What an other transaction actually is
An other transaction is an agreement that is not a procurement contract, not a grant, and not a cooperative agreement. DoD's authority to use one for prototypes sits at 10 U.S.C. § 4022, recodified from the old § 2371b when the FY2021 NDAA reorganized Title 10. A separate section, § 4021, covers research other transactions. The distinction matters because the FAR applies to procurement contracts. An other transaction is outside that definition, so the FAR does not apply on its own. Everything the FAR would have supplied (clauses, data rights defaults, cost principles, disputes procedure) has to be written into the agreement or it is simply absent.
The statutory scope is prototype projects directly relevant to enhancing the mission effectiveness of military personnel and the platforms, systems, components, or materials the department uses or is acquiring. For a software firm that is a lower bar than it sounds. A prototype need not be a physical article: DoD's own other transactions guide treats software, algorithms, data pipelines, and business processes as prototypes when they are being tested for a military application.
Two things follow. The work has to be a prototype, so an OT is the wrong instrument for buying a mature product off the shelf. And because the agreement starts near a blank page, the negotiation determines what a firm walks away owning.
Where a prototype OT tends to fit
Editorial weighting from public statute, DoD guidance, and practitioner reading. Illustrative rather than measured.
The four gates: how a project qualifies
A prototype OT is only available when one of four conditions in § 4022(a)(2) is met. At least one nontraditional defense contractor or nonprofit research institution participates to a significant extent. Or every significant non-federal participant is a small business or a nontraditional defense contractor. Or at least one third of total project cost comes from non-federal sources. Or the agency's senior procurement executive determines in writing that exceptional circumstances justify a business arrangement that would not be feasible under a contract.
DoD's guidance reads "significant extent" as supplying a new key technology or product, performing a significant share of the effort, or causing a material reduction in cost or schedule or a material increase in performance. A token sub on a large team does not open the gate. A firm doing the actual model development on a $4M prototype does.
For a small software firm the first two conditions are the realistic ones. The one-third cost share is expensive and rarely why a small firm gets in, and the exceptional-circumstances determination is slow and uncommon. Knowing which gate an effort runs through tells you who the government expects on the team.
What "nontraditional" actually means
This is the most misread term in the whole area. The definition at 10 U.S.C. § 3014, recodified from the old § 2302(9), says a nontraditional defense contractor is an entity that is not currently performing, and has not performed for at least the one-year period preceding the solicitation, any DoD contract or subcontract subject to full coverage under the Cost Accounting Standards.
The test is CAS coverage. It is not company size, not revenue, and not whether the firm has ever held a DoD contract. Under 48 CFR 9903.201-2, full CAS coverage attaches to a business unit that receives a single CAS-covered award of $50 million or more, or that received $50 million or more in net CAS-covered awards in its preceding cost accounting period. Contracts at or below the $2 million threshold are exempt from CAS entirely under 48 CFR 9903.201-1.
Run the arithmetic and the picture changes. A firm holding several SBIR Phase II awards, a few task orders, and a subcontract on a prime's program is almost certainly still nontraditional. Losing the status takes real scale. That matters in teaming conversations: a prime running a prototype effort under § 4022 needs a nontraditional participant to open condition (a)(2)(A), and small software firms are the usual answer.
The consortium model
Most DoD prototype dollars under this authority move through consortia. The government signs a base other transaction agreement with a consortium management firm, and member companies compete for individual project agreements under that base. Advanced Technology International and NSTXL manage several of the larger ones.
The names worth knowing include the Space Enterprise Consortium, the Information Warfare Research Project, the System of Systems Consortium, the Consortium for Command, Control and Communications in Cyberspace, the National Armaments Consortium, the Medical CBRN Defense Consortium, the Training and Readiness Accelerator, the Vertical Lift Consortium, and the Undersea Technology Innovation Consortium. Membership is generally open, with annual dues ranging from nothing to a few thousand dollars.
Consortium award sequence, typical durations
Direct award is the other path. The Defense Innovation Unit publishes Commercial Solutions Openings and awards prototype OTs itself, and CSO authority for DoD is codified at 10 U.S.C. § 3458. DIU has publicly described a target of roughly 90 days from posting an area of interest to award, and service innovation offices award directly as well. Through a consortium, the manager takes a fee, commonly a low single-digit percentage of award value, so read the base agreement before you build a price.
IP and data rights: the largest real difference
DFARS 252.227-7013 for noncommercial technical data and 252.227-7014 for noncommercial computer software are contract clauses. They attach when a contracting officer puts them in a contract. In an other transaction they are absent unless the agreement adopts them, which means intellectual property terms are negotiated fresh every time.
That cuts both ways. A firm that reads the agreement keeps its background IP and grants a narrow license to the delivered prototype. A firm that does not can sign away unlimited rights in code written years before the project started, because the opening template often says exactly that.
