Queue time and work time
Take a claim that took sixty days to resolve and add up the minutes anyone actually spent on it. Three touches, perhaps twenty minutes in total. The rest of those sixty days is queue time: the claim was workable and nothing was happening to it.
That ratio is worth calculating on real claims of your own, because the two halves respond to completely different interventions and most improvement effort goes to the smaller one.
What queue time is made of
Queue time is not one thing. It is at least four:
- Waiting for the worklist to reach it. The claim is workable and sitting behind others.
- Waiting for the payer. A response is outstanding and nothing can be done until it arrives.
- Waiting for the practice. A record, a correction or an authorisation is needed from the client.
- Waiting for the next available moment. The claim was picked up, partly worked and set down, because the next step could not be taken yet.
Only the second is genuinely outside your control, and even that one has a known distribution per payer. The other three are yours.
Why work time gets all the attention
Work time is visible. It has a person attached to it, it appears in productivity reporting, and it is what a manager watches. Queue time has nobody attached to it, appears on no report, and is measured by no standard system.
So improvement effort concentrates where the measurement is: templates, macros, shorter notes, faster screens. All of that is worth doing, and all of it competes for the smaller half.
The clock does not stop
Queue time is not free merely because nobody is being paid during it. The filing deadline runs. The appeal window runs. And the longer a claim sits between touches, the more context the next person has to rebuild, because whoever worked it last has forgotten.
That last effect couples the two halves: queue time inflates work time. A claim picked up twice in one day costs less on the second touch than the same claim picked up twice three weeks apart, even though the work is identical. Anything that shortens the interval between touches lowers the cost of the touch itself.
The shape of the fix is different
Adding capacity shortens the first kind of queue in proportion to how much you add. Double the people working a queue and the wait roughly halves. That is a real improvement, and a linear one.
Removing a queue is a different shape of change. Work that starts when its trigger fires — a remittance posts, a deadline approaches, a state changes — has no wait at all, and the improvement is not proportional to anything. It is the difference between a shorter line and no line.
Most operations have work in the first category that could be in the second. The way to find it is to ask which steps require a decision. A step that requires judgement needs a person and will queue. A step that requires only retrieval or a rule does not.
Measure it before arguing about it
Two timestamps make queue time visible:
- when the claim became workable — the remittance posted, the correction returned, the hold lifted
- when someone first touched it after that
The difference is the first kind of queue time, and it is the one most often assumed to be small. Sum it across a week and compare it against the recorded work time for the same claims. Whichever number is larger is where the next improvement belongs, and it is worth finding out rather than assuming.