Get named when your buyers ask AI.
We find the questions your buyers ask ChatGPT, Perplexity and Gemini, check every month whether the answers name you, and fix the pages and mentions that keep you out.
Your buyers ask ChatGPT before they search, and it answers with a competitor's name.
Four stages, run by us, reported to you every week.
Buyer questions
Reads your sales calls and support tickets to list the 20 to 40 things buyers want you known for, so we measure what your market actually asks.
The monthly run
Turns those into 60 to 120 real buyer questions and asks them every month, logged out, on ChatGPT, Perplexity and Gemini, so the score you see is comparable month to month.
The diagnosis
Splits those questions into the ones where AI has never heard of you and the ones where it knows you and names a competitor, because the two need different fixes.
Fixes and mentions
Rewrites the pages AI already reads, ships a comparison page and a case study, and gets you listed on the sites it trusts, so assistants have something of yours to quote.
Your visibility, scored the same way every month.
- Every prompt is shaped the way a buyer types it, persona plus segment plus use case, like a marine supplier in Singapore asking which tools automate quoting.
- The panel runs logged out on ChatGPT, Perplexity and Gemini, and records whether you are named, where you sit, who else appears and which sites they cite.
- The panel stays frozen for the quarter, so when your score climbs it is because the answers changed, never because the questions did.
Know which fix moves you before you spend on content.
- When AI has never heard of the capability, the fix is on your own pages: a direct answer above the fold, Q&A blocks that mirror the questions, schema the crawlers read.
- When AI knows you and still names someone else, more content does nothing. The fix is third-party citations, starting with the sites it already trusts.
- Every question that matters gets one named fix, and you pick the single question the guarantee sits on, in writing, before any of it ships.
The pages and mentions that get you quoted.
- Restructures five to ten of the pages already getting crawled: the answer in the first 80 words, Q&A blocks that mirror the panel's questions, schema the crawler can parse.
- Ships one comparison page and one case study with real numbers, because those are the two formats assistants cite most for a buying question.
- Asks for third-party mentions in friction order, people you know first, then directories that already list your category, cold outreach last, and logs every item with a date.
Every week, what shipped and what moved.
- A short update every Friday: what went live, which attribute moved on which assistant, and what ships next.
- A one-page report every month: visibility score and rank per attribute per assistant, citation share by domain, and the input log with every item dated.
- A question on your inbound forms, how did you first hear about us, with AI as an option, so the revenue side reads from your own data within about 60 days.
Setup, then the run.
The first twelve weeks build the instrument and ship the first fixes. From month three the panel runs monthly and the assets keep going out.
Setup
We get access, write your buyer questions, run the first measurement and ship the first round of fixes, so by week twelve you have a score and the pages to move it.
The run
The panel runs, a fixed quota of assets ships, you get the update every week and the report every month. The attribute you named reaches its agreed score by day 90, or we keep working free.
Ask AI about you before you talk to us
Email us your website and five buyer questions to request an AI visibility audit across ChatGPT, Perplexity and Gemini. We’ll reply to arrange the details.
Request by emailWhere AI SEO breaks
Can you guarantee traffic from ChatGPT?
No. AI referrals are a fraction of a percent of most sites' traffic, and one platform update can move that number overnight with nobody at fault. What we guarantee is visibility on the attribute you name, measured by the frozen panel. Revenue gets read from your own how-did-you-hear-about-us field.
Why is our baseline zero?
Because nobody has measured it before, and most brands start there. The baseline is the before picture. It tells us which attributes AI has never heard of and which it knows and skips, and those two get different fixes.
Why not just publish more content?
Because content is often the wrong fix. When AI already knows what you do and still names a competitor, the missing piece is trust, which comes from the third-party sites it cites. Six weeks of new pages will not move that. A listing on the right directory will.
How long until it moves?
A long-tail question can move in about three weeks. A category term with real competition takes four to six months. We tell you which kind each attribute is before you pick the one the guarantee sits on, so the date is honest.
Ready to be the answer?
We'll run five of your buyers' questions cold and show you who AI names today.
Book a Call