The dashboard you put in front of a board.
Most AI search work dies in a marketing meeting because nobody can show it moving. We track named prompts across 3-5 AI assistants, ChatGPT, Gemini, Perplexity, and Google AI Overviews, and report four KPIs, every assistant, every month.
From a prompt set to a monthly readout.
Named prompt set
The 20+ prompts your buyers actually use to evaluate your category, built with you.
Continuous tracking
Every prompt run across ChatGPT, Gemini, Perplexity, and Google AI Overviews on a fixed cadence.
Live dashboard
A live dashboard, not a quarterly PDF, with month-over-month movement and competitor context.
Monthly readout
A 30-minute readout that ends in decisions: what moved, why, and what to do next.
The method is published, not proprietary.
We make our entire measurement method public, including how we build prompt sets, which models we include, and how we score citation share and sentiment. Buyers can review our process before making a decision, and others in the industry can use it too.
When you work with us, you work directly with the team who created the method, not with someone just passing it along.
How we choose the prompts we track for a client.
We begin each engagement by creating a set of 20 to 50 specific questions. We select these questions together with the client and base them on actual buyer behavior. The questions stay the same across measurement cycles so we can track trends clearly. Instead of relying on just one question, we look at the whole set to get meaningful insights.
We group the questions into five types: branded, category, comparison, problem-solving, and recommendation-seeking. Branded questions usually make up 30 to 40 percent of the set, while problem-solving questions are more common in early-funnel categories. We review and update the set every quarter.
Which assistants we cover, how often, and what we capture.
By default, we track four assistants: ChatGPT, Gemini, Perplexity, and Google AI Overviews. These cover most consumer and enterprise AI searches in 2026 and each behaves differently, so it's important to include all four. Claude, Copilot, and Meta AI can be added if needed. Depending on your engagement and category, we track 3 to 5 of these assistants. We run prompts weekly, which is frequent enough to spot real changes without confusing normal model variation. Running them monthly risks missing important shifts, while daily tracking creates too much noise.
We run each prompt in a clean session with no history, personalization, or system instructions. This way, the answer shows how the model responds to an anonymous user by default. For every prompt and assistant, we collect five things: the full answer text, the cited sources and their order, whether the brand is mentioned, the sentiment of any mention, and where any product or feature appears in lists or comparisons. This set of data forms the basis for all our key performance indicators.
How each metric is scored.
Mention frequency
The share of tracked prompts, across all assistants, in which the brand is named at all. The baseline of AI presence: before citation or sentiment matter, the model has to mention you. Reported per assistant and blended.
Citation share
Of the sources an assistant cites, what share are yours, measured prompt by prompt and compared against named competitors. Where sources are listed in order, position is captured too, because the first citation outweighs the fifth.
Sentiment
How favorably each mention frames the brand, scored on a documented three-tier rubric (positive / neutral / negative). Reliable in English and Bahasa Indonesia; flagged in readouts when a prompt set spans other languages.
AI-attributed traffic
Clicks and leads that originate from AI surfaces: direct referrals plus modeled branded-search lift after a visibility win. We are explicit that this is correlational, not proof-grade causal attribution, and we write down the assumptions.
What this method cannot tell you.
Voice is not covered
The method runs on text-based prompts. Voice-only assistants use different retrieval logic and need a separate framework. Scoped for v2.0.
Multimodal is partial
We capture the text of answers and note image presence, but do not yet score visual brand visibility. Matters most in retail and creative categories.
Multilingual sentiment is harder
The three-tier rubric is reliable in English and Bahasa Indonesia. Other languages produce more disagreement, especially with strong honorific or indirect-speech norms.
Attribution is correlational
When branded search rises after a visibility win, we model the link and explain the assumptions, but we cannot prove causation. No public methodology we know of can.
Published, dated, and versioned.
We publish a version log every time the method changes. The current version is 1.0, dated May 2026. Future versions are marked with a date, a summary of what changed, and a link to the previous version. The current version always lives at this URL, so a buyer, a partner, or an AI engine can always cite the canonical method.
Questions, answered
What exactly do you measure?+
Which analytics platforms do you work with?+
Can you tie AI search back to revenue?+
Find out where your brand stands today.
Request an AI visibility audit and we will benchmark your brand and three competitors across the major AI assistants. You get a 15-page report and a 30-minute walkthrough.