Average Handle Rate
Definition
Average Handle Rate
Average handle rate is a term with two rival meanings. Some teams use it as a loose name for average handle time, the seconds a contact takes. Others use it as a true rate: contacts handled per agent hour. Check the units first.
The two readings aren’t compatible. One counts duration, the other counts throughput. The standard named metrics are average handle time and contacts per hour, so say which one you mean before anyone builds a target on it.
Here’s the quick test. If the number carries units of seconds or minutes, you’re looking at handle time. If it carries units of contacts per agent hour, you’re looking at a rate — and the rate reading is the one that feeds capacity planning.
Key takeaways
- Average handle rate is not a settled industry metric; two different teams will define it two different ways.
- Units decide the reading: seconds or minutes means handle time, contacts per agent hour means throughput.
- The two are reciprocals of each other, scaled to an hour, but they never move in a perfectly straight line.
- Occupancy and after-call work sit between them, which is why one figure can improve while the other stalls.
How it works
Average handle rate resolves into one of two calculations. Divide total handling seconds by contacts and you get handle time. Divide contacts by paid agent hours and you get throughput. Each one is the reciprocal of the other, scaled to an hour.
The reciprocal part is easy to picture. A six-minute handle time implies ten contacts in an hour of pure talk and wrap. Trim thirty seconds and the hourly figure rises.
It rises unevenly, though. Cutting duration at the short end of the distribution barely shifts the hourly count, while cutting it on the long tail moves the number sharply.
| Reading | Units | Formula | Best used for |
|---|---|---|---|
| Handle time | seconds or minutes | total handle seconds ÷ contacts handled | queue design, coaching, routing rules |
| Handled rate | contacts per agent hour | contacts handled ÷ agent hours | capacity planning, staffing, pricing |
| Adjusted rate | contacts per available hour | contacts handled ÷ productive hours | floor throughput, shrinkage analysis |
| Unlabelled figure | unstated | unclear | nothing, until someone labels it |
Two variables sit between duration and throughput, and both of them are yours to manage.
The first is occupancy rate. It decides how much of a paid hour an agent spends on contacts at all, rather than waiting in an idle queue for volume that never arrives.
The second is after-call work. It decides how much of the handled time never touches the customer, because notes, disposition codes and system updates all land inside the same clock.
Paid hours and available hours also part ways, and the gap is a legal one — not a scheduling preference.
Under 29 CFR 785.18, short rest breaks running from five minutes to about twenty are customarily paid, and they must be counted as hours worked.
So a throughput figure built on paid hours reads lower than one built on live, plugged-in time. Neither version is wrong. They answer different questions, and mixing them quietly wrecks the comparison.
The raw counter beneath both readings is calls handled. A rate is only ever as clean as the contact definition underneath it, so agree on what counts as one contact before you divide by anything.
That last point catches more teams than the arithmetic does. Transfers, callbacks and reopened tickets can be counted once, twice, or not at all — and each choice moves the rate without anyone touching performance.
Reporting cadence matters too. A rate averaged across a full shift hides the peaks, while the same rate cut into half-hour intervals shows where the queue and the roster stopped matching.
None of this makes either number the better one. Handle time answers a design question about the contact; the handled rate answers a money question about the hour, so pick the one matching your decision.
Examples
Four situations show the ambiguity doing real damage, and each one is fixable with a label rather than a project. Watch what the units say, then watch what the person reading the dashboard assumes they say.
The mislabelled dashboard. A client asks for average handle rate and gets a tile reading seven minutes and twelve seconds. Those are units of time, so it’s handle time wearing a rate’s name.
Every target built on that tile inherits the confusion. Ask for a 10 percent improvement and half the room hears “faster contacts” while the other half hears “more contacts per hour”.
The costing model. Throughput per agent hour turns straight into money once you attach a wage. The US Bureau of Labor Statistics reported a median hourly wage of $20.59 for customer service representatives in May 2024.
That figure only converts cleanly when the denominator is agent hours. Feed a handle-time number into the same model and cost per contact comes out inverted, a mistake that survives review because the output still looks plausible.
The staffing forecast. The same Bureau of Labor Statistics outlook projects employment for these roles to decline 5 percent from 2024 to 2034, with about 341,700 openings each year on average as workers leave the occupation.
A shrinking base with heavy churn makes throughput per hour the more useful planning number. You can’t hire your way out of every volume spike, so the rate you get from each remaining hour carries more weight.
The public-sector service target. The US federal customer experience programme, published at Performance.gov, sets service delivery expectations across government agencies.
Executive Order 14058, signed in 2021, established accountability through High Impact Service Providers and directed 17 agencies to take 36 specific actions. OMB Circular A-11, Part 6, Section 280 carries the annual guidance.
Programmes at that scale report duration and throughput separately. A faster contact and a higher hourly count are not the same claim about service, and conflating them hides which one actually moved.
Related terms
These five entries surround average handle rate on both sides of the reciprocal. Read them together and the disambiguation gets much easier, because each one names a piece of the calculation that the ambiguous phrase leaves unstated.
- Average Handle Time (AHT): the duration reading, measured in seconds per contact.
- Contacts Per Hour: the throughput reading, measured in contacts per agent hour.
- Calls Per Agent Per Hour: the voice-only version of the same throughput measure.
- After-Call Work: the post-contact admin time folded inside the handled clock.
- Workforce Management: the planning discipline that turns either figure into a roster.
FAQ
Is average handle rate a real metric?
Not a standardised one. It’s a colloquial phrase pointing at either average handle time or contacts per hour, and reporting tools rarely define it the same way twice.
How do I tell which reading someone means?
Look at the units. Seconds or minutes means handle time; contacts per agent hour means a rate.
Are handle time and handled rate just inverses?
Roughly, yes — one hour divided by handle time gives you contacts per hour. The relationship loosens once occupancy, paid breaks and after-call work take their share of the hour.
Which version should I use for capacity planning?
Use the rate. Planning works in agent hours, so a figure already expressed per hour drops into a roster without conversion. Handle time still matters, but it needs converting first.
Can both figures move in the same direction?
Yes, because when idle time falls, throughput can climb while handle time holds steady.
Source partners who want floor throughput measured on their own terms can start with the Outsource Accelerator BPO hubs.
Independent
