For more than a decade, ePacket was the invisible plumbing behind an enormous share of direct to consumer shipping out of China. A seller in Shenzhen could promise a customer in Ohio or Rotterdam a tracked parcel for a couple of dollars, and ePacket, not a dropshipping app or a marketplace algorithm, was the reason that price was possible. In 2026 that plumbing is visibly failing, and most sellers who still route parcels through ePacket by default have not rebuilt their shipping decisions around what the service actually is today.

At Zineps we watch cross border routing data across a large number of merchants, and the pattern this year is consistent. Parcels that used to clear in under two weeks are now taking three to four, the subsidised pricing that made ePacket the default cheap option has eroded, and the gap between what sellers expect from the service and what it actually delivers has widened enough to cost real conversion and repeat purchase rate. This is not a minor operational footnote. It is a structural shift in how low value cross border parcels move, and it deserves the same attention shippers already give to carrier surcharges, customs thresholds and marketplace delivery commitments.

What ePacket Actually Is, and Why It Got So Cheap

ePacket started in 2011 as a joint program between China Post, Hongkong Post and the USPS, designed to move small, light, tracked parcels from China into the United States faster and more reliably than ordinary international mail. It quickly became the backbone of the first wave of dropshipping and cross border e-commerce, because it offered tracking and reasonable speed at a price that had almost no relationship to the actual cost of moving a parcel across the Pacific.

That price gap existed because of how the Universal Postal Union has historically set terminal dues, the fees one country's postal operator pays another to deliver inbound mail. China was classified in a tier that let its postal operators pay far less than the real cost of last mile delivery in wealthier destination markets. Combine that subsidy with government support for outbound e-commerce logistics, and China Post could offer ePacket at rates that domestic carriers in the destination country could never match for a comparable parcel.

What Changed in 2026

Two forces converged this year and both were entirely predictable, which is exactly why so many merchants were caught unprepared.

The first is customs policy. The United States eliminated its de minimis exemption for low value imports across the board in 2025, closing the loophole that had let parcels under a certain declared value enter without formal duty collection. Every ePacket shipment now faces the same formal entry process as a commercial freight shipment, regardless of how small or cheap the item inside it is. The European Union followed with its own removal of the €150 duty free threshold, a shift we cover in detail separately, and the practical effect on ePacket specifically is that a service built around frictionless small parcel entry is now running through the same customs machinery as everything else.

The second force is the slower, less visible reform of the terminal dues system itself. The Universal Postal Union has been phasing in changes since its 2020 Congress that push self declared postal rates for major operators, China included, closer to actual cost. China Post and Hongkong Post have repriced ePacket multiple times over the past eighteen months as a result, and USPS itself briefly suspended then reinstated acceptance of inbound parcels from China during the messiest part of the policy transition, an episode many merchants still remember because it happened with almost no warning.

Put those two forces together and the arithmetic behind ePacket's old pricing simply does not hold anymore. Merchants who still think of it as a two to three dollar, two week shipping option are working from a memory of the service rather than its current state. Multiple fulfillment providers now report average delivery windows in the fifteen to twenty five day range for ePacket into the United States, stretching past thirty days during peak shipping periods, a long way from the seven to twenty day promise the service was originally marketed on.

What the Alternatives Actually Cost and Deliver Right Now

None of this means China origin dropshipping has become unworkable. It means the default choice needs to be re-earned rather than assumed, and the honest comparison between options looks different than it did three years ago.

• Yanwen remains the cheapest widely available option and the closest thing to a direct ePacket replacement on price. Industry benchmarks put economy transit at twenty to forty days with an on time rate around eighty percent, which is workable for low value, low urgency items but risky for anything a customer is actively tracking.
• 4PX typically prices ten to twenty percent above Yanwen on a comparable parcel, and in return delivers meaningfully better tracking granularity and dispute resolution support, which matters more than the price gap suggests once you account for support ticket volume.
• YunExpress has become the option most fulfillment providers point to for United States bound volume specifically, with reported transit times of seven to twelve days, daily tracking updates, a delivered duty paid option, and on time performance benchmarks above ninety five percent across roughly two hundred and twenty destination countries.
• Dedicated line or injection style services, where a parcel is flown in bulk and injected into a domestic postal or courier network close to the final destination, sit at the premium end but increasingly compete with ePacket on total landed cost once duty collection is unavoidable either way. We break down how this model works in our guide to injection shipping without a local warehouse.

