Queue Time (Wait Time)
The time a contact spends waiting in line after it has been routed to a queue but before an agent picks it up — reported on voice as Average Speed of Answer.
Average queue time = total time contacts spent waiting in queue / number of answered contacts (reported on voice as Average Speed of Answer)Queue time — also called wait time — is the slice of the customer's wait that sits entirely inside the queue: the clock starts when the contact is routed and parked, and stops the instant an agent connects. On voice it is almost always reported as Average Speed of Answer (ASA); in chat and messaging it shows up as time-to-first-agent. Whatever the channel, it is a pure read on one thing — whether you have enough people online to match the demand arriving right now. It is distinct from handle time, which is the clock that runs while the customer is actually being helped.
The industry shorthand for "good enough" is the 80/20 service level: 80% of contacts answered within 20 seconds. It is worth knowing that the 20-second figure came from balancing staffing cost against caller patience in the early telephony days, not from any study of what customers will actually tolerate — it spread because vendors shipped it as a default setting. Treat it as an operational convention, not a law of human patience.
The trap is that queue time is almost always reported as an average, and averages here are actively misleading. ASA blends the caller who connected instantly with the one who waited eleven minutes into a single reassuring number, so a comfortable 25-second mean can hide a long, angry tail. Report the distribution — or better, the service level (the share answered inside your threshold) — and the people you are failing stop disappearing into the arithmetic.
What it hides: queue time counts only the wait inside the queue, so everything the customer endured before it — IVR menus, a bot that couldn't help, hold music during a transfer — is invisible even though the customer counts all of it as waiting. It also counts only answered contacts: everyone who abandoned in queue drops out of the denominator, which means cutting staff can improve your average queue time while service actually gets worse. And a callback that swaps hold time for a delayed ring-back can zero out queue time on paper while the customer waits exactly as long.