Governance & Security|November 9, 2026|12 min read

The Debriefing Runbook: Turn a Loss Into Evidence in Three Days

FAR 15.506 gives you three days to request a post-award debriefing in writing. Here is the named-owner runbook: trigger log, question set built on evaluation factors, and protest-clock math.

Projectory team

It is 4:40pm on a Friday when the unsuccessful-offeror notice hits the capture inbox. The proposal manager is on PTO. The BD lead forwards it to three people with "thoughts?" in the subject line. By Tuesday, when someone finally drafts a debriefing request, the window has closed.

Here is the direct answer to the question most teams search for after a loss: on a FAR Part 15 negotiated procurement, you must submit a written request for a post-award debriefing within three days after receiving the unsuccessful-offeror notice, and the contracting officer may provide that debriefing orally, in writing, or by another method acceptable to the contracting officer [2]. The notice itself is governed by FAR 15.503, which is why the moment of receipt is the event you log, not the award announcement you saw on SAM.gov [1].

The teams that walk out of a debriefing with usable evidence are not better at arguing. They decided what to ask, who owns the clock, and where the findings go before the notice ever arrived. This is that runbook.

Three Days Is Not a Planning Window, It Is a Filing Window

Most capture teams spend the three days deciding whether a debriefing is worth requesting. That is backwards. The request is cheap and short. The decision about what to do with the debriefing comes later.

Your request does not need to argue the merits of your proposal. It needs to identify the solicitation and the requesting offeror, cite FAR 15.506 as the authority, and ask for the method and schedule of the debriefing [2]. Because the regulation permits oral, written, or another method acceptable to the contracting officer, asking which one applies is more productive than assuming you will get a meeting [2]. If you want a written debriefing because your technical lead is deployed, say so in the request and explain why.

The content requirements for the notice sit in FAR 15.503 [1]. Read the notice against those requirements before you respond. If an element you expected is missing, note it in your chronology and address it in writing rather than raising it in a phone call where no record survives.

One more structural point that costs companies real money: name an owner and a backup for the trigger. A notice that lands while the owner is on vacation is the single most common way a three-day right gets forfeited. The backup should have standing authority to send the request without waiting for a go-ahead from the pursuit lead.

The Trigger Log: What Gets Recorded the Hour the Notice Lands

Everything downstream of the notice depends on a chronology you can hand to counsel without reconstructing it from email threads. Build it as one document per loss, created the hour the notice arrives.

Record these fields at minimum:

  • Solicitation number and vehicle, including the authority the competition actually ran under, because that determines whether this runbook applies at all
  • Award notice date and the notice receipt timestamp, with the method of receipt and the recipient's name
  • FAR 15.503 content check, element by element, with any gaps noted in writing [1]
  • Request sent timestamp, the exact text of the request, and the delivery method
  • Debriefing method and date as designated by the contracting officer
  • Follow-up exchanges, each with date, channel, and the question text as submitted
  • As-submitted proposal package location, archived at submission time rather than at loss time

That last item deserves emphasis. If you archive the proposal only when you lose, you end up comparing the debriefing against a draft that has drifted. Lock the as-submitted volumes, the compliance matrix, and the final pricing narrative into a read-only location on the day of submission. Teams running requirement traceability through a single workspace get this for free; teams running on shared drives have to do it deliberately.

The chronology is also what lets you spot pattern failures across a portfolio. When six losses show an average of five business days between notice receipt and request sent, you do not have a debriefing problem. You have an intake problem, and it belongs in the same place you handle proposal intake and owner assignment.

Writing Questions Against the Evaluation Record, Not the Competitor

The fastest way to waste a debriefing session is to ask about the winner. FAR 15.506 expressly bars point-by-point comparisons between your proposal and competing proposals, and it protects certain competitor information from disclosure [2]. Every minute spent on questions the contracting officer cannot answer is a minute you do not get back.

Build the question set from the solicitation's own evaluation factors and subfactors. For each one, ask a version of the same four questions:

  1. What strengths were assigned to our proposal under this subfactor, and what proposal content supported them?
  2. What weaknesses or significant weaknesses were assigned, and what did the evaluators identify as the gap?
  3. Were any deficiencies identified, and against which solicitation requirement?
  4. How was our price or cost realism characterized in the terms the solicitation used?

Notice what these questions have in common: they are anchored to criteria, so the answers come back in evaluation language you can quote. "We were told our staffing approach carried a significant weakness because the phase-in schedule did not address the incumbent's Tier 2 coverage" is evidence. "They said the other guy was cheaper" is not.

If you are bidding DoD, check DFARS part 215 before you build the question set, because the enhanced debriefing provisions there address additional content and a follow-up question process [7]. Writing your questions without checking that part means you may leave entitled content on the table or submit follow-ups through the wrong channel.

