How to measure queue age and find work that is stuck

01

Define the queue and its clock

Choose one waiting point and define its entry and exit events. For example, an item enters when a complete request is accepted for review and exits when review starts. Decide whether the clock uses elapsed or business hours, how pauses are recorded, and who can change a timestamp.

Capture an item ID, entry time, current owner, priority, reason for any hold and next action. Keep returned items linked to the original request, but distinguish a new wait from the earlier one. Check a sample against the underlying records before trusting the report.

02

Read age alongside the count

At the same time each day, calculate the age of each open item from its entry time. Show the oldest item, a median and age bands chosen for your own commitments. Separate urgent, standard and complex work when their expected handling differs.

Hypothetical example: five requests have waited 1, 2, 2, 3 and 12 business days. The median is 2 days, while the oldest has waited 12. The median describes the middle item, but it does not explain why the oldest request is still open. A daily snapshot also excludes items completed before the snapshot; review those separately.

03

Give old items a route to resolution

Review items beyond the team's chosen age threshold with the people responsible. Check whether each is waiting for missing information, a decision, specialist capacity or a downstream release. Assign an owner and a dated next action, including a route for urgent exceptions.

Compare queue age, arrivals and completions over several cycles. A growing queue may mean arrivals exceed service capacity, but a blocked downstream step or a change in work mix can produce a similar picture. Investigate the cause before changing staffing or priority rules.

For an item on hold, record who is expected to act and the date to follow up. If a customer must supply information, keep the hold visible even if your internal service clock pauses. Repeated pauses can reveal an intake problem. Do not reset an old item’s original arrival date simply because its owner or priority changes.

At each review, choose whether to complete the item, request information, escalate a decision or close it under an agreed rule. Check closed items for premature closure. If a threshold produces more alerts than the team can investigate, review the reason and adjust the process or threshold with the owner of the service commitment.

04

Keep both dates for returned work

Keep two dates for returned work: the original request date and the date it re-entered the current queue. The elapsed time from the first date shows age since the recorded request, including any holds. The second shows the latest wait at this step. Earlier customer waiting may not be captured. If the queue has multiple service standards, report each class separately instead of applying one age threshold to all items.