Average Handle Time: Formula, Benchmarks, How to Reduce

Average handle time (AHT) is the average time an agent spends working a single customer contact from start to finish. For voice, you calculate it as total talk time plus total hold time plus total after-call work, divided by the number of calls. The 2026 cross-industry average sits near 6 minutes, but a good AHT is entirely relative to the type of contact: a "where is my order" call and a billing dispute should never share one target.

AHT is the efficiency metric support leaders reach for first and misread most often. It looks objective and it is easy to pull from any phone system or help desk, which is exactly why it gets turned into a target and then quietly ruins a support floor. Push agents to hit a low number and they will hit it, by rushing customers off the line, skipping the real fix, and generating the callback that never shows up in the AHT report. This guide covers what AHT actually measures, how to calculate it honestly for voice, chat, and email, what a good number looks like by industry in 2026, how it differs from average talk time, and how to bring it down in the one way that does not cost you quality.

What is average handle time?

Average handle time is the average duration of a complete customer interaction, measured across all the contacts your team handled in a period. For a phone call it covers the whole time an agent is occupied with that customer: the conversation itself, any time the customer spent on hold, and the after-call work the agent does once the call ends, such as logging notes, tagging the ticket, and updating an account. It is a workload and efficiency measure, not a quality measure. AHT tells you how long your team is tied up per contact, which is what you need for staffing and capacity planning. It tells you nothing on its own about whether the customer left happy or whether the problem was actually solved.

That distinction is the whole game. A falling AHT can mean a team is getting genuinely more efficient, with better tools and faster answers. It can equally mean agents are cutting customers short to protect their own average. The number looks identical either way, so AHT is only meaningful when you read it next to a quality metric, never alone.

How do you calculate average handle time?

For voice, the formula is straightforward. Add up everything an agent spends on the contact, then divide by the number of contacts.

AHT = (total talk time + total hold time + total after-call work) / total number of calls.

The piece teams forget is after-call work, sometimes called wrap-up time: the notes, tagging, and account updates a rep completes after hanging up. Leave it out and your AHT looks better than reality and your staffing math breaks, because the agent is not actually free the second the call ends. Include it, and AHT reflects the true time a contact removes an agent from the queue.

A worked example: over a shift an agent logs 180 minutes of talk time, 30 minutes of hold, and 30 minutes of after-call work across 40 calls. That is 240 total minutes divided by 40 calls, or an AHT of 6 minutes.

For chat and email the components change but the idea holds. Live chat AHT is active handling time (including the gaps while an agent works a concurrent chat) plus any wrap-up, divided by chats. Email is trickier, because a ticket can sit for a day between replies; measure the agent's actual working time on the ticket, not the calendar span, or your email AHT becomes a measure of your backlog rather than your effort. In every channel, be explicit about what counts, and keep the definition identical from month to month so the trend means something.

What is a good average handle time?

There is no universal good AHT, because the right number depends entirely on how complex your contacts are. A retail order-status call and an insurance claim are different jobs, and forcing both to the same target is how teams end up gaming the metric. That said, benchmarks give you a sanity check. Across all sectors the 2026 industry average lands around 6 minutes, and it varies widely by industry.

Read these as context, not targets. A SaaS support team handling complex integration questions should have a higher AHT than a retailer answering delivery questions, and forcing the SaaS team down to the retail number would just mean worse answers and more repeat contacts. Use the benchmark to ask a question, not to set a quota: if your AHT is far above the norm for your industry, something in your process or tooling is probably worth investigating; if it is far below, check that quality has not slipped. The useful comparison is almost always your own AHT over time within a single contact type, not your number against someone else's average.

What is the difference between average handle time and average talk time?

Average talk time is only one component of average handle time. Talk time measures the conversation itself, the minutes an agent is actively speaking with the customer. Handle time adds everything else the contact consumes: hold time and after-call work on top of the talk. So AHT is always equal to or greater than talk time, and the gap between them is revealing. A large gap usually means long hold times or heavy wrap-up work, both of which are fixable without touching the customer conversation at all. If your talk time is reasonable but your AHT is high, the problem is not your agents talking too long; it is dead time and paperwork around the call, which is the easiest and safest place to find savings.

