Launching a brand in Japan: the first 90 days of marketing

Most guides to Japan market entry marketing spend their length on culture: localise properly, earn trust, be patient. All fair, and none of it tells a new country manager what to do on Monday morning. This article is about the plumbing: the accounts, pages, listings, tags and dates that decide whether your first few million yen of ad spend teaches you anything.

It is written for international brands arriving through a new subsidiary, a distributor, a first store or a first marketplace listing, and for the head-office teams approving the budget from another time zone. There are four phases, each with its jobs and the numbers that show whether they worked.

It draws on Meridian's outside reviews of brands recently arrived in Japan.

Before launch: build what the ads will point at

Half of Japan's ad money is now digital: internet advertising passed half of all ad spend for the first time in 2025, at 50.2% (Dentsu, March 2026). Online sales are a smaller share than many head offices assume: e-commerce was 9.78% of physical goods sales in 2024 (METI, August 2025). Most customers will find you online, check two or three other places, and often buy somewhere else. Each of those places needs to answer in Japanese, in your words.

When we review brands in their first year here, four gaps come up again and again. Search the brand name plus どこで買える ("where can I buy it") and the top answers are a personal blog or a travel guide to a shop abroad. Search it with 保証 (warranty) or 評判 (reputation) and the only official result is a spec sheet. The page titles in search results are still in English. And the only tag on the site is Google's.

What to have ready before spending on ads

The legal page. Under the Act on Specified Commercial Transactions, anyone selling online to consumers in Japan must show the seller's name, address and phone number, a responsible person's name for companies, prices and shipping costs, payment method and timing, delivery timing and the rules for returns (Consumer Affairs Agency). The final order screen must also restate quantity, price, payment, delivery and cancellation terms, a rule in force since June 2022. A store without this page looks unfinished at best.

LINE and Yahoo! JAPAN measurement changed this year. On 1 April 2026, LINE Ads and Yahoo! JAPAN's display ads were merged into one platform, LINE Yahoo Ads (LINEヤフー広告). LINE Ads stops serving in late October 2026 and closes completely by March 2027, and the old LINE Tag does not work on the new platform: it needs the new measurement tag (LY Corporation, 2025 and 2026). Any launch checklist written before 2026 that says "install the LINE Tag" is out of date. With LINE passing 100 million monthly users in Japan at the end of 2025 (LY Corporation, January 2026), this is not a channel to measure later.

Reviews, done legally. A listing with zero reviews loses sales on every marketplace. The shortcut is closed: since 1 October 2023, paid or directed posts that hide that they are advertising count as misleading representation under Japan's Premiums and Representations Act, and that includes influencers and staff posting as ordinary customers (Consumer Affairs Agency). Use the sanctioned routes: Amazon's Vine programme (for brand-registered products shipped through FBA with fewer than 30 reviews), product seeding with a clear PR label on every post, and a polite review request after delivery.

Marketplaces. List where your category actually shops. Rakuten Ichiba now accepts merchants from 22 countries and regions with no base in Japan (Rakuten, September 2025), so a distributor is no longer the only door. Qoo10 is where much of Korean beauty and younger shoppers buy; ZOZOTOWN is where fashion gets browsed. Each has category rankings (ランキング), and a day in a top ten brings traffic no ad budget easily buys. Store approvals take weeks, so apply first.

What to measure before launch. Place a test order through every route and check it appears once, with the right value, in each ad platform and in analytics. Then search your brand in Japanese on a phone in Japan and note who answers each question. That is your baseline.

Weeks 1 to 4: capture the demand that already exists

Pick the launch date against the sale calendar. Rakuten Super SALE has run in March, June, September and December in recent years. Qoo10's Mega Wari (メガ割) runs four times a year; in 2026 the editions began in late February, late May and late August, with the fourth expected in November. Launching the week before one of these is a common mistake: shoppers hold purchases for the coupons, and you enter the biggest week of the quarter with no reviews and no ranking history. Launch four to six weeks earlier instead, build reviews, and join the sale with a listing that already has stars.

