LLMOcheki launches a public API, letting agencies build AI search measurement into their own dashboards

Itera, Inc. (head office: Shinjuku-ku, Tokyo; Representative Director: Takayuki Muto) has expanded the public REST API of LLMOcheki, its tool for measuring and improving LLMO, and begun offering five endpoints through which analysis data can be retrieved by external systems.

Five kinds of data — the overall GEO score, competitor comparison, citation source analysis and more — can be retrieved by API. Data for the clients under an agency account can be retrieved in bulk, for white-label delivery and automated reporting.

Background: the request to use measurement data on your own screen

As search moves toward experiences where generative AI produces the answer, the yardstick for marketing has widened from search ranking to how you are described inside an AI answer. We have provided our tool to agencies and businesses as a way to measure brand mentions inside AI answers over time.

What we heard most often was a wish to bring that measurement data into their own admin screens and BI tools, and to generate client reports automatically. For support companies in particular, transcribing the data for several clients by hand every month had become a burden.

The endpoints now available

Analysis data (three)

• Overall GEO score: a single score from 0 to 100 for brand strength in AI search, with a breakdown by axis — recall, presence, trust and so on
• Competitor comparison: mention rate and average mention position by AI engine, side by side with the competitors you have registered
• Citation source analysis: for the sources an AI answer cited, the proportion that are your own, competitors’ and third parties’, with the top sources

Results and operations (two)

• ROI summary: an estimated return on investment based on traffic and conversions arriving via AI
• Credit balance: the balance and usage of content generation credits

For agencies and OEM partners

An agency account can retrieve data for all the clients beneath it in bulk. A support company can build a dashboard under its own brand and deliver AI search measurement reports without LLMOcheki’s name appearing.

It can also be used to connect to BI tools such as Looker Studio, and to automate daily data retrieval.

Combined with the MCP integration

Alongside the API, the MCP integration is supported, so you can ask generative AI such as ChatGPT or Claude directly about where you stand in AI search. A question like “what is the gap in mention rate against our competitors this month?” can be answered by AI referring to the measurement data.

You can check how you look to AI by asking AI.

Security

Beyond API key authentication, there are access scopes, rate limits and usage logs.

How to start

Customers on the Pro plan or above, or with the MCP/API add-on, can issue an API key from “MCP/API integration” in the admin screen and begin straight away. The endpoint specifications are published in the REST API documentation on the same screen.

The ROI summary is an estimate based on the values you set, and does not guarantee results.

What comes next

Following these read endpoints, we plan to expand into operation endpoints such as running a measurement and publishing an article.

Comment from our Representative Director

“Measuring AI search is not something that finishes inside a single tool. For a support company, what matters is being able to build it into what they themselves deliver, and that requires the data to be in a form that can leave the product. If you are going to build the standard for measurement, being connectable comes first.”

LLMOcheki https://llmocheki.com/