Margins in 2026 are under pressure from clients expecting AI-speed delivery at the same scope. The number that tells you whether your team is paid for its time is utilisation, and most small agencies guess it.
Utilisation rate is billable hours divided by available hours, usually per person per week or month. Available hours are working hours after leave and holidays — not 24 × 7, not "whatever they worked". Billable hours are time logged against client work that is (or will be) invoiced, whether on time-and-materials, fixed fee or retainer. For a small agency the healthy range for delivery staff is roughly 65–80%; below 60% you are carrying cost, above 85% for months you are burning people and skipping the work (sales, process, training) that keeps next quarter's utilisation up. The number is useless as a quarterly estimate and valuable as a weekly fact, and the weekly fact comes from the same daily task log that feeds project profitability. Facts checked 14 October 2026.
| Available hours | Contracted hours for the period (e.g. 5 days × 8 = 40) minus approved leave and public holidays. Do not subtract meetings, admin or training — those are what utilisation measures against. |
|---|---|
| Billable hours | Time logged to a client project under a signed engagement, regardless of billing model. On fixed fee and retainer, still billable: it is the hours that fee bought. |
| Non-billable | Internal projects, sales and proposals, training, admin, unpaid rework, pro bono. Logged against an internal "client" so it is counted, not invisible. |
| Utilisation | Billable ÷ available. Per person per week; team average per month. |
| Realisation | Hours invoiced ÷ hours logged billable. Below 90% means you are logging billable time you then write off. |
The common mistake is measuring billable against hours worked rather than hours available: a designer who works 50 hours with 35 billable shows 70% on the wrong denominator and 87.5% on the right one — the second number is the one that predicts resignation.
Worked example: a 12-person agency, 8 delivery staff at 72% average, 4 others at 40%. Delivery utilisation is the number to manage; the blended 61% is the number to put in the pricing model, because it is what the fee has to carry.
If every task entry carries a client, a project and time spent, utilisation is a filter: sum time on client projects, divide by (contracted hours − leave − holidays) from the attendance and leave record, per person, per week. No separate timesheet, no Friday reconstruction. Export the month, and the same rows give project profitability, scope creep and the retainer renewal case. The structure that makes this possible is in clients and projects.
| Formula | Billable hours ÷ available hours |
|---|---|
| Available hours | Contracted hours minus leave and holidays |
| Billable hours | Time logged to a client project under an engagement, any billing model |
| Healthy range (delivery) | 65–80% |
| Warning | Under 60% sustained, or over 85% sustained |
| Realisation | Hours invoiced ÷ hours logged billable; aim above 90% |
| Cadence | Per person weekly; per team monthly |
| Source | Daily task log with client, project and time on every entry |
Divide billable hours by available hours for the period. Available hours are contracted working hours minus approved leave and public holidays; billable hours are time logged to client projects under a signed engagement, whether billed hourly, fixed fee or retainer. Compute it per person weekly and average per team monthly.
For delivery staff, roughly 65–80%. Under 60% sustained indicates unsold capacity or under-logging; over 85% sustained indicates burnout risk and neglected sales and internal work. Founders, leads and account managers run lower by design and should be reported separately.
Utilisation is billable hours over available hours — how much of the team's time went to client work. Realisation is hours invoiced over hours logged as billable — how much of that client work was actually charged. High utilisation with low realisation means hours are being given away through scope creep and write-offs.
Yes. The fee bought those hours; logging them as billable is what lets you see whether the fee covered the time. Treating fixed-fee work as non-billable hides the projects that lose money.
Because available hours include the time an agency needs for sales, proposals, training, process and rest. A team at 100% is doing none of those, so next quarter has no pipeline and the quarter after has no team. Plan capacity at about 75%.
Keep a daily task log where every entry has a client, a project and time spent — including an internal client for non-billable work — entered at the end of each day. Utilisation is then a filter on the log against available hours from the attendance and leave record, with no separate weekly timesheet.
Merik's daily task log carries client, project, time spent, status, blockers and proof links on every entry, and rolls up per project and per client; attendance and approved leave give the available hours. Utilisation per person per week is therefore a report over data the team already enters each day, and the same rows feed project profitability, estimates vs actuals and the scope-creep check. Task Insights shows where time went by client and project; CSV export covers anything you want to compute differently.
Merik does not label entries billable or non-billable automatically or compute realisation against invoices; set up an internal client for non-billable work and the distinction falls out of the project. See the tasks and clients modules.
Free, set up for you, every module included — see what the free workspace includes
Turn the hours your team already logs into a per-project margin.
Estimate vs logged hours per project, weekly, turned into a change request in time.
Same raw material, very different compliance and usefulness.
Request your free company workspace and bring employees, attendance, leave, payroll and tasks together.