De Minimis Ended for Amazon Sellers: Consolidate, Do Not Order Bigger | Inventory Hero
·18 min readSourcing & Suppliers
De Minimis Ended for Amazon Sellers: Consolidate, Do Not Order Bigger
De minimis ended for Amazon sellers, but if you ship containers with a customs broker your per-entry cost never moved. Here is who it actually hit, and why consolidating entries beats inflating order size.
T. Brian Jones is co-founder and CTO of Inventory Hero. He leads the engineering behind its Amazon data pipeline, demand forecasting, and the AI platform that lets sellers talk to their live inventory, sales, and supplier data in plain language.
I import full containers through a customs broker. Did de minimis ending change anything for me?
Almost nothing on the entry side. De minimis was an $800 per-person-per-day threshold. Any commercial shipment worth more than that always required a formal entry, a broker, and duty, so a seller importing pallets or containers was already paying the per-entry cost. Your ordering-cost term did not move, and your economic order quantity did not move with it. What did change for you is everything upstream and downstream of the container: samples, air top-ups, and any parcels you or your supplier were shipping direct to US customers under the threshold. Also check whether your broker repriced you in 2026 for reasons unrelated to de minimis.
Did de minimis really end for Amazon sellers, and is it still gone?
Yes. The $800 Section 321 de minimis exemption ended for shipments from China and Hong Kong on May 2, 2025 and for all countries on August 29, 2025 under Executive Order 14324. It is still suspended as of August 2026. CBP issued an interim final rule on June 24, 2026 indefinitely suspending the exemption, and on August 13, 2026 the Court of International Trade held in Axle of Dearborn, Inc. v. Department of Commerce that the President has IEEPA authority to eliminate it. Separately, the One Big Beautiful Bill Act repeals commercial de minimis by statute effective July 1, 2027, so even if the executive action were unwound, the exemption has a legislative expiry date.
Should I order bigger now that every shipment needs a customs entry?
Only if your fixed cost per order actually rose, and even then by far less than sellers assume. EOQ scales with the square root of the fixed cost per order, so a doubling justifies roughly a 41 percent increase in quantity, not a doubling. Because the EOQ total-cost curve is very flat near its minimum, holding your old quantity typically costs single-digit percentages more per year. If cash or FBA capacity is your binding constraint, keep your quantity and consolidate SKUs into fewer entries instead.
I ship DDP and my forwarder is the importer of record. Does any of this apply to me?
Not directly. If your forwarder or supplier is the IOR and quotes you one delivered-to-Amazon price per unit, the entry fee is buried inside that per-unit number and you never see it as a fixed cost per shipment. In that case the entry fee is not sitting in your ordering-cost term at all, and re-running EOQ over it is fiction. Ask your forwarder to break the quote into per-shipment fixed charges and per-unit variable charges before you model anything. Also note what DDP costs you beyond money: the customs relationship, the classification decisions, and any future duty refund belong to the IOR, not to you.
How small is too small for an inbound shipment now?
A workable operator rule of thumb: keep total entry overhead (broker fee, entry filing fees, your admin time) under roughly 2 percent of the shipment's goods value. Divide your quoted per-entry cost by 0.02 to get your own floor. At the $150 to $250 per formal entry commonly quoted in 2026, that floor is roughly $7,500 to $12,500 of goods value. Between 2 and 5 percent, question the shipment. Above 5 percent, you are paying a lot to clear a small box. The fix is usually not a bigger order of one SKU, it is putting several SKUs on the same entry.
Does the end of de minimis change my duty rate or just my paperwork?
Both, but they land in different decisions. The entry requirement adds a fixed cost per shipment, which is the ordering-cost term in EOQ and nudges order quantity up. The duty itself is a variable per-unit cost that raises your landed cost, which raises the capital portion of your holding cost and nudges quantity slightly back down. Duty is second-order for ORDER QUANTITY only. In dollars it is usually far larger than any entry fee, and it belongs in your pricing and product-viability decisions, not in your PO-sizing math. Your actual rate depends on your HTS classification, so pull it from your broker or a CBP Form 7501, never from a blog.