One discipline that pays off: map your question set to your own compliance matrix before the session. If you already know which Section L instructions you answered thinly, you can ask targeted questions about those exact subfactors. That is the same mapping logic behind an AI compliance matrix generator that shreds Section L and M, applied in reverse.

Running the Session: Two Notetakers and Same-Day Capture

Assign two notetakers with different jobs. One captures verbatim language, including partial sentences and hedges. The other maps each answer to a specific evaluation factor in real time. Reconcile the two sets within four hours while memory is still working for you.

The output should not be a narrative summary. It should be a findings table with one row per evaluation finding: factor, subfactor, finding type, the agency's exact wording, and the proposal location the finding appears to reference. A narrative summary reads well and proves nothing. A findings table can be audited, routed, and compared against your next submission.

Worked example. On a mid-size civilian agency recompete, a team received a significant weakness under Technical Approach for "insufficient linkage between the described surge staffing model and the transition milestones in Section C.3.4." Their findings table traced that language to a cross-reference: the surge model lived in Volume I section 2.3, the transition milestones in section 4.1, and no cross-reference connected them. Their compliance matrix had marked both requirements as addressed, because each was individually answered. Nothing in the matrix flagged the dependency between them. That is a matrix design fix, not a writing fix, and it only became visible because the note captured the agency's phrasing word for word.

Submit any permitted follow-up questions only through the method the contracting officer designates, in writing, with the chronology attached. Verbal follow-ups create ambiguity about what was asked and when, and ambiguity about timing is exactly what you cannot afford if a protest decision is still open.

Paraphrase destroys evidentiary value

"They thought our staffing was weak" is a feeling. "The evaluators assigned a significant weakness under Subfactor 2.2 because the proposal did not identify a named program manager for the first 30 days of transition" is a finding you can route to an owner, cite in a protest analysis, and verify as fixed on the next bid. Record the agency's exact wording before you interpret it, and keep the two visually separate in your notes.

The Protest Clock Is a Different Clock

Here is where teams get hurt. The three-day debriefing request deadline and the protest filing deadline are separate clocks running on separate rules, and satisfying one tells you nothing about the other.

GAO's bid protest regulations at 4 CFR part 21 set the filing and timeliness framework, including how timeliness is measured for protests that follow a required debriefing [6]. GAO's own bid protest resources describe the docketing process, the agency report, and the comment schedule that follows [8]. FAR 33.104 then governs what the agency must do when GAO notifies it of a protest, including withholding award or suspending performance in defined circumstances [4]. The practical consequence is blunt: file late and you may end up protesting a contract that is already in performance, with FAR 52.233-3 shaping what happens to that performance once a challenge is filed [9].

Run the timeliness analysis in parallel with the debriefing, not after it. Two things make that possible. First, separate your grounds by origin: challenges arising from the solicitation terms follow different timing logic than challenges arising from the evaluation, and conflating them produces a filing that is partly untimely. Second, remember the agency-level route under FAR 33.103 exists as a distinct option with its own decision-maker, record, and remedies [5].

Set a decision gate with a pre-named approver, counsel engaged before the debriefing rather than after, and a written no-protest rationale filed in the opportunity record when you decline. The rationale matters more than it looks. Six months later, when a similar loss shows the same evaluation pattern, you want to know why you passed the first time.

When the Runbook Does Not Apply: Task Orders, Schedules, and Simplified Buys

The post-award debriefing obligations live in FAR subpart 15.5, which attaches to Part 15 negotiated procurements [13]. If most of your revenue comes through multiple-award vehicles, the familiar runbook will not fit most of your losses.

Classify the competition first, then pick the request language. Citing the wrong authority is one of the most common reasons a feedback request gets declined outright.

Vehicle typeGoverning authorityWhat to requestWhat to avoid askingOwner
Part 15 negotiatedFAR subpart 15.5, FAR 15.506 [13][2]Full factor-by-factor debriefing with strengths, weaknesses, deficiencies, price evaluationPoint-by-point comparisons with competing proposals [2]Proposal manager
Order under multiple-award contractFAR 16.505 [10]Notification and post-award debriefing per 16.505 terms, plus the basis of selectionAssuming Part 15 debriefing content applies; ignoring the limits on protests of orders [10]Capture manager
Federal Supply Schedule orderFAR 8.405 ordering procedures [11]Basis of selection in the solicitation's own terms, such as LPTA or best-value tradeoffA formal Part 15 debriefing citing 15.506BD lead
Simplified acquisitionFAR part 13 [12]A short written explanation of the basis for awardA structured session with factor-level findingsBD lead
DoD negotiated procurementFAR 15.506 plus DFARS part 215 [2][7]Enhanced debriefing content and the designated follow-up question process [7]Submitting follow-ups outside the designated channelProposal manager

Use a single intake form across all five rows. Capture vehicle type, cited authority, dates, and the exact text of the request you sent. Once you have 20 of those records, you can see which request language gets answered and which gets declined, by agency and by vehicle. That is portfolio intelligence most companies never build because each loss is handled as a one-off.

