Billable Utilization Rate for Agencies: The 100% Trap

The same 280 billable hours can produce a 70%, 77.8%, or 87.5% result.

The arithmetic is correct in all three cases. The denominator is different. One report uses contracted hours, one removes approved leave and holidays, and one divides by hours people happened to record. Only the middle result answers a clean capacity question.

That denominator problem is why a billable utilization rate can look precise while sending an agency toward the wrong decision. A low result may be a demand or allocation problem, not an individual effort problem. A high result may hide missing non-billable time, fixed-fee overwork, or overtime. A clean 100% can even be correct for one short period without being a sensible permanent target.

This guide gives agency owners and operations leads a definition, worked example, role-setting method, and weekly decision table. The documentation and reports were checked on August 11, 2026. This is a desk review, not a hands-on product test or accounting, employment, or health advice.

The short answer

Use this formula for an agency capacity decision:

Billable utilization rate = approved billable hours / effective available hours × 100

For the same period and people:

Effective available hours = contracted working capacity - approved leave - public holidays - other scheduled non-working time

Keep three rules beside the formula:

1. Forecast and actual utilization are separate numbers. Forecast uses scheduled billable hours. Actual uses complete, reviewed time records.
2. Recorded hours are a completeness check, not the capacity denominator. Dividing billable hours by incomplete timesheets rewards missing non-billable entries.
3. There is no universal good rate. Derive a target from role responsibilities, pricing model, direct delivery costs, planned internal work, and the time horizon of the decision.

Use utilization as a management signal. Do not use it by itself as proof of productivity, project margin, work quality, or individual performance.

Start with the denominator, not the target

Most guides agree on the numerator: hours classified as billable client work. The disagreement begins under the line.

Current systems use several denominator concepts. Runn's metrics glossary defines effective capacity as contracted capacity minus time off. Teamwork's utilization report defines available time as total working hours minus unavailable time.

Those choices are not interchangeable. They answer different questions.

The recommended denominator for a capacity decision is effective available capacity. The recorded-hours ratio is still useful, but call it billable share or billability. It is not a substitute for a capacity-based utilization rate.

The same-hours, three-denominator test

Consider a hypothetical 10-person creative studio over one week:

• Contracted capacity: 400 hours
• Approved leave and holidays: 40 hours
• Effective available capacity: 360 hours
• Total time recorded: 320 hours
• Approved billable time: 280 hours
• Recorded non-billable time: 40 hours

The agency can now generate three plausible percentages.

Table: one numerator, three different management signals.

The 87.5% result looks strongest. It also carries the sharpest data warning: 40 effective hours are not represented in the timesheets. Those hours could be an unrecorded demand gap, internal work, a schedule error, or missing entries. The ratio cannot decide which one.

This is the first anti-gaming control: reconcile total recorded hours to effective capacity before interpreting billable utilization. If records are incomplete, mark the period incomplete. Do not fill the gap with an accusation.

Runn's current documentation makes the data risk concrete. Its reports can use scheduled hours when actual hours are missing, and the help page tells users to check completed timesheets first. That is a documented Runn behavior, not a claim about every product. It shows why an actual report needs a completeness gate.

Create an Agency Utilization Definition Record

A percentage should not leave the review meeting without its definition attached. Use one short record and version it when a policy changes.

Table: the Agency Utilization Definition Record.

The record prevents a quiet policy change from looking like performance improvement. If one month excludes leave and the next includes it, the two percentages are not comparable until the earlier period is restated or the break is disclosed.

The same rule applies to billable classification. A client workshop, revision, pitch, or research hour can move between categories without any change in work. Classification policy belongs in the definition, not in a manager's memory.

Separate forecast, actual, and completeness

One utilization column is not enough for a weekly agency review. Use three.

Planned and actual hours can diverge because a client delays a start, a task finishes early, scope expands, a person switches projects, or time is missing. The variance is not the verdict. It is the queue for review.

