The Job That Sits in Limbo: Killing Your Idle Queue
Most practice bottlenecks aren't busy jobs — they're idle ones. Here's how to find and clear the work that's stopped moving without anyone noticing.
Ask a practice owner where their capacity problem is, and they'll usually point at the busiest desks. But the real drag on turnaround is rarely the work that's actively being done. It's the work that's stopped. The BAS that's ready but not reviewed. The tax return sitting on 'waiting for client' three weeks after the client replied. The job that got started, hit a small snag, and quietly went cold.
This is the idle queue — jobs that are technically open, technically assigned, and moving nowhere. They don't show up as a fire because nobody's actively failing at them. They just sit. And in most firms, they account for far more lost time than any individual's workload ever does.
Idle time is invisible in most systems
Here's the trap with generic to-do lists and even a lot of account practice management software: they're built to show you what's assigned, not what's stalled. A job in a status called 'In Progress' looks identical whether someone touched it this morning or three weeks ago. The status tells you where the job is. It never tells you how long it's been there.
So the owner scans the board, sees jobs sitting in reasonable-looking columns, and moves on. Meanwhile a return that should have gone out in five days has been in 'Awaiting Manager Review' for eleven, because the manager cleared the urgent ones and this one never surfaced again.
The problem isn't effort. It's that idle time has no owner and no alarm.
Every status is a promise about time
The fix starts with reframing what a status means. A column on your job board isn't just a label — it's an implicit commitment about how long work should live there.
- Awaiting client information shouldn't run past your follow-up cadence — say, two chases over ten days.
- Ready for review shouldn't sit longer than a day or two, or your review step is the bottleneck.
- Signed, ready to start should never idle at all — that's the signed-but-not-started gap where engaged work quietly stalls.
- Ready to lodge or ready to invoice should clear same-week, because the work is done and only inertia remains.
Once you attach a rough time expectation to each status, the idle queue becomes measurable. Any job that's exceeded the expected time in its column is, by definition, stuck. Now you're not scanning a board and guessing — you're looking at a list of exceptions.
Where idle jobs actually accumulate
In client accounting work, the same three chokepoints appear again and again.
The 'waiting on client' that's no longer waiting
A job goes into a holding status when you request information. The client replies. But the reply lands in an inbox, not on the job — so the status never changes and nobody knows the blocker has cleared. The job is idle for a reason that no longer exists. This is why a client portal that ties responses back to the work item matters: when the answer arrives against the job, the job stops looking blocked.
The review queue that only one person can clear
Work piles up in front of a single reviewer, and because everything there looks equally 'ready', the reviewer works by memory and urgency rather than by age. The oldest jobs get buried under newer ones. A queue ordered by how long each job has waited fixes this instantly.
The finished job that never gets closed
The work is genuinely done, but the final step — lodge, send, invoice, mark complete — depends on someone remembering. These jobs are pure margin sitting idle: the effort is spent, the value just hasn't been captured. In WIP terms, they're finished work you haven't billed.
Make idle jobs surface themselves
You don't clear an idle queue by working harder. You clear it by making stalled work impossible to miss, then building a small rhythm around it.
Good accounting client management software should let you see, at a glance, not just where a job is but how long it's been there. In Finye, work items live on boards with clear statuses, and because responses come in through the portal and recurring jobs generate on schedule, the system knows when a job last moved. That turns 'which jobs are stuck?' from a question someone has to investigate into a view you can just open.
The workflow around it is simple:
- Run an idle review weekly. Not a full WIP meeting — five minutes on the jobs that have overstayed their status. Every one gets a decision: unblock it, reassign it, chase it, or close it.
- Give every status a single owner. A job in 'ready for review' is the reviewer's problem the moment it lands there. Ambiguous ownership is how a status nobody owns quietly kills your turnaround.
- Treat 'no movement' as the alarm, not 'past deadline'. If you wait for the deadline to react, you've already lost the buffer. A job that hasn't moved in a week is a warning; a job that's late is a failure.
Why this beats hiring for it
When turnaround slips, the instinct is to add capacity. But idle jobs don't consume anyone's hours — they consume calendar time. Adding a person to a firm full of stalled work just gives you more jobs to leave sitting. You're not short on effort; you're short on flow.
Clearing the idle queue is the cheapest capacity you'll ever find. The work is already largely done. Every job you unblock and push out this week is revenue you'd otherwise have recognised weeks later — or a client whose deadline you would otherwise have quietly drifted past.
Start with one column
You don't need to overhaul your whole board. Pick the status where work stacks up most — usually review or 'ready to lodge' — and sort it by how long each job has waited. Clear the oldest. Set a rough time limit for that column. Check it once a week.
Do that for one status and the idle queue stops being invisible. Do it for all of them, and turnaround becomes something you manage on purpose rather than something that happens to you.