How often should you
re-review HS classifications?
A classification code is not a one-time answer. Products change, suppliers substitute materials, and tariff schedules themselves get revised on a regular cycle. A code that was correct last year can quietly become wrong without anyone touching it. Here's how to think about review frequency, and which events should move a product to the front of the queue.
7 min read · Trade compliance upkeep
Why a correct code does not stay correct
Classification is often treated as a one-time task: research the product, assign the code, move on. That mindset works right up until something changes, and something almost always does, eventually. A code is only correct relative to a specific product, as it existed at a specific moment, under a specific version of a tariff schedule. Any one of those three things can shift without the code being touched.
The product itself can change quietly. A supplier substitutes one material for a cheaper one without flagging it as significant. A manufacturing process shifts. A bundled product gets a new component added. None of these show up on a customs form, but any of them can move a product into a different heading.
The schedule can change too, and on a much more predictable cycle. The World Customs Organization revises the shared six-digit Harmonized System roughly every five years, and individual countries revise their own extensions on their own schedules, sometimes annually. A code that was correct against last year's schedule can simply cease to exist, or get redefined, without any change to the product at all.
Event-based triggers, not just a calendar
The most reliable way to catch a stale code is to tie a review to specific events, rather than relying purely on a fixed calendar. A few of these matter more than others.
| Trigger event | Why it matters |
|---|---|
| Material or component change | Even a minor substitution can shift which heading applies, especially for products classified partly by composition. |
| Supplier or country of origin change | A new supplier may describe or manufacture the same nominal product differently enough to affect the classification. |
| New destination market | A code correct for one country's schedule is not automatically correct for another, even when the shared six digits match. |
| Tariff schedule revision | Codes can be renumbered, split, merged, or redefined during a scheduled update, independent of anything the company did. |
| Customs inquiry or audit | A question about one code is often a sign worth checking related codes, since the same misunderstanding tends to repeat across similar products. |
Any one of these events is a better reason to review a code than the simple passage of time. A product that has not changed in three years and sells into a market with a stable schedule may genuinely not need re-checking on a fixed cycle. A product that just swapped suppliers needs a look regardless of when it was last reviewed.
Setting a baseline review cadence
Event-based triggers catch most of what matters, but they rely on someone reliably flagging the event in the first place, which does not always happen. A baseline cadence acts as a backstop for whatever the trigger-based process misses. The right cadence depends less on the product and more on how much is at stake if the code is wrong.
| Risk tier | Suggested cadence | Typical examples |
|---|---|---|
| High duty exposure | Every 6–12 months | High-volume products, high duty rate categories, products near a heading boundary |
| Regulated or licensed goods | Every 6–12 months | Products subject to export controls, permits, or quotas |
| Standard catalog items | Every 12–24 months | Stable, well-established products with a clear, uncontested heading |
| Low volume, low duty | On schedule revision only | Infrequently shipped items where the cost of a full review outweighs the exposure |
None of these numbers are fixed rules. They are a starting point for allocating limited review time toward the products where an error is most expensive, rather than spreading review effort evenly across a catalog where most products carry very little risk.
What an actual review should check
A useful review is not simply re-reading the existing code and confirming it still sounds right. It re-verifies the same three things that made the original classification correct in the first place.
- The product description on file still matches the product as currently manufactured and sold, not just its original spec.
- The heading and subheading originally chosen still exist and mean the same thing under the current version of the schedule.
- Any national extension past the shared six digits is still correct for every destination market the product currently ships to, including markets added since the last review.
- The documented reasoning behind the code is still on file and still makes sense against the current product and schedule.
If any of these four have changed since the last review, the code needs to be re-derived, not just re-confirmed.
Who should be doing the reviewing
Ownership matters more than most companies plan for. A review process with no assigned owner tends to happen only in response to a problem, which defeats the point of reviewing proactively in the first place.
Whoever owns supplier relationships or product data is usually best positioned to flag a material or supplier change, since they see it first. That flag should route directly to whoever owns classification, not sit in a general change log nobody checks.
Tracking tariff schedule updates across every destination market a company ships into is a specialized, ongoing task. It is one of the areas where automated monitoring earns its keep, since it removes the dependency on one person remembering to check.
The periodic backstop review is the easiest to let slip, since nothing forces it. It needs an actual owner and a place on a calendar, the same way any other recurring compliance task would.
Common pitfalls
Reviewing the code without reviewing the product
Confirming a code still exists in the schedule is not the same as confirming it still matches the product. The product side of the check gets skipped more often than the schedule side.
Only reviewing after a customs inquiry
Waiting for a question from customs to trigger a review means the company is always reacting, never ahead of the problem, and by then the exposure may already cover years of past shipments.
Applying the same cadence to every product
A flat review schedule across the whole catalog wastes review time on low-risk products while under-reviewing the small number of products that actually carry meaningful exposure.
Not tracking new destination markets as a trigger
Shipping an existing, already-classified product into a new country is treated as routine, when it actually requires the same national-extension check as classifying a brand-new product.
Getting started
Putting a review cadence in place does not require reviewing the entire catalog immediately. It works better sequenced:
- Identify which products carry the most duty or regulatory exposure, and review those first, regardless of when they were last checked.
- Set up a way to catch material, supplier, and new-market changes as they happen, rather than waiting for a periodic review to surface them.
- Assign an owner for tracking tariff schedule updates across every destination market currently in use.
- Put the remaining backstop cadence on an actual calendar, tiered by risk, rather than leaving it as an informal intention.
The goal is not to review every product constantly. It is to make sure the products where being wrong is expensive are never far from a check, while the rest of the catalog gets a sensible, lower-effort backstop.
Let classifications
stay current automatically.
Enthron monitors tariff schedule updates across dozens of countries and flags products for review the moment a material, supplier, or destination market changes, so codes never go stale unnoticed.