AHT is a budgeting metric, not a coaching metric

Your best rep just got dinged in a one-on-one for an average handle time thirty seconds over target. She's the one who volunteers for the gnarly billing disputes nobody else will touch, and she has just learned that the quickest route to a clean number is to stop volunteering.

That is the whole problem with average handle time in a single scene: a genuinely useful number aimed at exactly the wrong target.

What AHT is actually for

Average handle time is a capacity input. It is the raw material of staffing math, and it is very good at that job.

To answer the only question a scheduler really has — how many people do I need on the floor at 10am on Tuesday to keep the queue moving — you need three things: how many contacts will arrive, how long each one takes to handle, and the service level you have promised to hit. AHT is that middle term.

Multiply a benchmark like that by forecast volume, run it through the queueing math, and you get headcount. The service level it is feeding usually looks something like this:

Answer 80% of contacts inside 20 seconds, and to do that you need enough agents on shift — a number you cannot calculate without knowing how long the average contact ties one up. That is AHT earning its keep. Notice what the calculation never asks: which individual handled which ticket. AHT is a property of the queue, not the person sitting in it.

Why it breaks the moment you point it at a person

An average is only meaningful across a representative mix. One rep's handle time is an average over whatever happened to land in their queue this week, and that is never the same mix as anyone else's.

So when a team-sizing number gets printed on an individual scorecard, it stops measuring the work and starts shaping it, in two predictable and equally unwelcome directions. The first is rushing: the fastest way to cut your handle time is to close before the problem is actually solved. The second is cherry-picking: quietly steering toward the password resets and away from the account migrations. The rep who takes the five-minute tickets looks like a star. The rep who takes the forty-minute ones looks like a problem.

Rushing has a bill, and someone else pays it

Speed pressure does not happen in a vacuum. It collides, first, with resolution.

Closing a ticket before it is truly done does not delete the work — it relocates it to a second contact, and drags first-call resolution down as it goes. AHT and FCR are meant to sit in tension; coach only the first and you quietly sabotage the second. It collides, too, with quality — the thing a stopwatch cannot see.

That is the number that actually opens the ticket and reads it: whether the answer fit, whether the customer felt heard, whether "resolved" was true. None of it shows up in seconds elapsed. And the reason any of this matters sits one step downstream.

A rushed, half-resolved ticket is precisely the bad experience that sends people to a competitor. Shaving ninety seconds off an average is a poor trade for the customer who never comes back.

Use it where it works

None of this makes AHT a bad metric. It makes it a misfiled one. Put it back where it belongs and it is invaluable; leave it on the scorecard and it slowly teaches your best people to do worse work.

Keep it at the aggregate level, for forecasting and staffing — track it, trend it, feed it into the capacity model. Take it off individual targets. If a rep's handle time comes up in a one-on-one at all, treat the high number as a question rather than a verdict: go look for the workflow friction, the missing macro, the knowledge base where answers go to hide. Then coach on the things that survive scrutiny — resolution, quality, and the customer's actual outcome.

AHT tells you how many people to hire. It was never built to tell you whether any one of them is good at the job. If you want that answer, read the real cost of a ticket it feeds into, and judge individuals against the benchmarks that measure resolution and quality — not the clock.