The Department of Defense is the largest SBIR funder, and treating it as a single program is the most common way founders waste months. DoD SBIR is run by components — the Army, Navy, Air Force and Space Force, plus defense agencies such as DARPA, MDA, DLA and DHA — and each writes its own topics, sets its own evaluation practices and runs on its own cycle. There is a shared submission system and a common release calendar, but the customer you are actually selling to is a component.

Topics, not open ideas

With limited exceptions, DoD funds against topics that a government technical point of contact wrote because a program needs something. Your job is to respond, not to propose a new direction. Read the topic literally: the objectives, the phase descriptions and the references tell you who wrote it and what they already tried. If you find yourself explaining why the topic should have been written differently, you are applying to the wrong topic.

DoD awards are contracts, not grants. That means a statement of work, a contracting officer, deliverable dates and acquisition cost rules — a different world from NSF or NIH, and covered in more detail alongside the agency comparison. Registration in the federal contracting systems takes longer than founders expect and should be started well before you plan to submit.

The pre-release window is where proposals are won

DoD topics appear in a pre-release period before submissions open, and during that window you may contact the topic author directly with questions. Once submissions open, that channel closes and questions move to a moderated public forum. This is the most valuable and most underused feature of the whole program. Use it to learn what problem sits behind the topic, what has already failed, and what a useful Phase I result would look like to the person who wrote it. The dates and the exact rules for contact change with each cycle — confirm them in the current announcement, and read how to handle that conversation before you place the call.

Transition is the real prize

Phase I and Phase II money is worth having, but the outcome that changes a company is transition: your technology getting adopted into a program of record, or bought at scale through a Phase III award, which can be sole-sourced without further competition. That requires a program office that wants your capability and has budget lines to put it in. Companies that win repeat Phase II awards without ever transitioning are running a research shop, not building a business — a distinction worth weighing against the tradeoffs in grants versus investors.

Practical implication: start building the customer relationship during Phase I, not at the end of Phase II. Ask the technical point of contact who else in the component has this problem, and get in front of them.

Who DoD suits

DoD fits companies with hardware, software or materials that map onto a stated military need, that can tolerate procurement administration, and that can handle security and export-control obligations where they apply. If your only market is defense, plan for long sales cycles and a lot of unpaid relationship-building. If you have dual-use potential, say so — it strengthens both the proposal and the company. See also selling a new product to the government.

Projects House builds and tests the hardware behind defense-adjacent proposals. Start through the contact form.