If you import by sea in containers or on pallets and you already work with a customs broker, start here: de minimis ending did not change your per-entry cost. De minimis was an $800 threshold. A $6,000 shipment always required a formal entry, always needed a broker, and always paid duty. Your fixed cost per order went from roughly $250 to roughly $250. Nothing in your reorder math moved, and most of what you have read this year has been written as if it did.
What did change sits at the edges of your operation: samples, small air top-ups, and any parcels moving direct to US customers. This article is about those, about the shipment-size floor they imply, and about the one move that helps every importer whether or not they ever used de minimis, which is putting more SKUs on fewer entries.
Answer this before you touch a single purchase order. There are three profiles and the change lands very differently on each.
Your import pattern
Did your fixed cost per shipment move?
What to do
FCL or LCL ocean, pallets, existing broker
No. You always filed a formal entry. $250 to $250.
Nothing per SKU. Skip to consolidation.
Sub-$800 parcels: samples, air top-ups, China-direct DTC
Yes, hard. From roughly $0 in clearance cost to a full entry per parcel.
Set a shipment-size floor and a sample policy.
DDP where your forwarder is the importer of record
Invisible. It is bundled into your per-unit price.
Ask for the quote split before modeling anything.
That first row is the modal Amazon private-label seller, and for that seller the honest headline is that de minimis ending was a duty-rate event and a supplier-price event, not a reorder-cadence event. If a competitor's article told you to double your POs because of customs entries, and you have been filing formal entries for three years, that advice was written for someone else.
Yes, and the legal picture is more settled than it was a year ago.
Date
What happened
May 2, 2025
The $800 de minimis exemption ended for shipments from China and Hong Kong.1
July 4, 2025
The One Big Beautiful Bill Act (section 70531) repealed commercial de minimis by statute, effective July 1, 2027, and added civil penalties of up to $5,000 for a first improper Section 321 claim and up to $10,000 for each subsequent one, effective August 3, 2025.2
August 29, 2025
Executive Order 14324 suspended duty-free de minimis treatment for all countries.3
February 20, 2026
The Supreme Court decided Learning Resources, Inc. v. Trump and Trump v. V.O.S. Selections, Inc., invalidating the IEEPA tariffs at issue.4
February 20 to 21, 2026
Within a day of that decision, the President signed Executive Order 14388 continuing the de minimis suspension, on the reasoning that ending an exemption is not the same as imposing a tariff.5
June 24, 2026
CBP issued an interim final rule indefinitely suspending the de minimis exemption and moving it into regulation.6
August 13, 2026
The Court of International Trade, sitting as a three-judge panel, granted summary judgment for the government in Axle of Dearborn, Inc. v. Department of Commerce, holding the President does have IEEPA authority to suspend the exemption.1
One honest caveat: the CIT ruling is a trade-court decision on a contested question and the appellate path is not closed. It barely matters for planning, because the statutory repeal on July 1, 2027 arrives regardless.2 Plan as if $800 duty-free entry is permanently gone.
Section 321 de minimis
The provision that allowed shipments valued at $800 or less per person per day to enter the United States free of duty and without a formal customs entry. It is currently suspended for all countries and is repealed by statute for commercial shipments effective July 1, 2027.
Three places. None of them is your main container.
Test orders and samples. A 50 unit test batch that used to slip in under $800 now needs an entry. At $150 to $250 that is $3 to $5 per unit of pure clearance overhead, enough to make a viable product look dead on a landed-cost sheet. The fix is accounting discipline, not a bigger test. In your spreadsheet, keep two landed-cost columns per candidate SKU: cost per unit at test quantity and cost per unit at first production quantity, and only ever make the go or kill decision on the second one. Put the entry fee on a separate row in an ops or validation tab, dated to the month you spent it, not inside the SKU's per-unit cost. Better still, piggyback the samples onto an entry you were already filing, so they ride a fixed cost you were paying anyway.
