You cycle count without stopping the floor by counting a small subset of locations on a rolling schedule instead of the whole warehouse at once, which keeps inventory accuracy high while orders keep shipping. Shipider makes this practical: every count that does not match the system feeds straight to a checker, who approves the correction or adjusts it, and the change lands in the SKU movement history with a timestamp.
Why continuous counting beats the annual shutdown
The once-a-year physical inventory is a familiar ritual: freeze the warehouse, pull staff off their normal jobs, count everything over a long weekend, and hope the number reconciles. It feels thorough. In practice it carries three problems that quietly hurt a growing operation.
The first is disruption. Shutting down means no receiving, no picking, no dispatch for the duration. For a warehouse that grew up on Excel and is now juggling more customers than a spreadsheet can track, a full stop is expensive and stressful, and it usually happens at the worst time of year.
The second is the stale number. An annual count tells you what was true on one day. The moment the floor restarts, the count begins drifting again. For the other fifty-one weeks you are quoting stock from a figure you no longer fully trust, which is exactly the situation cycle counting is meant to prevent.
The third, and the most damaging over time, is the missing root cause. A big-bang count produces a giant reconciliation at the end, a pile of variances you have to swallow all at once. By then the trail is cold. You cannot tell whether a shortage came from a mis-scan at receiving, a pick error, or a location that was never put away correctly. You just adjust the books and move on, which guarantees the same errors return.
Continuous cycle counting flips all three. Nobody stops working. The accuracy figure stays fresh because you are always sampling. And because you count small batches often, each discrepancy is recent enough to investigate while the trail is still warm.
How to run cycle counting inside a maker-checker system
Cycle counting works best when a count is not the same thing as an edit. In many spreadsheet-run warehouses, whoever notices a mismatch just changes the number, and the record of what really happened disappears. Shipider separates the two roles on purpose. A processor (the maker) counts and proposes; a checker confirms or adjusts. That single split is what turns counting into an audit trail instead of a guessing game.
A practical rolling count in Shipider looks like this:
- Pick the locations for the day. Instead of counting everything, you select a handful of storage locations to check. Prioritise by velocity and value: the SKUs that move fastest and cost the most are where errors do the most damage, so they earn more frequent counts.
- Count what is physically there. The processor walks to the location and scans the barcodes using the camera on a phone or tablet. No dedicated scanner gun, no docking cradle, no separate hardware to charge. Whatever device is in hand becomes the scanner, and the rack map shows exactly which location to stand in front of.
- Compare against the system. Shipider holds a running remaining quantity for every SKU. The count either matches that figure or it does not. A match confirms the location and moves on quietly. A mismatch becomes a discrepancy.
- Flag the discrepancy to a checker. The processor does not silently overwrite the number. The discrepancy is raised for a checker to review. The checker decides whether to approve the adjustment (accepting the physical count as correct) or hold it for a closer look.
- Record the outcome. Once the checker acts, the adjustment is written into the SKU movement history with a timestamp and the running remaining quantity updated. The record shows who counted, who approved, and when.
The discrepancy view has a checked and unchecked filter, so a supervisor can see at a glance which counts still need a decision and which have been resolved. Nothing falls through the cracks, and no correction happens without a second set of eyes.
Why the second signature matters
The value of maker-checker is not bureaucracy for its own sake. It is that every stock change carries a reason and an owner. When a customer later asks why their recorded quantity changed on a Tuesday, you can point to the exact count, the person who ran it, and the checker who signed off. That is the difference between "the number moved and we do not know why" and a defensible, timestamped trail.
An ABC-style counting frequency
Not every location deserves the same attention. A common approach is to sort SKUs into bands by a mix of movement velocity and value, then count the busiest, most valuable band more often than the quiet, low-value band. The idea is simple: spend your counting effort where a mistake costs the most.
The table below is a starting frame you can adapt to your own product mix. Treat the frequencies as a pattern, not a prescription.
| Band | Typical profile | Suggested count frequency | Why |
|---|---|---|---|
| A | High velocity and/or high value; the SKUs that drive most of your shipping | Most often | Errors here hit fulfilment and customer trust immediately, so catch them fast |
| B | Moderate movement and value | Regular but less frequent | Worth steady attention without crowding out the A band |
| C | Slow movers and low value; the long tail | Least often | Low impact per error, so a lighter touch keeps effort proportional |
Because Shipider prices on usage and keeps live analytics on movement, you can look at which SKUs actually turn over rather than guessing from memory. That lets you assign bands from real activity and revisit them as your mix shifts through the year. A SKU that was a slow mover last quarter may have become an A-band item, and the count schedule should follow.
