Arcvue
Business development & pricing—four sources on one number
By the time you send it, there is nothing in the rate you cannot defend.
Arcvue BD & Pricing builds a bid rate from your own cost structure—labor categories, indirect rates, and competitor and market benchmarks—and keeps the basis for every number in it. The award comes back against that record, so win rate by rate posture is a number you have rather than an opinion you hold. This page follows Arcvue pricing one labor category on one bid, through nine steps.
Specimen Systems LLC · an example 120-person contractor · all figures synthetic
A solicitation names a position, and at that moment nobody can defend a single number about it.
Nothing is inherited yet. An RFP arrives and asks for a Senior Systems Engineer. There is a rate that has to go in a box, and today that number typically comes out of a copy of a master spreadsheet somebody saved for this bid.
The model in that spreadsheet is usually good. The workflow around it is where the time goes and where the errors come in: positions added by hand, vehicle rates and LCATs typed in from a separate source, indirect rates taken from wherever the current version lives. Finding the formula behind an output means pressing F2 on cells until you locate it, and version control is whoever saved last.
What this page counts. Not the rate—it barely moves. The number of independent sources that say the rate is right. It is zero here. By stop 06 it is four, and the fourth one is the one that decides recompetes.
Synthetic. An invented firm bidding an invented recompete. Follow it down—the rate never changes after stop 02. What changes is how much you know about it, and the last thing you learn is worth more than the first.
The first source is your own cost structure, and it was computed last night.
The empty slot from stop 01 gets its first defensible number. This is a T&M bid, so the rate you propose is the rate you are paid for the life of the contract—the indirect rates are a costing assumption inside it, not something billed separately. The wage is $84.10 an hour, from payroll. The burden is your own pools, carried with the basis that set them and watched all year against what the pools are actually producing.
And this business is running above the rates it filed. Fringe, overhead and G&A are all landing higher year to date than the FY2026 provisionals. On the filed rates the position costs $148.26 an hour and the bid earns 9.7%. On what the pools are actually producing it costs $153.72, and the same bid earns 6.4%.
Which basis you price on is a choice, and the proposal records the one you made. A rate set carries a basis—the provisional you filed, year-to-date actuals, prior-year actuals, or an estimate—and the bid is costed on it. If you believe the gap closes, price the provisional. If you do not, price the wrap you are actually running, and 6.4% becomes a number you decided rather than one you find out at year end.
Screen Arcvue proposal pricing, filed provisional rates with their basis recorded. Basis the hourly cost is the burdened month divided by its hours—$24,908.35 over 168.0—so the proposal and the ledger cannot drift apart. A pricing model maintained separately from the ERP drifts the day somebody gets a raise. Synthetic sample.
| Rate basis | Fringe | Overhead | G&A | Cost per hour | Margin at 164.20 |
|---|---|---|---|---|---|
| Filed provisional, FY2026 | 28.4% | 22.7% | 11.9% | 148.26 | 9.7% |
| Year to date, actual | 30.2% | 24.9% | 12.4% | 153.72 | 6.4% |
Screen Arcvue proposal pricing, with the rate basis recorded on the proposal. Read this one row by row. Two alternatives, not components of a total. Synthetic sample.
The second source is the schedule you are bidding under, and it caps this position at $178.48.
A defensible cost came out of stop 02. It is not yet a legal rate: the contract vehicle sets a ceiling, and exceeding it is not a pricing decision but a compliance failure discovered after award.
Vehicle libraries carry ceiling rates for GSA MAS, OASIS+, and the rest, and each position maps to the correct Labor Category Alignment Tool entry—so the match reflects how GSA actually classifies the work rather than how similar two job titles look. At $164.20 against a cap of $178.48 you are pricing $14.28 an hour under the ceiling, and that is visible at the position level rather than discovered in aggregate.
Room under the ceiling is information, not an instruction. $14.28 an hour across the 168.0 hours this position works in a month is $2,399.04 you are not charging. That may be exactly right on a price-sensitive recompete—but it should be a decision somebody made, not a number nobody looked at.
The third source comes from outside your company, and it is a check rather than an input.
You price from cost. You validate against market.
Cost and cap are both facts about you. Neither says whether $164.20 is a sensible number in the market you are bidding into. Published schedule data shows where your rate sits against competitors billing the same labor categories on similar contracts—here, a median of $169.75.