Small air top-ups. Air-freighting a few hundred units to cover a gap now carries the same clearance toll as a container. Air is still usually right when you are about to run out, because a stockout costs rank and ad momentum that a few hundred dollars of entry fees does not touch. Just size the top-up so the overhead disappears into it rather than sending two 200-unit rescues three weeks apart.
China-direct legs on your other channels. If any part of your business was shipping direct to US customers from a Chinese warehouse under the threshold, that model is over, not more expensive. That is a channel-economics decision, not a reorder-quantity one, and it is the genuinely destroyed use case in this whole story.
Use a percentage, not a fixed dollar rule, because your entry cost is yours and not the one in this article.
Between 2 and 5 percent, question the shipment. Above 5 percent, you are paying a lot to clear a small box, and the answer is almost never "order more of that one SKU."
The government's own piece of this is small, for scale: merchandise processing fee on a formal entry is 0.3464 percent of value with an FY2026 floor of $33.58 and a ceiling of $651.50, and informal entry MPF is a flat $2.69 to $12.09.8 The broker's fee is the part that matters, and it is a market rate you negotiate.
This section is for the second row of that table: the seller who genuinely lived under the threshold. If you are the container importer, your inputs did not change and you can read this as background.
D is annual demand in units, S is the fixed cost of placing and receiving one order, H is the cost of holding one unit for a year. The key property: EOQ scales with the square root of S. Double the fixed cost per order and the optimal quantity rises about 41 percent. Triple it and it rises about 73 percent, not 200 percent.
H swings EOQ harder than S does, and most sellers hand-wave it with a generic 25 percent. Build it. For a small standard unit at 0.05 cubic feet with a $3.00 landed cost:
H component
Math
Per unit per year
FBA storage
0.05 cu ft x ((9 months x $0.78) + (3 months x $2.40))
$0.71
Cost of capital
$3.00 landed x 20 percent
$0.60
Markdown and aging allowance
$3.00 x 5 percent
$0.15
H
$1.46
Note that peak storage months are three times off-peak, so a unit that sits through Q4 carries most of its own storage cost in one quarter. If your item is bulky, slow, or seasonal, your real H is higher than a generic rate and your real EOQ is smaller. Our inventory carrying cost breakdown walks the components.
A 3,000 unit per year SKU, $3.00 landed, H of $1.46, previously topped up by air courier. Under de minimis the seller could not ship more than $800 of goods per parcel, which is 266 units. That is the part most coverage misses: de minimis was a cap on order size, not only a fee waiver. Old S was about $75, all courier handling and admin time, with no broker and no entry.
Under de minimis
After, same cadence
After, re-optimized
Fixed cost per order (S)
$75
$325
$325
Order quantity
266 (capped by $800)
266
1,156
Orders per year
11.3
11.3
2.6
Ordering + holding cost per year
$1,040
$3,859
$1,687
The number a bookkeeper will ask for is in the middle column: keeping the old cadence means 11.3 entries a year at $250 each, which is $2,820 of new entry fees on one SKU. That is the actual bill, and it is why this seller has to change something.
Re-optimizing to 1,156 units recovers most of it. But notice where the recovery comes from: the cap lifting, not clever EOQ tuning. And notice the new quantity, 1,156 units at $3.00, is a $3,468 shipment that still fails the 2 percent test on its own.
Order 800 units instead of the optimal 1,156, a 31 percent shortfall, and your annual cost is about $1,803 against $1,687. That is 7 percent worse for 31 percent less cash tied up. The EOQ total-cost curve is famously flat near its minimum, which is the single most useful fact in this article:
If cash and FBA capacity are not binding, move toward the new EOQ.
If cash is binding and you are choosing which SKU to fund this cycle, deliberately under-order. The shadow price of your cash is far above the cost-of-capital rate baked into H, and a few percent of ordering-plus-holding cost never justifies freezing another thousand dollars in cycle stock.