The pattern across all four is the same trade off between price, speed and reliability that shippers already navigate with domestic carriers, just compressed into a smaller, faster moving market where the reputation each option earned two years ago is no longer a safe guide to how it performs today.

Why Betting on a Single Postal Channel Is a Structural Risk

Sellers who route one hundred percent of cross border volume through a single postal channel are exposed to the same concentration risk we have written about in the context of domestic last mile carriers. When Bol consolidated same day delivery onto a narrower set of partners earlier this year, the businesses that felt it hardest were the ones with no fallback option built into their routing logic. The same principle applies here: a single channel dependency means a repricing announcement, a customs policy change or a temporary suspension turns into an operational emergency rather than a routing adjustment.

That risk compounds when a parcel is headed toward a marketplace with a strict delivery performance requirement. Amazon's enforcement of its ninety percent on time delivery rate for FBM sellers has already forced many merchants to rethink how they route third party fulfillment, and a cross border parcel stuck inside a degraded ePacket lane is exactly the kind of shipment that quietly drags an account's performance metrics down without anyone noticing until the marketplace flags it.

A Simple Framework for Deciding Where Each Parcel Should Go

Rather than picking one service and defaulting every parcel to it, the merchants handling this well are routing on a small set of variables that can be evaluated automatically for every order.

• Declared value and duty exposure. Higher value items should generally move on a delivered duty paid service so the customer never receives a surprise customs bill after checkout.
• Weight and dimensions. Letterbox sized, lightweight items still benefit from postal economy channels, while anything approaching volumetric weight limits usually costs less on a commercial courier or injection service.
• Destination country reliability. Some postal administrations process inbound China origin volume far more smoothly than others, and that variance matters more than any single carrier's average performance number.
• Customer tier and delivery promise. A first time buyer with no delivery expectation set can tolerate a slower, cheaper channel more easily than a repeat customer who was quoted a specific delivery window at checkout.
• Return and dispute likelihood. Categories with higher return rates justify the extra tracking granularity and dispute support that a service like 4PX or a dedicated line option provides, even at a higher per parcel cost.

How Zineps Automates This Instead of Leaving It to Guesswork

This is precisely the kind of decision Zineps was built to make continuously rather than periodically. As the operating system for shipments, Zineps ingests live rate and transit performance data across postal and commercial cross border channels, and applies routing rules per parcel automatically based on declared value, weight, destination and customer tier, rather than relying on whichever service happened to be cheapest when a merchant last checked.

Smart shipping rules inside the platform mean that when a channel like ePacket degrades on a specific lane, routing shifts toward the next best option immediately rather than after a support ticket backlog reveals the problem.

The same multi-carrier shipping strategy that protects domestic delivery against a single carrier's bad week applies just as directly to cross border postal channels, treating ePacket as one input among several rather than a fixed default.

Carrier invoice auditing closes the loop by comparing what merchants are actually billed against contracted rates across every channel, including the postal services that rarely get the same scrutiny domestic couriers do, which matters more than ever now that terminal dues reform keeps shifting the underlying cost base.

The Underlying Lesson

ePacket was never really a shipping method. It was a temporary subsidy dressed up as one, and 2026 is the year that subsidy finally caught up with reality. The merchants who treat this as a one time inconvenience will keep defaulting back to the cheapest label on the screen and keep absorbing the consequences every time policy or pricing shifts again. The merchants who treat cross border routing as a live decision, evaluated per parcel against current data rather than last year's assumptions, will barely notice the next change when it comes, because their infrastructure was already built to route around it.

The Universal Postal Union publishes its terminal dues reform schedule and remuneration system details directly, and it remains the most reliable primary source for understanding where this pricing is headed next, rather than relying on secondhand summaries.

If your cross border shipping strategy still assumes ePacket behaves the way it did in 2022, talk to the Zineps team about building live, channel neutral routing that adjusts the moment a postal partner's economics change again.