Inventory Optimisation Playbook

Most warehouses are simultaneously overstocked and understocked. They have too much of the wrong things and not enough of the right things. This is not a contradiction — it is the natural result of managing hundreds or thousands of SKUs with the same blunt instruments: a uniform reorder point, a standard safety stock formula, and a buying team that responds to stockouts by ordering more and raising minimums.

The financial cost of excess inventory is well understood: working capital tied up in stock, storage costs, product obsolescence, and write-offs. Less well understood is the hidden cost of understocking, which shows up not just in lost sales but in emergency freight costs, customer credits, and the management time spent firefighting. Facilities that measure both costs consistently find that the total cost of poor inventory management is 15–25% of their annual inventory value.

The playbook in this report is based on a three-stage process: segment your inventory to understand what you actually have; recalibrate your replenishment parameters based on real demand and supply data; and build a systematic process for identifying and resolving dead stock before it becomes a write-off problem.

ABC analysis ranks products by revenue contribution: A items are your top 20% of SKUs that generate roughly 80% of revenue, B items are the next tier, and C items are the long tail. This is useful but incomplete, because it tells you nothing about demand variability. A product can be high-revenue (A) but extremely unpredictable (X, Y, Z) — and the inventory strategy for a predictable A item and an unpredictable A item should be completely different.

XYZ analysis adds the demand variability dimension. X items have stable, predictable demand (coefficient of variation below 0.5). Y items have moderate variability. Z items are erratic or intermittent — they might sell 50 units in one week and nothing for the next three. Z items are where most safety stock waste lives, because buyers tend to maintain stock to cover peak demand rather than average demand.

When you overlay ABC and XYZ, you get nine segments: AX, AY, AZ, BX, BY, BZ, CX, CY, CZ. Each segment warrants a different replenishment strategy. AX items should be tightly controlled with lean safety stock — you know when you will need them. AZ items require either a higher service level commitment from your supplier (shorter lead times, more frequent deliveries) or acceptance that you will occasionally run short.

A static reorder point — reorder when stock falls to X units — was appropriate when you reviewed inventory once a week. Modern WMS systems track stock levels in real time, which means you can recalculate reorder points dynamically based on actual consumption and current supplier lead times rather than fixed assumptions.

The standard reorder point formula is: Reorder Point = (Average Daily Demand × Lead Time) + Safety Stock. The safety stock formula, in turn, is: Safety Stock = Z-score × Standard Deviation of Demand × √Lead Time, where the Z-score reflects your target service level (1.28 for 90%, 1.65 for 95%, 2.05 for 98%).

In practice, the critical input that most operations get wrong is lead time. They use the nominal lead time from the supplier contract (say, five days) rather than the actual lead time from their receipt history (which might average 6.2 days with a standard deviation of 1.8 days). Using the actual lead time distribution, and recalculating it monthly from real receipt data, typically reduces safety stock requirements by 10–15% while maintaining the same service level.

Dead stock — product that has not moved in 90, 180, or 365 days depending on your sector — ties up capital, occupies storage space, and in some categories deteriorates in quality. Most businesses know their dead stock problem is larger than they admit, but they do not have a systematic process for dealing with it.

Start by defining dead stock for your business. A fashion retailer might define it as anything unsold after two seasons. A food distributor might define it as anything with less than 30 days of shelf life remaining. An industrial distributor might use 12 months of no movement. Once you have a definition, run the analysis monthly and report the value at cost.

For each dead stock SKU, you have four options: return to supplier (if the contract allows); discount and sell; donate or recycle; or write off. The right option depends on the value, the remaining shelf life, and the supplier relationship. The important thing is that a decision is made. Dead stock left to sit does not improve over time.

A UK-based FMCG distributor running Warewiser across three sites conducted a full ABC-XYZ segmentation exercise in Q3 2024. Prior to the exercise, they were managing 4,200 active SKUs with uniform 14-day reorder points and a 15% safety stock buffer across all product lines.

The segmentation revealed that 68% of their safety stock value was held against Z-category items — products with erratic, unpredictable demand. For these items, the safety stock was not preventing stockouts (because the demand spikes were not forecastable) but was generating significant holding costs. By moving Z-category items to a make-to-order model where possible, and reducing safety stock to a minimum viable level on the remainder, they freed up £1.2 million in working capital over six months.

Total carrying cost reduction was 22%, with no material change in service levels. The project took approximately eight weeks of analyst time spread across a four-month period.