Managing inventory in a warehouse rests on four things: a location system that identifies every storage position, scanning at every touch point, cycle counting instead of annual stocktakes, and slotting by velocity rather than by product family. Software records what you scan, so the discipline has to exist before the system can report anything useful.
The foundations, in the order they need building:
- A location naming system
- Scanning at receiving, putaway, picking and dispatch
- Cycle counting by value
- Velocity-based slotting
- Reorder points and safety stock
Start With Location Naming
Every storage position needs a unique identifier, and the scheme needs to be readable by someone who has worked there for a day.
A standard format works across most layouts:
- Aisle, numbered sequentially
- Bay, numbered along the aisle
- Level, numbered or lettered from the floor up
- Position, where a bay holds multiple pallets side by side
That produces a label such as A-12-C-2, which tells an operative exactly where to go without local knowledge.
Three rules make the scheme work:
- Label every position physically, with barcodes, not just in the system
- Keep the sequence consistent, so aisle numbering does not reverse direction halfway along
- Never reuse a retired location code, which creates phantom stock in historical records
Without this, no inventory system can tell anyone where anything is, and accuracy problems become untraceable.
Scan at Every Touch Point
Inventory accuracy is a function of scanning discipline rather than software quality.
Four points require it:
- Receiving. Scan the item against the purchase order, recording quantity and condition, which is the first control on any order fulfilment London site
- Putaway. Scan item and destination location together, which creates the link the system relies on
- Picking. Scan item and location, confirming the right stock came from the right place
- Dispatch. Scan against the order, catching errors before they leave the building and reach the transport services London leg
A single unscanned step breaks the chain. If putaway is not scanned, the system knows stock arrived but not where it went, which guarantees a discrepancy later with no way to trace it.
Barcodes handle this for most operations. RFID allows bulk reading without line of sight, which suits high-volume or high-value goods where the tag cost is justified.
Cycle Count, Do Not Stocktake
An annual stocktake finds errors up to twelve months after they happened, by which point the cause is untraceable and the damage is done.
Cycle counting counts a portion of locations continuously, so errors surface within days.
Set frequency by value using ABC analysis:
- A items, the small proportion of lines generating most consumption value, counted most frequently
- B items, moderate value, counted at an intermediate interval
- C items, low value, counted least often
The principle is that counting effort should follow value. Counting every line at the same interval over-invests in lines where an error costs little.
Investigate, do not adjust. An adjustment that balances the system hides the process failure that caused the variance. Trace each discrepancy to its cause, which is usually mis-picking, unrecorded damage, putaway error or a receiving mistake.
Track accuracy as a measured figure over time. A single count tells you little; a trend tells you whether the process is improving.
Slot by Velocity
Slotting determines how far pickers walk, and travel time is the largest component of picking labour, which makes it the main lever on warehouse picking speed and accuracy.
Fast movers belong close to dispatch and at waist height. Slow movers belong high, low and far from dispatch.
Re-slot periodically, because velocity changes with season, promotion and product lifecycle. A slotting plan set once and never revisited gradually becomes wrong.
Consider pick weight and size alongside velocity. Heavy items belong at or below waist height regardless of how fast they move, both for safety and for pick speed.
Group items frequently ordered together. Where order data shows two lines repeatedly appearing on the same order, slotting them near each other cuts travel per order.
Set Reorder Points & Safety Stock
Stockouts and overstock are both inventory management failures, and the controls are the same.
A reorder point triggers replenishment when stock falls to a defined level. It needs to account for supplier lead time, so ordering only when stock hits zero guarantees a gap.
Safety stock covers variability in both demand and lead time. Set it against how much each varies, not as a flat percentage across all lines. A line with reliable supply and steady demand needs far less than one with erratic supply.
Review both when conditions change. A supplier lead time extension, a demand shift or a new seasonal pattern invalidates figures set under previous conditions.
Carry costs make overstock expensive rather than safe. Stock sitting in a prime location is occupying space, tying up capital and risking obsolescence, whether the space is your own or rented warehousing services Birmingham.
Stock Rotation
Rotation method depends on what you store.
FIFO, first in first out, suits most goods and prevents stock ageing in place.
FEFO, first expired first out, is required for anything with an expiry date, including food, pharmaceuticals and chemicals. FEFO differs from FIFO whenever a later delivery has an earlier expiry date, which happens regularly.
LIFO, last in first out, suits a narrow set of non-perishable bulk goods where access dictates the sequence.
Enforce rotation through the system rather than relying on operatives choosing correctly. A system that directs the pick to a specific location removes the decision.
What a WMS Does & Does Not Do
A warehouse management system directs putaway, picking and replenishment, holds location-level stock records and reports accuracy, and contracted warehousing services London providers run one as standard.
What it does not do is create accuracy. It records what gets scanned. An operation with poor scanning discipline implementing a WMS typically finds accuracy unchanged, because the software was never the constraint.
Fix in this order:
- Location naming and physical labelling
- Scanning discipline at all four touch points
- Cycle counting and discrepancy investigation
- Slotting
- Then software
What to Measure
Five figures tell you whether inventory management is working:
- Inventory accuracy, comparing system records against physical counts
- Stock turn, how many times inventory sells through per year
- Pick accuracy, the proportion of picks completed without error
- Stockout rate, lines unavailable when ordered
- Obsolete stock value, stock unsold beyond a defined period
Review obsolete stock quarterly. It occupies prime space, costs money to hold, and the longer it sits the less it recovers on disposal.
Frequently Asked Questions
What is the best inventory management method for a warehouse?
Cycle counting with ABC classification and velocity-based slotting works across most operations. The specific method matters less than consistent scanning at every touch point.
How do you achieve high inventory accuracy?
Scan at receiving, putaway, picking and dispatch, count continuously rather than annually, and trace every discrepancy to its cause rather than adjusting the record.
What is the difference between FIFO and FEFO?
FIFO ships the oldest delivery first. FEFO ships the earliest expiry date first, which differs whenever a later delivery expires sooner, and is required for dated goods.
How much safety stock should a warehouse hold?
Set it against demand variability and supplier lead time variability for each line, rather than applying one percentage across all stock.
Do you need a WMS to manage warehouse inventory?
Smaller operations manage with spreadsheets and disciplined scanning. A WMS becomes necessary as location count, SKU count and order volume rise beyond manual tracking.

One Response