Product
Cycle counting that fits inside a coffee break.
One annual stocktake stops being the control the whole business depends on. A site counting a zone every day holds its accuracy between counts instead of restoring it once a year, and no part of a shift is released to do it.
The shift
From one big count a year to a small count every day.
A cycle count, or a rolling stocktake in UK terms, only works if it is cheap enough to repeat. Manual counting is not. At 50 to 100 pallets per hour, a meaningful sample costs a shift, so the count slips and the annual event grows back.
Sentispec Inventory reads up to 3,000 units per hour, up to 30x faster than manual counting. That changes the arithmetic of frequency rather than the arithmetic of a single count. A run that would have taken most of a day takes 20 minutes, so it can be scheduled into a break instead of into a plan.
At the three warehouse hub in Eindhoven, daily 20 minute scans of 500 to 1,000 pallets run during breaks. The two annual wall to wall counts still happen, but the site holds 99%+ accuracy between them rather than rediscovering the truth twice a year. Read the wall to wall page for how the two work together.
What a daily run involves
- One driver, one forklift, no extra labour. The unit is picked up by a forklift the site already operates and put back afterwards.
- Normal travel speed, one shelf level at a time. Nothing about the drive is special, which is why two to three days of training is enough.
- Results while the run is still going. Each aisle is processed as soon as its recording finishes.
- A discrepancy list, not a stock report. You see the exceptions, with the recording behind each one.
Why frequency beats effort
A discrepancy found today costs less than the same one found in December.
- Errors are corrected while their cause is still visible. A pallet in the wrong location found the same day is a put away to redo. Found six months later it is a write off, an out of stock, or a customer complaint.
- Accuracy stops being seasonal. Ten live sites report 99% and above after go live, with several holding 99.99%. How that number is measured is set out on the accuracy page.
- Picking gets faster because the data is right. One global 3PL measured a 6 to 8% productivity improvement across its Northern European warehouses, within a general range of 5 to 10%, alongside a 5 to 10% labour cost reduction.
- Where accuracy protects output, the case is not about labour at all. A wind turbine manufacturer loses €7,000 for every minute a line stops, so 99%+ accuracy is a production requirement rather than a cost saving.
Counting more often is normally the expensive option. When a full run costs 20 minutes of one driver's time, counting more often is the cheap option.
Scale
100,000 pallets scanned per month at one 3PL.
A global 3PL serving retail and fashion clients across Northern Europe runs comprehensive audits of 30,000 to 100,000 pallets for clients who demand them frequently.
- 100,000 pallets scanned per month across the network, on continuous counting rather than scheduled events.
- Stock accuracy increased to 99.97%.
- Missing pallet related labour down 95%, covered in more detail under lost pallet search.
The full story is in the 3PL case study, and the cycle counting frequency guide works through how often a given stock profile needs counting.
Questions
What managers ask about cycle counting.
How long does a cycle count take?
A 20 minute run covers 500 to 1,000 pallets. That is short enough to drive during a break or between tasks, which is how the sites running daily counts do it without releasing any part of a shift for stocktaking.
Is a cycle count the same as a stocktake?
A cycle count is a partial count of a zone or a sample, repeated on a cycle. A stocktake, in the UK sense, usually means the full count of the whole building. Sentispec Inventory runs both with the same unit and the same drive down the aisle, so the only difference is how much of the warehouse you cover in one go. Terms are defined in the glossary.
Does continuous counting replace the annual stocktake?
It replaces its purpose rather than the event itself. Counting continuously holds 99%+ accuracy all year, so the annual wall to wall confirms a position you already trust. Most sites keep one because auditors and end customers expect it.
How much work does our WMS team need to do to start counting?
None. It works out of the box on a stock file exchange, so import a stock file and you get a full discrepancy report back. A standard API is available when you want corrections to flow back automatically, and that decision can wait until months after go live. See integrations.
Next step
See a 20 minute run over an aisle like yours.
A 30 minute demo covering what a daily count would look like on your zones, your racking and your label format.