LLMOcheki extends AI search measurement to 16 markets, and begins rolling out a multilingual admin screen

Itera, Inc. (head office: Shinjuku-ku, Tokyo; Representative Director: Takayuki Muto) has extended the markets measured by LLMOcheki, its tool for measuring and improving LLMO, to 16. How a brand appears in AI search in overseas markets can now be seen in one place, from a Japanese-language admin screen.

The market and language can be set per project and per keyword, and both the question put to AI and the analysis of its answer run in the local language. Measured values for existing projects in the Japanese market are unaffected.

Background: the same brand is described differently from market to market

Generative AI refers to different sources depending on the language and region of the question. A question in Japanese draws on Japanese articles and reviews; a question in German draws on German-language media and local comparison articles.

The result is that the same brand is described differently by market. A company introduced correctly in Japan may be described as a product in a different category in the market it is entering, or may not come up as a candidate at all. Understanding the situation overseas has meant asking local agencies case by case, which made comparing markets against a common standard difficult.

Feature 1: setting the market per project and per keyword

The market and language for measurement can be set for each project. That setting is reflected in everything that depends on the market: AI search measurement, search rankings, search volume, and the extraction of related keywords.

The market can also be specified for individual keywords, so several markets can be measured in parallel within one project. Where no market is set on the keyword, the project setting applies.

Feature 2: measuring in the local language

Set a market other than Japan and both the instruction given to AI and the analysis of its answer run in that market’s language. Measurement happens under conditions closer to the answer a local user actually receives.

Feature 3: comparing across markets

Even measured market by market, the definitions of the metrics are common. Appearance rate, citation rate and competitor comparison are calculated to the same standard, so differences between markets can be compared directly. Where you have exposure and where you are thin can be seen on a single screen, informing how budget is allocated.

On existing measured values

Existing projects and keywords all retain the Japanese market and Japanese language settings. The instruction text used for measurement in the Japanese market is unchanged, so this release causes no movement in measured values.

We place weight on the continuity of ongoing measurement, and take it as a principle that a change in specification should not break the existing time series.

A multilingual admin screen

The admin screen can now be displayed in twelve languages: Japanese, English (US and UK), Chinese (traditional and simplified), Korean, Thai, German, Swedish, French, Hindi and Arabic. Arabic is supported with right-to-left display.

Multilingual display is being extended starting from the main screens, so some screens currently still display in Japanese. Coverage will continue to expand.

Comment from our Representative Director

“A company introduced correctly in Japan can be described as something else entirely by AI overseas — this actually happens. Ask a local agency and you learn fragments, but you could not compare markets by the same yardstick. Being able to line them up on the same metrics is, I think, what a decision about expanding abroad requires.”

LLMOcheki https://llmocheki.com/