Landed cost: the number that should drive supplier choice

Every procurement conversation eventually collapses into a single number: the factory price. It is the number on the quotation, the number the buyer reports upstream, the number that anchors the negotiation. It is also, on its own, close to useless.

A supplier in Yiwu quoting 2.10 USD per unit can be more expensive than one in Ningbo quoting 2.45. A Turkish factory at 3.90 EUR can beat an Italian one at 3.40 once you count the rejects. The number that decides your margin is landed cost: everything it took to put one saleable unit in your warehouse, ready to ship to a customer. Anyone selecting suppliers on factory price alone is running on incomplete data.

What actually belongs in a landed cost

The textbook definition, product + freight + duty + insurance, is the minimum, not the model. In practice, a landed cost that actually predicts your real per-unit cost needs to include every line that would disappear if the order disappeared.

• Ex-works product cost, net of any tooling amortization allocated across the order
• Inland freight to the origin port and export clearance at origin
• Ocean or air freight, including BAF, CAF, and low-sulphur surcharges that show up on the final invoice
• Marine insurance (usually a small fraction of a percent of cargo value, and worth having for anything above a few thousand euros)
• Import duty and, where applicable, anti-dumping duties. The second category is what can blow up the budget on goods such as ceramics, bicycles and steel products from China
• VAT financing cost: you pay it at clearance and recover it months later, and that gap has a real cost of capital
• Destination port fees, THC, demurrage risk, customs broker
• Inland trucking to your warehouse, including any ADR or fragile handling
• Quality-related loss: rejected units, rework, returns from customers caused by manufacturing defects
• Financing cost on the deposit (a 30% upfront payment is common, tied up for the weeks or months between deposit and delivery)

The last three are the ones most buyers omit, and they are the ones that most often flip a decision. A 3% reject rate on a low-margin SKU is not a rounding error: it is the difference between the supplier who wins the tender and the one who should have.

The incoterm trap

The single most common mistake in supplier comparison is comparing quotes on different incoterms. FOB Shanghai vs. DAP Milan vs. EXW Guangzhou are three completely different numbers, and normalising them mentally almost never works, the buyer underestimates the freight leg, overestimates the duty, or forgets clearance entirely.

The rule is simple: force every supplier to quote on the same incoterm, and let it be the incoterm closest to your warehouse. If the supplier refuses DAP or DDP, that itself is a signal, they either do not have the logistics maturity to price it, or they are hiding volatility. Both are useful to know before signing.

The supplier with the lowest FOB and the highest reject rate is often the most expensive one in the room. It just takes months and a returns spike to prove it.

Where Italian-produced goods actually win

Landed cost math is often used to dismiss European production out of hand. It is worth running the numbers properly, because for several categories the gap closes far more than expected once the full model is applied.

On an order of promotional textiles, an Italian producer may quote well above a Chinese equivalent at the factory gate. Add ocean freight at current rates (which swing widely by lane and season), import duty (many apparel codes carry around 12% in the EU, so check the exact HS code), VAT financing over a few months and a realistic quality loss, and the gap narrows noticeably. On short runs, the Italian option can win outright: the MOQ premium and the tooling amortization on a Chinese run shift the math.

This is not an argument that Made in Italy is always cheaper. It is an argument that the decision requires a landed-cost model, not a factory-price reflex. Half the sourcing conversations that end in "too expensive" would end differently with a proper spreadsheet.

A working template

The landed cost model does not need to be sophisticated. It needs to be complete and it needs to be applied consistently to every supplier under evaluation.

Build it in three blocks

• Block 1 (product): ex-works price, tooling per unit, packaging, inspection cost per unit
• Block 2 (logistics): origin handling, freight allocated per unit (CBM × rate ÷ units per CBM), insurance, destination fees, duty, VAT financing, inland delivery
• Block 3 (risk-adjusted): expected reject rate × unit cost, expected customer return rate × cost of return handling, currency hedging cost if paying in a non-home currency

Run it for each shortlisted supplier, on the same volume, on the same incoterm, on the same delivery window. The winner is rarely the one with the lowest Block 1. It is almost always the one with the tightest Block 3.

The number that gets forgotten

One line missing from most landed-cost models: the cost of the buyer's own time. A supplier who requires four rounds of sample revisions, two escalations to fix a spec drift, and a physical trip to the factory to unblock production is not a cheap supplier. That cost sits in headcount, not in the PI, which is why it is invisible on the spreadsheet, and why the same three suppliers keep getting reordered from year after year even when a cheaper option is on the table.

The purpose of a landed-cost discipline is not to build a bigger spreadsheet. It is to protect the buyer from the seductive simplicity of a single number on a quotation. Factory price is where negotiations start. Landed cost is where decisions get made. This is the kind of check Opifix builds into every RFQ, with freight and duty included in the quote so offers can be compared like for like.