Pick work stalls when every order is assigned to a named person, because an absence freezes that person's whole queue while the rest of the floor waits for something to do. The fix is to offer picks to a role instead, so whichever picker is free takes the next one and nothing sits behind a sick call.

Why does one absent picker stall a whole warehouse shift?

Because named assignment turns a person into a single point of failure. When a supervisor hands out orders at the start of the shift, each picker owns a fixed list. If one picker is out, that list does not move until somebody notices, reads it and reassigns it by hand.

The cost is rarely visible on the shop floor. It shows up later as orders that miss the courier pickup, a rise in "where is my order" calls and a pack bench that goes quiet at 11 am and then floods at 3 pm. Owners often read this as slow staff. It is usually a queue shape problem.

If you already batch orders into pick runs, the problem gets sharper, not softer. A batch assigned to one absent person strands many orders at once. Our guide on batch picking and walking time covers how to build the runs. This article covers who gets to take them.

What is a pooled pick queue?

A pooled pick queue offers each pick to a role, such as "picker", instead of to one named person. Anyone in that role can see the open picks, take one, and put it back if they cannot finish it. The task then shows who took it.

This is how Commmerce handles work the app raises: a pick or a stock count is offered to the role, and whichever picker is free takes it. There is no list to rebuild when someone calls in sick, because nobody owned the list in the first place.

Three small behaviours make the pool work in practice:

How to move from named assignment to a pooled pick queue

You can change over in a week without retraining the floor. The steps below assume you already generate pick lists from orders.

  1. Write down how picks are assigned today: by person, by zone, or by order source. Note who does the assigning and how long it takes each morning.
  2. List the three or four tasks that genuinely need a specific person, such as handling high-value jewellery or serial-numbered electronics, and keep only those restricted.
  3. Define a picker role and add every person who walks the floor to it, including temporary staff on their first day.
  4. Switch pick tasks from a named person to the picker role, then let the team take picks from the open list for one full shift while the supervisor watches the queue instead of editing it.
  5. Review the end-of-day board: how many picks were taken, how many were put back, and which orders crossed the courier cut-off.
  6. Tune the order cut-off and batch size using that review, not a guess.

Pair this with a sensible order cut-off time. A pool shortens the queue, but the cut-off decides which orders enter it.

What happens when two pickers take the same pick?

The earliest claim wins, whatever order the claims reach the server in. That is the rule people expect, because it is the one they will each remember: whoever tapped first got it.

It matters because warehouse wifi is patchy. In Commmerce, a picker can claim a task with no internet, carry on with it, and have the claim confirmed when the connection returns. If somebody else had claimed it earlier, the person who lost is told by name at that point, instead of discovering it at the shelf.

Losing a claim you have already started is the unavoidable cost of working offline. Ordering claims by when somebody tapped, and not by when their phone reconnected, keeps it rare. Re-claiming a task you already hold simply succeeds, which is what a retry after a dropped connection looks like.

How should you staff pick work in the Diwali rush?

Add temporary pickers to the pool rather than to a supervisor's list, and keep the supervisor's job to watching the queue. Festive weeks bring more orders and more disruption together, and a pool absorbs both without any reshuffling.

Retail demand in India peaks around the festive season, a pattern industry bodies such as the Retailers Association of India track every year, and the India Brand Equity Foundation publishes sector context on how large and fast moving the market has become. Neither tells you how to run your own floor, but both are a reminder that the last week before the festival is the wrong time to find out your pick process depends on one person turning up.

Which numbers show that picking is no longer stalling?

Four numbers tell you, and all of them come from the pick queue itself. Track them for two weeks before the change and two weeks after.

MeasureWhat it tells youHealthy sign
Time from order release to pick takenHow long work waits for a free pair of handsShrinks after the switch
Picks put backWhere items, bins or stock records are wrongFalls as you fix those bins
Orders past the courier cut-offWhether the queue is actually reaching the truckClose to zero on absence days
Picks per picker per shiftWhether work is spread or piled upEvenly spread

The put-back count is the one most owners ignore, and it is the most useful. A bin that keeps getting released is a bin where the system says there is stock and the shelf says otherwise. Fixing it often points back to warehouse slotting or a stock count that is overdue.

What mistakes should you avoid with a pooled queue?

The common mistakes are keeping the old named lists as a habit, skipping the review and pooling work that really needs one person. A pool only works if people trust the open list over the supervisor's notebook.

The practical takeaway

If orders sit unpicked on days when someone is absent, you do not have a staffing problem first, you have an assignment problem. Look at who owns the queue before you hire anyone, and let the role, not the person, hold the work.