Average Handling Time (AHT)

Average handling time is the mean total duration an agent spends on a customer contact. It includes three components, and leaving any of them out is the most common way the figure gets misreported:

AHT = (Total talk time + Total hold time + Total after-contact work) ÷ Number of contacts

After-contact work — the notes, system updates and follow-up actions the agent completes once the customer has gone — is the part most often excluded, which flatters the number without changing the actual cost of serving.

What AHT is genuinely useful for

Capacity planning. If you know your contact volume and your AHT, you can work out how many agents you need to meet a given service level. That is a real and important use, and it is the reason the metric exists.

Spotting process problems. A sudden rise in AHT for one issue type usually means something upstream broke — a system is slow, a policy changed, a form is confusing. AHT is a good smoke alarm.

Comparing like with like. AHT by issue type, segmented properly, tells you which categories of work are expensive.

Why it is a poor performance target

Set AHT as a target for individual agents and you get exactly what you asked for: shorter contacts. The ways agents achieve that are mostly bad for the business — ending contacts before the issue is finished, transferring work that could have been handled, skipping the notes that the next agent will need, and declining to check the thing that would have prevented a callback.

The result is lower AHT and higher total contact volume, because the work comes back. First call resolution drops, cost per resolved issue rises, and the dashboard looks better.

The sound approach is to treat AHT as a capacity input and a diagnostic, and to set targets on resolution and quality instead. Where AHT genuinely needs to come down, the lever is removing work from the agent — better information, fewer systems, clearer policies — not asking them to go faster.

Segmenting it

A single organisation-wide AHT is close to meaningless, because it averages a password reset with a disputed transaction. Useful cuts:

  • By issue type, which is where process problems show up.
  • By channel — chat, phone and email have structurally different durations.
  • By tenure, which shows how long new agents take to ramp.

Avant One℠ CX records handling time against case type and channel on the same record as the resolution outcome, so AHT can be read alongside whether the issue was actually finished.

Frequently asked questions

What is included in average handling time?

Talk time, hold time and after-contact work. Excluding after-contact work is the most common reporting error.

What is a good AHT?

There is no universal figure — it depends entirely on issue complexity. A simple balance enquiry and a disputed transaction are not comparable, which is why AHT should be read by issue type.

Should AHT be an agent target?

Generally no. Targeting it at individual level reliably damages resolution rates and increases repeat contacts. Use it for capacity planning and diagnostics.

How is AHT different from AWT?

AHT is how long the contact takes. Average wait time is how long the customer waits before it starts.

Resolve more on first contact

Case and complaint management with every channel on one customer record, and OneIQ℠ AI assisting the agent rather than measuring them harder.