The short answer
If you sell to government, the Federal Acquisition Regulation does not forbid you from talking to your buyers. It actively encourages it. FAR 15.201(a) states that "exchanges of information among all interested parties, from the earliest identification of a requirement through receipt of proposals, are encouraged," and 15.201(c) lists nine techniques for doing it, including one-on-one meetings with potential offerors.
What the rules do is something stranger, and much less discussed. They change what a finding is worth after you have it. Two provisions govern the back half of the process:
- You may be required to give your finding to your competitors. Under FAR 15.201(f), when specific information about a proposed acquisition that would be necessary for preparing proposals is disclosed to one or more potential offerors, "that information must be made available to the public as soon as practicable... in order to avoid creating an unfair competitive advantage."
- Obtaining the wrong thing is itself the offense. FAR 3.104-3(b), implementing 41 U.S.C. 2102, provides that "a person must not, other than as provided by law, knowingly obtain contractor bid or proposal information or source selection information before the award of a Federal agency procurement contract to which the information relates."
Put together: in the commercial world, research produces two prized things, a proprietary edge and the unexpected disclosure. In federal procurement, the first is nullified by regulation and the second is a hazard rather than a gift. Your research program has to be designed around that, and the design is mostly about timing.
The window is genuinely open, and it is wider than most vendors use
The single most common mistake in government-facing research is excessive caution. Teams assume contact is restricted and stay away, then write a proposal against a requirement they never tested. The FAR says the opposite.
FAR 15.201(b) explains the purpose plainly: exchanging information improves "the understanding of Government requirements and industry capabilities, thereby allowing potential offerors to judge whether or how they can satisfy the Government's requirements." Paragraph (c) says agencies "are encouraged to promote early exchanges of information about future acquisitions," and then enumerates the techniques:
| Technique | FAR 15.201(c) item |
|---|---|
| Industry or small business conferences | (1) |
| Public hearings | (2) |
| Market research, as described in part 10 | (3) |
| One-on-one meetings with potential offerors | (4) |
| Presolicitation notices | (5) |
| Draft RFPs | (6) |
| Requests for information (RFIs) | (7) |
| Presolicitation or preproposal conferences | (8) |
| Site visits | (9) |
Paragraph (c) also names what those exchanges are for, and the list reads like a research plan: resolving concerns about "the feasibility of the requirement, including performance requirements, statements of work, and data requirements," and "the suitability of the proposal instructions and evaluation criteria."
Three further points widen the window. FAR 15.201(e) confirms RFI responses "are not offers" and that "there is no required format for RFIs." FAR 15.201(f) opens by stating that "general information about agency mission needs and future requirements may be disclosed at any time," with no cutoff. And FAR 3.104-4(e)(3) confirms the procurement integrity rules do not restrict "individual meetings between a Federal agency official and an offeror or potential offeror," subject to the condition discussed below.
The practical implication is that the pre-solicitation period is not merely permitted research time. It is the only period in which the research is both unrestricted and still able to change the requirement.
What closes, and exactly when
The hinge is the release of the solicitation. FAR 15.201(f) is explicit: "after release of the solicitation, the contracting officer must be the focal point of any exchange with potential offerors."
That is not a prohibition on contact. It is a re-routing of it. Every question now goes through one person, in writing, on the record, alongside every competitor's questions. The consequences for research are concrete:
- You lose the interview format. A contracting officer answering written questions is not a participant in a conversation. There is no probing, no follow-up on an interesting answer, no rapport.
- You lose the population. The end users whose workflow you actually need to understand are no longer available to you as respondents on this acquisition.
- You lose timing control. Answers arrive on the agency's schedule, in an amendment, usually shortly before the proposal is due.
So the research calendar and the acquisition calendar are the same calendar, and the useful half of it ends at a date the agency publishes in advance. Everything you want to know about how the work is actually done has to be learned before that date.
