Is it cheaper to keep a Japan marketplace store dormant than to close it?

It depends entirely on which platform, and the difference between them is larger than most brands expect. On Amazon Japan a genuine dormant state exists: the Individual plan costs 100 yen per item sold rather than the 4,900 yen monthly fee of the Professional plan, and Amazon's own pricing page states that changing or cancelling a selling plan is possible at any time. A brand that stops advertising, moves to Individual and lets a handful of orders trickle in pays close to nothing to hold the position. Rakuten Ichiba is structurally different: the store plans carry a stated contract period of one year, and the monthly plan fee is not billed monthly — Rakuten's own fee page says it is paid either in two six-month instalments or as a single annual payment depending on the plan, at 25,000, 65,000 or 130,000 yen per month before tax by tier. Money already paid for a period you intend to sit out is not recoverable by going quiet. Yahoo! Shopping used to be the cheapest place to stay dormant because there was no monthly fee at all; from September 2026 Yahoo!'s own announcement introduces a 10,000 yen monthly system usage fee (before tax) and a 2.5% sales royalty, so dormancy there acquires a price it did not previously have. Work out your own answer per platform before you assume that doing nothing is free.

Will I lose my Japanese trademark if I stop selling in Japan?

Not automatically, but you become vulnerable to losing it, and the mechanism is harsher than most foreign brands realise. Article 50 of Japan's Trademark Act provides that where a registered trademark has not been used in Japan for the designated goods or services by the owner, an exclusive licensee or a non-exclusive licensee for three consecutive years or more, any person may petition for a trial to cancel the registration. The clause that matters most is in the second paragraph: unless the respondent — that is, you, the registrant — proves use of the registered trademark within the three years before the petition was registered, the owner cannot escape cancellation. The burden of proof sits on the trademark holder, not on the party attacking it, and the statute allows an escape only where the respondent shows a justifiable reason for non-use. The third paragraph closes the obvious workaround: use started within three months before the petition, after you learned the petition was coming, does not count as use — unless the respondent shows a justifiable reason for having made that use. The practical consequence for an exit decision is that a plan to leave now and re-enter later has a clock attached to it, and the paperwork proving you kept selling something in Japan is the thing that protects your re-entry option. We are not a law firm and this is not legal advice — whether any specific arrangement counts as use is a question for a Japanese benrishi.

Do I still have to deal with Japanese consumption tax after I close my store?

Very likely yes, and the registration does not quietly lapse because you stopped selling. If you are registered under the qualified invoice system, cancelling that registration is a filing with a deadline attached: the National Tax Agency states that to make the registration lose effect from the first day of the next tax period, the notification must be submitted by the day falling 15 days before that first day, and that a submission made after that point takes effect from the first day of the period after next instead. Ceasing business entirely is a separate notification from cancelling invoice registration, so check which one your situation actually calls for. Separately, a foreign company with no office in Japan is still a consumption tax payer where it makes taxable transfers in Japan and its base-period taxable sales exceed 10 million yen, and the tax agent (nozei kanrinin) you appointed to file on your behalf has to remain in place for the final filings rather than being released the day the store closes. We are not a tax agent or a licensed tax professional — sequence this with your zeirishi, and see our guide to Japan consumption tax for foreign sellers for the registration side.

If I close a Rakuten or Amazon store, can I reopen later and keep my reviews?

Treat the answer as no unless your own contract says otherwise. Reviews, ratings and sales history attach to the store account rather than to your brand, which is the same mechanic that governs what you keep when you change agencies — if you keep the account and only change who operates it, the history survives; if a new account has to be opened, you start from zero regardless of how much history the old one had. What we could not establish is the official position on what happens to a Rakuten storefront and its accumulated reviews after a store closes: Rakuten's merchant terms are not published publicly, and we found no first-party statement on post-closure treatment of store pages or review history. Several Japanese agency blogs describe both the notice period and the closure mechanics confidently, but that is secondary commentary and we are not going to present it as policy. The honest instruction is to get the answer from your own contract and from your platform contact in writing before you close anything, not from an English article — including this one.

How long should a Japan marketplace store run before I decide it has failed?

