Conventional wisdom on ordering sounds like: "the bigger the order, the lower the per-unit cost, so order more." It's true at the factory quote level. It's false at the annualized P&L level, because the money that is not in your quote is the cost of keeping inventory around after you've paid for it.
Inventory carrying cost is the dollar expense, per year, of holding unsold inventory. For most consumer goods brands, that number is 25–40% of the value of unsold inventory, annualized. A brand sitting on $100,000 of product at the end of a quarter is burning $6,250–$10,000 a quarter on that inventory without selling a unit.
This article is the counterbalance to the "order more" instinct.
The four carrying-cost components
1. Cost of capital (10–15% annualized)
Money in inventory is money not doing anything else. At a 10% cost of capital — either borrowed money at an interest rate, or equity capital that could earn that return elsewhere — every $100 sitting in unsold product is $10 a year in opportunity cost. Brands using SBA loans or merchant-cash-advance financing are often at 14–20%, making this cost higher.
For a bootstrapped brand, this line feels invisible — you're not paying a loan interest bill on your inventory. It's still real; the cash you parked in inventory is cash you don't have to run the business on.
2. Storage, insurance, handling (5–10%)
Annualized 3PL storage fees: $15–$35 per pallet per month. A pallet holds 600–1,200 units of folded apparel, so roughly $0.25–$0.60 per unit per month of storage, or $3–$7 per unit per year if it sits all year.
Insurance on stored inventory: 0.5–1.5% of value annually.
Handling fees (receiving, counts, audits): 1–3% of value annually.
On a $20-cost product, that's $0.50–$1.00 per unit per year. Annualized, roughly 5–10% of product value, higher for bulky goods.
3. Obsolescence and shrinkage (3–8%)
Fashion-forward product: obsolescence is high. A seasonal collection that doesn't sell through loses style value every month past its window.
Evergreen product: obsolescence is low. A plain organic cotton tote is as sellable in 2027 as in 2026.
Shrinkage (damaged, lost, miscounted inventory) runs 1–3% for most operations.
4. Markdown risk (5–10%)
The most invisible and most expensive line. When inventory sits past its sell-through window, the brand eventually discounts to clear it. Common cascade:
- Week 12: 20% off
- Week 18: 30% off
- Week 24: 50% off
- Week 36+: clearance at 70% off
Each markdown destroys margin, and — worse — trains the customer base to wait for the next discount. A brand that markdown-cascades aggressively pays twice: once on the unsold inventory, and once in eroded full-price conversion on future product.
How to size an order properly
Three quick frameworks:
Framework A: Weeks of supply
Calculate average weekly sell-through (units per week) from the last quarter. Set a target weeks-of-supply — 12–20 weeks is reasonable for most mid-size brands. Order that number times your weekly run rate.
Example: selling 60 units per week, 16 weeks of supply target = 960 units. Round to 1,000.
Framework B: Lead-time-plus-buffer
Calculate your reorder lead time (typically 10–16 weeks for custom apparel from India). Order enough to cover lead time plus 4 weeks of buffer.
Example: 14-week lead time, 60 units/week, 4 weeks buffer = 18 weeks × 60 = 1,080 units.
Framework C: Cash cycle
Calculate your cash conversion cycle: inventory days + accounts-receivable days − accounts-payable days. Order in quantities that don't blow up cash cycle.
If your terms with the factory are "50% deposit, 50% on shipment" and you finance 60-day terms with retailers, the order size that keeps cash flowing has to balance against what cash you have to bridge the production → sell-through window.
The per-unit savings trap
Factories typically show you a volume-discount curve: 500 units at $X, 1,000 at $0.85X, 2,500 at $0.75X, 5,000 at $0.68X. The temptation is to order at the 5,000-unit price because "we'll sell through eventually."
If your sell-through rate says that 5,000 units will take 30 months to clear, the per-unit saving of $0.17 ($0.85X → $0.68X) is obliterated by 30 months of carrying cost on the unsold portion.
On a $10 cost item:
- 1,000 units at $8.50 = $8,500. Sell through in 12 months. Carrying cost on avg-inventory: $4,250 × 25% × 1 year = ~$1,000. Total cost: $9,500.
- 5,000 units at $6.80 = $34,000. Sell through in 30 months. Carrying cost on avg-inventory: $17,000 × 25% × 2.5 years = ~$10,625. Total cost: $44,625.
On a per-unit basis: $9.50 vs $8.93. Not dramatically different, and you locked up 4× the capital for 2.5× the time.
The lesson is not "always order small." It's "the carrying-cost math must be in the decision alongside the per-unit price."
What "the right order size" actually means
Define it as the order that minimizes total landed cost per unit sold — not per unit produced.
Total landed cost per unit sold includes:
- FOB cost from factory (your quote)
- Freight, duty, 3PL (landed cost)
- Carrying cost on the portion that doesn't sell in the first 90 days
- Markdown losses on whatever clears at less than full price
- Reorder penalty (sampling, setup, amortized) if you under-order and re-run
Model that total for three or four order sizes. The curve usually has a clear minimum. That's your order.
The uncomfortable truth
Most brands over-order on their first or second production run because the factory quote makes it look cheaper to do so. Then, 18 months later, they're clearing out SKUs at 60% off and wondering where the margin went. Ordering at the quantity that sells in one cycle — not the quantity that minimizes FOB — is the discipline that compounds.
Frequently asked questions about inventory carrying cost
Is 25–40% annual carrying cost really typical?
Yes, for mid-size consumer brands. The range decomposes as: 10–15% cost of capital (opportunity cost of cash tied up in inventory), 5–10% storage and handling (3PL, insurance, warehouse), 3–8% obsolescence and shrinkage, and 5–10% markdown risk. Different categories have different ranges — fashion runs high because of markdown risk, commodity goods run lower because markdowns are less aggressive.
How do I know if I'm over-ordering?
Track weeks-of-supply: current inventory ÷ average weekly sell-through. For most consumer apparel brands, 12–20 weeks of supply is healthy; 30+ weeks is a warning sign you'll be marking down or sitting on dead inventory. If you're heading into a peak season the metric flips — 30 weeks on-hand in September for a November peak is fine; 30 weeks in March for a summer product is a problem.
What's the break-even between ordering less and re-ordering more often?
Every reorder has a fixed setup cost (pattern review, QC revisit, sample rerun, factory re-onboarding of the SKU). For most cut-and-sew custom apparel, the reorder setup cost is $800–$2,500 per SKU. If your weekly sell-through is strong, the carrying cost of over-ordering quickly exceeds the setup cost of two or three smaller orders per year. For steady-state SKUs, two production runs a year (roughly 6-month inventory cycles) is a common optimum.
How do markdown cascades work?
When inventory sits past its seasonal window, the price reductions compound: 20% off at week 12, 30% at week 18, 50% at week 24, 70% at clearance. Each markdown destroys margin and signals to customers to wait for the next discount on your next product. The quiet cost of over-ordering isn't just the unsold units — it's the pricing behavior you teach customers on the units that did sell. Brands that rarely discount protect future margin more than they lose on today's markdown avoidance.
Related reading
- When Low-MOQ Is Actually More Expensive: The Amortization Trap — the other side of the sizing question.
- Apparel MOQs Explained: Why Factories Set Minimums — why the factory quote curve looks the way it does.
- Sampling to Bulk: A Realistic Timeline — the reorder-lead-time variable that feeds into framework B.
Found this useful?
Share it with your network
