Most support dashboards lead with the size of the queue, and size on its own is close to meaningless. A busy team clears 300 open conversations by Friday. A quiet team can sit at 40 open with one of them untouched since the 2nd, and the second team has a problem the first does not.
Measure the oldest untouched conversation
The number worth putting on the board is the age of the oldest conversation with no agent reply since the customer’s last message. Oldest open is the wrong filter, because it counts threads where you are waiting on the customer, and those are not your backlog. Once you exclude them, most teams find their real oldest item somewhere they were not looking: a queue with no owner, or a channel somebody muted in March and nobody unmuted.
Age buckets beat an average
Report the count in four buckets — under 4 hours, 4 to 24 hours, 1 to 3 days, over 3 days. An average hides the shape completely: 200 fresh tickets and 5 ancient ones average out to something reassuring, and the 5 are the ones generating the complaint that reaches your director. The over-3-days bucket should be small enough that you can name every conversation in it. If you cannot name them, it is too big.
Why old tickets get older
Age compounds. A 9-day-old conversation is harder to pick up than a 1-day-old one, because whoever opens it has to read 14 messages before they can say anything. So it gets skipped, and a week later it is 16 days old and needs 20 minutes nobody has. Tomasz Wierzbicki handled this with a standing rule: everything over 5 days goes to one named person for that day, cleared or closed with an explanation, and the queue stopped growing a tail.
Where to put it
One line, visible without clicking anything: the age of the oldest untouched conversation, and the count over 3 days. Two numbers. The rest of the queue dashboard can live behind a link, because those two are the ones that make somebody do something before lunch.
Leave a Reply