Utilization Targets by Role: A Framework for Agencies and Consultancies

Every services firm's utilization report has the same tell. There is one number at the top, usually somewhere between 70% and 80%, and everyone stares at it as if it means something. It does not. A single blended target hides both the delivery team you are overworking and the leadership bench you are underpricing, and it never matches the way the P&L is actually generated.

The fix is not a better dashboard. It is a role-by-role target set that reflects what each role is actually supposed to produce. This is the framework.

Why is a single utilization target wrong?

A firm-wide utilization target is a reporting convenience masquerading as a management metric. It treats the sales pipeline principal, the delivery engineer, and the project manager as if they were making the same trade-off between billable and non-billable time. They are not.

• The principal. Their non-billable time is where next quarter's revenue comes from. Under-utilizing them is fine if they are closing. Over-utilizing them starves the pipeline.
• The delivery engineer. Their non-billable time is training, interviews, and internal projects. Under-utilizing them is a coordination failure. Over-utilizing them is a burnout risk.
• The project manager. Their non-billable time is coordination, PMO work, and account grooming, most of which the firm requires but the client does not pay for directly.

Averaging those together gives you a number that hides all three problems at once. If the average is 78%, is delivery under-billed or is the leadership team over-billed? You cannot tell.

What targets should you set for each role?

The specific numbers depend on your firm's model, but the shape is stable across services firms with 30 to 500 people. Use this as a starting point, then calibrate to your actual non-billable expectations.

Two things to notice. First, the range is intentional. A firm that skews toward custom engineering will pull senior IC targets to the top of the range. A firm that runs mostly retainers will pull them down slightly to make room for account work.

Second, the highest-paid people have the lowest targets. That is not a mistake. It is the model.

How do you calculate a role target from first principles?

You do not need a benchmark study. You need a clear view of the role's non-billable time.

1. Start with available hours. Take annual working hours, usually 2,080, subtract PTO (typically 15 to 20 days plus holidays), and subtract expected sick time. Most services firms land around 1,900 available hours per person per year.
2. Estimate structural non-billable time. Interviews, training, internal reviews, all-hands, and PMO work. For a senior engineer this is usually 10 to 15% of available hours. For a principal, 45 to 60%.
3. Subtract and normalize. Take available hours minus non-billable hours, divide by available hours. That is the ceiling of the target. Round down to the nearest 5% to leave a small margin.

If your senior engineer has 1,900 available hours and 300 hours of structural non-billable time, their ceiling is 84%. The target sits at 80 to 85%, depending on how tight you want to run.

What happens when actuals drift from target?

The gap between target and actual is a signal, not an aesthetic complaint. Read the direction and duration to know what to do.

• Actual below target for one to two weeks. Normal variance. No action required.
• Actual below target for three or more weeks in one role. Coordination or pipeline issue in that role. Investigate before treating it as a demand problem.
• Actual above target for two or more weeks in one role. Overallocation. Fix the plan before it becomes a retention problem.
• Actual consistently below target across all roles. Pipeline problem. Sales and marketing own the fix.
• Actual consistently above target across all roles. Understaffed. Accelerate the hiring plan.

The mistake to avoid here is chasing weekly noise. Two-week rolling averages are the right resolution.

How do you avoid gaming the utilization number?

Gaming is inevitable when the metric is monitored but the definitions are loose. Three specific rules close the most common gaming patterns.

• Internal projects do not count as billable. Even if they are formally billed to a "firm building" pseudo-project. The rule is: if a real external client would refuse to pay for these hours, they are not billable.
• Sales support counts as non-billable for delivery, and as target-work for principals. Otherwise every scoping call becomes a way for delivery to inflate their number.
• Time under 10% allocation to a project is rolled up as coordination. Otherwise the plan collects sub-5% assignments to make bench disappear on paper.

Without these rules, utilization becomes a number people write, not a number that reflects reality.

How does the target flow into staffing decisions?

The target is not just a report. It is a planning input. Every allocation decision should reference the target.

• Bench threshold. Anyone more than 15 points below target for two weeks is a coordination flag.
• Overload threshold. Anyone more than 5 points above target for two weeks is a burnout flag.
• Hiring trigger. When a role is consistently at 5 or more points above target across the practice, the next hire is queued in that role.
• Redeployment trigger. When a role is consistently at 10 or more points below target, the next reallocation opportunity goes to that role first.

These are not soft guidelines. They are the mechanism by which the target shapes the operating rhythm.

What actually matters

A utilization target is a statement about how you expect each role to spend its time. A single blended target says the firm has not thought hard about that question. A role-by-role target set says you have. Set the numbers from first principles, hold the definitions tight, and treat drift as a signal to investigate, not a number to explain away. Every meaningful staffing decision downstream, from hiring to escalation to bench management, gets easier once the target is right.