Operations

The Invisible Queues Running Your Business

Work in your business spends most of its life waiting — in inboxes, approval stacks, and someone's mental to-do list. Those waits are queues, and unmanaged queues set your delivery speed.

The Lobbi Delivery Team
July 6, 20264 min read

The Lobbi Delivery Team

Operational Systems Engineering

Time one job through your business — any job. Stamp the moment it arrives and the moment it ships. Then add up the minutes anyone actually touched it.

The gap between those two numbers is the most important operational fact about your business, and almost nobody measures it. A 4-hour job delivered in 11 days means the work spent over 95 percent of its life waiting. Not being processed. Not being improved. Waiting.

Where it waits is in queues you cannot see, because nobody built them on purpose.

Where the Invisible Queues Live

A queue is anywhere work sits between effort. Once you adopt that definition, you find them everywhere.

  • The shared inbox is a queue with no order discipline — newest on top, oldest buried.
  • The approval request in a manager's email is a queue item competing with 200 unrelated messages.
  • The handoff between two people is a queue where work waits to be noticed, because nothing announces its arrival.
  • The personal to-do list is a private queue, invisible to everyone else, ordered by memory and mood.
  • The weekly meeting is a scheduled queue: decisions wait up to seven days for their slot.

None of these look like operational infrastructure. All of them set your delivery speed. The work time in your process is bounded and fairly stable. The wait time is unbounded — it grows silently with load, and nobody is accountable for it because nobody can see it.

The Arithmetic of Waiting

Queueing has real math behind it, and one result matters for every operator: as utilization climbs toward 100 percent, wait times do not grow linearly — they explode. A resource that is 95 percent busy generates dramatically longer queues than one at 80 percent. This is why your busiest people are your longest queues, and why stacking more work on the most reliable person makes the whole system slower.

The second result that matters: queue length predicts delivery time. If 40 jobs are open and you complete 4 a day, the average new job is looking at roughly 10 days regardless of how urgent anyone feels. Cutting the number of open jobs is not cosmetic — it is the most direct lever on how fast anything ships.

What you seeWhat it actually isFirst measurement
Full shared inboxUnordered queueAge of oldest unopened item
Pending approvalsQueue inside emailAverage approval wait in days
Busy team memberOverloaded server, growing queueOpen items assigned to them
Everything in progressExcess work-in-progressCount of open jobs vs weekly completions

Three Moves That Shrink the Wait

You do not need new software to act on this. You need three operating decisions.

  1. Make queues visible. Every wait point gets a place where its contents and ages can be seen — a board, a list, a view. The measure that matters is item age, not item count. An inbox with 30 items all under a day old is healthy. One with 8 items and a two-week-old at the bottom is not.
  2. Cap work in progress. Decide how many jobs can be open at once, and finish before starting. This feels wrong to every team that tries it and shortens delivery time for every team that sticks with it. Starting work early does not make it finish early; it makes everything finish late.
  3. Set queue rules. Each queue gets an order discipline (oldest first, or an explicit priority tier) and a service expectation (opened within 4 business hours, approved within 1 business day). Now a queue can be in breach, which means someone can notice and act.

The Approval Queue Deserves Special Attention

Of all the invisible queues, approvals are the most expensive per item, because an approval usually blocks everything behind it. A job waiting on a 30-second yes is a job whose entire remaining process is frozen.

Two fixes carry most of the weight. First, pre-authorize: define the conditions under which approval is automatic — under a money threshold, matching a standard pattern — and reserve human review for genuine judgment. Second, give approvals their own channel with its own service expectation, instead of letting them swim in general email. A dedicated approvals view with a one-business-day rule routinely removes days of latency from a process without anyone working a minute more.

The work in your business is probably fine. The waiting is the problem — and waiting is a design choice once you can see it.

You do not need people to work faster. You need work to wait less. In most operations, the wait is 90 percent of the calendar.

Sources

Topic clusters

Ready to see where the friction is?

The Lobbi's Operations Discovery maps your workflows, identifies your highest-impact bottlenecks, and gives you a clear picture of what's possible.

← All insights