INVENTORY CONTROL
- Original Article: Visit Link
Waste is not just the problem.
It is also the signal.
Modern retail has sophisticated capabilities to prevent waste. Retailers invest heavily in forecasting and, when forecasts are imperfect, can influence purchasing through price reductions and promotions. Together, these provide powerful tools to manage waste. Yet despite decades of technological and operational refinement, waste remains a persistent commercial challenge.
What causes waste?
Excess inventory is an obvious explanation. At the end of a product’s commercial life, unsold units are, by definition, excess to realised demand. But this describes the outcome, not necessarily how it arose. Some excess inventory may result from lower thanforecast demand; some may be an economically rational consequence of balancing availability against waste. Perishability may also appear an obvious explanation: perishable products expire and therefore become waste. But perishability is a property of the product, not the cause of it remaining unsold. It imposes a finite commercial life within which demand must become sales, determining when an unresolved commercial problem becomes irreversible – not why the product remained unsold. Academic research supports a broader explanation. In their 2018 study, Retail Store Operations and Food Waste, Christoph Teller, Christina Holweg, Gerald Reiner and Herbert Kotzab examined 28 stores across supermarket, hypermarket, discount and convenience formats. They identified multiple causes of food waste, including unpredictable demand and customer behaviour, inefficient store operations and elevated product requirements, with causes differing across formats and product categories.
Commercial time is demonstrated more directly by Rob Broekmeulen and Karel van Donselaar’s 2019 study, Quantifying the Potential to Improve on Food Waste, Freshness and Sales for Perishables in Super markets. Across three product categories and 27 stores operated by three large European retailers, a measure combining casepack size, average demand and available store shelf life explained 42% of the variation in waste. Their findings on time were striking. Increasing available store shelf life by one day reduced model led waste by 43.1%, while improving freshness. Changes in supplychain and replenishment practices also substantially reduced waste, with improved execution especially beneficial in medium and large stores.
Storeworks - Services - Products - (Contact Storeworks)
Multiple causes
Multiple potential causes create a further difficulty. There is no reason to assume they operate independently, nor sufficient evidence to know how they interact in every retail environment. If one factor’s effect depends on another, correcting a single cause may not produce the expected reduction in waste. The cost of intervention may there fore be relatively easy to estimate while its effect – and prospective return – remains uncertain. This complexity and uncertain return may explain why discussions of waste gravitate towards simpler explanations. When causes are difficult to isolate, expensive to investigate and uncertain to resolve, there is an incentive to focus on those readily observed, measured and acted upon. Excess inventory, fore casting error and perishability provide such explanations. They are not necessarily wrong; their simplicity can make them appear more complete than they are.
The complexity of waste creates another problem. If multiple factors can produce the same out come, identifying the underlying cause is difficult. Even investigation or correction of retailer data may not reveal why a product failed to sell. Correcting a record and explaining the failure are not necessarily the samething. Retail systems ultimately observe transactions. Where execution failure prevents consumer demand from becoming a sale, sales history records the lower sales, not the demand that was lost. Unless recognised, this distortion may influence subsequent forecasting and replenishment as evidence of lower demand.
Modern forecasting systems can compensate for many known disruptions and anomalies, but sophistication does not imply perfection. The residual risk lies in unidentified execution failures: a system cannot necessarily correct a distortion it has not recognised.
Consequences
For suppliers, the consequence may extend beyond the immediate lost sale or eventual waste. A supplier can potentially pay twice: first when existing demand fails to become a sale, and again if the lower sales history contributes to reduced future replenishment. This creates a particular hurdle for suppliers of perishable products. Ambient products can suffer the same execution failures and distortion of recorded demand, but do not face the same irreversible commercial deadline. An unsold ambient product may still sell later; a perishable product eventually loses that opportunity. Perishability therefore limits the time available to recover from execution failure.
