How long a conversation takes to work — which is not the same as how long it takes to answer.
Average handle time. Average handle time (AHT) is the average amount of working time a support conversation consumes from pickup to wrap-up. In call centres, average handle time is talk time plus hold time plus after-call work, divided by calls handled. In email and chat support, it is the active time agents spend per conversation.
AHT was born in the phone world, where the boundaries are crisp: a call starts, the customer may go on hold, the call ends, the agent writes it up. Sum those three phases across all calls, divide by the call count, and you have the metric.
Asynchronous channels break that crispness, and this is the part teams get wrong. An email thread can stay open for three days and contain six minutes of actual work; its elapsed time says nothing about its cost. Handle time on async channels means active time — reading, researching, writing — summed across every touch of the conversation, by every agent who touched it. If your tool only records elapsed time, you are looking at resolution time wearing AHT’s name badge.
AHT is the unit cost of support. Multiply it by conversation volume and you get the team’s real workload in hours — which is how you forecast hiring honestly instead of by feel, and how you spot that one topic quietly eating a disproportionate share of the week.
It is also the most dangerous metric to set as a target. Agents graded on handle time learn to end conversations rather than finish them, and the cost reappears downstream as repeat contacts and falling first contact resolution. Use AHT for planning and diagnosis; grade quality with satisfaction and resolution metrics instead.
AHT formula. AHT = (total talk time + total hold time + total after-call work) ÷ number of conversations handled. For email and chat, substitute the sum of active working time across all touches of each conversation. An AHT of 8 minutes across 900 monthly conversations is 120 hours of handling work.
Segment before you react: an average blends thirty-second password resets with hour-long refund disputes into one meaningless middle. AHT per topic or per intent is where the signal lives — a rising AHT on one topic usually means a process or documentation gap, not slower people. And when drafting tools shorten the writing phase, verify the saved minutes show up as capacity, not just as a nicer chart.
Give your customers faster answers and your team their evenings back.
Desktop and mobile. Free for solo.