If your H is understated, fix H first. It moves the answer more than S does.
Run your own numbers in the EOQ calculator rather than trusting these inputs.
Quite a lot, and honesty about this is the difference between a model and a decision. Three omissions each outweigh a $250 entry fee for most private-label sellers:
Supplier price breaks. Textbook EOQ assumes unit cost is constant. It is not. Tiers at 500, 1,000, and 2,500 units routinely move unit cost by 5 to 15 percent, which on a 1,000 unit order is hundreds or thousands of dollars, and that is what actually determines PO size for most private label brands. If crossing a break is worth more than your whole ordering-plus-holding cost, the break wins and EOQ is a sanity check, not the answer.
Carton, pallet, and CBM quantization. "1,342 units" is not an orderable quantity. Your master carton might be 48 units, your pallet might be 30 cartons, and your forwarder prices by cubic meter with a minimum. Round to the packaging, then check the cost impact, which the flat curve tells you is small.
The LCL to FCL step. Freight is not linear. Moving from loose LCL cargo to a full 20 foot container is a step change of thousands of dollars in either direction, and it is a fixed cost that dwarfs entry fees. If a quantity increase pushes you across that line, price the freight before you price the entry.
And to be explicit about the ranking: duty is second-order for order quantity only. In dollars it is frequently the largest single line in this entire discussion, and it belongs squarely in your pricing, margin, and product-viability decisions. It just does not tell you how many units to buy.
This is the higher-leverage move for nearly every seller, including the container importer for whom nothing else here applied. The broker's entry fee is charged per entry, not per SKU, and additional HTS lines on the same entry typically carry a small per-line charge instead of a second full fee. So:
S per SKU = (entry fee + brokerage fixed charges) / number of SKUs + per-line charge + your own admin time
In the worked example, spreading a $250 entry across four SKUs with a $10 per-line charge and $75 of your own admin time takes S from $325 to about $148, which cuts that SKU's annual ordering-plus-holding cost from $1,687 to about $1,139. Roughly $548 a year saved on one SKU, financing zero extra units. Multiply by your catalog.
The execution is where this gets real. Eight SKUs from four factories do not become ready on the same day.
Batch to a calendar, not to each SKU's reorder point. Pick a window, every four to six weeks, and gather every SKU that hits its reorder point inside it. A SKU ordered slightly early loses a little holding cost. A separate entry loses the whole fixed fee.
Decide where goods wait, and who pays. Consolidation means finished goods sit somewhere until the slowest factory is ready. Your forwarder's China warehouse will usually hold free for a limited window, commonly around 7 to 14 days, then charge per CBM per day. Get that free-time window and the overage rate in writing, because a two-week hold on four factories' output can quietly exceed the entry fee you were saving. If the wait exceeds the free window, the saving is gone and you should ship what is ready.
Have a slip rule written down before a factory slips. The default should be: if the late SKU is more than the free-time window away from ready, ship the rest and let the straggler ride the next batch, unless that SKU would stock out first. Deciding this in the moment is how sellers end up paying for both the delay and the second entry.
Plan the Chinese New Year batch backwards. Production slots close for weeks and every factory slips at once. Consolidation windows around CNY should be set from the latest factory's committed slot, with a hard cutoff date after which you ship without stragglers.
Then check the three-way collision, because these constraints fight each other:
MOQ sets a floor per SKU per order. Calendar batching can push a SKU below its MOQ, in which case you order the MOQ and let it ride longer. See MOQ and how to work with it.
Restock limits and the low-inventory-level threshold push the other way: batching means larger, lumpier inbound quantities that can bump your FBA capacity and restock limits, while under-ordering to stay inside them can drop you below the historical-days-of-supply threshold that triggers the low-inventory-level fee.
Your consolidation window has to fit inside both. When it cannot, split the difference by sending the batch to a 3PL or AWD and drip into FBA, which decouples the customs entry from the FBA inbound entirely.