There is no official or industry-standard answer, and any article that gives you a confident number is inventing it. What we can say is what makes a Japan-specific judgement different from a generic one. Japanese marketplace demand is heavily event-shaped — Rakuten Super Sale, Prime Day, the year-end and New Year period — so a store that has not traded through at least one full cycle of the events that matter in its category has not yet produced a readable number, and a store judged only on its launch quarter is being judged on the period when its ranking, review count and repeat-purchase base are structurally at their weakest. Against that, waiting is not free: you are paying a fixed floor every month, and on Rakuten paying it in advance. The framing we use with brands is not a duration at all but a decision date set in advance — pick the point at which you will run the arithmetic in this article, define what result would make you stop, and write both down before the emotional cost of the decision has accumulated. The broader market is not the thing failing: Japan's BtoC ecommerce market reached 26.1 trillion yen in 2024, up 5.1% year on year, so a channel that is not working is generally a fit or cost-base problem rather than a shrinking-market one.

When to Exit a Japan Marketplace: A Shutdown Decision

A framing note before the detail. A Japan channel that is not working is usually not a story about a shrinking market. Japan's BtoC ecommerce market reached 26.1 trillion yen in 2024, up 5.1% year on year, with an ecommerce penetration rate of 9.8% in the goods category, according to the Ministry of Economy, Trade and Industry's market survey published in August 2025. That removes one comfortable explanation. If the channel is underperforming, the cause is your fit with it, your cost base inside it, or the local operating attention it is getting — all things a decision can act on, and one of them fixable without leaving at all.

"Exit" Is Four Different Decisions, and Brands Conflate Them

Almost every conversation we have about this starts with a brand saying it wants to pull out of Japan, and ends with the discovery that it meant something much narrower. The four decisions below have wildly different costs and wildly different reversibility, so name yours before you price it.

Stop feeding it. Advertising off, no more inbound inventory, catalogue left standing. Reversible in days, costs you the fixed platform floor and nothing else.
Downgrade to the cheapest maintained state. The plan tier drops, the store stays open, the account and its history stay yours. This is the option that gets skipped, and it is the subject of its own section below.
Close one channel. One storefront closes; the Japan business continues elsewhere — your own site, a distributor, another marketplace. Your tax position and your trademark exposure may not change at all.
Leave the Japanese market. Every channel closes. This is the only one of the four that triggers the tax, records, trademark and data obligations in the second half of this article, and it is the one brands most often say when they mean one of the first three.

This distinction earns its place first because the cost of the fourth decision is routinely used to justify staying in the first. "Closing Japan is expensive" is true of a full market exit and largely untrue of dropping one underperforming marketplace, and conflating them keeps brands paying a fixed monthly floor for a channel they have already privately given up on.

The Arithmetic That Says Stop

There is no industry benchmark that tells you when a Japan store has failed, and anyone quoting one is making it up. What exists instead is a set of calculations you run on your own numbers. Do them in this order, because each one can end the exercise.

1. Find the floor you pay whether or not you sell anything

This number is knowable exactly, and a surprising number of brands have never written it down. As of August 2026, the published fee structures are:

Rakuten Ichiba — three store plans at 25,000 / 65,000 / 130,000 yen per month before tax, plus a 60,000 yen initial registration fee. Rakuten's own fee page states a contract period of one year, and — this is the part that decides the exit maths — that the monthly store fee is not paid monthly: it is settled either in two six-month instalments or as a single lump sum depending on the plan.
Amazon Japan — 4,900 yen per month before tax on the Professional plan, or 100 yen per item sold on the Individual plan, with Amazon stating that changing or cancelling a selling plan is possible at any time.
Yahoo! Shopping — no initial fee, no monthly fee and no sales royalty today; from September 2026, per Yahoo!'s own announcement, a 10,000 yen monthly system usage fee (before tax) and a 2.5% sales royalty apply, while the mandatory 1.5% campaign contribution borne until August 2026 ceases.

The Rakuten prepayment structure is the most under-appreciated fact here. On a plan billed six months ahead, going quiet in month two does not save you months three through six — that money is spent. A Rakuten exit decision therefore has a date attached to it, and making it two weeks before a billing period rolls is worth materially more than making it two weeks after.