A real agency example shows why the workflow matters. DPDK's CEO told the Society of Digital Agencies that the agency reviews planned versus actual hours and revenue each week. DPDK also ran a nearly year-long experiment in which Operations prefilled timesheets, then revised the process using what the team learned. The SoDA interview does not prove one correct method. It shows that the capture process itself can be tested instead of treated as fixed.

If the actual column is low and completeness is also low, close the time-record problem first. If actual is low with complete records, investigate demand, allocation, blockers, and role design before looking at an individual score.

What is a good billable utilization rate for an agency?

A good rate is one that meets the agency's economic requirement while leaving explicit capacity for the role's necessary non-billable work, with margin, delivery quality, and workload still inside the agency's limits.

That answer is less portable than a 75% target. It is also defensible.

The 2025 SPI Professional Services Maturity Benchmark reports a 68.8% average billable utilization in a broad professional-services project-margin table. The same report covers different organization sizes and markets, including agencies, consultancies, architecture and engineering firms, and software services. Treat the SPI average as context, not as an agency, team, or role target.

The agency market itself is not one operating model. Promethean Research's 2026 Digital Agency Industry Report combines survey data from 1,452 agency leaders, 3,172 employee positions, and observational data from more than 200,000 agencies. It distinguishes business models, team structures, pricing methods, and service mixes. Those differences change how much client work a role can carry and what one billable hour means financially.

Build the target in this order:

1. Set effective capacity for the role and period.
2. Budget required internal work such as management, sales support, training, quality review, and process maintenance.
3. Reserve a named operating buffer for normal variation. Do not hide the buffer inside unpaid overtime.
4. Calculate the revenue or contribution needed from billable work.
5. Test whether the economic floor and the role's time budget can both be true.
6. If they cannot, change price, scope, service mix, cost, or role design. Raising the target does not create more contracted hours.

A creative-studio target derived from its economics

Use this labeled hypothetical for a five-person studio over one month:

• Effective available capacity: 760 hours
• Staff and allocated operating cost for the period: $45,000
• Expected realized revenue per billable hour: $125
• Desired operating margin in this simplified model: 20%

Break-even revenue is $45,000. At $125 of realized revenue per billable hour, the studio needs 360 billable hours, or 47.4% utilization, to cover the modeled cost.

To model a 20% margin, required revenue is $45,000 / (1 - 0.20), or $56,250. That requires 450 billable hours, or 59.2% utilization.

This is not a benchmark or an accounting rule. It is a management model. Change the cost allocation, realized rate, margin definition, or capacity and the target changes. The useful output is the relationship between the assumptions, not 59.2% as a number to copy.

The model also exposes a decision. If the role budget says only 420 hours can be billable without dropping required management and quality work, a 450-hour economic requirement will not fit. The agency must adjust the business model instead of pressuring the timesheet.

The 100% trap has three forms

A correct short-period result

Suppose a person is contracted for 40 hours, has eight hours of approved leave, and records 32 approved billable hours. Effective capacity is 32 hours, so billable utilization is 100%.

That is mathematically clean. Productive's current capacity and availability guide documents that time off reduces availability and that utilization is based on availability. A correct 100% for one period does not prove overwork.

A permanent plan with no internal capacity

A standing 100% target is a different claim. It assigns every effective hour to billable work. If the role still owns internal meetings, coaching, training, proposals, process work, or administration, the plan has left those duties without capacity.

The American Institute of Architects makes this boundary explicit for architecture firms. Its KPI guidance says necessary indirect work includes training, support, management, and other non-project activity, so utilization should be optimized rather than maximized. Architecture is not digital agency work, but the time-budget principle transfers cleanly.

Avoid turning that planning error into an automatic health claim. UK HSE workload guidance says organizations should set adequate and achievable demands relative to agreed hours, while current NIOSH fatigue guidance identifies extended hours, stress, and demanding work among several fatigue factors. Neither authority names a billable utilization threshold. Inspect actual hours, recovery time, workload, rework, quality, and employee concerns.

