Queue psychology: the wait your customer feels, not the one you measure
Two waits, and you only measure one
The planner’s world runs on measured time: average speed of answer, service level, seconds in queue. But the customer doesn’t experience seconds — they experience a feeling, and the two can diverge sharply. A well-handled ninety-second wait can feel shorter than a badly-handled forty-second one. Since the things the business actually cares about — abandonment, satisfaction, whether the customer comes away willing to deal with you again — track the felt wait far more closely than the measured one, a planner who optimises only the seconds is optimising the wrong variable.
This isn’t an argument to stop measuring ASA; it’s an argument to understand what sits between the number you measure and the outcome you want. Decades of queueing-psychology research — much of it traceable to the work of David Maister on “the psychology of waiting lines” — converge on a few robust findings, and each one has a planning implication.
What makes a wait feel worse
Four effects do most of the work. Uncertain waits feel longer than known ones — not knowing how long you’ll be holding is more stressful than a longer wait you were told about up front. Unexplained waits feel longer than explained ones — “we’re busier than usual after a system issue” buys patience that silence does not. Anxious or unoccupied waits feel longer — dead air drags; a sense of progress or useful information passes. And unfair waits feel far longer — the customer who suspects others are being answered ahead of them for no good reason abandons fast and remembers it. None of these change the seconds on your dashboard, but all of them change the abandonment curve and the satisfaction score the seconds are supposed to predict.
The levers, and what they cost the plan
The good news for the planner is that several of these levers are cheap relative to staffing. Telling customers their expected wait or position turns an uncertain wait into a known one and flattens the early-abandonment cliff. An honest reason during a spike buys real patience. A callback offer — “keep your place, we’ll ring you back” — is the strongest lever of all, because it removes the wait entirely from the customer’s experience while letting you serve the contact later, in a trough, smoothing the very peak that caused the queue. That last point makes callbacks a genuine capacity tool, not just an experience nicety: they move work from the moment of highest demand to a moment of spare capacity, which is exactly what a planner spends all day trying to do.
There is a discipline to it, though. Expected-wait messaging that is consistently wrong is worse than none, because it destroys the trust the message depends on. A callback you don’t honour on time is a new failure, generating a chase. So these levers belong in the plan, sized and resourced — the callbacks still have to be staffed, in the trough you promised them into — not bolted on as a real-time reflex.
What it changes for the planner
Keep measuring ASA and service level — you need them, and they remain the engine of the staffing maths. But hold them loosely as proxies for the thing you actually care about, and watch abandonment and customer-reported experience as the truth-tests. When a spike is coming that you can’t fully staff, reach for the felt-wait levers — set expectations, explain, offer the callback — because they buy patience and shift load in ways that more agents, often, cannot do in time. The planner who understands that the customer is timing a feeling, not a clock, has a whole second set of levers the ASA-only planner never picks up.
Pair this with abandonment and caller patience, is service level a dead KPI?, and the real-time lever menu.