2. Subtract the mandatory take, not the headline rate

Every one of these platforms has a compulsory variable layer beyond the commission most brands quote to themselves. On Rakuten that includes a system usage fee that varies by plan and device, point programme funding, an affiliate reward, a mall-safety system fee set at 0.1% of monthly sales, and payment processing. On Yahoo! the mandatory minimums stack into a range we work through in detail in the Yahoo! Shopping operations guide. Amazon's referral fees and the wider comparison sit in our breakdown of Japan ecommerce pricing models. The test is not "is my gross margin positive" but what is left after the compulsory take, the fixed floor and the operating hours the channel consumes — and whether that residue is bigger than what the same hours would earn deployed anywhere else.

3. Test the trend, not the month

Three questions, each of which needs at least one full seasonal cycle of data to answer honestly:

Is contribution improving, flat or decaying? A channel losing money while improving each quarter is an investment. The same loss on a flat line is a subscription to a mistake.
Has the cost base moved under you? Fee changes are exogenous — the September 2026 Yahoo! change is a live example — and a channel that penciled out under the old schedule may not under the new one. Rerun the floor calculation on the new numbers rather than on the ones you signed up to.
What is the channel consuming in attention? Japanese-language customer messages, review responses, campaign entries and platform compliance work do not scale down with revenue. A small store can consume the operating hours of a large one, and those hours have a price even when no invoice names them.

4. The reopen test

The cleanest question in the set, and the one that cuts through sunk cost: if this channel were not already in your account list, would you sign up for it today, at today's published fee schedule, knowing what you now know about your category's performance in it? If the answer is no, you are not deciding whether to exit — you have decided, and the only remaining question is how to do it without paying more than you need to. The September 2026 Yahoo! change hands every seller on that platform this exact question, whether they wanted it or not.

Running these numbers on a Japan store right now? Tell us which platforms you are on, what the last twelve months look like and what the store costs you in hours — we will tell you which of the four decisions we think you are actually facing.

Talk to LAUNOVA

The Middle Option Nobody Prices for You

Between "keep funding it" and "close it" sits a state that is cheap on some platforms and expensive on others, and the asymmetry is not obvious from the outside. Platforms advertise how to join. None of them advertise how cheaply you can stay.

Amazon Japan: a real dormant state exists. Moving from the Professional plan to Individual replaces a 4,900 yen monthly fee with a 100 yen per-item charge, and Amazon states that plan changes and cancellations are possible at any time. Separately, the Amazon Services Business Solutions Agreement states that a seller may terminate its use of the selling services immediately at any time by notifying Amazon, while Amazon itself must give 30 days' notice to terminate for convenience. Your side of that exit has no notice period.
Rakuten Ichiba: dormancy is not a product. The one-year contract term and advance billing mean the cost of an idle store is the cost of an active one. We could not find any first-party Rakuten statement of the notice period required to close a store; the merchant terms are not published, and the only public trace we could confirm is that Rakuten's own transparency-law disclosure page references clauses of those terms by number, including a clause on cancellation and termination. Japanese agency blogs state a specific notice requirement with confidence, and we are deliberately not repeating the figure, because we could not source it to Rakuten. Read your own contract, and get the answer in writing from Rakuten before you plan around any number.
Yahoo! Shopping: the dormant state is being repriced. Free-to-hold until August 2026; from September 2026, holding it costs 10,000 yen a month before tax plus royalty on whatever it does sell.

One asset does not have a dormant state on any platform: inventory. If your stock is in FBA, deciding to go quiet does not stop the meter. Amazon Japan's fee schedule sets removal and disposal at the same per-unit price — 60, 90, 130 or 200 yen per unit for standard-size items by weight band, and 170 to 870 yen for large and oversize, tax included — so there is no saving in abandoning stock rather than shipping it back, and the choice is a pure logistics one. Long-term storage surcharges also start earlier than most sellers assume: Amazon's help documentation puts the first surcharge band at 271 days, assessed on an inventory snapshot taken on the 15th of each month. One discrepancy we are not going to paper over: Amazon's own public pricing page still carries a footnote referring to inventory stored beyond 365 days, while the seller help page describes the 271-day band. Both retrieved 13 August 2026 — confirm the current bands in your own Seller Central before planning a removal schedule. The wider FBA-versus-warehouse trade-off is in our FBA versus 3PL comparison.