That is a validation layer against pricing you have already built, not a substitute for it. The distinction matters more than it sounds. A tool that priced from market data would produce a rate with no relationship to what the work costs you, and the first time your cost structure differed from the median you would be bidding a loss with great confidence.
The fourth source is the contractor you are trying to displace.
Stop 04 said the market sits at $169.75. That is an average of strangers. On a recompete there is a far more specific number available: what the current contractor’s award implies they are charging on this work—total obligated value over the headcount, which you supply. No public dataset carries what a competitor is charging on THIS contract; this is an estimate built from real obligated dollars.
Look the predecessor contract up by identifier or by recipient, and compare position by position. Here the incumbent’s implied rate is $171.50—$7.30 an hour above the proposed rate, and $6.98 below the schedule cap. It is only as good as the headcount you entered, and it moves if that estimate does. It is still the most specific reading available on a recompete, and it is available before submission rather than inferred from the debrief.
Four independent readings of the same hour, and the rate has not moved since stop 02.
Each of the four stops above produced a number about $164.20 without any of them consulting the others. Cost comes from your ledger, the cap from the contract vehicle, the median from published schedule data, the incumbent's rate from federal obligation data. That independence is what makes the set worth anything—four numbers from one source is one number stated four times.
| Reading | Comes from | Rate | Against your $164.20 |
|---|---|---|---|
| Your cost | Your own ledger, burdened at the rates you bill at | 148.26 | $15.94 of margin—9.7% |
| Market median | Published schedule rates, same labor category | 169.75 | You are $5.55 under |
| The incumbent | Federal obligation data for this contract, over the headcount you enter | 171.50 | You are $7.30 under |
| Schedule cap | The contract vehicle you are bidding under | 178.48 | Priced $14.28 an hour under it |
Screen Arcvue pricing workspace, one position on one proposal. Read this one row by row. The four rates are alternative readings of the same hour, not components of a sum, so no total rule is drawn. Basis the ordering is deliberate—cost first, cap second, outside data last. A workflow that consults the market before it knows its own cost has already lost the ability to say no. Synthetic sample.
Before it goes out, somebody signs for it—and the review flags where the words and the numbers disagree.
Four sources agree at stop 06. That is a well-priced position; it is not yet a submission. The pricing audit walks a structured compliance review—questions specific to the contract type—and runs a systematic comparison that flags discrepancies between the assumptions you have stated and what the numbers actually show.
Then a formal attestation with digital signoff. Not a compliance stamp. It is a governance step that makes pricing and leadership agree, in writing, on what is being submitted and why—which is the document you want to have when a debrief goes badly eighteen months from now.
The audit trail was never assembled. Assumptions are versioned as they are made, so every derivation is visible in the interface rather than reconstructed by pressing F2 on cells. A trail that has to be assembled at submission time is a trail that reflects what somebody remembered, and remembering is the part that fails.
One bid is not a plan, and the only useful question about a pipeline is whether it is enough.
The proposal from stop 07 is submitted. It is one bid. Most pipeline tools will tell you what is in the pipeline; almost none will tell you whether it covers what the company has already committed to.
Coverage is measured against the actual growth target—the one the forecast made explicit—broken down by division, year by year across the forecast horizon. The $3,434,400.00 of go-get revenue in that plan is precisely the number this pipeline has to cover, and stating the two side by side is the difference between a pipeline report and a growth plan.
Opportunities are scored, not listed. Federal solicitation sources are monitored daily and scored against your capabilities and past performance, so what surfaces is what plausibly fits—and everything else stays out of the way. A feed that shows everything is a feed nobody reads.
The outcome comes back, and the thing you were building was never the bid.
It was the record of the bid.
Months later the award is announced. Won or lost, the outcome is captured against the proposal that produced it—the rates bid, the assumptions made, the competitor rates visible in the award data, and who signed for them at stop 07.
Stop 01 opened with a slot in an RFP that nobody could say anything about. The next time that labor category appears in a solicitation, it does not start at zero sources. It starts with what this firm bid, what it cost, what the market said, what the incumbent billed, and whether it won—which is the difference between a BD function with institutional memory and one that re-derives the same answer every eighteen months, usually differently.
BD performance measured, not estimated. Win rate by division, by contract type, by rate posture relative to the incumbent. The question “does bidding under the incumbent actually win us work?” has an answer in your own data. Arcvue produces it. Almost nothing else does.