Fractional CFO for Engineering Firms: Backlog, Utilization, and Project Cash

Eleven months of backlog can still leave an engineering firm with an empty team next month.

The backlog may sit in the wrong discipline, depend on a delayed notice to proceed, represent a task-order ceiling rather than authorized work, or require people the firm cannot hire. Meanwhile, a fixed-fee project can consume senior-review hours faster than the billing schedule releases cash.

A fractional CFO for an engineering firm should connect backlog quality, discipline-level capacity, utilization, labor multipliers, project forecasts, billing, collections, and cash. The role is not to manage designs or replace project managers, the controller, the CPA, or contract counsel. It is to make the financial consequence of the firm's project and staffing choices visible before payroll is committed.

Backlog needs an authorization status

Engineering backlogs often combine work with very different certainty:

• Executed and authorized work already in production.
• Contracted work awaiting notice to proceed, funding, permitting, or client scheduling.
• On-call or indefinite-delivery contract capacity without a funded task order.
• Awarded work still subject to commercial terms.
• Proposals and likely recompetes that are not backlog at all.

Keep those categories separate.

For each contracted project or task order, record:

1. Remaining net service revenue.
2. Authorization and funding status.
3. Expected start and monthly burn.
4. Required discipline, level, location, and project leadership.
5. Contract type and fee risk.
6. Subconsultant commitments.
7. Billing trigger and expected collection timing.

ACEC Research Institute's Q2 2026 survey found a median backlog of 11 months among responding engineering firms, with nearly half reporting at least a year. That is useful market context, not permission to hire against one firm-wide number. Backlog only supports a specific hire when authorized work converts into that person's available hours in the right period.

Translate backlog into discipline-level hours

Assume a civil group has $2.6 million of remaining net service revenue. Management expects it to convert over the next 12 months, but the work is not equally firm:

Backlog layer	Remaining NSR	Timing confidence	Planning amount
Active and authorized	$1,450,000	100%	$1,450,000
Contracted, funding pending	$700,000	60%	$420,000
On-call ceiling without task order	$450,000	20%	$90,000
Total	$2,600,000	—	$1,960,000

The $640,000 difference between stated and timing-adjusted backlog is not a write-off. It belongs in upside scenarios until authorization and timing become clearer.

Next, convert the planning amount into hours by role using the project's expected revenue per direct labor hour and subconsultant profile. A water-resources backlog cannot automatically fill transportation capacity. Nor can four junior engineers replace a project manager's review and client responsibilities merely because total available hours match.

The CFO should challenge timing and financial assumptions while operations owns the technical staffing plan.

Utilization is a capacity signal, not a target to maximize

Define the denominator before comparing utilization:

Utilization = direct project hours ÷ practical available hours

Practical available hours should reflect paid leave, holidays, training, management duties, business development, and other approved nonproject work. Firms may use different definitions, so external comparisons require a definition bridge.

Suppose 80 employees have 1,720 practical available hours each:

80 × 1,720 = 137,600 practical hours

The operating plan assumes 68% direct utilization:

137,600 × 68% = 93,568 direct hours

Actual utilization is running at 62%:

137,600 × 62% = 85,312 direct hours

The gap is 8,256 direct hours. At a hypothetical $145 of net service revenue per direct hour, that represents about $1.197 million of annualized revenue capacity:

8,256 × $145 = $1,197,120

That is not automatically lost revenue or profit. The work may not exist, the skill mix may be wrong, leaders may be investing time in winning future projects, or the plan may have used an unrealistic denominator. The calculation tells management what to investigate.

Deltek's 47th annual Architecture & Engineering study reported utilization just under 60% and operating profit of 16.7% across its respondents, alongside softening backlog and limited staff growth. Those figures are external reference points, not Bennett targets. Discipline mix, public versus private work, contract type, leadership time, geography, and definitions all change the economics.

Want to test this against your own numbers? Book a free 20-minute Profit & Tax Leak Check to pinpoint the first profit, tax, cash-flow, or financial-structure issue worth fixing.

Connect the labor multiplier to the project forecast

Utilization answers how much practical time reaches projects. The labor multiplier answers how much net service revenue the firm earns relative to direct labor cost:

Net labor multiplier = net service revenue ÷ direct labor cost

Assume a fixed-fee engineering project has:

• $750,000 of contract revenue.
• $110,000 of direct subconsultant cost.
• $640,000 of net service revenue.
• $190,000 of planned direct labor.
• $20,000 of other direct project cost.

The planned net labor multiplier is:

$640,000 ÷ $190,000 = 3.37

Planned project contribution before firm overhead is:

$640,000 − $190,000 − $20,000 = $430,000

Mid-project, the firm has incurred $125,000 of direct labor and the project manager now expects another $95,000 to finish. Forecast final direct labor is $220,000, $30,000 over plan.

The forecast multiplier falls to:

$640,000 ÷ $220,000 = 2.91

Forecast contribution falls to $400,000 before any other cost change. The financial response is not “increase utilization.” Management needs to know why the estimate changed: rework, review load, scope, fee, staffing mix, schedule, or a client decision.

