MSP Margin Get the Monday Close
A row of servers in a rack with status lights on

Measuring utilization without gaming it

Utilization shows whether your team has room for more clients or is running out of capacity, but only if you define it carefully and measure it from clean time entries.

Cost of service
Guide
4 min read
By MSP Margin Editorial

Utilization is the share of a technician's time that goes to client work. It sounds simple, and it is one of the most commonly misread numbers in an MSP. The confusion usually starts with the definition, because a managed services shop has more than one kind of client work.

Three kinds of time, two useful ratios

Every hour a technician is paid for falls into one of a few buckets:

  • Billable time is work you invoice separately: time and materials, project hours, overages, or hours drawn from a prepaid block.
  • Agreement time is client work covered by a flat monthly fee. It produces no separate invoice, but it is the labor your recurring revenue pays for.
  • Non-billable time is internal: meetings, training, internal projects, admin, and gaps between tasks.

Then there is time that is paid but never logged, which belongs in its own bucket rather than being quietly ignored.

From these you can build two ratios. Billable utilization is billable hours divided by available hours. Productive utilization is billable plus agreement hours, divided by available hours. Available hours means paid hours minus PTO and holidays.

For an MSP that earns most of its revenue from flat-fee agreements, billable utilization alone can look alarmingly low even when the team is fully booked. Productive utilization is usually the better measure of capacity. Billable utilization is still useful for watching project and overage revenue.

A worked example

These numbers are illustrative only, picked to make the arithmetic easy.

A technician has 160 available hours in a month. Their PSA time entries show 30 hours of billable project work, 85 hours on agreement tickets and 25 hours of internal work. The remaining 20 hours have no entry.

  • Billable utilization: 30 / 160 = 18.75%
  • Productive utilization: (30 + 85) / 160 = 71.9%
  • Internal time: 25 / 160 = 15.6%
  • Unaccounted time: 20 / 160 = 12.5%

Check that the buckets add up: 30 + 85 + 25 + 20 = 160 hours. If yours do not add up to available hours, the report is missing something.

The billable figure alone would suggest this person has a lot of spare capacity. The productive figure says otherwise. The unaccounted 20 hours is the line to ask about, because it may be real slack, or it may be agreement work that was done and never logged. The second case makes your agreements look more profitable than they are.

What a realistic target looks like

There is no single correct utilization target, and published benchmarks rarely describe a shop like yours. A realistic target is one you set from your own history, after accounting for the time your business genuinely needs people to spend on internal work.

Targets also differ by role. A help desk technician working a ticket queue has a different pattern from a project engineer, and a service manager who also takes tickets should not be measured on the same scale as either. Set targets by role, and make sure they leave room for training, documentation and the kind of proactive work that prevents tickets later.

Treat sustained productive utilization above your target as a capacity signal. If the whole team runs hot for several months, response times and quality usually suffer before anyone files a complaint, and it is time to think about hiring or repricing.

Measuring from PSA data without gaming

Utilization measured from time entries is only as good as the entries. A few practices keep the data honest:

Measure against paid hours, not logged hours. If the denominator is hours logged, untracked time disappears and everyone looks busy. Use available hours from payroll so the unaccounted bucket shows up.

Watch minimum increments. If your PSA rounds every entry up to a fixed increment, a technician who closes many short tickets will show more time than they spent. That inflates utilization and makes agreements look more expensive than they are. Compare entries with actual ticket durations now and then.

Classify work types consistently. Agreement versus billable depends on the work type and the agreement's coverage settings in the PSA. If those mappings drift, time lands in the wrong bucket. Review the setup when you add a new agreement type.

Be careful tying pay to utilization. When compensation depends heavily on logged hours, people log more hours. Some will pad entries; others will skip internal work that matters. If you use utilization in reviews, pair it with quality measures such as reopened tickets or client feedback.

Run the report monthly, by technician and by team, and look at trends over a quarter rather than reacting to a single month.

This guide is general information for operational planning, not accounting or legal advice. If you use utilization data for compensation decisions or financial reporting, confirm the approach with your CPA and, where relevant, an employment advisor.