The rule that makes your finding public
Now the part that has no commercial analogue. The rest of FAR 15.201(f) reads:
When specific information about a proposed acquisition that would be necessary for the preparation of proposals is disclosed to one or more potential offerors, that information must be made available to the public as soon as practicable, but no later than the next general release of information, in order to avoid creating an unfair competitive advantage.
Consider what this does to the economics of a research finding. You run a well-designed study, you surface something specific and material about the proposed acquisition that nobody else understood, and it is genuinely necessary for preparing a proposal. That is the definition of a successful research outcome, and it is precisely the circumstance in which the information must be published.
The finding cannot be kept. Not because it leaked, and not because you were careless, but because the regulation is built to prevent exactly the asymmetry that commercial research exists to create. Fairness to other offerors is the governing value, and your competitive advantage is the thing being regulated away.
The same logic runs through the Paperwork Reduction Act on the government's side of the table. When an agency publishes a Federal Register notice for a customer feedback collection, its instrument and burden estimate go out for public comment before fielding, and, in the words of the National Science Foundation's 20 August 2026 renewal notice, comments "will become a matter of public record." We cover that regime in our guide to government customer research and the ten-person rule.
This does not make research worthless here. It relocates where the value sits. Since the content of an acquisition-specific finding may not stay yours, the durable advantage is in everything upstream of it: understanding the mission, the operational reality, and the user's day, none of which is acquisition-specific and none of which triggers the publication rule. FAR 15.201(f) itself draws that line by allowing general mission-needs information to be disclosed at any time.
Obtaining is the offense
The second provision reshapes how an interview must be conducted. FAR 3.104-3 has two parallel prohibitions implementing 41 U.S.C. 2102. Paragraph (a) prohibits knowingly disclosing contractor bid or proposal information or source selection information before award. Paragraph (b) prohibits knowingly obtaining it.
That second one lands on the researcher. In every other research setting, an unplanned disclosure is the best thing that can happen in an interview. The participant volunteers something you did not know to ask about, you follow the thread, and that is where the insight lives. Following the thread is the core interviewing skill.
Under the procurement integrity rules, the same moment is a hazard. You cannot control what a government official volunteers, and the prohibition attaches to obtaining. FAR 3.104-4(e)(3) permits individual meetings only "provided that unauthorized disclosure or receipt of contractor bid or proposal information or source selection information does not occur." Receipt, not solicitation.
The practical protocol that follows is unglamorous and worth writing into your discussion guide:
- State the boundary at the top of every session. Say plainly that you are not seeking any bid, proposal, or source selection information, and ask the participant to steer away from it.
- Interrupt rather than probe. If a participant heads toward evaluation criteria, competitor pricing, or where a proposal ranks, stop the thread. This is the one research context where curiosity is the wrong instinct.
- Keep a contemporaneous record. A timestamped transcript showing what was asked and how a stray disclosure was handled is the best evidence that receipt was neither sought nor knowing.
- Know which conversations are safely general. FAR 3.104-1 provides that an official is generally not participating personally and substantially in a procurement merely through "the performance of general, technical, engineering, or scientific effort having broad application not directly associated with a particular procurement."
Designing the program around the calendar
Three rules cover most of it.
Front-load everything. Treat solicitation release as a hard research deadline. Any question requiring a conversation with an end user must be answered before it. After release, plan only for written exchanges through the contracting officer.
Classify every question by which side of the line it lives on. Before fielding, sort your discussion guide into mission and workflow questions, which are durable, publication-safe, and reusable across every future bid, versus acquisition-specific questions, which are time-boxed and may have to be published. Invest most of your effort in the first category, because it compounds and the second does not.
Run continuous research rather than bid-triggered research. If your only contact with government users happens when an acquisition is live, you are doing all of your research in the narrowest and most legally constrained window available. Understanding of the mission built over years, outside any particular procurement, is both the safest and the only kind that stays yours.
