“I want to sell my products overseas. With cross-border e-commerce, we can start small, right?”
I hear this from Japanese founders all the time. Sake-based drinks, cosmetics, specialty foods. Japan is full of good products, and every time I listen to one of these founders, I think: this could work abroad.
My answer is always half yes. What has changed recently is that I can no longer say yes to the other half.
Let me be clear up front. This is not a “cross-border e-commerce is dangerous” post, and I am not here to scare you with regulations. I supported a company selling amazake (a traditional Japanese sweet fermented drink) overseas, I am in the middle of a formal cosmetics import registration right now, and we are building an e-commerce site for Singapore that will carry 300 Japanese products. What those projects taught me is a sorting exercise: what you can fix later, and what is decided at the entry point.
“Start small” is still good advice. With one catch
The standard advice in cross-border e-commerce is: don’t go big on day one, start small. That part still holds. When founders ask me, I still say: run a small test first.
The catch is that this phrase often carries an expired assumption. If “start small” means “ship individual parcels and just start selling,” that approach is running out of road.
Vietnam is the clearest example. On July 1, 2026, its new E-commerce Law (Law No. 122/2025/QH15) took effect, moving the supervision of cross-border e-commerce from decree level up to statute level. Platforms now carry stronger seller-verification duties, and the conditions for cross-border selling are spelled out.
There was a time when quasi-personal-import selling passed through a gray zone with rules that existed mostly on paper. From where I sit, that era is closing.
Here is the sorting line that matters most:
Clearing customs and being legal to sell commercially are two different things.
The parcel arrived, so we’re fine. That is not how it works. If you sell continuously, regulators and platforms see commercial distribution.
So I never use individual shipping as a sales channel. For me it is a sampling tool. It exists to test reactions and decide whether to take the next step.
Three things are decided at the entry point
So what exactly is “decided at the entry point”? In my experience it comes down to three things.
Let me first list what you can fix later: pricing, page design, how you run ads, which tools you use. All of that can change while you run. The entry point cannot.
1. Who takes responsibility in that country
To sell cosmetics in Vietnam, a responsible entity inside Vietnam has to place the product on the market, and each product needs a registration number (the cosmetic product proclamation). According to JETRO’s guidance, that number is valid for five years, and the application needs an ingredient list, a letter of authorization from the manufacturer, and proof of the local company’s registration.
After five years you re-apply. So this is not a one-time formality. It is a renewal cost you build into the long-term plan.
Which means the first question is not “can we ship from Japan?” It is “whose name carries the responsibility inside Vietnam?” A local import agent, a distributor, or your own legal entity. Until that is settled, everything downstream stays suspended: selling on your own store, listing on TikTok Shop, all of it.
A side note. We saw the same thing with influencer-led live commerce. Vietnamese influencers also want to confirm that responsibility sits with a domestic company before they touch your product. They do not want to get banned.
Responsible-entity design is protection for your partners as much as for you.
2. What crosses the border, as what
The same product can land in different HS codes (the international classification for traded goods), and the code determines the tariff.
So you do not leave this to chance. You design, in advance, what the product crosses the border as. If a cosmetic claims functional benefits, its classification shifts and different permits kick in. Amazake is similar: market it as a functional health drink and you are suddenly talking to the health ministry about advertising permits.
In the amazake project, one property saved us: the drink is non-alcoholic. In Vietnam, once a product falls into the alcoholic-beverage bucket, both tariffs and paperwork get heavy fast. Same “Japanese fermented drink,” completely different difficulty depending on what it enters as. This is the same pattern I wrote about in how exporting Japan-spec quality unchanged puts you on hard mode in Vietnam.
Logistics is the quiet third factor here. For chilled products like amazake, Vietnam has no equivalent of Japan’s nationwide cool-chain delivery. Whether the product can survive ambient-temperature distribution often decides how wide your channel can go, and that is a product-design question, not a marketing one.
3. How you say it (claim-wording design)
This one is live in our cosmetics project right now. In Vietnam, the moment you print “antibacterial,” “deodorizing,” or “improves” on a cosmetic, the product drifts toward pharmaceutical treatment. Registration gets much heavier.
Does that mean you cannot say anything? Not quite.
- “Antibacterial” becomes “balances the skin”
- “Deodorizing” becomes “odor care” (with a note that fragrance provides the cover)
You redesign the wording to fit inside the regulation. “Improves” gets the same treatment, shifting to “care” or “approach.”
You are not abandoning the appeal. You are translating it. Whether you do this wording design at the entry point changes every ad and every page that comes after.
One more entry-point job: negotiating ingredient disclosure. Registration requires concentration data, preservative percentages for example, and for OEM manufacturers those formulas are trade secrets. By default they will not disclose.
