Samples arrive by courier at absurd per unit freight, get waved through outside the normal entry process, and then distort the cost model or vanish from it entirely. This post covers how to treat samples honestly: expensed as development, kept out of unit costs, and documented for customs like any other import.
A courier shipment of six prototype units can carry freight that works out at more per unit than the finished product will retail for. If that freight lands in the landed cost model, one of two bad things happens. The model output becomes nonsense and the team stops trusting it, or someone quietly deletes the shipment and the model becomes fiction of a tidier kind.
Samples exist to make decisions: approve the material, check the fit, test the market. Their cost belongs with the other costs of deciding, in development or sourcing budgets, where a manager can see that three rounds of prototypes cost what they cost and ask whether round four is necessary.
Small does not mean invisible. Samples are imports, with a declarable value even when the supplier invoices them at zero or a nominal amount, because customs value is about the value of the goods, not the generosity of the supplier. Marking commercial samples down to nothing on the paperwork is an undervaluation habit that outlives the sample stage and contaminates the relationship with your broker.
Some reliefs exist for samples in defined circumstances and quantities, and a broker can advise where they genuinely apply. The dividing line is intent: samples used up in testing are one thing, free stock that gets sold is production wearing a sample label, and that relabelling is a finding in any audit.
Production landed cost should be built only from production shipments, at production freight rates, in production quantities. Sample costs sit in a development ledger by project and supplier. At quarter end the two ledgers answer different questions: what does a sold unit cost us, and what did it cost to get this range ready to sell.
Blending them flatters neither answer. The range that needed five prototype rounds looks identical to the range that needed one, and next season sourcing decisions get made without the most useful fact about the last season.
The grey zone is the sample shipment that sells: the photography set sold off, the trial quantity listed to test demand. The moment a unit is sold, its costs belong in cost of goods, at a fair allocation, and the duty and freight records should be reclassified with it. Decide that rule now, in writing, before a successful test makes the question expensive.
Keep reading: How do currency swings change landed cost between order and arrival?, Landed cost for small parcels versus pallets: why the maths changes, and Insurance and landed cost: counting cover without double counting cover the nearby cases in this series.
Not generally. Samples are imports with a customs value, and duty applies unless a specific relief covers the circumstances. Zero value invoices do not make goods valueless for customs purposes, and declaring them so creates undervaluation risk.
Because courier freight on a handful of prototypes bears no relation to production freight per unit, and including it poisons the cost model that pricing depends on. Samples are a cost of developing the range, and belong in the development budget where that spending can be managed.
At a value reflecting the goods themselves, typically what they would cost as merchandise, declared honestly even if the supplier charged nothing. Your broker can advise the accepted method for your market; the principle is that a gift invoice is not a valuation method.
Their costs move into cost of goods, including a fair share of freight and any duty paid, and the records should show the reclassification. Selling samples as a quiet channel while costing them as development distorts both ledgers and invites questions in an audit.