Role
Warehouse stock counts
How long does a count take before the warehouse is back to normal work? That number decides whether the count is a control tool or a cost repeated once a year.
Intro call
Four questions we would start with
How long before the warehouse is back to normal work after a count?
If it is three days, the count costs more than the differences it finds.
What do you put off first when a count drags on?
Usually explaining the differences - and then the count ends in an accounting adjustment instead of an improved process.
Who does the counting?
If a zone is counted by the person materially accountable for it, the count is formally flawed - the accountable person may be present while it is counted, but should not be part of the counting team. That is the first thing an auditor checks.
Whose desk does a difference that cannot be explained end up on?
And whether anybody checks whether the same line also differed a year earlier.
We take over the organising and the settling, you keep the judgement on differences
Preparing the count, assigning areas, recording the counts and summarising the differences can be taken off the desk. Judging the cause of a difference and deciding on an adjustment stay with a person.
- The count is recorded at the rack. - No sheets retyped in the evening, which is the most common source of false differences.
- The difference is calculated straight away. - Lines that disagree go back for a recount the same day, while the goods are still standing in the same place.
- Cycle counts instead of one big count. - Counting spread across the year by turnover and value - the warehouse does not stop and the stock is checked more often.
Scope
What can be taken off warehouse stock counts
Twenty example tasks. Each one has a human approval point. Below is a selection, and the scope we fit to your counting rhythm.
Preparing the count
- Dividing the warehouse into counting areas
- Assigning areas to counting teams (decision: the specialist)
- Holding movements in the area being counted
- Count sheets without the system quantities
- A list of lines needing particular attention
Counting and recording
- Recording the result at the rack
- A signal when a line is counted twice with different results
- A signal when a location is skipped
- Marking an area as counted
Differences and their explanation
- A summary of differences as soon as an area closes
- Sending the lines that disagree for a recount
- Linking a difference to the movements of the recent period
- Pointing out lines that also differed in previous counts
- Judging the cause of a difference (judgement: the specialist)
Settling the count
- Differences summarised by value and by range
- A draft report with the differences (approval: the specialist)
- Passing the result to accounting
Cycle counts
- A counting schedule by turnover and value
- A signal when a line has not been counted within the set period
- Stock accuracy over time
A day's work
Today
- A count means stopping the warehouse for a day or two
- Sheets are filled in by hand and retyped into the system
- You see the differences once all the sheets are entered
- Recounting the lines that disagree happens a day later
- You do not explain the causes of differences, because there is no time
- You do not know whether the same line also differed a year earlier
With AI assistance
- The count is recorded at the rack, with no retyping
- A skipped location and a double count report themselves
- You see the differences as soon as an area closes
- The recount happens the same day, on the same stock
- The difference is linked to the movements of the recent period
- Lines that differ repeatedly are pointed out
- How long a count takes
the warehouse goes back to work sooner
counting and recording happen in one step
- The size of the differences
fewer differences caused by a recording error
the result is not retyped from a sheet
- Disruption to the warehouse
cycle counts instead of one stoppage
counting is spread across areas and periods
- Differences that can be explained
more differences with an established cause
the difference is set against the period's movements
We start with recording at the rack, because that is where most false differences come from.
- Step 1
Counting without retyping
The result recorded where the counting happens, with signals about skips and duplicates.
- Step 2
Differences the same day
A summary once an area closes and the lines sent for a recount.
- Step 3
Linking differences to movements
What happened to the line in the recent period, before you call a difference unexplained.
- Step 4
Cycle counts
A schedule by turnover and value. Last, because it needs the counting from step 1 to work smoothly.
Common questions
Questions that come up at the first meeting
How is this different from a fixed asset count?
There you count assets that stand in one place, once a year, and the problem is identification and location. Here you count goods in constant movement, and the problem is that the stock changes while you count. Both need records, but the rhythm and the method are different.
Will a count need fewer people?
A count needs as many hands as there are lines to count, and we will not change that. What changes is the time: the team does not retype sheets and does not come back to the warehouse two days later to recount what disagreed. That means a shorter stoppage, not fewer people counting.
Does this fall under the AI Act?
Stock counting is not a high-risk use. The only point needing care is that the summaries of differences concern areas and ranges rather than the people counting or accountable for a zone - otherwise the count turns into a tool for assessing employees.
Still have a question that is not here? Book a call
Next step
How many lines from your last count were left with no established cause for the difference?
We will show you what a count looks like when the counting is not retyped and a difference goes back for a recount the same day.