Which ingredients need exact concentrations? Is “above or below 1%” enough? There is no standard answer to what that 1% separates. It gets settled case by case.
If you skip this during product selection, you will hit the wall later.
So what reaction tells you to go next?
That covers the entry. You send samples. What do you watch to decide on full entry?
Trade shows work as a place to find partners. Meeting distributors and wholesalers still starts at trade shows more often than anywhere else. If you are planning a market visit around one, the seven phrases that win trust during a Vietnam business inspection may help.
As evidence for an entry decision, though, I do not trust trade-show reactions much. At a trade show, everyone says nice things. “What a wonderful product,” business cards change hands, and then nothing happens. It is remarkably common.
What I watch instead is the front line of distribution: retail stores. Will a physical store say “we will put this on our shelf,” and mean it with a purchase order rather than lip service?
Equally important, is the Japanese side ready to produce the paperwork, the letters of authorization and ingredient lists, quickly? We check this at the start of every project. Manufacturers rarely stall out of bad faith. Internal approvals and supplier relationships slow the documents down, and document lead time becomes your launch lead time.That check is the first thing we run in a market entry and research engagement, because document lead time is the part of the schedule you can measure before you commit.
Building the distribution network followed the same order. Courting wholesalers who hold inventory goes nowhere at first. A retailer places an order, and only then do wholesalers move. So the sequence becomes: test small via cross-border e-commerce, find stores that will stock you, and then distribution starts moving.
Designing the entry this thoroughly does slow you down compared with just shipping parcels. Registration numbers and authorization letters mean waiting, and the waiting lands on your timeline.
Is slower worse? I would argue the opposite. Doing it properly looks like a cost, but an ambiguous responsible entity generates invisible coordination costs that bleed for years and end up more expensive. And in Vietnam, online shoppers are seriously wary of counterfeits, so being able to say “officially imported” is itself a selling weapon on the shelf.
Why we build on Shopify
I will keep the promotional part short. We build this “start small, grow into official sales” entry on Shopify. Right now we are building a Singapore-facing e-commerce site carrying 300 Japanese products on the same foundation.
The reason is simple: market-level settings for currency, pricing, and shipping come built in, so a small sampling test and later official sales can run on one platform without rebuilding the site in between.
The honest weak point: the monthly fee is due even in months before you sell anything. It is not a free-to-try foundation, so we treat it as the fixed cost of never rebuilding from test to production. Tool talk can wait until the entry design is done.
One question to take home
This got long, so I will compress the takeaway into a single question.
When someone (including yourself) says “let’s start cross-border e-commerce,” pause once and ask: “Who takes responsibility in that country?”
If the answer comes instantly, your entry design is probably done. If the question stalls, start there, before the product and before the website.
In the amazake project and the cosmetics project alike, there were moments we simply could not advance without clearing it. My field feel at this point: something like eight-tenths of the outcome is decided before you sell anything.
Frequently asked questions
Q. Where should I start when checking cross-border e-commerce regulations?
First, who can be the responsible entity in the destination country. Then the product’s HS code classification and applicable tariff agreements. Last, whether your claims fit inside local regulation. In my experience, projects that run this order backwards stall in the second half. Carrying Japan-spec products in unchanged makes everything heavier, a pattern I wrote about in the Japanese-quality trap in Vietnam expansion.
Q. Is test-selling via individual parcels illegal?
It depends on the country and the product. But clearing customs and being recognized as commercial distribution are two different things. Vietnam’s E-commerce Law, effective July 2026, tightened supervision of cross-border e-commerce, and continuous selling is safer to design on the assumption that it counts as commercial distribution. We limit individual shipping to sample distribution. On choosing a partner for local regulatory work, see how to choose a consulting firm for Vietnam market entry.
Q. How do I judge whether my product suits cross-border e-commerce?
Can you instantly answer “who takes responsibility in that country”? Can the product survive ambient-temperature distribution? Can your side produce documents (authorization letters, ingredient lists) quickly? Those three give you a working answer. Seeing the retail floor with your own eyes beats spreadsheets, too. If you visit, seven phrases that win trust during a Vietnam business inspection may help. Still unsure? Use the session below and we will map it together.
If you want a first read on where the entry point sits for your own product, use the session below.
Want a sparring partner for your entry design?
I work with founders selling Japanese products into Vietnam and Southeast Asia: responsible-entity design, regulatory mapping, and building the store on Shopify. Straight talk, from someone running a company on the ground.
Book 30 minutes →
Shogo Harada原田 祥吾
CEO · Linnoedge Inc. · LinkedIn↗
Operating IT offshore development and overseas expansion support businesses across two bases: Tokyo and Vietnam. A leader who believes in “Systems over Spirit,” structuring cross-border businesses that often tend to be opaque. Committed to providing “reproducible quality” to organizations and clients rather than relying solely on individual skills.