find the end user, and propose the technology. A focused open topic gives you two of the three. The problem is named, often down to performance thresholds and cost targets, and the customer is already behind it. Only the solution is left open. That middle option is usually the easier one to win, because the path from Phase II to a Phase III contract is visible before you start writing.
Can I win more than one SBIR award at the same time?
Yes, and companies do it regularly. The SBA Policy Directive acknowledges that firms often submit similar proposals to more than one soliciting agency, and submitting broadly is not prohibited. The restriction applies to funding rather than submission. At award you certify that you do not have essentially equivalent work funded by another federal agency, and essentially equivalent work cannot be funded twice. The practical implication is that your submissions need to be planned as a portfolio, with each one covering a distinct piece of your R&D roadmap. If two awards land on the same work, you will be choosing one and giving up the other.
How many SBIR proposals can I submit now that caps are in effect?
That depends on the agency, and you need to check rather than assume. The 2026 reauthorization requires every participating agency to set a cap on how many Phase I proposals a single firm may submit per fiscal year. Agencies had to establish those limits at least 90 days before the fiscal year began, and each chooses its own structure, per company, per solicitation, or per topic. Waivers are limited to no more than 5 percent of an agency’s topics in any year. Published numbers are still scarce, so check the solicitation or the agency portal for each program you are targeting before you plan a year of submissions.