What a Full Exit Costs That Entry Never Warned You About

These apply to the fourth decision — leaving the market — and to a surprising extent to the third. They are the reason "we will just close it and think about Japan again in a few years" is not the cost-free option it sounds like.

Your Japanese trademark has a three-year clock

This is the most consequential item on the list and the one least often connected to an exit decision. Under Article 50 of Japan's Trademark Act, where a registered trademark has not been used in Japan for the designated goods or services by the owner, an exclusive licensee or a non-exclusive licensee for three consecutive years or more, any person may petition for a trial to cancel that registration. The second paragraph places the evidential burden on the registrant: unless the respondent proves use within the three years preceding registration of the petition, the owner does not escape cancellation, the only exception being where the respondent demonstrates a justifiable reason for the non-use. The third paragraph forecloses the obvious fix — use begun within three months before the petition, after learning it was coming, does not count, unless the respondent shows a justifiable reason for having made that use.

Read against an exit plan, that means the option to come back later is not free-standing. It depends on maintaining and documenting genuine use in Japan, and the documentation burden falls on you rather than on whoever challenges you. Brands that intend to return frequently keep a minimal channel alive for exactly this reason — which is another argument for pricing the downgrade option properly before treating closure as the only alternative to full funding. Whether any particular arrangement constitutes use is a legal question for a Japanese benrishi; we are not a law firm and do not advise on it.

The tax registration does not close itself

Stopping sales does not end a consumption tax obligation. If you hold a qualified invoice registration, the National Tax Agency requires the cancellation notification to be filed by the day falling 15 days before the first day of the tax period from which you want the registration to stop applying; file later than that and it takes effect from the start of the period after next instead — a full extra period of being a registered business. Ceasing business altogether is a separate notification from cancelling invoice registration. And the tax agent a non-resident seller appoints does not become unnecessary the day the store closes: the final filings still have to be made by someone, and that someone is the appointed agent. The registration side of all this is in our guide to Japan consumption tax for foreign sellers. We are not tax advisers; sequence the filings with your zeirishi.

Records outlive the store by years

Japanese corporate tax rules require books and documents to be retained for seven years from the day after the filing deadline for the relevant fiscal year, extending to ten years for a fiscal year in which a loss arose on a blue-form return. Electronic transaction data carries its own retention requirements under the Electronic Books Preservation Act. If your Japanese order, invoice and settlement data lives inside a platform back office or on an agency's systems, export it before the accounts are closed, not afterwards — a request for historical data from a platform you no longer sell on is a much weaker position than a download while you still have login rights.

Customer data, stated accurately

Japan's personal information protection law provides that a business shall keep personal data accurate and up to date within the scope necessary for the purpose of use, and shall endeavour to erase it without delay once it is no longer necessary. The strength of that wording matters: it is a duty to endeavour, not an absolute deletion mandate, and it sits alongside — not in place of — the retention duties above. That tension is exactly what a wind-down has to resolve deliberately: decide what is retained for tax purposes, what is deleted, and who does the deleting, before the accounts go dark.

Obligations that survive the closure

Terminating a marketplace contract does not retroactively cancel the transactions made under it. Amazon's Business Solutions Agreement makes this explicit, providing that obligations arising from transactions concluded before termination continue to apply. As a general matter of contract, the returns, warranty and after-sales commitments you published to Japanese customers are commitments to those customers, not to the platform — closing the storefront does not discharge them. We looked for an official Japanese statement specifically on post-closure returns and warranty handling and did not find one, so treat that as the general legal shape rather than as a cited rule, and get your own position confirmed. While you are still advertising at all, the Specified Commercial Transactions Act's disclosure requirements for mail-order selling continue to apply to what you display.

