Broad Agency Announcement

A Broad Agency Announcement (BAA) is the instrument a federal agency uses to buy basic and applied research. FAR 35.016 governs it. A peer or scientific review panel evaluates each proposal against the criteria the announcement states. Proposals do not compete head to head, because a BAA carries no common statement of work.

FAR 35.016(a) sets the boundary of that scope. A BAA covers basic research, applied research, and the part of development that does not relate to a specific system or a hardware procurement. An agency that wants a known deliverable built to a known specification must use a different instrument.

What the FAR requires a BAA to contain

FAR 35.016(b) states four requirements. The announcement, with any supporting documents, must do all four.

• Describe the research interest. The interest covers one program requirement, or a broadly defined area that spans the full range of what the agency wants.
• Describe the selection criteria, their relative importance, and the method of evaluation.
• Specify the period of time during which the agency will accept proposals.
• Contain instructions for the preparation and the submission of proposals.

FAR 35.016(c) requires the agency to publicize the availability of the BAA through the Governmentwide point of entry. Subpart 5.5 can also authorize publication in scientific, technical, or engineering periodicals.

One consequence catches people out. Under FAR 35.016(f), the agency does not have to synopsize the individual contract actions that come out of a BAA. The original notice satisfies that requirement. Awards under a BAA therefore do not generate the individual announcements you would watch for under a conventional procurement.

Why a BAA is still a competition

A BAA does not look like a competition, and contractors regularly assume it is a sole-source route or a grant program. It is neither.

FAR 6.102 lists the competitive procedures. FAR 6.102(d)(2) makes competitive selection of basic and applied research a competitive procedure when the award results from both of the following:

• A BAA that is general in nature. It identifies areas of research interest. It includes criteria for the selection of proposals. It solicits the participation of all offerors that can satisfy the government's needs.
• A peer or scientific review.

The two conditions are cumulative. The general announcement alone is not sufficient, and the review alone is not sufficient. FAR 35.016(a) points to 6.102(d)(2) directly, which is why the peer review is a structural requirement rather than an agency preference.

How the evaluation works without a Section M

This is the part that surprises people, and it is the reason the term deserves a page.

A negotiated procurement under FAR Part 15 gives every offeror the same statement of work. Section M then states the evaluation factors, and the source selection authority compares proposals against each other on those factors. That comparison is the whole mechanism.

A BAA removes the shared work statement. Each offeror proposes its own research. FAR 35.016(d) draws the consequence in one sentence:

"Written evaluation reports on individual proposals will be necessary but proposals need not be evaluated against each other since they are not submitted in accordance with a common work statement."

A BAA proposal is therefore assessed on its own merits, not ranked against the field. The practical effects follow from that:

• There is no single winner by default. An agency can fund several proposals under one announcement. It can also fund none.
• A written evaluation report covers each proposal individually. The FAR requires that report.
• Reviewers are scientific or technical. The panel judges research merit, not proposal craftsmanship.
• The criteria carry a stated relative importance, per FAR 35.016(b)(2). Announcements commonly express that ordering in words rather than in percentages. "Listed in descending order of importance" is the standard phrasing, and it is a real instruction about weight.
• Funding availability is part of the decision. FAR 35.016(e) makes the primary basis for acceptance technical merit, importance to agency programs, and fund availability. Cost realism and reasonableness also count, to the extent appropriate. A technically excellent proposal in a depleted research area does not get funded.

The contrast with Section L matters too. Section L exists because a Part 15 procurement needs every proposal in the same shape for comparison. A BAA has less reason to enforce that shape, so its instructions are usually shorter. Shorter is not the same as optional. Read them with the same care.

BAA against RFP

The two-step white paper

Many BAAs run in two stages. Offerors submit a short white paper first. The agency reviews it and tells the offeror whether it encourages a full proposal. Only then does the offeror write the full proposal with a cost volume.

This two-step process is agency practice, not a FAR requirement. FAR 35.016 does not use the words "white paper" and does not describe a two-step submission. Agencies build the white paper gate into their own announcements, so the procedure varies from one announcement to the next. Read the announcement you actually received, and do not assume the shape of the last one.

The stage is worth the effort even so. A white paper costs a fraction of a full proposal, and a discouraging response is genuine information. Teams that treat "not encouraged" as a rejection miss the point. It is a cheap, early, honest signal about fit, delivered before the expensive work starts.

Open BAAs, calls, and the closing date

FAR 35.016(b)(3) requires the announcement to specify the period during which the agency will accept proposals. It sets no minimum length. Two shapes are common.

• A single closing date. The announcement behaves like a conventional solicitation with one deadline.
• An open or rolling BAA. The announcement stays open for a long period, frequently a year or more, and accepts proposals throughout. Agencies often issue periodic calls, topic calls, or amendments against that open announcement. Each call carries its own research topic, its own deadline, and sometimes its own funding ceiling.

The second shape is the one that gets mishandled. An open BAA has no urgency of its own, so it sits in a pipeline for months. Its calls do have urgency, and the call is the thing you actually respond to. An amendment to an open BAA is frequently the most useful signal it emits, because it shows where the agency puts money now.

Where the Commercial Solutions Opening and the Advanced Research Announcement sit

Three announcement families sit near each other, and contractors confuse them constantly.

Broad Agency Announcement. Research. FAR 35.016. Peer or scientific review. Covered above.

Commercial Solutions Opening (CSO). A general solicitation route for innovative commercial products and commercial services. The statutory authority is 10 U.S.C. 3458.

