LLMOcheki launches automatic design of up to 100 measurement questions, read from your own and competitors' sites

Itera, Inc. (head office: Shinjuku-ku, Tokyo; Representative Director: Takayuki Muto) has begun offering Prompt Design, a new feature in LLMOcheki, its tool for measuring and improving LLMO.

AI reads what your own site and your competitors’ sites say and, from the criteria buyers compare when making a decision, designs up to 100 measurement questions automatically. Enter a URL and the criteria buyers compare are extracted, then combined with question types to generate the questions. The passage each criterion came from, and the criteria filled in from industry knowledge, are shown separately.

Background: the quality of measurement is decided by how the questions are written

To measure how your brand is introduced in AI search, you need questions people would plausibly put to an AI. However precise the measurement itself, if the questions diverge from how people actually go about deciding, the resulting numbers will not reflect reality.

The difficulty is that the angles a set of questions must cover differ by industry. What buyers compare when choosing cannot be enumerated without knowing the field. Covering every criterion on which competitors are being judged, in particular, has depended on accumulated experience.

What it does

1. Reading your own and competitors’ sites

Enter a URL and AI works out, from what each site says, the purchase factors for that industry — the criteria buyers compare.

2. Showing where each criterion came from

For every criterion extracted, you can see which site and which passage it came from. Criteria not stated on any site, filled in from industry knowledge, are marked as such.

Keeping what was read from the sites separate from what AI supplied lets the user decide which criteria to keep.

3. Editing criteria and generating in bulk

Criteria can be selected, discarded or added as chips. Combined with question types — informational, comparative, problem, purchase — up to 100 natural questions are generated automatically.

Register only the ones you want as measurement prompts, and daily AI search measurement can start from there.

4. Saving and reusing an analysis

An analysis is saved, and editing the criteria or regenerating the questions costs nothing further.

Availability

Available from today on every plan that includes AI search measurement. Site analysis is charged by credit usage.

Comment from our Representative Director

“There is a great deal of talk about making measurement more precise, but it means nothing if the question being asked is the wrong one in the first place. What buyers actually compare differs by industry, and filling that in has been the most laborious part of the job.

We show what was read from the sites separately from what AI supplied because we believe the choice between them belongs to the user.”

About LLMOcheki

LLMOcheki is a tool for measuring and improving LLMO. It measures mentions and citations of your brand daily across eight AI search engines including ChatGPT, Gemini, Perplexity and Google AI Overview, and covers analysis, article creation and carrying out improvements in a single tool.

LLMOcheki https://llmocheki.com/