Guide · Japanese search entry points

Where do Yahoo! JAPAN and docomo search results come from? Do Japanese indie sites need to optimize for them separately?

Updated 2026-11-13 · kagentx

Short answer

docomo's dメニュー search results come from Google, as its official explanation page states (read in full on 2026-11-13); Yahoo! JAPAN announced on 2010-07-27 that it would adopt Google's search engine, and whether that is still entirely so today we have not read in any Yahoo! statement: not verified. So there is no evidence that a Japanese indie site needs a separate effort for these two entry points; first get the Japanese pages indexed and written well in Google. Over the last 7 days one Japanese site had 7 sessions from yahoo / organic, 2 from docomo's referral, and 134 from google / organic.

Where do docomo search results come from?

From Google. docomo's official explanation page for dメニュー search (service.smt.docomo.ne.jp, read in full on 2026-11-13) says the service is provided based on search results supplied by Google and content information from each service provider; the partner for web, image and video search is given as Google; video search calls the YouTube API. NTT docomo also states that it is not responsible for the accuracy of search, the displayed results or the content of link targets, and that ranking and result counts change at any time. The page shows no update date.

The same page also says dメニュー search supports only Android OS 6 and later and iOS 10 and later.

Where do Yahoo! JAPAN search results come from?

It has used Google's search engine since 2010. In its 2010-07-27 announcement, Yahoo Japan Corporation said web, image, video and mobile search and search-linked ads would use Google; the Yahoo! brand stays, and Yahoo! JAPAN processes the results and offers them as its own service; Yahoo! JAPAN also supplies Google with auction and shopping data so that this content reaches results faster (Impress Watch's report of the same day, which we read in full; the Yahoo press release PDF can be fetched but our tools could not read its text, so we did not read the original).

Note that this is an announcement from 2010. Whether Yahoo! JAPAN's search page still works entirely this way, we have not read in a current Yahoo! page; our two requests from this network for search.yahoo.co.jp's robots.txt and search page on 2026-11-13 both timed out, so the Yahoo! side was not tested, and we write “not verified”.

How to reachWhat we verifiedWhat we did not verify
dメニュー search (docomo)The official page says results are based on search results supplied by Google (read in full on 2026-11-13)The page's update date
Yahoo! JAPAN searchThe 2010-07-27 announcement adopting Google's search engine (Impress Watch report)Whether it is still entirely so today; an index submission route for site owners; robots.txt

A real example: a Japanese site's sessions from these two entry points over 7 days

MochiStamp is a Japanese site we use as a public case study (with the operator's consent). In its GA4, sessions by source / medium for the seven days 2026-11-06 to 2026-11-12 (read from the dashboard on 2026-11-13):

Source / mediumSessions
google / organic134
(direct) / (none)110
chatgpt.com / ai-assistant46
yahoo / organic7
service.smt.docomo.ne.jp / referral2

These two entry points add up to 9 sessions, a fraction of google / organic. The sample is small: seven days, one site. Whether yahoo / organic is Yahoo! JAPAN, GA4 cannot tell; for a Japanese site it very probably is, not verified; the 2 docomo sessions have the domain dメニュー lives on as their source, and we did not check them one by one. All we can say is that these two entry points are real and small, and most likely move together with how well you rank in Google.

Japanese pages' impressions and rankings in Google are in Japanese pages in search and AI answers: real data from two sites.

Should an indie site optimize for these two entry points separately?

Going by what we verified, no separate effort is needed: docomo's results officially come from Google, and Yahoo! JAPAN has used Google's engine since 2010. Doing Japanese pages well is more likely to affect all three at once.

  1. 1. First get the Japanese pages indexed in Google

    A Japanese page needs its own URL, a Japanese title and description, and a place in the sitemap; how to get indexed is in How do you get a new website indexed by Google? How many indexing requests can you send a day?.

  2. 2. Get hreflang right

    Japanese pages and the other language versions point to each other and carry x-default; how to write it is in How do you write hreflang on a multilingual site? How does AI search treat it?.

  3. 3. Track the Japanese entry points separately in GA4

    Pick “Session source / medium” as the dimension, note how many sessions yahoo and service.smt.docomo.ne.jp each have, and check once a week. When traffic is small the numbers jump around; look at the direction over several weeks.

  4. 4. Do not reconcile GA4's yahoo sessions directly with Search Console clicks

    Search Console data is Google Search data; whether Yahoo! clicks are counted in it, we found no official explanation.

  5. 5. Check separately whether Yahoo! JAPAN has a site owner entry point

    We have not verified. If there is one, go by Yahoo!'s own help pages; the Yahoo! search help home page has a “ウェブ検索について” category and a “検索結果情報の更新” article, and we did not read their text.

FAQ

Does Yahoo! JAPAN search use Google's engine?

On 2010-07-27 Yahoo! JAPAN announced that it would adopt Google's search engine and search-linked ads, covering web, image, video and mobile; the Yahoo! brand stays and the results are processed and shown by Yahoo! JAPAN. This is a 2010 announcement; whether it is still entirely so today, we have not read in a current Yahoo! statement: not verified.

Whose search is docomo's?

The official explanation page for dメニュー search says: the service is based on search results supplied by Google and content information from each service provider; the partner for web, image and video search is given as Google, and video search uses the YouTube API. Read in full on 2026-11-13.

Should you optimize separately for Yahoo! JAPAN and docomo?

There is no evidence that a separate effort is needed. Both entry points' results are tied to Google, so first get the Japanese pages indexed and written well in Google. Whether Yahoo! JAPAN has an index submission route for site owners, we have not verified.

Why does my GA4 have sessions from yahoo / organic?

It means someone clicked through from Yahoo's organic search results. Whether it is Yahoo! JAPAN, GA4 cannot tell; for a Japanese site it very probably is, not verified. One Japanese site had 7 such sessions in the last 7 days, against 134 from google / organic.

What is service.smt.docomo.ne.jp / referral?

This is the domain dメニュー at docomo lives on, and the source is recorded as referral. The official page says dメニュー search results come from Google, so these sessions should be clicks from dメニュー search result pages, but we did not check them one by one. One Japanese site had 2 in the last 7 days.

Can Search Console show data from Yahoo! JAPAN?

Not verified. Search Console data is from Google Search, and we saw no official statement on whether Yahoo! JAPAN clicks are counted; do not reconcile GA4's yahoo sessions directly with Search Console clicks.

What else does a Japanese site need to be found by Japanese AI and search?

How to write the Japanese pages themselves, hreflang and Japanese titles are in the guide on Japanese pages; look at the Japanese entry points' search sources separately by source / medium in GA4 instead of only google / organic.

Sources

kagentx

Now you know people arrive from AI.
Want to know what AI says about you?

An evaluation asks ChatGPT, Perplexity and Claude your buyers' questions (with web search on) and records who they recommend, which pages they cite and whether you are among them. Basic $1, results usually in 2–3 minutes.