The service-business job-costing framework explains loaded labor and direct cost. Engineering adds contract authorization, discipline capacity, project review, subconsultants, and public-sector or client-specific billing rules.

Rework and quality effort need separate visibility

Quality control is part of delivering professional engineering work. It should not be treated as waste merely because it consumes hours.

The useful distinction is among:

• Planned quality assurance and technical review.
• Normal iteration required by the approved scope.
• Client-requested change with a recoverable fee.
• Internal correction or rework.
• Unfunded work performed before authorization.

If every excess hour is coded to the original task, management cannot distinguish a fee problem from a delivery problem. If teams avoid recording review time to protect a metric, the project forecast becomes fiction.

Set project codes and review practices that capture the truth without creating a punitive timesheet culture. Then use the pattern to improve scoping, staffing, review budgets, and change control.

Contract type changes the risk

One dashboard should not interpret every engineering contract the same way.

Contract structure	Primary financial exposure
Fixed fee	Hours and other cost exceed the remaining fee
Time and materials with a ceiling	Work continues after authorized capacity is nearly exhausted
Cost plus	Incurred cost or indirect-rate treatment is not recoverable under the agreement
Unit price	Production volume and field conditions differ from the pricing assumption
On-call or IDIQ	Contract ceiling is mistaken for funded, scheduled work

Government and public-sector work may add indirect-rate, allowability, audit, retainage, or invoice-support requirements. The controller, contract specialist, CPA, and counsel should define those obligations. The CFO should quantify their effect on margin, billing, and cash.

Project profit and project cash move on different schedules

Assume a project forecasts $120,000 of final contribution but has the following six-week cash path:

Six-week project cash	Amount
Client receipts	$180,000
Payroll and burden	($142,000)
Subconsultants	($58,000)
Travel, software, and other project cash	($14,000)
Net project cash	($34,000)

The project consumes $34,000 during the window even though it is expected to finish profitably. A billing milestone, client approval, invoice-support issue, or slow payment creates the gap.

The thirteen-week cash forecast should connect project invoices and collections to payroll, subconsultants, tax, debt, and owner payments. For material projects, show the next billing trigger, responsible owner, invoice date, expected cash date, and downside date.

The engineering-firm CFO decision matrix

Signal	Reporting answer	CFO decision
Backlog is high but a discipline is underused	Show total backlog	Separate authorization, timing, discipline, and confidence
Utilization misses plan	Report one percentage	Find whether demand, skill mix, management load, or execution moved
Project margin appears at close	Show actual cost	Reforecast estimate to complete while fee remains
Hiring is justified by the pipeline	Show proposal value	Underwrite authorized hours, role need, start timing, and downside cash
Project is profitable but cash is tight	Show accrual profit	Connect billing triggers, A/R, payroll, and subconsultants by week
Rework is debated from memory	Challenge timesheets	Separate planned review, scope change, correction, and unfunded work

ACEC's Q2 2026 survey also reported that firms constrained by staffing were becoming more selective, including turning away less profitable work. That is the right kind of response to capacity pressure: choose work from contribution, strategic fit, risk, and cash—not simply top-line backlog.

The first 90 days

Days 1–30: make project data trustworthy

• Reconcile the project system, time records, billing, A/R, and general ledger.
• Define net service revenue, direct labor, utilization, and multiplier consistently.
• Classify backlog by authorization, timing, discipline, and contract type.
• Identify old unbilled work, exhausted fees, and unreconciled subconsultants.

Days 31–60: connect projects to people

• Reforecast estimate to complete and final margin on material projects.
• Convert timing-adjusted backlog into role-specific monthly hours.
• Separate planned review, scope change, rework, and unfunded effort.
• Model hiring, overtime, contractors, and delayed-start scenarios.

Days 61–90: connect the portfolio to cash

• Build the rolling thirteen-week cash forecast.
• Add project billing triggers, receipt dates, and subconsultant commitments.
• Establish approval gates for hiring, pursuits, fee changes, and capital spending.
• Run a monthly project-portfolio review that ends with named actions and dates.

A fractional CFO cannot repair missing timesheets, unreliable estimates to complete, inconsistent project codes, or an unreconciled billing system. Project accounting, operations, controller, contract, or compliance work may need to come first.

Sources

• ACEC Research Institute: Engineering Business Sentiment Q2 2026
• ACEC Engineering Inc.: Engineering Firms Have More Work Than People
• Deltek: 47th Annual Architecture & Engineering Industry Study findings
• Zweig Group: 2026 Financial Performance Report of AEC Firms

Once the records are dependable, fractional CFO support should turn project information into staffing, pursuit, pricing, collection, and liquidity decisions. Architecture practices can see the related but phase-specific treatment in the architecture-firm CFO guide. The Profit & Tax Leak Check can identify whether project margin, utilization, overhead, billing, cash timing, tax structure, or owner distributions create the first constraint.