The history is not portable

Reviews, ratings, ranking history and repeat-customer relationships attach to the store account. If you close and later reopen under a new account, none of it comes back, which is the same mechanic that determines what you keep when you change operators — worked through in who owns your store when you switch or exit an agency. If an agency holds the account rather than you, resolve that question before you begin a wind-down, not during one.

A Wind-Down in the Right Order

Sequence matters here, because several of these steps are gated on access you lose when an earlier step completes. This is the order we would run it in:

Decide which of the four decisions you are making, in writing, and price it before announcing it internally.
Confirm who holds the account and what your contract actually says about term, billing schedule and notice — from the contract, not from a blog.
Time it against the billing cycle, particularly on Rakuten, where fees are paid in advance.
Stop the spend before the store: advertising, campaign entries, inbound inventory. This is reversible and buys you a clean month of data on what the channel does unsupported.
Plan the inventory: removal orders take time, and removal and disposal cost the same, so the decision is logistical rather than financial.
Export everything while you still have login rights: orders, customer messages, settlement reports, product content, images and review data.
Serve the open obligations: pending orders, returns windows, warranty commitments.
Give notice to the platform and, separately, to any agency operating on your behalf.
Handle the registrations: consumption tax cancellation to its deadline, tax agent continuity, records retention set up to survive the closure, and a decision on the trademark.

What LAUNOVA Does — and Where We Are Not Neutral

Read this section knowing what we sell. LAUNOVA works exclusively with overseas brands selling in Japan, and running marketplace stores day to day is one of the things we are paid for — so we have an obvious commercial interest in you keeping and funding your Japan store. That is precisely why the arithmetic sits at the top of this article rather than the bottom, and why the downgrade option gets a section of its own: the honest answer for some brands is that the channel is fine and the operating layer under it is not, and the honest answer for others is to stop.

Where we are useful: rebuilding the cost picture per channel so the decision runs on real contribution rather than platform revenue; running the Japanese-language operating layer if the channel is worth another cycle with better execution; and, as part of Japan market entry and channel work, sequencing a wind-down so the exportable assets and the reopen option survive it. We scope and price this against your situation rather than from a rate card, so there is no number to quote here.

What we do not do: we are not a law firm and do not advise on trademark strategy, cancellation trials or contract disputes — those go to a Japanese benrishi or lawyer, and we route you rather than improvise. We are not tax advisers and do not file consumption tax notifications; that is your zeirishi's work. We do not sign or terminate platform contracts on a brand's behalf. And we will not tell you a Japan channel is salvageable when the arithmetic says it is not.

If you are working through this now, the most useful things to send us are which platforms you are on, twelve months of contribution by channel, and how many hours a week the Japan operation actually consumes. Get in touch and we will tell you which of the four decisions we think you are facing.

Switching or Exiting a Japan EC Agency

Changing who operates the store, rather than closing it — and what you keep either way.

Yahoo! Shopping Store Management

The September 2026 fee change in full, and the keep-or-close arithmetic specific to Yahoo!.

Japan Consumption Tax for Foreign Sellers

The registration and tax-agent obligations that a store closure does not end.

