Capacity utilisation compares a defined productive output or time measure with explicitly available capacity for the same period. Separate paid hours, planned absence, operational availability and productive delivery. Show the denominator bridge and exclusions, because changing availability rules can improve the percentage without increasing useful output.
Choose the capacity question and measurement unit
A team can be paid for forty hours per person while having fewer hours available for customer delivery. Leave, training, equipment downtime and required administration can all affect the denominator. Start with the decision: staffing, machine scheduling and billing need different measures. Define whether utilisation uses time, physical output or another capacity unit, and keep numerator and denominator compatible.
Specify the productive event. Billable time is not automatically productive time, and productive internal work may not be billable. A machine running on defective output is occupied without generating the intended useful output. Record those distinctions where they matter. Avoid measuring one team's delivered hours against another team's theoretical capacity simply because the workforce totals are readily available in one spreadsheet.
Build a visible availability bridge
Start with a named baseline such as contracted hours or scheduled machine hours. Subtract only the categories defined as unavailable under the metric. Preserve those categories separately and prevent overlapping deductions. If a training session occurs during a machine outage, both may be relevant operational facts, but subtracting the same hour twice understates available capacity. State whether overtime extends the denominator and how partial-period staff are handled.
Treat unrecorded time as unknown rather than automatically idle. A missing timesheet can lower the productive numerator without showing whether capacity was actually unused. Report coverage and reconcile recorded hours to the relevant schedule. Where a constrained skill or machine limits delivery, an aggregate staff-hour ratio may conceal the bottleneck; show the constrained population separately instead of assuming every available hour is substitutable.
Interpret utilisation alongside outcomes
High utilisation can leave little room for variability and recovery, while low utilisation may reflect intentional reserve capacity or a demand shortage. Pair the ratio with service levels, quality and backlog where those affect the decision. A target should follow the operating model rather than a universal claim that full utilisation is ideal. Check whether rework is counted as productive time and make that choice visible.
The BLS concepts page provides background on distinguishing inputs and output in productivity measurement. The internal utilisation rule below is independently defined and fictional. Keep the numerator, each denominator component and source coverage in the report so a changed percentage can be explained. An improvement caused solely by excluding more unavailable time is a definition or availability movement, not automatically increased delivery.
For cross-team comparisons, verify that mandatory support and training responsibilities are treated consistently. A team assigned more essential non-delivery work should not appear less capable merely because its agreed duties are missing from the denominator explanation.
Fictional delivery team: paid time is not available time
A fictional scenario to explain the method, not a customer result or a market benchmark.
A fictional team has 1,000 paid hours in a month. Its internal rule excludes 100 hours of planned leave and 50 hours of required training from available customer-delivery capacity, leaving 850 hours. Recorded productive delivery is 680 hours, so utilisation is 680 divided by 850, or 80%.
Using paid hours instead would give 68%, a different measure. Excluding another 50 hours of mandatory administration would raise the ratio to 85% without adding a single delivered hour. The analyst therefore shows the full bridge and keeps the administration treatment stable across comparisons. Thirty hours of missing timesheets are reported separately rather than called idle.
These invented values demonstrate denominator choices, not an efficiency judgement about real staff. The team also reviews service quality and an aged backlog before drawing a planning conclusion. The percentage alone cannot show whether the staffing level or utilisation target is appropriate.
TAKE THIS INTO YOUR NEXT CONVERSATION
A practical next-step checklist.
- Define productive work and compatible units.
- Bridge baseline hours to available capacity.
- Prevent overlapping exclusions and show coverage.
- Review quality, bottlenecks and backlog alongside utilisation.
Common questions.
Is billable utilisation the same as productive utilisation?
No. Productive work can be unbilled, and billing rules may include time that does not represent useful delivery.
Does a higher percentage prove more output?
No. A smaller denominator can raise utilisation while delivered work stays unchanged.
Further reading & source context.
These primary references provide background. They are not evidence for the fictional example above and do not endorse Accuralytix.
Found an error or a missing qualification? Send an editorial correction.
Check my data readiness.
Use a practical tool to turn this idea into a clearer starting point.
Check my data readiness