That last point is worth sitting with. The customs entry and the FBA inbound do not have to be the same event, and once you separate them, consolidating entries stops fighting your restock limits.
If you ship containers with a broker: change nothing about your order quantities on account of de minimis. Instead, count your entries for the last twelve months, and if that number is larger than the number of ordering windows in your calendar, you are paying for entries you could have combined.
If you were living under the $800 threshold: your ordering cost really did rise, so rebuild H properly, re-run EOQ, and then deliberately under-order relative to it if cash is tight, because the curve is flat. Set a shipment-size floor from your own entry quote divided by 0.02, and move samples onto entries you were already filing.
If you ship DDP: ask your forwarder to split the quote into per-shipment and per-unit charges. Until you have that, you cannot tell whether any of this applies to you, and you are trusting someone else's classification decisions with your money.
The customs change moved one term in one formula, for some sellers. It did not repeal supplier price breaks, freight step functions, or your cash constraint, and those were always the bigger levers.
Axle of Dearborn, Inc. v. Department of Commerce, U.S. Court of International Trade, Slip Op. 26-94, decided August 13, 2026 by a three-judge panel granting summary judgment for the government, holding the President has IEEPA authority to suspend the de minimis exemption because doing so subjects low-value goods to existing congressionally authorized rates rather than creating new duties. Slip opinions are published at https://www.cit.uscourts.gov/slip-opinions. The same litigation record confirms the China and Hong Kong suspension took effect May 2, 2025 and the worldwide suspension took effect August 29, 2025 under EO 14324. ↩↩2
One Big Beautiful Bill Act, Pub. L. 119-21, section 70531 (H.R. 1, 119th Congress): https://www.congress.gov/bill/119th-congress/house-bill/1. Repeals the commercial de minimis entry privilege for all countries effective July 1, 2027, and establishes civil penalties of up to $5,000 for a first violation and up to $10,000 for each subsequent violation involving improper Section 321 entry, effective August 3, 2025. ↩↩2
Learning Resources, Inc. v. Trump and Trump v. V.O.S. Selections, Inc., decided February 20, 2026, holding the IEEPA tariffs at issue unlawful. Opinion available through the Supreme Court's slip opinion index for the October Term 2025: https://www.supremecourt.gov/opinions/slipopinion/25↩
Executive Order 14388, "Continuing the Suspension of Duty-Free De Minimis Treatment for All Countries," signed within a day of the February 20, 2026 Supreme Court decision: https://www.federalregister.gov/executive-order/14388. Sources differ on whether it was signed February 20 or February 21, 2026 (Supply Chain Dive reported February 21; law-firm trade advisories report the same day as the opinion). The order states that the de minimis elimination is not affected by changes to the validity of previously imposed IEEPA tariffs. ↩
"Indefinite Suspension of the De Minimis Exemption for Merchandise Arriving Through All Modes Other Than the International Postal Network," interim final rule, Federal Register document 2026-12670, published June 24, 2026: https://www.federalregister.gov/d/2026-12670↩
Directional market rates compiled from published 2026 U.S. customs-brokerage pricing pages: roughly $150 to $250 per standard formal entry, higher for complex entries with many HTS lines, lower for informal entries (available for shipments not exceeding $2,500). These are third-party commercial rates, not published government fees, so there is no authoritative URL to cite. They vary by broker and volume commitment and they move. Get a written quote for your own account. ↩
U.S. Customs and Border Protection FY 2026 customs user fees under 19 CFR 24.23, effective October 1, 2025: merchandise processing fee for formal entries is 0.3464 percent ad valorem with a minimum of $33.58 and a maximum of $651.50, plus a $4.03 manual filing surcharge; informal entry MPF is a flat $2.69, $8.06, or $12.09 depending on entry class. Regulation text: https://www.ecfr.gov/current/title-19/part-24/section-24.23. Annual fee adjustments are announced at https://www.cbp.gov/trade/basic-import-export/user-fee-table↩