The CSO is a Department of War instrument. It is not in the FAR. The implementing rule is DFARS subpart 212.70, "Defense Commercial Solutions Opening." DFARS 212.7000 states that the subpart implements 10 U.S.C. 3458 for the acquisition of innovative commercial products or commercial services through a general solicitation known as a commercial solutions opening. A civilian agency has no equivalent route under this authority, and that limitation is the fact most contractors get wrong.

Two further points are worth carrying.

• A CSO is a competitive procedure, but the designation lives somewhere other than the CSO subpart. DFARS 206.102-70 states that competitive selection of proposals through peer review under a commercial solutions opening is a competitive procedure.
• The CSO borrows the BAA evaluation model. DFARS 212.7004(c) mirrors FAR 35.016(d) almost word for word. Proposals need not be evaluated against each other. Offerors do not submit them in response to a common performance work statement or statement of work.

The phrase "commercial solutions opening" does not appear in the text of 10 U.S.C. 3458 itself. The statute speaks of general solicitation competitive procedures, and the DFARS supplies the name.

Advanced Research Announcement (ARA). An announcement vehicle that some agencies, notably Air Force Research Laboratory components, issue for research topics. It functions much like a BAA in practice, often with the same white paper first step.

We cite no statute for the ARA, and that omission is deliberate. The FAR does not define "Advanced Research Announcement," and it is not a term of art in Title 10. Live ARAs state that a resulting award takes the form of a FAR-based procurement contract or of an other transaction. The announcement lists several distinct authorities and the applicable one depends on what the agency finally awards. Any single statutory citation for "an ARA" would therefore be wrong more often than right. Treat the ARA as an agency announcement format, and read the authority that the specific announcement names.

The mistake that makes this term matter

A BAA arrives looking like a solicitation, so it gets read like one. Somebody opens it looking for Section L and Section M, does not find them, and concludes the document is incomplete or informal. The team then does what it always does: it writes a proposal, in the house template, against the best guess at a requirement.

That guess is the failure. There is no requirement to guess at. The agency described areas of research interest and asked what you would propose. A proposal that answers a requirement the agency never wrote is answering nobody. It reads as generic to a reviewer who is a working scientist and who is reading it for research merit, not for compliance.

The second half of the mistake is the evaluation model. Teams assume that a stronger proposal beats a weaker one, because that is how every other competition they have entered worked. Under a BAA it does not. Your proposal is not measured against the field. It is measured against the criteria and against the money available in that research area. A team that spends its effort trying to out-write competitors it cannot see has misread the document at a structural level.

What goes wrong in practice

Teams treat a BAA like an RFP and lose the whole premise. The document does not tell you what to build. It tells you what the agency is interested in. The proposal has to supply the idea. Reusing a Part 15 proposal skeleton produces a document that answers questions nobody asked.

Teams skip the white paper stage or treat it as a formality. Where an announcement offers that stage, it is the cheapest information in the process. A short paper that draws a discouraging response has saved the cost of a full proposal, which is a good outcome, not a bad one.

Teams ignore the stated order of importance. A BAA rarely publishes percentages. It states the criteria in an order and says what that order means. That order is an instruction about where the effort belongs. A team that treats the criteria as equal puts its pages in the wrong place.

Teams underweight the technical point of contact. BAAs frequently name a program officer and invite technical discussion before submission. A Part 15 procurement usually forbids that contact. The habit carries over, and the conversation that the announcement invites never happens.

Teams miss the calls under an open BAA. The parent announcement sits open for a year and looks like it has no deadline. The calls issued against it have real deadlines and real money. Monitoring A team that monitors the parent and not the calls monitors the wrong object.

Teams misread "not selected" as "not competitive." With no common work statement, a decision not to fund can reflect the research area budget rather than the quality of the work. Ask. Agencies that run research programs will often tell you.

Teams propose development work under a research announcement. FAR 35.016(a) excludes development of a specific system or hardware procurement. A proposal that is really an engineering build in research language is a scope mismatch a scientific reviewer will see immediately.

What to do

1. Confirm the instrument first. Find the authority the announcement cites before you plan anything.
2. Read the areas of research interest as the actual subject matter. There is no other requirement.
3. Find the evaluation criteria and the statement of their relative importance.
4. Allocate the proposal effort to match that stated order.
5. Check whether the announcement uses a white paper stage. Submit at that stage where it exists.
6. Check the announcement for an open period or for a single closing date.
7. Track every call and every amendment issued under an open announcement.
8. Contact the technical point of contact where the announcement permits it.
9. Write for a scientific reviewer. State the hypothesis, the approach, and the risk plainly.
10. Address cost realism. Research budgets are small, and an unrealistic cost is a real discriminator.

Do not reuse a Part 15 proposal template without rebuilding the outline from the announcement.

What a Broad Agency Announcement is not

A BAA is not a grant. It is an acquisition instrument, and an award under it is usually a contract, although agencies also use other instruments.

A BAA is not a sole-source route. FAR 6.102(d)(2) treats it as a competitive procedure, and the peer or scientific review is the reason.

A BAA is not an RFP with the sections missing. The absence of a common statement of work is the design of the instrument, not an omission from the document.

A BAA is also not a promise of funding. FAR 35.016 lets an agency accept proposals over a long period and fund whatever the research portfolio and the budget support. The result is several awards, one award, or none at all.

Silas™ classifies a solicitation into its announcement family from the notice text and its attachments, and reads which regulatory regime the document cites. It detects whether the announcement asks for a white paper or for a full proposal. It extracts the stated evaluation criteria and their order of importance, and links the calls issued under an open announcement back to their parent.