Routing Findings So They Change the Next Submission

Debriefing notes die in folders. The fix is a findings register with named owners, not a better folder structure.

Normalize every finding into a register row with these fields: evaluation factor, subfactor, the agency's exact language, your root-cause assessment, the owner, the due date, and the next bid where the fix gets verified. That last field is the one most registers omit, and it is the one that turns a note into a change.

Then sort findings into two piles, because they go to different owners:

  • Writing defects are library and drafting problems: unclear narrative, missing compliance response, a requirement answered in a volume the evaluator was not reading, a cross-reference that was never made. These belong to the proposal content owner and get fixed in the content reuse library and the structured drafting workflow.
  • Evidence defects are performance record and teaming problems: experience that was judged insufficiently relevant, references that were too old, a contract value too small to signal scale. No amount of rewriting fixes a thin record.

Evidence defects have an upstream remedy most teams skip. FAR subpart 42.15 covers contractor performance information, including the contractor's opportunity to submit comments, rebutting statements, or additional information on an evaluation [14], and CPARS is the system where those evaluations and responses live [15]. If a debriefing tells you your past performance was judged weak on relevance, open the active evaluations in CPARS and use those response rights inside the stated window rather than after the record closes [15][14]. Check how the solicitation and FAR 15.305 framed past performance evaluation [16], then build your relevance matrices to match that framework instead of to your own org chart. Teams doing this at volume treat it as an evidence pipeline, which is the thinking behind CPARS-aligned past performance narratives.

Audit the register at every go/no-go. A weakness that appears in three consecutive debriefings for the same agency is not a proposal problem anymore. It is a capture risk, and it should surface in your go/no-go scoring as a scored factor rather than as a surprise during Red Team.

Frequently asked questions

How long do I have to request a post-award debriefing?

On a FAR Part 15 negotiated procurement, you must request it in writing within three days after receiving the unsuccessful-offeror notice [2]. Log the receipt timestamp, not the award announcement date.

Can the agency refuse to give me a debriefing?

FAR 15.506 conditions the debriefing on a timely written request and describes what the debriefing must and must not include, including the bar on point-by-point comparisons with competing proposals [2]. Content limits are not refusals. A request citing the wrong authority on a non-Part 15 buy, however, is commonly declined.

Does requesting a debriefing extend my protest deadline?

Do not assume it does. GAO's timeliness rules at 4 CFR part 21 govern that question, and timeliness after a required debriefing is measured against the debriefing itself rather than the award notice [6]. Run the analysis with counsel in parallel with the session, not after it.

Do task orders get debriefings?

Orders under multiple-award contracts run on FAR 16.505, which carries its own notification and post-award debriefing provisions and its own limits on protests of orders [10]. Schedule buys follow the FAR 8.405 ordering procedures instead [11].

What about preaward notices?

Preaward debriefings are a separate provision at FAR 15.505 with its own request timing, so a preaward exclusion is not handled with the post-award runbook [3].

Why are some debriefings far more specific than others?

Much of it is decided before the session. FAR 15.308 requires the source selection authority's decision to be documented, including the rationale for any business judgments and tradeoffs made [17], and FAR 15.305 sets the framework for how proposals are evaluated [16]. FAR 15.506 then limits what that record may disclose [2]. A record that ties each strength and weakness to a solicitation criterion and a proposal location supports a specific debriefing; one that restates ratings does not. You cannot control the agency's record, so ask factor-by-factor questions that invite criterion-level answers.

Can we respond to a past performance rating we think is wrong?

Yes, within the stated window. FAR subpart 42.15 covers contractor performance information, including the contractor's opportunity to submit comments, rebutting statements, or additional information on an evaluation [14], and CPARS is the system where those evaluations and responses live [15]. Because past performance information feeds later proposal evaluations under FAR 15.305 [16], an uncontested rating becomes standing evidence against you. Monitor open evaluations rather than discovering them during a bid.

Do this in the next seven days

  1. Stand up the trigger log as a template in your workspace, with the FAR 15.503 content checklist built in [1].
  2. Name the owner and the backup for debriefing triggers, in writing, with authority to send the request without approval.
  3. Pre-draft the request letter so the only variables are solicitation number, offeror name, and date.
  4. Archive your next submission as a locked, retrievable package on submission day.
  5. Open your findings register and backfill the last three losses, even from memory, so the first real entry is not the first entry.

Track one metric this week: the percentage of losses in the last 12 months where a written debriefing request went out inside the FAR 15.506 window [2]. If that number is below 80 percent, your problem is the Friday-afternoon notice, not your evaluation score. The teams that walk out with usable evidence decided what to ask before the notice arrived.

References

Related articles