Trading Company, Distributor, or Direct: Selling Into Japan

Once a foreign manufacturer understands how a Japanese buyer decides, the next question is mechanical: through whom do you actually sell? For most technical B2B companies the honest comparison is trading company vs distributor in Japan, set against the third option of building your own entity. The choice sets your margin, how much control you keep, how fast you reach a first sale, and, most consequentially, who owns the relationship with the end customer. This piece explains why the famous general trading houses are probably not your route, what a specialized trading company is and how to recognise the one you are talking to, and how the three routes compare when you put them side by side.

The route you probably do not need: the sogo shosha

The first thing most people picture is a sogo shosha (総合商社), a general trading company. Japan has seven majors: Mitsubishi, Mitsui, Itochu, Sumitomo, Marubeni, Toyota Tsusho and Sojitz. They are not distributors in any sense a Western exporter would recognise. They act at once as traders, investors, project developers and risk managers, and they own stakes in the mines, gas fields and plants they trade from, which is precisely why they cannot easily be cut out of a supply chain. The scale is easiest to see in one of their own filings: Marubeni’s company profile lists 121 branches and offices in Japan and overseas as of 1 April 2026, sitting on top of 324 consolidated subsidiaries and 153 affiliates as of 31 March 2026.

That scale is the reason a mid-size supplier rarely belongs on their desk. A supplier turning over EUR 20-300M is, to an organisation carrying several hundred subsidiaries across every sector from energy to food, not a counterparty worth a dedicated relationship, however good the product. The point does not turn on any single figure; it turns on the gap between your order book and the smallest transaction that gets a general trading house’s attention. There is one real exception, and it is the bridge to the next section: many specialized trading companies are themselves subsidiaries of a sogo shosha, so you may end up inside that group anyway, but through a category arm that operates at your scale rather than through the parent.

The realistic partner: the specialized trading company (senmon shosha)

The partner that fits a technical B2B manufacturer is usually a senmon shosha (専門商社), a specialized trading company that concentrates on one field or product category. The distinction is not cosmetic. A specialized house has category expertise, existing relationships with the specific buyers you are trying to reach, and a book small enough that a new line can matter to it.

What a senmon shosha actually is

A senmon shosha is generally defined as a trading company where a single field or product accounts for more than half of revenue. That concentration is the asset you are buying. Where a sogo shosha spreads across everything from energy to food, a specialized house lives inside, say, industrial fasteners, analytical instruments, or a class of electronic components, and it already sells to the engineers and purchasing departments you would otherwise spend two years finding. It is far more accessible to a smaller foreign supplier than a general trading house, because your line is closer to the centre of its business rather than a rounding error at its edge.

Size is worth thinking about in relative rather than absolute terms. You will see specific revenue thresholds quoted for the “right size” of trading partner, but the ones in circulation trace back to marketing material rather than to any authoritative source, so we do not repeat them here. The useful test is relational, not numerical: the partner should be large enough to have genuine, active relationships with the buyers you want, and small enough that adding your line is a decision someone there will actually champion rather than file. A house where you would be one small account among giants gives you reach without attention; a house too small to reach your target buyers gives you attention without reach. The fit you want sits between those, and you find it by asking who they already sell to, not by asking their turnover.

How to tell which one you are talking to

Specialized trading companies come in three broad types, and knowing which you are dealing with tells you a lot about its incentives and reach. Some are subsidiaries of a sogo shosha, which gives them a large parent’s financing and logistics behind a category focus. Some are manufacturer-affiliated, tied to a particular maker, which can mean deep channel access but also a house view about which products fit alongside their parent’s. And some are independent, owing allegiance to no group, which usually means the widest freedom to champion your line and the most direct read on whether they will actually push it. When a Japanese counterpart introduces themselves as a shosha, the useful question is which of these three they are, because a manufacturer-affiliated house and an independent will treat your product very differently. The question that separates them fastest is not about structure but about accounts. Ask which buyers in your category they currently supply, who at those buyers they deal with, and which competing or adjacent lines they already carry. A house that answers with named accounts and named functions is describing a business it actually runs. A house that answers with coverage claims and market size is describing one it would like to run.

What a trading company does that a Western distributor does not

It is easy to look at a trading company’s margin and see a middleman markup. That misreads what you are paying for. A Japanese trading company typically does several things a Western distributor often does not bundle together. It extends trade credit, financing the gap between when goods move and when the end customer pays. It carries inventory and currency risk on its own book. It handles customs, logistics and the import mechanics that a first-time entrant underestimates. And in technical categories it sometimes provides engineering or after-sales support to the end customer on your behalf. The margin, in other words, is the price of financing, risk-bearing and category access, not a toll for passing your product along. Whether that price is worth it depends on how much of that work you could realistically do yourself from abroad, which for a company with no Japanese-speaking staff is usually not much. A useful way to test the margin is to list each function the partner performs and ask what it would cost you to replicate it, in cash and in management time; the ones you cannot replicate at all are the ones you are genuinely buying, and the margin looks very different once that list is in front of you.

The three routes side by side

Set the options next to each other and the trade-offs become concrete: a specialized trading company, a direct distributor agreement, or your own entity selling directly. The EU-Japan Centre’s guidance for European companies is blunt about the starting point: most smaller firms cannot justify a branch office and therefore operate through Japanese partners, and those partners usually cover a defined territory, market or industry, so you may end up contracting more than one.

Margin and who gets paid for what

A trading company or distributor takes the largest cut, because it carries the most, credit, inventory, risk and access. A pure agent, who introduces customers for a commission without taking title to the goods, takes less but also carries less and gives you less cover. Your own entity keeps the whole margin and pays for it in fixed cost and management attention. There is no free option here; there is only which costs you are willing to convert from variable to fixed.