A result above 100%

If the same person records 34 billable hours against 32 effective hours, the rate is 106.25%. That could reflect two hours of overtime, a late leave change, duplicated time, or another mismatch.

Do not cap the report at 100% and hide the exception. Preserve the hours, flag the result, and resolve the cause. A result above 100% is a review signal, not a performance badge.

Pair utilization with realization and project margin

Billable classification does not guarantee that the agency earns the expected money.

For a time-and-materials project, approved billable hours may flow directly toward an invoice. For a fixed-fee project, another 20 client-project hours can raise utilization while the fee stays fixed and direct delivery cost rises. The team looks more billable while the project becomes less profitable.

Use a three-number diagnostic:

1. Billable utilization: approved billable hours divided by effective capacity.
2. Realization: recognized or invoiced revenue divided by the standard value assigned to those billable hours. Define the agency's exact method because firms use the term differently.
3. Project margin: net project revenue less direct delivery cost, divided by net project revenue.

This is why activity cannot rescue an unclear economic model. A keyboard or application signal does not prove that work was billable, accepted, in scope, or profitable. The activity is not productivity framework gives managers a broader set of output, quality, customer, delivery, rework, and trust measures.

Use the weekly five-branch decision table

Review the metric at team or role level first. Use several comparable periods before a staffing move. A rolling four weeks is one practical starting point, but the right window depends on the sales cycle and project length.

Table: a weekly utilization decision record.

Every row has a management action. None says, "Tell the person to work harder."

Before using an individual's rate in a review, confirm six controls: enough billable demand existed; the work was allocated; role responsibilities were budgeted; leave and schedule were right; billable categories were applied consistently; and time records were complete. If one fails, the number belongs to the operating system before it belongs to the person.

Put the method into a 30-minute weekly review

The review can stay small:

1. Freeze the prior week's leave, schedule, and time approvals.
2. Publish completeness first. Keep incomplete teams out of comparison.
3. Compare forecast and actual by role or discipline, not only agency-wide.
4. Add realization or project margin for the work that drove the result.
5. Route every exception through one branch of the decision table.
6. Record an owner, action, due date, and definition version.
7. Revisit hiring only after several comparable periods and a pipeline check.

Do not buy software before the definition works in a simple sheet. Once the agency knows the records and decisions it needs, the wider time-tracking buying checklist can test whether a product preserves the denominator, approvals, history, exports, and access rules. The real cost guide helps add manager review and correction work to the seat price.

For teams with overnight or client-separated work, the BPO evidence-chain checklist is useful even outside a call center. It keeps scheduled time, captured time, approved time, client evidence, and work quality from collapsing into one record.

Where KordanoTime fits

KordanoTime is not live. Access is planned to begin on December 1, 2026.

The initial product is being built with planned online and offline time tracking, project and task attribution, schedules, attendance, approvals, leave and break tracking, reports, and payroll-ready exports. Those planned records could support the capacity, completeness, and approval parts of this method. They will not choose the agency's denominator, classify work correctly, or prove project margin by themselves.

There are no live-product screenshots or trial results to show yet. When access begins, an agency should still test one leave change, one missing timesheet, one fixed-fee overrun, and one post-approval correction before trusting the report.

If a clear weekly record fits the way your agency wants to run capacity reviews, join the Founding list.

Make the next percentage explain itself

Do not ask whether 75% is good until the report can explain 75%.

Name the people, period, billable rule, capacity source, leave treatment, overtime rule, missing-time rule, approval state, owner, and effective date. Keep forecast separate from actual. Pair the result with realization or project margin. Route the pattern to a management action.

Then the billable utilization rate can do its real job: show where paid capacity meets client demand, and where the agency needs to sell, unblock, re-scope, rebalance, price, or hire.