Sources

• Rakuten Ichiba store plans — three tiers at 25,000 / 65,000 / 130,000 yen per month before tax, 60,000 yen initial registration fee, a stated contract period of one year, and the statement that the monthly store fee is paid in two six-month instalments or as a lump sum by plan rather than monthly. First-party, Rakuten official store-opening pages (rakuten.co.jp/ec/plan/ and /ec/plan/cost_detail/), retrieved 13 August 2026.
• Rakuten store-closure notice period and post-closure treatment of the storefront and review history: no official source found. Rakuten's merchant terms (出店規約) are not published publicly; the only public trace we could confirm is Rakuten's Transparency Act disclosure page referencing clauses of those terms by number, including a termination clause. Japanese agency blogs state a specific notice period, and we have deliberately not reproduced it. Verify against your own contract.
• Amazon Japan selling plan fees — 4,900 yen per month before tax on the Professional plan, 100 yen per item on the Individual plan, with plan changes and cancellations possible at any time. First-party, Amazon Japan seller pricing page (sell.amazon.co.jp/pricing), retrieved 13 August 2026.
• Termination rights — the seller may terminate use of the selling services immediately at any time by notifying Amazon, Amazon may terminate for convenience on 30 days' notice, and obligations from transactions concluded before termination survive. First-party, Amazon Services Business Solutions Agreement (Japan), section on term and termination, via Seller Central help.
• FBA removal and disposal fees — identical per-unit pricing for both, at 60 / 90 / 130 / 200 yen per unit for standard sizes by weight band and 170 to 870 yen for large and oversize, tax included. First-party, Amazon Seller Central Japan fee documentation, retrieved 13 August 2026.
• Long-term storage surcharge — first band at 271 days, assessed on an inventory snapshot taken on the 15th of each month. First-party, Amazon Seller Central Japan help documentation. Discrepancy disclosed: Amazon's own public pricing page still carries a footnote referring to inventory stored beyond 365 days. Both pages retrieved 13 August 2026; confirm current bands in your own Seller Central.
• Yahoo! Shopping fees — no initial fee, monthly fee or sales royalty as of August 2026; from September 2026, a 10,000 yen monthly system usage fee before tax and a 2.5% sales royalty, with the mandatory campaign contribution borne until August 2026 ceasing. First-party, Yahoo! Shopping business pages (business-ec.yahoo.co.jp/shopping/), retrieved 13 August 2026. Yahoo!'s announcement states "from September 2026" without naming a specific day; secondary reports giving 1 September are not confirmed by the official page and are not relied on here.
• Trademark non-use cancellation — Article 50 of Japan's Trademark Act: three consecutive years or more of non-use in Japan by the owner or a licensee allows any person to petition for cancellation; the respondent bears the burden of proving use within the three years before registration of the petition, subject to a justifiable-reason exception; use begun within three months before the petition after learning of it does not count, again subject to a justifiable-reason proviso. First-party, statutory text via e-Gov (Act No. 127 of 1959, Article 50); the Japan Patent Office also publishes guidance on proving use. Not legal advice.
• Qualified invoice registration cancellation — the notification must be submitted by the day falling 15 days before the first day of the tax period from which the registration is to cease to have effect; submitted later, it takes effect from the first day of the period after next. Ceasing business is a separate notification. First-party, National Tax Agency (nta.go.jp), invoice system procedural guidance.
• Foreign-business consumption tax liability and tax agent — a foreign corporation without a Japanese office is a consumption tax payer where it makes taxable transfers in Japan and base-period taxable sales exceed 10 million yen, with a tax agent notification required. First-party, National Tax Agency tax answer 6635, stated as current as of 1 April 2025.
• Record retention — books and documents retained for seven years from the day after the corporate tax filing deadline, extended to ten years for a fiscal year in which a loss arose on a blue-form return; electronic transaction data subject to the Electronic Books Preservation Act. First-party, National Tax Agency tax answer 5930 and the NTA's electronic books guidance.
• Personal data — the duty to keep personal data accurate and up to date within the scope necessary for the purpose of use, and to endeavour to erase it without delay when it is no longer needed. First-party, Act on the Protection of Personal Information, Article 22, statutory text via e-Gov; the Personal Information Protection Commission publishes the official guidance. Stated as an obligation to endeavour, which is what the statute says — not an absolute deletion mandate.
• Mail-order advertising disclosure requirements — the items a distance seller must display under Article 11 of the Specified Commercial Transactions Act, including business name, address and telephone number, and the conditions under which some items may be omitted. First-party, Consumer Affairs Agency guidance and the statutory text via e-Gov. We found no official statement specifically on post-closure returns and warranty handling; the point above is stated as a general contractual position, not as a cited rule.
• Japan BtoC ecommerce market size — 26.1 trillion yen in 2024, up 5.1% year on year, with a goods-category ecommerce penetration rate of 9.8%. First-party, Ministry of Economy, Trade and Industry market survey on electronic commerce, published 26 August 2025. Used only as market context; we found no official statistics on foreign brands exiting Japan and have not implied any.
• No benchmark for "how long before a Japan store has failed" is quoted in this article, because no credible one exists. Nothing here is legal, tax or investment advice, and no case, client or figure in this article is a LAUNOVA number.

When Should a Foreign Brand Exit a Japan Marketplace?