How patterns feed root-cause work
A single discrepancy is a correction. A pattern of discrepancies is a lead. This is where rolling counts earn their keep beyond simply keeping the books tidy.
When you count small and often, the variances arrive as a stream rather than a flood, and each one still has context around it. Over a few weeks you start to see shapes: a particular location that keeps coming up short, a SKU family that is always over, a shift or a step in the flow where numbers reliably drift. Those shapes point at causes.
Common threads worth watching for:
- The same storage location repeatedly off, which can hint at a labelling or rack-map problem rather than a counting error.
- A SKU that is consistently short after receiving, which points back upstream. Tightening the receiving-to-put-away flow often removes a whole category of downstream mismatches.
- Two similar SKUs that seem to swap quantities, a classic sign of a pick or put-away mix-up between neighbouring locations.
Because every adjustment lives in the SKU movement history with its running remaining quantity, you can walk a suspicious SKU backwards through its movements and see where the line diverged. That is the raw material for a real investigation. We wrote a fuller guide on how to find the root cause of SKU discrepancies that picks up exactly where a flagged count leaves off.
The point of cycle counting, done this way, is not only to keep the number right. It is to keep giving you fresh, well-documented evidence so you can fix the process that produced the error, not just paper over the result.
The payoff: numbers you can quote any day
When counting runs continuously and every correction is checked and recorded, a few things change about how the warehouse feels to run.
You stop dreading the annual event, because there is no single high-stakes count to get through. The work is spread thin across the calendar and folded into normal days. Staff are not pulled off their jobs for a marathon weekend.
You can quote stock with confidence on any given day, not just in the week after a physical inventory. When a customer or a colleague asks whether you can fulfil an order, the remaining quantity you are looking at has been kept honest by recent counts, and you can see when a location was last verified.
And for a 3PL running many customers under one roof, that trust compounds. Each customer's stock sits in its own tenant, counted on its own rolling schedule, with an audit trail you can show them. The conversation shifts from defending your numbers to simply sharing them.
None of this needs a warehouse-wide shutdown, a special scanner fleet, or an ERP migration. It needs a schedule, a camera, a running quantity, and a checker. That is the calm, capable version of stock accuracy, built for warehouses that outgrew the spreadsheet.
Frequently asked questions
How many locations should I count each day?
Enough that every SKU gets counted at a frequency matched to its band over your cycle, and few enough that a processor can finish them properly without rushing. Start small, watch how long a batch actually takes on the floor, and adjust. The goal is a steady rhythm you can keep, not a heroic daily total you abandon after a week.
Do I still need an annual physical inventory if I cycle count?
Many operations find that a well-run rolling count keeps accuracy high enough that the annual event becomes a formality rather than a rescue mission. Whether you keep a full count for accounting or audit reasons is a business decision, but the day-to-day trust in your numbers comes from the continuous counting, not the once-a-year one.
What stops a processor from just fixing the number themselves?
The maker-checker split. In Shipider the processor proposes a count but does not commit the adjustment; a checker reviews the discrepancy and approves or holds it. The checked and unchecked filter shows which counts are still waiting, so corrections cannot quietly slip through without a second signature.
Do I need special scanning hardware to count this way?
No. Shipider uses the camera already on a phone or tablet to scan barcodes, so any device on the floor can run a count. There is no scanner fleet to buy, charge, or maintain, which is part of what makes rolling counts easy to start.
How does counting help me find the cause of errors, not just fix them?
Every adjustment is written into the SKU movement history with a timestamp and a running remaining quantity, so patterns become visible over time: a location that keeps coming up short, a SKU that is always off after receiving. Those patterns are the starting point for real root-cause work rather than a one-off correction.
Ready to keep your stock honest without ever stopping the floor? Create your Shipider account and start counting on a schedule that fits your warehouse.
Related reading: Picking Strategies for Small Warehouses: Which One Actually Fits Your Floor
Related reading: How to Reduce Mis-Ships in a Warehouse: A Packing Workflow That Actually Works