Should you try to reduce average handle time?

Sometimes, and carefully. Lowering AHT is worthwhile when the time is being lost to friction that does not help the customer: an agent hunting through five systems for an answer, a clunky after-call form, avoidable hold time while someone finds a supervisor. Cutting that is pure gain, because the customer gets the same outcome faster and the agent handles more contacts. Lowering AHT is actively harmful when you do it by pressuring agents to end interactions sooner, because the time you save on the call comes straight back as a repeat contact, a lower first contact resolution rate, and a worse experience. The honest goal is never "lower AHT." It is "remove the wasted time inside the handle," which lowers AHT as a side effect while quality holds or improves.

How do you reduce average handle time?

Real AHT reduction comes from making it easier for agents to do the work, not from telling them to hurry. These are the levers that shorten handle time without pushing agents to rush, roughly in order of impact.

1. Put answers at the agent's fingertips. The single biggest source of wasted handle time is an agent searching for information mid-contact. A well-organized internal customer service knowledge base that surfaces the right article without a hunt cuts both talk and hold time directly, because the agent is answering instead of looking.

2. Attack after-call work. Wrap-up time is often the quietest 20 to 30 percent of AHT and the easiest to cut. Pre-filled disposition codes, auto-logged call notes, and a help desk that writes the summary for the agent remove minutes per contact that the customer never sees and never benefits from.

3. Route to the right agent the first time. A contact that lands with someone who cannot solve it burns handle time before it is transferred to someone who can. Skills-based routing that matches the customer to a capable agent on the first try removes the most expensive kind of wasted time. Getting this right depends on how well you run your ticketing system and the queues underneath it.

4. Reduce hold time. Hold time is handle time the customer experiences as pure waiting. Give agents the authority and the reference material to resolve common issues without consulting a supervisor, and the holds shrink. Every avoidable hold you remove drops AHT and improves the interaction at the same time, which is the rare win with no tradeoff.

5. Deflect the simplest contacts entirely. Self-service does not just lower ticket volume; it changes the mix of what reaches your agents. When customers resolve the two-minute questions themselves, the contacts left for agents are the substantive ones, and while that can raise average AHT, it does so for the right reason. Judge AHT against the contact mix, not in a vacuum.

6. Coach with call review, not stopwatches. The productive version of coaching is reviewing real interactions to find where time genuinely leaks, a repeated fumble through a tool, a verification step that could be shorter, and fixing that specific thing. The unproductive version is putting a timer on agents, which teaches them to rush and to close contacts prematurely. One improves AHT and quality together; the other trades quality for a number.

Why average handle time is not a quality metric

Average handle time measures effort and cost, not customer outcomes, which is why treating it as a scorecard for agents backfires so reliably. The metric it fights against is first contact resolution, the share of issues solved in a single interaction. Push AHT down in isolation and agents learn to close contacts fast rather than solve them fully, first contact resolution slips, customers call back, and your total workload rises even as each individual handle time looks better. That is the AHT trap: the number improves while the operation gets worse. The fix is to never look at AHT alone. Read it beside first contact resolution and CSAT, and only celebrate a falling AHT when those quality numbers hold or rise with it. A lower AHT next to a lower resolution rate is not efficiency; it is a warning.

Making average handle time a metric you can trust

Average handle time earns its place in capacity planning because it tells you honestly how much agent time each contact consumes, which is what you need to staff a queue. It stops being useful the moment you turn it into an agent target and start rewarding speed for its own sake. Measure it consistently, including after-call work, keep the definition identical across months, and compare your number to your own history within a contact type rather than to someone else's industry average. Then read it the way it should always be read: next to first contact resolution and CSAT, and paired with first response time so responsiveness and effort are both in view. Handled that way, AHT becomes a planning tool that helps you staff and streamline a support floor, rather than a vanity number that quietly makes the customer experience operation worse while looking like progress.