A consolidation of 340 low-value parcels lands at Chicago O'Hare on a Tuesday morning. Eighteen months ago most of that freight would have cleared under the $800 de minimis threshold in a few hours, with no entry filed and nobody watching. Now every piece needs a formal or informal entry. The broker is missing an HTS classification on eleven lines, and the commercial invoice values on another thirty do not match what the buying system says was actually paid. The freight crossed 7,000 miles in two days. It sits at the gateway for four more.
Nothing went wrong in transport. What went wrong happened weeks earlier, in the data.
Customs stopped being a background process
Two dates changed the arithmetic this year. On June 24, 2026, CBP published a rule in the Federal Register indefinitely suspending de minimis treatment for merchandise arriving through any mode other than the international postal network, which means those shipments now require standard formal or informal entry procedures. On August 13, the Court of International Trade upheld the suspension, rejecting a challenge from importer Detroit Axle on the reasoning that withdrawing an exemption is legally distinct from imposing a new duty.
The operational consequence is simpler than the legal one. A category of freight that used to clear itself now files paperwork, and the queue it files into is busier. Brokers and forwarders report exam rates on flagged lanes running 15 to 25 percent above 2025 baselines, alongside CBP's stated FY2026 priorities of more air cargo examinations and tighter scrutiny of shipment valuations. The gridlock has been severe enough that DHL temporarily froze business-to-consumer shipments into the US above $800 rather than keep feeding a backlog.
For most teams this shows up as a new class of exception that their systems were never set up to catch. Ocean and air tracking tells you the freight arrived. It does not tell you the entry was rejected.
A customs hold is usually a data failure wearing a customs uniform
Strip out the genuinely random exams and the causes of a hold are repetitive:
Documents that arrive after the freight does. The commercial invoice, packing list, and certificate of origin exist somewhere, usually in an inbox. They reach the broker when someone remembers to forward them, which is often the morning after the flight landed.
Values and classifications that disagree across systems. The ERP carries one purchase price, the supplier's invoice carries another, and the HTS code in the item master was set for a product variant that changed two seasons ago. Any one of those mismatches is enough to stop a line.
Partner government agency requirements nobody flagged at order entry. FDA, USDA, and FCC requirements attach to the product, not the shipment, so they are knowable months before the goods move. They usually get discovered at the border instead.
Exams. Some are unavoidable. Their cost is not, because what makes an exam expensive is rarely the exam itself. It is the days that pass before anyone internally notices the shipment stopped.
The first three are all versions of the same problem: the facts needed to clear a shipment live in different systems, in different formats, owned by different teams, and nobody reconciles them until a broker asks. This is the gap a command center like Orkestra is built to close, by normalizing order, shipment, and document data from ERP, TMS, and carrier feeds into one record per shipment so classification, value, and origin can be checked before the freight moves rather than after it stops. Bridgestone used that normalization to hold $11.6K in weekly savings across 22 shipments, largely by making the underlying data accurate enough to act on.
The clock you are actually racing
A customs hold is rarely billed as a customs hold. It is billed as everything that accrues while the box cannot move.
Free time at the terminal keeps running during a hold. So does the chassis. Detention and demurrage start on their own schedules, and by the time the entry clears you are three days into charges that nobody authorized and no one can now avoid. We broke down how demurrage and detention actually accrue separately, because most teams discover the meter only when the invoice arrives. The same delay compounds into the surcharge stack on the freight invoice six to eight weeks later, at which point reconstructing what caused what is close to impossible.
This is the argument for treating a hold as an event to be worked in hours, not a status to be reported weekly. The duty owed does not change if you clear on day one or day five. Everything else does.
Give the broker one place to look
Most document chasing is not a diligence problem. It is an addressing problem. The certificate of origin is not missing; it is in a different person's email, attached to a message about something else.
Pinning every document a shipment needs to the shipment itself, where the broker, the forwarder, and the importer of record all read from the same record, removes an entire category of delay. Orkestra's document management works this way, holding commercial invoices, packing lists, certificates, and customs paperwork against the shipment record so the broker requests nothing and the compliance trail assembles itself. When an entry is rejected, the correction goes back to one place instead of into a thread.
Build the document set at order confirmation, not at departure. By the time freight is on a plane, every hour you spend gathering paperwork is an hour of dwell you have already committed to. That same upstream discipline pays off again the day a trade rule changes, because finding every affected shipment depends on classification and origin already sitting on the order line.
Escalate on the hold, not on the delivery date
The structural failure here is timing. Roughly 62 percent of supply chain exceptions are detected only after they have already affected the customer, and a customs hold is close to the purest example: the shipment is stationary, the cause is knowable, and nobody looks until someone downstream asks a question.
Watching for the hold itself changes what the team can do about it. When a status turns to hold or an entry comes back rejected, that is the moment to assign an owner, notify the broker, and start the correction, not the moment the promised delivery date passes. Orkestra's exception management detects those events as they land, prioritizes them by what they will actually cost, and routes each one to a named owner with the entry data and documents already attached. OIA Global cut exception resolution time by 60 percent in a single quarter working this way.
The same discipline applies past the border. Clearance is one handoff in a chain of them, and teams that lose shipments at customs usually lose them again at drayage, which is why tracking a container after it arrives at the port tends to be the next gap to close. In regulated freight the stakes compound faster still, as anyone keeping visibility on a pharma shipment across a border can attest, where a six-hour hold and a temperature limit interact badly.
Two numbers worth reporting
Most teams cannot say how often their shipments are held or how long clearance takes, which makes the problem impossible to argue about with a budget attached.
Track hold rate by lane, supplier, and product category, and track clearance cycle time from arrival to release. Those two numbers turn a diffuse complaint into a list of specific suppliers sending bad documents and specific classifications that keep getting questioned. Fix the top three and the rest of the tail usually shrinks with them.
What to do Monday
Pull the last 90 days of shipments that were held. Sort by cause. If the biggest bucket is documents or valuation, the fix is upstream in your data, not in your broker relationship, and it is achievable this quarter. Then set a rule that a hold raises an exception with an owner within the hour, the same way a missed cutoff would.
The de minimis window is not reopening. What separates teams that absorb that from teams that keep paying for it is not customs expertise. It is whether the data needed to clear a shipment is correct and assembled before the shipment arrives, and whether anyone finds out the same day when it is not. That is precisely the discipline Orkestra was built to make routine: see every shipment and exception in real time, across every carrier and system, early enough to act.
Book a demo to see how it works on your lanes.