Control and who owns the end customer

This is the trade-off that matters most in the long run. Through a distributor or trading company, the end-customer relationship is largely theirs. They hold the account, they field the questions, and if the arrangement ends, the customer often stays with them. The EU-Japan Centre’s report on structuring these deals draws the line clearly between a distributor, who buys and resells on its own account, and an agent, who introduces on commission while you keep title and, with it, more of the direct relationship. Your own entity gives you the customer outright, at the cost of building everything behind it. Decide early how much you are willing to hand over the relationship, because it is far harder to take back than to give.

Speed to first sale

Speed runs the other way from control. A specialized trading company already inside your category can produce a first conversation with a real buyer in weeks, because it has the relationships you would otherwise build from zero. A direct distributor is a little slower, because you must find and qualify it first. Your own entity is slowest of all: incorporation, a first hire, and a market presence built from nothing, before the first order. If the strategic goal is to test whether Japanese demand is real before committing capital, the partner routes buy you that test cheaply.

When your own entity is the right answer anyway

The own-entity route is not the loser in this comparison; it is the endgame for a company that has decided Japan is a market it intends to hold rather than test. It becomes the right answer once the volume justifies the fixed cost, once you need to own the customer relationships outright rather than rent them, and once losing control of the technical conversation to an intermediary would cost you more than the overhead of running your own presence. The sequencing question is what matters. Many companies use a partner route to prove demand, then build an entity once the numbers are in, folding the early relationships into their own operation where the agreements allow it. That transition is a large enough decision on its own that we treat it separately in our guide to Japan market entry; the point here is only that the three routes are often a sequence, not a permanent fork.

Exclusivity, territory, and the real multi-line problem

A common worry is that a Japanese distributor will carry your competitors alongside you. In practice the opposite structure is common for foreign brands. The Practical Law country guide to distribution in Japan describes exclusive distribution as both common and permissible, and clauses restricting a distributor from handling competing products during the term of the agreement as enforceable, subject to reasonableness review under the Antimonopoly Act. So a direct competitor on the same shelf is generally not the risk.

The real multi-line problem is subtler. A distributor’s portfolio is broad even when nothing in it directly competes with you, and your line is one of many claims on a finite sales force’s attention. Exclusivity is also commonly tied to annual performance targets, which means underperformance can cost you the territory, or trigger a review of the exclusivity, without ending the whole agreement. The question to ask before you sign is therefore not only “will you carry a competitor” but “where does my product sit among everything else you represent, and what happens to my territory if the first year is slow.” Those terms are a matter for your own legal counsel to negotiate and draft; the point here is to know which levers exist before you reach that table.

Who has the technical conversation once a partner sits in the middle

Every route that puts a partner between you and the buyer creates the same structural question: who conducts the specification-level conversation that actually decides a technical sale? Once a trading company or distributor is the face of your product, the detailed exchange with the customer’s engineers is mediated by someone who is not you and who may not be a specialist in your category. That is the same failure that explains why a right-priced quote can still lose in Japan: the technical argument that would have won never reaches the people scoring it. As inference from practice rather than a documented rule, your shosha or distributor contact is frequently the person who has to trigger or write up the internal proposal inside the end customer, the document we cover in how the ringi approval process works. If that is so, then arming your partner is not optional. What they can put in front of the buyer, in Japanese, is only as good as what you give them, which is the subject of what Japanese buyers want in your technical documentation.

This is the part that separates a route decision from a route disaster. Choosing a partner does not remove your technical responsibility; it changes who carries your technical case into the room, and it puts a premium on giving that partner material strong enough to survive being quoted without you present. A partner who understands the commercial terms but cannot answer an engineering question, and whom you have not equipped to get the answer quickly, will lose deals you would have won face to face.

What to do now

• Rule the sogo shosha in or out honestly. Unless you are already inside one through a category subsidiary, assume the general trading houses are the wrong scale and focus on specialized houses in your product category.
• Identify which of the three senmon shosha types you are talking to, a sogo subsidiary, a manufacturer-affiliated house, or an independent, before you read anything into their enthusiasm.
• Price all three routes on the same four axes: margin, control, speed, and ownership of the customer. Do not compare a distributor’s margin to your own without also pricing what the distributor carries that you would have to.
• Before signing an exclusive, settle where your line sits in the partner’s portfolio and what happens to your territory if year one is slow. Take the drafting to your own legal counsel.
• Whichever route you choose, prepare the Japanese-language technical case your partner will carry, and name who answers an engineering question and how fast.

Key takeaways

• The sogo shosha is usually the wrong route: the seven majors operate at a scale where a EUR 20-300M supplier is not a material counterparty, unless you reach them through a category subsidiary.
• The senmon shosha is the realistic partner: a specialized trading company brings category expertise and existing buyer relationships, and your line is meaningful to its book.
• Trading-company margin is not a pure markup: it pays for trade credit, inventory and currency risk, customs and logistics, and sometimes engineering support.
• Control and speed pull in opposite directions: partners are faster and cheaper to start but own the customer; your own entity keeps the relationship at a much higher fixed cost.
• The multi-line risk is attention, not a competitor: exclusivity with a non-compete is standard; the real question is where your line sits in the portfolio and what performance targets attach to your territory.

What to read next

• Japan Market Entry: The Complete Guide for 2026
• Doing Business in Japan: A Practical Guide for Foreign Companies
• Nemawashi: What Happens After Your Japanese Inquiry Goes Quiet