If you're importing goods and tracking landed cost, one question comes up fast: when a container arrives with ten different SKUs, how do you split the duty bill across them?
It sounds like a bookkeeping detail. It isn't. The method you choose determines whether each product's unit cost reflects what you actually paid to land it, or whether it's carrying someone else's burden. Get it wrong consistently and your margin reports lie to you. Pricing decisions follow bad data. Profitability by SKU becomes fiction.
There are six allocation methods in common use. None of them is universally correct. Here's what each one does, when it makes sense, and where it breaks down.
Customs duties aren't assessed on a shipment as a whole. U.S. Customs and Border Protection calculates duty at the 10-digit HTS code level, which means every product classification has its own rate. One SKU might face a 0% duty. Another in the same container might face 25%. A third might be assessed at cents per kilogram rather than a percentage of value at all.
If you're pooling all duties and splitting them by a single rule, you're almost certainly moving cost between products in ways that don't reflect economic reality.
ASC 330, the GAAP standard governing inventory valuation, is explicit on this point: if duties apply only to certain items at a specific customs value, only those items' costs increase by that duty. Duties should not be allocated across unaffected items.
That's the standard. The question is how to get there practically.
Apply each line item's own HTS rate directly to that item's dutiable value.
Formula: Item Duty = Quantity x Unit Cost x Currency Exchange Rate x HTS Rate
This is how CBP actually calculates what you owe, so it's the most defensible method under ASC 330. It handles mixed-rate shipments correctly and doesn't transfer cost between products. It's also the most demanding to operate: it requires up-to-date HTS classification for every SKU you import, and breaks down when classifications are stale or tariff schedules shift mid-year (which has happened repeatedly since 2025).
If your catalog is stable and your HTS codes are maintained, this should be your starting point for duty allocation.
Distribute the total duty bill proportionally by each item's declared commercial value.
Formula: Item's Share = (Item Value / Total Shipment Value) x Total Duty
This is the most common fallback and the industry default for residual duty charges, brokerage fees, and government-assessed fees like the Merchandise Processing Fee (0.3464% of shipment value) and Harbor Maintenance Fee (0.125% for ocean shipments). It aligns with how CBP measures dutiable value on a CIF basis, so it's conceptually coherent.
The limitation shows up in mixed shipments. If you're importing high-value compact electronics alongside low-value bulky packaging, value-based allocation over-burdens the electronics even if their HTS rate is lower. For shipments where value and duty exposure track reasonably well together, this method works fine. For outlier combinations, it can distort unit costs meaningfully.
Allocate costs proportionally by physical weight.
Formula: Item's Share = (Item Weight / Total Shipment Weight) x Total Charge
Weight-based allocation makes sense for freight. Ocean container rates, drayage, and ground freight are driven by mass, so distributing those costs by weight mirrors the actual cost driver. It becomes distortive for duty allocation, because CBP generally assesses duties on value, not weight (specific duty rates charged in cents per kilogram are the exception, not the rule).
A concrete example: a shipment carrying 2,000 kg of steel brackets worth $20,000 and 200 kg of electronic control boards worth $40,000. Allocating freight by weight gives the brackets 91% of freight costs, which is accurate. Allocating duties by weight gives the brackets 91% of duty costs, which is likely wrong if their HTS rates differ.
Weight is the right tool for freight. It's the wrong tool for duty.
Split costs equally across all units.
Formula: Per-Unit Charge = Total Cost / Total Units
This is the simplest method and the weakest for mixed-SKU loads. It treats every unit as if it carries the same logistics burden, which is only true for homogeneous shipments. If you're receiving a single product in a single size with a single HTS classification, quantity-based allocation is fine. If you're receiving ten products across three duty categories, it produces costs that have no relationship to what any individual SKU actually generated.
Quantity-based allocation is appropriate for per-piece inspection fees or handling charges assessed on a unit basis. It should not be the default for duties.
Allocate by cubic volume (CBM).
This method applies when a container fills on cube before it fills on weight, which is common for low-density finished goods like apparel, furniture, or packaged consumer products. If your freight cost is driven by how much space your goods occupy rather than how much they weigh, volume-based allocation is the accurate choice for freight.
For duty allocation specifically, volume is rarely the right driver. Customs valuation is based on commercial value, not cubic meters. Volume-based duty allocation is mostly seen in categories where goods are both low-density and low-value, where the distortion from using it is minimal in dollar terms.
