Duty optimization: the four levers, and how to know which one applies
Lowering duty is not one move. It is four: preference, deferral and elimination, recovery, and accuracy. What each lever does, and why knowing which applies is a data question.
Key takeaways
- Duty optimization is not a single tactic; it is four distinct levers that each target a different part of the duty.
- Preference: claim FTA treatment on goods that qualify. Deferral and elimination: use an FTZ or bonded warehouse. Recovery: reclaim duty through drawback. Accuracy: fix classification, valuation, and sourcing.
- Which lever pays off depends entirely on your flow, so the decision starts with duty broken down by product, origin, and tariff type.
- The biggest mistake is applying a generic tactic instead of the one your data points to.
- Finding and sizing the levers is an analytics problem the import data already supports.
When duty rose through 2025 and 2026, "how do we pay less duty" became a board question. The usual answer, a single tactic someone read about, is the wrong shape. Duty optimization is not one move; it is four distinct levers, each targeting a different part of the duty, and the one that pays off depends entirely on your flow. Pulling the wrong lever wastes effort on savings that are not there. The right starting point is not a tactic at all; it is seeing where your duty actually sits.
Lever one: preference
The first lever is claiming free-trade-agreement treatment on goods that qualify. Under USMCA and other agreements, originating goods can enter duty-free, but only if you certify them against the rules of origin and actually claim the preference. Many operations leave money here, either by not claiming on eligible value or by not knowing which goods would qualify with a small sourcing change. The lever is measured by the FTA qualification rate: eligible value claimed versus left on the table.
Lever two: deferral and elimination
The second lever is timing and structure. A Foreign-Trade Zone or bonded warehouse defers duty until goods enter commerce, eliminates it on re-exports, and can allow an inverted-tariff benefit where the finished-good rate is lower than the components. As duty rises, the cash-flow value of deferral and the savings on re-exports grow with it. This lever pays off for operations with large or slow-moving inventory, significant re-exports, or manufacturing from imported inputs.
Lever three: recovery
The third lever reclaims duty already paid. Duty drawback refunds up to 99% of duty on imported goods that are later exported or destroyed. For an operation that both imports and exports, it can be a meaningful recovery, and it grows with the duty rate. The catch is that not all special tariffs are drawback-eligible, so the recoverable amount has to be calculated carefully rather than assumed. The lever is sized by matching import duty to subsequent exports.
Lever four: accuracy
The fourth lever is not a program; it is getting the basics right. Wrong HTS classification overpays or underpays duty, and a mis-built customs value moves the base the rate applies to. Accuracy is the lever that saves money without any special structure, just by ensuring you are not paying duty on a wrong code or an inflated value. It is also the one that reduces risk at the same time, because the same errors that overpay can underpay and create liability.
Knowing which lever applies is a data question
The reason generic duty advice fails is that the levers are flow-specific. The only way to know which to pull is to see duty broken down by product, origin, and tariff type: where it concentrates is where a lever can bite. High unclaimed FTA-eligible value points to preference; large duty on long-held or re-exported goods points to a zone or drawback; overpayment from inconsistent codes points to accuracy. That decomposition, and sizing each lever against the actual flow, is customs analytics, and it is what turns duty optimization from a pitch into a prioritized plan.
If duty is a growing cost, the first step is not choosing a tactic but seeing where your duty sits. Tell us what you import, or see how the platform models duty and the levers that lower it.
Frequently asked questions
What is duty optimization?
Duty optimization is the deliberate work of lowering the duty an operation pays, legally, by using the mechanisms customs law provides. It is not avoidance or misdeclaration; it is claiming preferences you qualify for, deferring or eliminating duty through zones and warehouses, recovering duty through drawback, and making sure classification, valuation, and sourcing are correct so you are not overpaying. Because each mechanism targets a different part of the duty, optimization is really about knowing which ones apply to your specific flow and how much each is worth.
What are the main ways to reduce import duty?
There are four broad levers. Preference: claim free-trade-agreement treatment on goods that meet the rules of origin. Deferral and elimination: hold goods in a Foreign-Trade Zone or bonded warehouse to defer duty, avoid it on re-exports, and sometimes pay a lower finished-good rate. Recovery: use duty drawback to reclaim up to 99% of duty on goods later exported or destroyed. Accuracy: correct classification, valuation, and sourcing so you are not overpaying on wrong codes or values. Each targets a different piece of the duty stack.
How do you know which duty-saving strategy applies?
By looking at your own data, not a generic checklist. The starting point is duty broken down by product, origin, and tariff type: that shows where the duty concentrates and therefore which lever bites. High FTA-eligible value that is unclaimed points to preference. Large duty on goods held long or re-exported points to a zone or drawback. Overpayment from inconsistent codes or values points to accuracy. The analysis sizes each lever against your actual flow, so you pursue the ones that are worth the effort and skip the ones that are not.