Occupancy vs. Utilization: What Each Metric Tells You
Occupancy and utilization both describe how busy your agents are, but they divide by different time. Occupancy asks: of the time an agent was logged in and available, how much went to contacts? Utilization asks: of the time you paid for, how much went to productive work? Same agent, same shift, two different percentages, and two different decisions.
This guide covers the formulas, a worked example, realistic target ranges, the mistakes that make these numbers lie, and how to use each metric on a real floor. It is written for operations managers running contact centers, insurance agencies, healthcare intake desks and sales teams on phones.
TL;DR
- Occupancy = contact-handling time ÷ logged-in available time. It is a real-time, intraday metric for staffing and routing.
- Utilization = productive time ÷ total paid time. It is a planning metric for budgets and capacity.
- The denominator is the whole difference. Shrinkage (breaks, training, absence) hits utilization but not occupancy.
- Sustained occupancy above roughly 90% is a burnout signal, not a win. Read it next to first call resolution and attrition.
- Never compare numbers across teams until you confirm both use the same formula.
Table of Contents
- The short definitions
- The formulas side by side
- Why a high occupancy can sit next to a modest utilization
- What good looks like
- Occupancy is not a score to maximize
- Utilization is not a score to maximize either
- When to use which metric
- Five mistakes that make the numbers lie
- Making the metrics usable in your tools
- Our perspective: define it, then watch the pair
- How RevRing fits
- A simple monthly routine
- FAQ
- Sources
The short definitions
Occupancy is the percentage of an agent’s logged-in time spent on contact-related activity. Call Centre Helper counts talk time, hold and after-call work (ACW) as contact-related. Waiting for the next call is the part that is left over.
Utilization is the share of an agent’s total paid or scheduled time that goes to productive work. WFM Labs frames it as how much of the payroll budget turns into available service capacity.
One is about behavior under load. The other is about scheduling efficiency. That is why the two get treated as different tools by workforce management specialists, and why mixing them up causes bad staffing calls.
The formulas side by side
| Occupancy | Utilization | |
|---|---|---|
| Numerator | Talk + hold + ACW (handle time) | Productive time (handle time at minimum, sometimes plus coaching or training) |
| Denominator | Logged-in available time | Total paid or scheduled time |
| Shrinkage included? | No | Yes |
| Time horizon | Intraday, real time | Weekly, monthly, planning cycles |
| Main lever | Staffing level and routing | Shrinkage control and scheduling |
| Main risk when too high | Burnout, attrition, lower FCR | Little room for training, coaching and absence |
The technical split is the denominator. WFM Labs describes occupancy as using only logged-in available time, and utilization as using total paid time, including breaks, training and shrinkage.
A worked example
Take one agent on an 8-hour paid shift: 480 minutes.
- Breaks, team meetings, a coaching session and a system outage take 144 minutes. The agent is logged in and available for 336 minutes.
- Of those 336 minutes, 285 are talk, hold and ACW. The other 51 are waiting for work.
- Occupancy = 285 ÷ 336 = 84.8%.
- Utilization (handle time ÷ paid time) = 285 ÷ 480 = 59.4%.
Now count 30 minutes of the training as productive work, as many WFM teams do. Utilization becomes 315 ÷ 480 = 65.6%. Occupancy does not move at all.
The agent looks stretched on one metric and underused on the other. Both are true. The first says the floor is busy when agents are on. The second says a big slice of paid time is not available for customers.
Why a high occupancy can sit next to a modest utilization
WFM Labs gives the same pattern: an agent can show 85% occupancy and 70% utilization when a significant part of the shift is lost to shrinkage (source). Shrinkage is the paid time agents are unavailable: breaks, vacation, sick days, training, coaching. The same glossary puts typical shrinkage at 30 to 38% of paid hours, with a common utilization target of 65 to 75% (WFM Labs glossary).
That has a practical meaning. If you cut shrinkage, utilization rises and you get more capacity for the same payroll. If you push occupancy up, you get more handled contacts per logged-in hour, but you pay for it in recovery time.
What good looks like
Targets depend on channel, call length and contact volume. Treat these as starting points, not promises.
- Occupancy: Many operations aim for the high 70s to mid 80s. Call Centre Helper cites a commonly used warning ceiling of 85 to 90%, and sustained levels above 90% are tied to burnout, attrition and weaker first call resolution (Call Force benchmarks).
- Utilization: 65 to 75% is a typical target, which fits 30 to 38% shrinkage (WFM Labs).
Small teams behave differently. With a handful of agents, the queue is spiky, so the same average staffing produces wider swings in occupancy. Large pools can run hotter safely because the load smooths out. If you run a 10-seat insurance desk, do not copy a 500-seat benchmark.
Occupancy is not a score to maximize
The easy mistake is reading higher occupancy as better. Zero idle time means zero recovery between contacts. Agents who run flat out all day take worse calls by afternoon, and the cost shows up later as turnover.
There is a second trap. Workforce management author Ger Koole points out that metrics are linked: high occupancy means little if first call resolution is low. When issues are not resolved, customers call back, which adds workload and lifts occupancy. The metric improves while service gets worse.
So never read occupancy alone. Pair it with:
- First call resolution
- Average handle time and ACW trend
- Abandonment rate and service level
- Attrition and absence by team
If you run outbound or blended queues, abandonment has a regulatory side too. Our U.S. checklist on the 3% abandoned-call ceiling covers how dialer pacing and occupancy interact.
