A supplier who moves from twelve units a carton to ten, or shrinks the unit inside the same shelf box, changes freight per unit, duty allocation, and warehouse space without changing the invoice price per piece. This post shows where pack drift hides and the checks that surface it in the month it happens.
Supplier price per unit is the number every system watches. Pack configuration is a line in a spec sheet nobody owns. So when a supplier repacks twelve to a carton as ten to a carton, perhaps to suit another customer line, the purchase order still matches, the invoice still matches, and the goods arrive correct in every way the checks measure.
Underneath, freight per unit just rose, because the shipment now needs more cartons, more cube, and more handling for the same unit count. Duty allocation by weight or value shifts with the new gross weight. The warehouse feels it first as shelf space that no longer fits the plan, which is to say the warehouse feels a costing problem as a storage problem and reports it in the wrong meeting.
The harder variant is the unit itself: a bottle that goes from 500 to 450 millilitres in the same shelf facing, a textile that loses weight, a multipack that drops a piece. Retail price holds, unit cost holds, and the customer eventually notices what the cost model never did. For landed cost the effect is real either way: freight and duty are being paid to move packaging and air in proportions the original model never priced.
None of this requires bad faith. Suppliers optimise their own lines constantly. The failure is a purchasing setup where specification changes travel by rumour instead of by notice.
Units per carton, gross weight per unit, and cube per unit, recorded per shipment and compared with the previous shipment of the same SKU. Any movement beyond a small tolerance gets a question to the supplier while the goods are still in the warehouse, not a discovery at year end. These are receiving dock numbers. They cost nothing to capture because the dock already handles every carton.
Add a goods in photo for the first shipment of any SKU after a gap. Spec drift loves a gap: the reorder after six months is the shipment where the new pack quietly debuts, because everyone receiving it assumes it matches a memory instead of a document.
One clause does most of the work: no change to pack configuration, unit size, materials, or country of origin without written notice before shipment, with unnotified changes at the supplier cost where they raise landed cost. Suppliers sign it readily in a calm negotiation. Its value is not the penalty. It is that the clause makes the change a decision someone must announce, and announced changes get costed.
Keep reading: Supplier quotes versus actual landed cost: closing the estimate gap, Landed cost variance reports: explaining a gap to finance, and Landed cost dashboards: the five numbers worth checking weekly cover the nearby cases in this series.
Because freight, handling, and often duty allocation are driven by cartons, weight, and cube, not by the unit price line. Fewer units per carton means more cartons and more cube for the same quantity, so the cost of landing each unit rises while the invoice looks unchanged.
By recording units per carton, gross weight, and cube per shipment and comparing them with the last shipment of the same product. A step change in any of the three is visible immediately at receiving if anyone is looking at the comparison.
Usually not. Suppliers adjust packaging for their own production reasons and often assume it does not matter to the buyer. It becomes a buyer problem because most purchase terms never made pack configuration a notifiable specification. One contract clause fixes that.
Quantify the landed cost change per unit, raise it with the supplier against the specification or the notice clause, and update the cost model and any price decision that relied on the old figure. Then add the SKU to a watch list for the next two shipments.