The SBIR Phase II Cost Volume, Demystified

Engineers treat the cost volume as paperwork. A contracting officer treats it as a polygraph. An inconsistent, padded, or cross-subsidized cost volume tells them they cannot trust a single number you wrote, including the ones in your technical volume.

Build it like this.

The cost ladder

Build it in this order, every time:

Direct Labor
  + Fringe
= Total Direct Labor
  + Other Direct Costs   (cloud, inference, materials, travel, subcontracts)
  + G&A
  + Fee / profit
= Total

At small-business scale a two-tier indirect structure (Fringe and G&A) is the common, defensible default. The governing rule is not a specific number of pools; it is a consistent, allocable, auditable structure. Add a separate overhead pool only if it genuinely reflects how the business operates and is applied consistently. An unsupported extra pool double-counts facilities and invites audit questions. Boring wins here.

Indirect percentages vary by firm. Do not copy a number from a blog (including this one). Use your actuals.

The trap that reads as a red flag

There is a defensible way and a dangerous way to justify a low G&A rate.

Right: "Our G&A is low because we are a lean, founder-run firm with no separate finance department." That is a true statement about your operations.

Dangerous: "Our G&A is low because the cost is largely absorbed by our other contracts / covered by our commercial revenue." That is a cross-subsidy assertion. It raises a cost-allowability problem under the Federal Acquisition Regulation (the cost-allocability principles in FAR Part 31), and a contracting officer reads it as a warning sign about every other number in your volume. It is not always an automatic rejection, but it is an audit-and-negotiation problem you created for free.

Same low rate. One framing is a story about discipline. The other is a liability. Know which one you wrote.

What the reviewer hears

Substantiate labor rates, do not invent them

Labor rates must be actual, reasonable, and supportable. If you have payroll history, that is the basis. For a role you have not hired yet, a published benchmark (for example the Bureau of Labor Statistics occupational wage for the relevant code) is one defensible reference, and the conservative side of it is easier to defend in negotiation. The rule is supportability, not a particular database. A rate you can source is a rate that survives a cost review.

Hit the ceiling clean

Two arithmetic disciplines that are not optional:

• Land within a few dollars of the ceiling, with no cents. The government dislikes cents and dislikes leaving ceiling unaccounted. A total of $1,499,999.97 looks like you did not control your own budget.
• Do not pad to reach the number. Tiny, obviously-make-work line items to consume ceiling are transparent and they cost you trust.

Reconcile with the schedule

The per-milestone cost allocation and the milestone table in the work plan must show the same totals. This is the single error a contracting officer can catch with zero domain knowledge, so it is the one they always catch. Build the milestone numbers once and reference them in both places. The milestone-table piece covers the shape.

A note on billing reality

The dollar figure on a milestone is the proposal's allocation. How you actually get paid after award depends on the contract type: a cost-type contract bills incurred cost against provisional then final indirect rates; firm-fixed-price and grants follow their own payment terms. Knowing which one your topic uses, before you propose, keeps you from designing a cost volume the award vehicle cannot actually execute.

Commercialization dies when it sounds like intent

While you are being concrete with money, be concrete in the commercialization section too. A winning one is structured: company and team, customer and competition with third-party demand evidence (a cited market study, a real LOI in negotiation), a sized market broken by segment, specific IP, a real Financing subsection (capital invested, rounds raised, named investors, the path to Phase III), named partners, and named customers on both the defense and commercial side. A Financing subsection and a DoD- customers subsection are legitimate and expected. What is weak is a paragraph of intent, an agency list as the opener, or "we will raise a round" standing in for a customer.

Get the cost volume consistent, allowable, and clean and you remove a whole category of quiet rejection. Back to the main playbook.