Utilization is not a score to maximize either
A utilization figure near 100% means no time for training, coaching, meetings or normal absence. Those do not disappear. They get squeezed out, or they show up as unplanned shrinkage. A plan that assumes 90% utilization when your real shrinkage is 33% will leave you chronically short.
The WFM Labs write-up makes the related point: using occupancy as the main capacity-planning indicator leads to chronic understaffing because it ignores unavoidable shrinkage (source). Plan with utilization and shrinkage. Steer the day with occupancy.
When to use which metric
Use occupancy for intraday decisions
- Moving agents between queues or channels
- Deciding when to hold offline work like email or callbacks
- Tuning dialer pacing and routing rules during the shift
- Spotting a queue where agents are running hot or sitting idle
A real-time view matters here. Teams that route by skill or geography need live occupancy by group, not a monthly average. Our six-step rollout for skills-based routing shows how to set that up in 4 to 6 weeks.
Use utilization for planning decisions
- Forecasting headcount and hiring lead times
- Budgeting payroll against expected volume
- Judging whether shrinkage is eating your capacity
- Comparing scheduling efficiency across teams, once formulas match
For a fuller staffing model that ties these together, see our guide on an AI-backed call center staffing model. Healthcare teams can also use a 60-day test of staffing decisions.
Five mistakes that make the numbers lie
- Not writing the formula down. Call Centre Helper warns that comparing utilization figures without defining the formula is the most common error. A 70% in one center can equal 85% in another if one counts vacation and holidays in the denominator and the other does not (source).
- Treating breaks inconsistently. The same 80% occupancy means different things if breaks are shrinkage in one report and available time in another.
- Using one team’s target on another. Inbound insurance quoting and outbound appointment setting have different handle times and idle patterns.
- Reading daily averages only. An 80% daily occupancy can hide a 98% morning peak and a 55% afternoon trough.
- Ignoring adherence. If agents are not in the state the schedule expects, both metrics drift. Tracking adherence for regulated teams closes that gap.
Making the metrics usable in your tools
None of this works if the data sits in three systems. Occupancy needs agent states and handle time per interval. Utilization needs those states plus the schedule and paid hours. You need a dashboard that shows both by team and by hour, with state definitions that match your formula. If your phone, CRM and reporting tools are patched together, it is worth reviewing hybrid work technology priorities before adding another system.
Clean call logging also matters. If dispositions and wrap-up times are inconsistent, ACW is wrong, and occupancy is wrong with it. Our note on CRM call logging fields covers the five fields worth getting right.
Our perspective: define it, then watch the pair
We see the same pattern in the regulated teams we work with, in insurance, healthcare, legal and real estate. Managers pick one number, usually occupancy, and manage everything to it. Then attrition climbs, or the schedule never has room for coaching, and nobody can say why.
Our view is simple:
- Write the formula into your reporting. Name the numerator, the denominator and every state that counts.
- Run occupancy as a band, not a target. Set a floor and a ceiling, and treat breaches as a signal to rebalance, not to push harder.
- Run utilization with shrinkage. Forecast shrinkage by category so training and compliance time are planned, not stolen.
- Watch outcomes beside both. Speed-to-contact, resolution and compliance tell you if the efficiency is real. On sales floors, speed-to-lead often matters more than a tidy utilization figure.
- Let the dialer work for the metric. Pacing, routing and call scoring should remove idle time without squeezing recovery time out of the day.
The goal is infrastructure that scales: the same definitions at 12 agents and at 180.
— Marc
How RevRing fits
RevRing connects dialing, CRM, routing, AI call scoring, workforce management and reporting in one platform, with compliance built in for TCPA, DNC and HIPAA BAA needs. That means agent state, handle time and call outcomes live in the same place, so you can review occupancy and utilization from consistent data. If you run predictive dialing, read our guide on keeping abandon rate under 2%. Plans start from 39.99 USD per month per seat on Starter. See RevRing pricing or how it works, then schedule your live demo.
A simple monthly routine
- Week 1: Document both formulas and agent states. Get sign-off from ops and finance.
- Week 2: Pull occupancy by interval and queue for 30 days. Mark the hours above your ceiling and below your floor.
- Week 3: Break shrinkage into categories. Compare planned to actual.
- Week 4: Adjust schedules and routing. Re-check FCR, abandonment and attrition before declaring a win.
Repeat monthly. Change one lever at a time, or you will not know what worked.
FAQ
What is the difference between occupancy and utilization in a call center?
Occupancy divides contact-handling time by logged-in available time. Utilization divides productive time by total paid time. Because utilization counts shrinkage, it is almost always lower than occupancy for the same agent.
What is a good occupancy rate?
Many centers aim for the high 70s to mid 80s. Benchmarks cite 85 to 90% as a warning ceiling, with sustained levels above 90% linked to burnout and attrition (Call Force). Smaller teams usually need a lower target because their volume is spikier.
What does 75% utilization mean?
It means about three-quarters of paid time went to productive work under your definition. The rest was breaks, absence, meetings or idle time. Check how your report defines the denominator before judging it, since 75% on total paid hours is a different result than 75% on scheduled hours.
How do I calculate agent utilization?
Divide productive time by total paid time for the period, then multiply by 100. Decide upfront whether training and coaching count as productive. Keep that choice the same in every report.