The RFP lands on a Thursday. The capture tracker is green. Then the writing room opens a shared drive and starts reconstructing how the company actually works: a solutioning deck from two quarters ago, a staffing chart with people who left, a past-performance note with no owner. Three days into a two-week calendar, the team is doing archaeology. The evaluation factors are still waiting.
We treat that scramble as a capture problem, not a heroics problem. Pursuit week should be execution against a solicitation. If the first days are spent hunting for how delivery runs, who decides, and what you are allowed to claim, the volume will be generic where it needed to be specific.
Capture-ready is a packet, not a pipeline status
A pursuit can look “in capture” and still be unreadiness wearing a color. Pipeline stages and a named capture lead are necessary. They do not give a proposal team something they can write from when the clock starts.
We look for a capture-ready packet: the operating picture, the technical artifacts that describe it, and the decision notes that constrain what the volumes may say. Without those, writers invent a method from memory or imply proof that lives in someone else’s inbox.
Three tests decide whether a team is actually ready:
Current. Delivery would still recognize how work runs. Last year’s org chart is not current. A method that names roles, handoffs, artifacts, and what the government would see — and that an operations lead would still sign — is.
Decided. Someone has recorded the choices writers should not relitigate in week one: whether this is a go, what we will not invent, which teammates are in, which discriminators we can defend, and which claims are parked until evidence exists. Unwritten “we’ll figure it out when we see the RFP” is a delay with a due date.
Findable. The packet has owners and dates. If the first action after release is “who has the latest…?”, the archive is a pile, not a system.
This is not a promise that early artifacts win the bid. Evaluation is fact-specific, and the RFP will still change the outline. The useful question is narrower: if this dropped tomorrow, would we be writing to the factors — or reconstructing ourselves?
We do not invent client names, past performance, or win histories to fill a thin packet. If proof is thin or not yet assembled, the decision log should say so. Method, staffing logic, and controls have to carry what you cannot cite.
When we sit with a capture and proposal pair before release, we look for:
- An operating sequence delivery would still recognize: intake, execution, review, correction, and what the customer sees.
- A short artifact index: what exists, who owns it, when it was last confirmed, which volume it usually feeds.
- Decision notes in writing: go/no-go logic, teammates, claims we will not make, open questions the RFP must close.
- An evidence index that is honest about what can be requested, from whom, and on what timeline.
What we typically check and do not treat as ready: a pipeline CRM with empty artifact fields; a boilerplate capability catalog; a technical approach recycled from a different PWS.
Practical moves
These are checks we use when capture wants to be ready before release. They are not a template for every vehicle, and they are not legal or procurement advice.
1. Write the operating picture now, while delivery can still argue with it.
Ask people who run the work to describe a normal week: how work enters, who is accountable, which artifacts move, where quality is seen, and what breaks. Capture that as a short sequence with roles — not a marketing one-pager. If delivery cannot recognize the picture, proposal cannot defend it. Update it on a named cadence so pursuit week is not the first time anyone notices the org chart is fiction.
2. Keep a decision log the writing room is allowed to trust.
Record the choices that otherwise get re-litigated under deadline: why this pursuit, what we will not claim, which teammates are in, which discriminators we can show, and which questions are parked for the RFP. A log is a constraint set so volume authors do not invent a strategy in paragraph four. If a decision is still open, name the owner and the date it must close.
3. Index evidence before you need it.
List the proof typical volumes will ask for: candidate past-performance vehicles, technical artifacts, certifications you actually hold, staffing constructs, and who can release each item. Note gaps in the same list. An honest gap — this claim has no cite yet; do not write it — is more useful than a late scramble nobody can stand behind in orals. We do not manufacture proof to close a cell.
4. Produce capture-ready technical artifacts with owners and dates.
The packet we mean is boring on purpose: current-state method notes, toolchain descriptions delivery still uses, interface and transition sketches, quality and risk language tied to how you actually run, role descriptions that match labor you can staff. Each item needs an owner and a last-confirmed date. Tie artifacts to the volumes they usually feed so mapping after release is matching, not a treasure hunt.
5. Run a pre-RFP dry run: if it dropped tomorrow, what would we still hunt?
Take two hours. Open a blank outline for a Volume II–style technical approach and a management narrative. See how far the packet gets you before someone has to ping a person, reconstruct a process, or guess a discriminator. The leftover list is the capture work remaining. A dry run will not predict evaluation. It will show you whether pursuit week is execution or archaeology.
None of this replaces a color team after release. It gives that team a method and a constraint set worth marking, instead of a first draft that is still discovering the company.
Who this is for
This is for federal contractors whose capture and proposal teams share a calendar but not a packet — pipeline green, artifacts stale, decisions still in conversation. It is not a guarantee about ratings, and it is not advice on whether to bid.
If you are in that bind — RFP not out, writing room already destined to reconstruct delivery — we help teams produce capture-ready technical documentation so release week is execution. Tell us the vehicle, the constraint, and how much of the packet already exists. We will say whether a collaboration is a fit.