A common misconception in demand forecasting is that low-volume SKUs should simply be aggregated away. Operationally, that often works until a low-volume product represents disproportionate financial exposure.
In our case, some ultra-premium luxury SKUs sold fewer than 100 units per year, yet a single transaction could represent several thousand euros in revenue.
From a supply-chain perspective, these products looked statistically insignificant.
src="/articles/forecasting/sku-paradox.png" alt="The ultra-premium SKU paradox. A 100% stacked horizontal bar chart compares two views of the same SKU catalogue. Top bar — Revenue contribution: 80% bulk SKUs in gray, 20% ultra-premium SKUs in Artometrix green. Bottom bar — Transaction volume: 95% bulk SKUs in gray, 5% ultra-premium in green. Below the chart, two columns of commentary separated by a hairline divider. Left, Supply chain view: 'Statistically insignificant — volume is dominated by bulk SKUs, the natural instinct is to aggregate the small slice away.' Right, Finance / CFO view: 'Impossible to ignore — 5% of transactions carry 20% of revenue, every miss is several thousand euros of exposure on a single transaction.' Headline above the chart: 'Ultra-premium SKUs carry four times their weight.'" caption="5% of transactions, 20% of revenue — same SKUs, two operational lenses. Tap to enlarge." width={3262} height={1826} />
À propos de l'auteur
Cyril Noirot
Lead Data Scientist
Data scientist freelance. Je conçois et déploie des systèmes de décision — prévision, pricing, marketing measurement, optimisation.