Once that time expires, the sales opportunity disappears while the lower recorded sales history remains. If that history influences future demand estimates, the consequences may extend beyond the product’s commercial life. Simplified explanations of waste also assume that consumer demand had a fair opportunity to become reported sales throughout the product’s commercial life. Consumer demand and reported sales are not the same. Sales are demand filtered through execution and time:
ISB Optimus - Solutions - (Contact ISB Optimus)
Sales = Demand × Execution × Time.
Demand is consumer purchase intent for the product being analysed. Execution combines data integrity, logistics efficiency and instore execution. Time is the commercial period during which demand can become a reported sale. This is a conceptual relationship, not a mathematical identity: reported sales reflect the interaction of demand, execution and commercial time. An execution failure occurs when existing consumer demand is prevented from becoming a sale by a failure in data integrity, logistics or instore execution. Demand that fails to become a sale cannot be observed directly, but consumer buying intent can be inferred from sales patterns. Broken patterns may indicate interrupted demand conversion rather than absent demand.
The distinction matters: a nosale event and a no demand event are not necessarily the same thing. The principle is familiar in retail. A nosale event’s significance depends on whether it is consistent with the product’s established sales pattern. No sales for a hig-hvolume staple carries a different signal from the same event for a slow moving premium product. Analysis formalises this familiar judgement and applies it consistently across large numbers of SKU-store observations. Sales patterns can break for countless reasons. A break therefore does not establish execution failure. The challenge is to distinguish, with operational grade certainty, those warranting investigation – and quantify the financial incentive to investigate them.
Identifying, detection & intervention
Identifying where intervention is warranted does not require every cause to be diagnosed first. Detection and diagnosis are different problems: the immediate objective is to identify interrupted demand conversion with sufficient confidence to justify intervention; precise diagnosis may follow. Why would retailers not already do this across their entire SKU range? The answer lies partly in incentives and risk. When execution failure prevents one product’s sale, substitution may preserve much of the retailer’s category revenue while the supplier loses the sale. Distinguishing no demand from execution induced no-sale events across millions of SKU-store-day observations also creates risk. Even small errors at scale could increase inventory and working-capital requirements – unattractive in a low margin business. Retailers have an incentive to improve execution, but operate within an enormous operational universe and finite resources. Suppliers have a concentrated interest in their products and, particularly in South Africa, often fund merchandising resources capable of intervention. What they may lack is sufficiently precise information to direct those resources towards the failures that matter most.
A second paradox emerges: products receiving the greatest operational attention also generate the most visible waste. Perishable products are intensively managed because their limited commercial life leaves little opportunity for recovery. Yet they continue to generate disproportionate waste. Expiry makes unresolved problems visible, forcing attention through markdowns, writeoffs or disposal. Nonperishable products provide no equivalent short term signal. They may instead remain as slowmoving inventory, tying up working capital, depressing supplier sales and masking poor execution. The commercial consequence remains but is no longer recognised as waste.
Henro Biscoplus - Products - (Contact Biscoplus)
It has become hidden underperformance
Waste therefore performs an important diagnostic function. Not every instance results from execution failure. But where execution failure prevents existing demand from becoming sales, the finite commercial life of a perishable product eventually forces its consequence into view, creating an opportunity for investigation and correction. Ambient products provide no equivalent short term signal. Execution failure can remain hidden, persist or recur without attracting intervention, allowing its commercial consequences to accumulate. The absence of visible waste should therefore not be mistaken for the absence of execution failure – or a smaller commercial problem.
This distinction is not merely theoretical. Across data-sets analysed over the past decade, including non-perishable products, we have identified commercial opportunities associated with execution failure ranging from 15% to 135% of recorded sales. An opportunity exceeding 100% does not mean more than 100% of sales were lost; it expresses the estimated opportunity relative to recorded sales. Each dataset determines the scale of opportunity it reveals. The methodology for identifying individual failures is beyond the scope of this article; the important point is that execution failure can impose substantial commercial costs without producing the visible signal created by perishable waste.