Attach a background IP schedule. List every component, model, library, and dataset that predates the project. Anything unlisted risks being treated as developed under the agreement.
Define the license grant per deliverable. Government purpose rights in the integration layer with restricted rights in the core model is a normal, negotiable outcome. A single project-wide grant is not.
Name the third-party and open-source components and their licenses. A copyleft dependency found inside a delivered container after acceptance is a bad week for everyone.
Address data as its own asset. Training data, fine-tuned weights, evaluation sets, and prompt libraries have separate value, and a clause written for "technical data" often fails to say which of them it covers.
Cost accounting, pricing, and audit
FAR Part 31 cost principles do not apply by default, and neither do the Cost Accounting Standards. Certified cost or pricing data is not required, because that requirement attaches to contracts. DCAA gets no automatic audit rights; access to records is a negotiated term.
What replaces all of it is milestone payment. A well-written agreement defines payable milestones tied to demonstrable technical events, and the government pays on a demonstrated result rather than on incurred cost. For a small software firm that is often the largest operational difference from a cost-reimbursement FAR award, which requires an accounting system judged adequate for accumulating costs by contract line.
The advantage is real but temporary. A prototype OT can run on ordinary commercial books. A follow-on production contract, or an SBIR Phase II, puts a compliant accounting system back in scope. Firms that skip that build discover the gap at the worst moment.
An OT is an exception to the FAR, not to fiscal law
Appropriations still have a purpose, a time, and an amount. The Anti-Deficiency Act still applies. The False Claims Act reaches claims for federal money regardless of instrument. Export control, classification, and CUI handling follow the data and the work, and program offices routinely write NIST SP 800-171 style requirements straight into the agreement because the FAR clause that would have done it is absent.
Protest rights: what you give up
GAO's bid protest jurisdiction comes from the Competition in Contracting Act and reaches solicitations and awards of procurement contracts. An other transaction is not one, so GAO will generally decline to review the award itself. It will take up the narrower question of whether an agency improperly used other transaction authority in place of a procurement contract; MorphoTrust USA, LLC, B-412711 (May 16, 2016) is the usual citation.
The Court of Federal Claims is the other door. In Oracle America, Inc. v. United States, 975 F.3d 1279 (Fed. Cir. 2020), the Federal Circuit addressed a challenge to a follow-on production award made under the prototype authority and confirmed that the court can hear protests brought in connection with a procurement. The surface exists. It is narrow, slow, and expensive.
Plan for both sides. Win an OT and the award is far more durable than a Part 15 award: no 100-day GAO clock, no automatic CICA stay under 31 U.S.C. § 3553(d). Lose one and you likely have no cheap remedy. Debriefs are discretionary and often thin, and feedback after a consortium competition is a courtesy rather than a right.
Speed: what changes and what does not
The genuine savings are specific. No Part 15 source selection plan, no full trade-off documentation to write and defend, no CAS or certified-cost-data negotiation cycle. The first round is a short white paper instead of a full proposal, which compresses the offeror's effort and the evaluation together.
The parts that do not compress usually control the schedule. The money still has to exist and be the right color of appropriation. The requirement still needs a sponsor who wants it. Security paperwork and system access take as long as they take. The agreements officer has a queue. A consortium prototype commonly runs 60 to 120 days from release of the request to award when funding is in hand, and considerably longer when it is not.
An honest comparison has to say where the FAR wins. A small, well-defined software buy under simplified acquisition procedures, with the simplified acquisition threshold at $250,000 under FAR 2.101, can beat any OT on speed. The OT advantage is real in the range where the alternative would have been a full Part 15 competition.
The follow-on production pathway
This is the provision that makes prototype OTs strategically interesting. Under § 4022(f), a follow-on production contract or transaction may be awarded without further competition when competitive procedures were used to select the prototype participants and those participants successfully completed the project. The original solicitation and the agreement have to state that a follow-on may result.
There is no statutory ceiling on the follow-on. A competitively selected $3M prototype can become a production award many times its size without another competition. That is why competitors watch these awards closely.
Two sentences in the agreement are worth more than most of the rest of the document. Does it say a follow-on production award is contemplated? And does it define "successfully completed" in measurable terms, tied to acceptance criteria such as a detection rate at a stated false-alarm rate, a latency budget under a stated load, or a completed integration against a named system? Satisfaction language leaves that determination to whoever is in the chair at the end.
How this compares with SBIR Phase III
SBIR has its own sole-source pathway. 15 U.S.C. § 638(r)(4) directs agencies, to the greatest extent practicable, to issue Phase III awards including sole-source awards to the firm that developed the technology under Phase I or II, and FAR 6.302-5(b)(4) supplies the authority for other than full and open competition. Phase III has no dollar ceiling, no time limit, need not be funded with SBIR money, and can be awarded by an agency other than the one that funded the earlier phases.