Use different allocation methods for different cost components, matching each charge to the economic driver that actually generated it.
| Cost Component | Recommended Allocation Basis |
|---|---|
| HTS-assessed duties (ad valorem) | Per-item HTS rate x value |
| HTS-assessed duties (specific rate) | Per-item HTS rate x quantity or weight |
| Residual duty pool, MPF, HMF | Value-based |
| Ocean or air freight | Weight or volume |
| Drayage and last-mile | Weight |
| Cargo insurance | Value |
| Customs brokerage fees | Shipment-wide, value-based or equal split |
| Inspection and documentation fees | Manual or equal split |
This is the approach that every major ERP platform and logistics practitioner recommends for production use. The central insight is that no single allocation driver is correct for all cost types. Freight is a physical problem: weight and volume drive it. Duties are a valuation problem: the declared value of each item drives them. Treating these the same way is where distortion creeps in.
The added complexity is real but manageable. Once allocation rules are configured by cost type, the method applies consistently across every shipment without case-by-case judgment calls.
Beyond the duty itself, most imports carry a stack of additional charges that need to find their way into unit cost.
Customs broker fees typically run $125 to $500 per entry and cover clearance, regulatory compliance, and entry filing. Because broker work applies to the entry as a whole, these are most commonly allocated shipment-wide by value, unless a specific line item required specialized handling (FDA notification, USDA coordination, etc.), in which case that cost should be assigned directly to the relevant SKU.
MPF and HMF are calculated by CBP as percentages of shipment value, so value-based allocation mirrors their actual basis and is the natural choice.
Cargo insurance premiums are proportional to shipment value by design, so value-based allocation precisely reflects how the coverage is priced.
Late-arriving invoices are the operational challenge that trips up most teams. Freight bills and broker charges often arrive days or weeks after goods have already been received and made available for sale. The best practice is to book an estimated landed cost at receipt using historical rates or vendor quotes, then true-up when actuals arrive. The variance gets split between inventory (for units still on hand) and cost of goods sold (for units already moved), according to documented policy. This matters more than it sounds: the U.S. Bureau of Transportation Statistics reported that the producer price index for freight and cargo transportation rose 9.8% in January 2025 versus January 2024. Stale estimates compound into real valuation noise fast.
Under ASC 330, inventory must be stated at its complete acquisition cost, including all expenses necessary to bring goods to their present condition and location. Import duties, freight-in, customs brokerage, and cargo insurance all belong on the balance sheet as part of inventory cost until goods are sold, at which point they flow through cost of goods sold.
Expensing these costs immediately instead of capitalizing them understates inventory assets and overstates current-period expenses. For audited financials, that's a problem. For financial decision-making, it's also a problem, just a quieter one.
For companies above the small business threshold ($31M average annual gross receipts for 2025), IRC Section 263A (UNICAP) aligns with ASC 330 for tax purposes, requiring capitalization of both direct and indirect costs associated with inventory, including tariffs.
One additional consideration worth flagging in the current tariff environment: if duty increases push your inventory cost above what goods can realistically be sold for, ASC 330's lower of cost or net realizable value (LCNRV) rule requires an impairment write-down. Whether customer pricing can absorb tariff-driven cost increases is a judgment call, but it's one that needs to happen at the inventory valuation level, not just in a sales meeting.
The right allocation method depends on what you're allocating and what your shipments look like.
For teams importing a narrow, stable catalog with well-maintained HTS codes, per-item rate application is worth the operational investment because it's the most accurate and the most defensible. For teams managing a wide SKU range with frequent classification updates, a value-based default for duties plus weight-based for freight gets most of the way there with much less maintenance overhead.
The pattern that causes the most problems is using a single method for everything: quantity-based allocation applied to duties in a mixed-SKU shipment, or value-based applied to freight when cube, not value, is what fills the container. The distortions compound over time and show up as margin variance that's hard to trace.
Hybrid allocation, with documented rules by cost type, is the approach that holds up across changing product mix, shifting freight rates, and evolving duty schedules. It takes more upfront configuration, but it produces cost data you can actually trust.
Tracking landed unit cost accurately across a mix of products, shipments, duties, and additional charges is exactly the kind of problem that breaks down in spreadsheets. The data lives in too many places: customs entries, freight invoices, broker bills, receiving records. Pulling it together manually every month is slow and error-prone.
Fuse Inventory Control handles landed cost allocation at the shipment line-item level, so every product carries the duty and freight costs it actually generated. It connects those unit costs directly to your financial reporting, which means month-end close isn't a manual reconciliation exercise. Your COGS is current, your margins are accurate, and your team isn't spending a week every month stitching spreadsheets together.
If you're importing and your landed cost process still lives in Excel, it's worth seeing what that looks like when it's automated.