Inventory holding cost (H)
Holding cost H is the annual cost of carrying one unit in inventory. It decomposes into five components: cost of capital (5 to 10 percent), storage and warehouse (2 to 5 percent), services (insurance, IT, 1 to 2 percent), risk (shrinkage, damage, 1 to 3 percent), and obsolescence (variable, 2 to 20 percent depending on category). Defensible SMB total: 20 to 30 percent of unit cost for stable goods[1].
Cost of holding one unit in inventory for one year, in $ per unit per year.
The five components
- Capital: the opportunity cost of cash tied up in inventory. Typically your weighted-average cost of capital or your incremental borrowing rate. SMB range 5 to 10 percent in mid-2026.
- Storage: warehouse rent, racks, utilities, labour to put-away and pick. If you use a 3PL, this is the pallet-storage fee per month annualised.
- Services: insurance, IT (WMS license), inventory-tax in some US jurisdictions.
- Risk: shrinkage (theft, damage, miscounts), spoilage where applicable.
- Obsolescence: write-down expected at end of cycle. Negligible for staples, dominant for fashion / electronics.
How to express H
Two equivalent forms. Per-unit-per-year: H = $12/unit/year. Percent-of-unit-cost:i = 30 percent, with H derived as i * C inside the Wilson formula. The percent form is usually easier to defend to a CFO because each component is a natural percent (capital cost, warehouse cost as percent of inventory value, and so on)[2].
How to defend H without overstating
Show the decomposition explicitly. "Our 25 percent carrying cost is 8 capital + 4 warehouse + 2 service + 2 risk + 9 obsolescence" is defensible. "Industry standard" is not. The carrying-cost percentage page walks the component-by-component build-up.