In the first month we would put most of the budget on intent and little on reach:

• Brand search campaigns on Google and LINE Yahoo search ads, covering the name in katakana, in Latin letters and the common misspellings.
• A small category search budget on the Japanese words people actually type, such as ワイヤレスイヤホン おすすめ, sent to a Japanese landing page.
• Marketplace ads inside Amazon and Rakuten, where buyers already are.
• A PR TIMES release on launch day. It costs ¥30,000 per release before tax (PR TIMES, checked October 2026).
• Retargeting on Meta and LINE Yahoo display, limited to site and listing visitors.

What to measure. Brand search impressions week by week, in Search Console and in the ad platforms. Cost per click and conversion rate on brand campaigns. How many of the top results for your brand name are your own pages. Marketplace session-to-order rate and review count. If brand searches are flat after a launch announcement, the problem is awareness, and more search budget will not fix it.

Weeks 5 to 8: add reach, one channel at a time

Now test the channels that create demand rather than catch it: Instagram and Facebook, LINE Yahoo display, YouTube, or creator seeding with disclosed posts. Add one at a time, each with its own budget and a number agreed in advance. Three channels launched in the same week cannot be told apart.

If your product is bought more than once, open a LINE official account now. Collect friends from packaging inserts, order confirmation pages and store counters, and write the first messages before anyone joins. Then rewrite landing pages using the search terms report: the words customers type in Japanese are rarely the words in the global brief.

Measurement gets harder here. Someone watches an Instagram video, searches the brand on Amazon three days later and buys: Instagram reports nothing and Amazon takes the credit. A few ways to read it honestly:

• Watch brand searches and marketplace sales in the days after a reach channel starts or pauses.
• Hold out one region or one audience for a few weeks and compare it with the rest.
• Track cost per new customer across all channels together, not return on ad spend inside each dashboard.
• For LINE, track friends added, block rate after each message and purchases from message links.

Weeks 9 to 12: decide what to scale, fix or stop

Put every channel on one page, marketplaces included, and rank them by cost per new customer. The bottom channel stops, unless there is a specific, fixable reason it underperformed. The top one gets more money in steps of 20 to 30% a week, because cost per customer usually rises with spend and you want to see where it starts climbing.

Then plan the next quarter: the next big marketplace sale, your category's seasonal peaks (gifts in December, new-life demand in March and April) and the one channel to test next. Head office gets it in English, with a one-page Japanese summary for the local team or distributor, so everyone works from the same numbers.

What to measure. Blended cost per new customer: total marketing spend divided by new customers across every channel. Share of sales coming from brand searches against category searches, which shows whether you are still living off launch awareness. Repeat purchase at 60 days, an early signal but a useful one. And the baseline search from before launch, repeated, to see who now answers "where can I buy it".

A worked example

Take a consumer electronics brand, wireless earbuds for instance, launching on Amazon, Rakuten and its own Japanese site with ¥3 million a month for ads. This is an illustration, not a client.

Two months before launch, the Rakuten application goes in and the Amazon brand registration starts. The site gets Japanese titles, a where-to-buy page, the legal page and a support page with the warranty terms written out in Japanese. Tags go in for Google, LINE Yahoo Ads and Meta, and a test order is checked in all three. Thirty units go to Amazon Vine.

The launch date is set for late October, about five weeks before the December Rakuten Super SALE. Month one splits roughly ¥1.2 million on search (brand and category), ¥900,000 on Amazon ads, ¥500,000 on Rakuten's in-store ads and ¥400,000 on retargeting, with a PR TIMES release on day one. Month two adds Instagram with tech creators labelled as PR, and the brand joins the Super SALE with reviews in place. In month three the team compares cost per new customer across the five budget lines, cuts the weakest and plans the March sale and the new-life season.

Nothing in that plan is clever. What matters is the order: pages and tags, then demand capture, then reach.

The mistakes we see most

• Ad accounts opened in the agency's name, so the brand cannot see or keep its own data.
• Measuring Google only, then cutting LINE or Meta because they "don't convert".
• A global site with a Japanese banner on top and English page titles underneath.
• Launching the week before a big sale, or with no reviews at all.
• Paying for reviews or undisclosed influencer posts, which has been illegal since October 2023.
• Leaving "where can I buy it" and warranty questions for blogs and resellers to answer.

Most of them come from planning Japan from abroad, where Japanese search results, marketplace rankings and LINE are hard to see.

If your launch is in the next six months, Meridian's free pre-launch review looks at your category the way a Japanese buyer does: who shows up, who already sells your products, and the three things we would fix before launch day.