Ask a principal at a CM or A/E firm what a proposal costs and you'll get a confident answer. Ask them to show the math and you'll get a different one. The gap is almost always larger than people think, because the first number only counts the hours somebody logged.
The number most firms quote
It's usually the marketing hours: "our coordinator spends about a week on a SOQ, two on a full proposal." Multiply by a blended rate and the number looks manageable. That's what gets logged, if anyone logs it.
That number is wrong in three directions, and they all stack.
What's missing: the hours nobody logs
The coordinator's hours are the visible ones. The expensive ones belong to the people the agency wants to see on the org chart: the principal in charge, the project manager, the technical leads who write the approach. They rework the executive summary, sit through the color reviews, rehearse the interview, and finish the approach section at night. Almost none of it gets charged to the pursuit.
Count it honestly and the principal alone usually lands at 40 to 80 hours on a full proposal. Add the project manager's and technical leads' hours, and a designer's week on layout and production, and the real hour count is a multiple of the one in the estimate.
What's missing: what those hours would have billed
Those senior hours came out of somewhere: billable project time, client relationships, or evenings. None of those are free, and the billable ones are the costliest, because the firm paid the salary and never sent the invoice.
So a principal's proposal hour has two costs. The loaded cost, meaning salary plus benefits, is what payroll shows. The margin that was never billed is what payroll hides. Together they come to roughly the hour's billing rate, which for senior staff is typically two to three times the loaded cost.
Across a full proposal, with marketing doing the production and principals doing the thinking, true labor cost runs 1.8 to 2.4 times the loaded cost. Most estimates count the first number and stop.
What's missing: the crunch
Proposal cycles concentrate. The last week before submission takes roughly twice the effort per page of the weeks before it, and most of that effort is overtime. Call it a 25% premium on the final-week hours.
On its own that premium is modest, a tenth or so of the labor cost. The larger bill arrives afterwards:
- Diminished output the following week — recovery time isn't logged, but it's real.
- Quality drift on live projects — the job the project manager was running gets less attention during proposal week.
- Talent risk — the senior staff doing the all-nighters are also the ones recruiters call.
None of that shows up on the pursuit's cost line. It shows up in turnover a year later.
What's missing: the pursuits you didn't chase
The cost that hurts most never appears on a timesheet. Pull last year's go/no-go list and sort the no-gos into two piles: declined for fit, and declined because the team was already under water on another submittal.
The second pile isn't random. Firms decline the pursuits that are different: a new agency, a new delivery method, a new geography. Which means the firm keeps optimizing for the work it already knows how to win and rarely gets to the work that would grow it.
The simple model
Here's the math any principal can run in an afternoon. Run it for your last 12 months. "Loaded rate" throughout means salary plus benefits.
- Hours per pursuit, by role, × each role's loaded rate × pursuits per year — the obvious cost, once principal and PM hours are actually counted.
- + Billable-role hours × (billing rate − loaded rate) — the opportunity cost. Coordinators and designers don't carry this; principals, PMs, and technical leads do.
- + 25% premium on the final-week hours — the crunch.
- + Pursuits declined for bandwidth × your historical win rate × expected fee × 0.5 — the work you didn't chase. The 0.5 is a haircut for fit, and for the fact that you'd have pursued them stretched. Use margin instead of fee if you want the strict version.
The number that comes out is uncomfortable. For a firm where a principal carries 40 to 80 hours of every full proposal, it typically lands at three to four times what leadership assumed. The number most firms quote is missing two-thirds or more of the actual cost.
Knowing the real number doesn't tell you what to spend on fixing it. It tells you what the status quo is already costing you.
What this changes
Two things, usually. First, the conversation about fixing the process gets much shorter: once the true cost is on the table, any serious fix gets compared against a number you are already paying. Second, go/no-go changes. Once the "declined for bandwidth" pile is visible, principals start asking why it exists.
Run the model. Print it out. Bring it to your next principals' meeting. The conversation that follows is the most useful one your firm will have this quarter.