Where Koji fits
Research under these rules has an unusual set of requirements: it must move fast, because the window closes on a published date; it must be documented, because you may need to show what was asked; and it must be tightly scoped, because the interviewer has to stay away from defined categories of information.
Koji is built for exactly that. A study is defined by structured questions in six explicit types (open_ended, scale, single_choice, multiple_choice, ranking, and yes_no), so the boundary of the conversation is written down before anyone is interviewed, reviewable by counsel, and identical for every participant. That is a materially better compliance position than a human moderator improvising in a live call, where an off-guide probe is one follow-up question away. See structured questions in AI interviews and how to customize interview questions.
Every session produces a complete transcript automatically, which is the contemporaneous record the protocol above depends on. See how to record customer interviews and interview recording consent laws.
And because AI-moderated interviews run asynchronously and in parallel, a study that a moderator would need six weeks to schedule can field in days, which matters more here than anywhere else: the deadline is not internal, it is published in the solicitation. Thematic analysis and one-click reports mean the finding reaches your capture team while the window is still open. Compared with legacy platforms built for panel surveys or moderated labs, this is the difference between research that informs a bid and research that arrives after the requirement is frozen.
Further reading: expert interviews, B2B buyer journey and buying committee research, and enterprise security for AI research platforms.
Frequently asked questions
Can vendors talk to government buyers before a solicitation?
Yes, and the FAR encourages it. FAR 15.201(a) states that exchanges "from the earliest identification of a requirement through receipt of proposals, are encouraged," and 15.201(c) lists nine techniques including one-on-one meetings, RFIs, draft RFPs, and site visits. All exchanges must remain consistent with the procurement integrity requirements at FAR 3.104.
What changes after the solicitation is released?
The contracting officer becomes the single point of contact. FAR 15.201(f) provides that "after release of the solicitation, the contracting officer must be the focal point of any exchange with potential offerors." Contact is re-routed rather than banned, but interviews with end users effectively stop for that acquisition.
Do I have to share what I learn with competitors?
Sometimes, yes. Under FAR 15.201(f), specific information about a proposed acquisition that would be necessary for preparing proposals, once disclosed to one or more potential offerors, "must be made available to the public as soon as practicable" so as not to create an unfair competitive advantage. General information about mission needs and future requirements may be disclosed at any time and does not carry that consequence.
What is the risk if a government participant volunteers something sensitive?
FAR 3.104-3(b), implementing 41 U.S.C. 2102, prohibits knowingly obtaining contractor bid or proposal information or source selection information before award, and 3.104-4(e)(3) permits individual meetings only where unauthorized receipt does not occur. Because the prohibition attaches to obtaining, the safe practice is to state the boundary up front, stop the thread rather than probe it, and keep a transcript.
Is market research allowed during an active procurement?
Market research is expressly listed at FAR 15.201(c)(3) as a technique for early exchanges, and FAR 3.104-1 indicates that general technical or scientific effort "having broad application not directly associated with a particular procurement" is generally not personal and substantial participation in that procurement. Acquisition-specific inquiries after solicitation release should route through the contracting officer.
How should a capture team schedule customer research?
Backwards from solicitation release. Treat that date as the deadline for all conversational research, run mission and workflow studies continuously and outside any live acquisition so the understanding is durable and publication-safe, and reserve the post-release period for written questions to the contracting officer.
Do your government research before the window closes
The rules that govern selling to government are not a reason to avoid customer research. They are a reason to do it earlier, faster, and with a written scope. The vendors that lose here are not the ones who ask too much; they are the ones who ask too late, after the requirement is frozen and the only remaining channel is a written question in an amendment.
Koji lets a capture or product team field a scoped, fully transcribed study in days instead of weeks: structured questions that define the boundary before anyone is interviewed, AI-moderated voice interviews that run asynchronously and in parallel, automatic thematic analysis, and one-click reports. No moderator bias, no scheduling, no research expertise required. From question to insight in hours, not weeks.
Start a free Koji study and get your findings while the window is still open.