The IP position is stronger by default. SBIR data rights are asserted under DFARS 252.227-7018 with a 20-year protection period under the current SBIR/STTR Policy Directive, and they travel with the technology into Phase III rather than being renegotiated.
Against that, Phase III is a FAR contract with everything that implies at scale: CAS exposure past the thresholds, cost or pricing data where applicable, DCAA involvement, and the full DFARS data rights apparatus. A prototype OT trades that machinery for negotiated terms and a production pathway you have to win a competition to reach. For a software firm the two fit together. Prove the capability under SBIR with clean data rights, then pursue either a Phase III with a program office that has money or a prototype OT where the agency wants speed and a wider integration scope.
Side by side on the axes that decide
| Axis | FAR Part 12 / Part 15 contract | Prototype OT (10 U.S.C. 4022) | SBIR Phase III |
|---|---|---|---|
| Governing rules | FAR and DFARS apply in full | FAR does not apply; every term is written into the agreement | FAR and DFARS apply, plus SBIR-specific clauses |
| IP and data rights | DFARS 252.227-7013 and -7014 defaults; unlimited rights where the government funded development | Negotiated from close to a blank page; the background IP schedule is the control | DFARS 252.227-7018 SBIR data rights, 20-year protection period, carried forward |
| Cost accounting | FAR Part 31 cost principles; CAS above the thresholds; DCAA likely on cost-type work | No CAS and no certified cost data by default; payable milestones; records access is negotiated | Same obligations as any FAR contract; often awarded fixed-price |
| Protest exposure | GAO under CICA plus the Court of Federal Claims; automatic stay available | GAO generally declines the award itself; COFC available on narrow grounds | GAO and COFC; a sole-source award can still be challenged |
| Time to award | Commonly 6 to 12 months for a competitive Part 15 buy | Commonly 60 to 120 days through a consortium when funding is ready | Limited mainly by the program office and the contracting shop; no competition step |
| Path to production | Recompete, or exercise priced options | Non-competitive follow-on under § 4022(f) if the criteria and the agreement language line up | Phase III is itself the production vehicle; no ceiling |
Common objections, answered plainly
Doesn't an OT let the government skip the rules?
It skips the FAR. Fiscal law, the Anti-Deficiency Act, the False Claims Act, export control, classification, and security requirements all still apply. In practice most prototype agreements end up containing terms that read like FAR clauses, because the program office wants those protections and has to write them in by hand.
Does joining a consortium get you awards?
It gets you on the distribution list and makes you eligible to bid. Selection is still competitive, often against a member base in the hundreds or thousands, and a request that draws sixty white papers is ordinary. The firms that win usually have a program relationship before the request appears.
Does an OT mean we keep all our intellectual property?
Only if the agreement says so, and government templates frequently open with broad rights language. The firms that come out well arrive with a background IP schedule already drafted.
Bottom line
The choice comes down to three questions: how much the firm's existing intellectual property matters, how much protest protection it will give up, and whether a program sponsor with money wants the prototype to become production. A prototype OT is strong where background IP is valuable and the sponsor is real. A FAR award is stronger where the requirement is well defined, the buy is small, or the firm wants the certainty of standard clauses. SBIR Phase III is strongest where the technology already exists because the firm built it under an earlier phase.
None of it survives a weak technical case. The instrument controls the terms and the timeline. It does not make anyone want the prototype.
Frequently asked questions
It is a binding agreement, but it is not a procurement contract as the FAR defines one. That is the source of most of its differences: the FAR, the Cost Accounting Standards, and the standard DFARS data rights clauses do not apply unless the agreement adopts them.
Under 10 U.S.C. § 3014, an entity that is not currently performing and has not performed for at least one year before the solicitation any DoD contract or subcontract subject to full Cost Accounting Standards coverage. The test is CAS coverage, so most small and mid-size firms qualify regardless of how much DoD work they have done.
Rarely at GAO, which generally declines jurisdiction over the award itself while agreeing to review whether the agency could use the authority at all. The Court of Federal Claims has taken challenges brought in connection with a procurement, including follow-on production awards, though the route is narrow and costly.
No. A non-competitive follow-on under § 4022(f) requires that competitive procedures were used to select the prototype participants, that the participants successfully completed the prototype, and that the original solicitation and agreement contemplated a follow-on. Check that the agreement defines "successfully completed" against measurable acceptance criteria.
They answer different problems. SBIR funds a firm to build and prove a capability, with statutory data rights and a sole-source route to Phase III. A prototype OT gives speed and negotiated terms but assumes a program sponsor already wants the thing.