See how AI talks about your business — before your customers do.
Geode asks ChatGPT, Claude, Gemini, Perplexity and Grok the questions your customers ask — through their APIs, many times over — then scores how often they name you, where you land, and whether they get your details right.
Scan, score, fix — then scan again
A measurement, not an opinion. Every number on your report traces back to answers we can show you, from questions you can read.
We ask what your customers ask
“Best plumber in Round Rock.” “Who repairs tankless water heaters near me.” We put those questions to every assistant through its own API, several times each, because the same question does not always get the same answer.
We score what comes back
Every answer is read and counted: how often you are named, how high you land in the list, how much of the conversation your competitors own, and whether the facts stated about you are actually correct. One number out of 100, with the working shown.
You get a plan, then we measure again
A printed report and a 90-day checklist in priority order — profiles, sources, reviews, the pages on your site. Rescan next month and you can see exactly what moved.
What this is, honestly
- These are API answers, not the consumer apps. What a logged-in user sees in a chat app can differ from what the same model returns to us.
- We report rates, not rankings — “named in 7 of 9 answers, usually second or third”, never “you rank #2 on ChatGPT”.
- Nobody can guarantee a placement in an AI answer, and we never sell one. We sell the measurement and the work that tends to move it.
Results reflect AI API responses captured at scan time. AI outputs vary between runs and change as the underlying models are updated. No ranking, placement, or revenue outcome is expressed or implied.
Pick the level of help that fits
Every option starts from the same measurement. Move up when the numbers say it is worth it, not before.
This report, fresh every month: your score, who AI recommends instead of you, and what changed — delivered print-ready.
- Monthly rescan across every covered AI assistant
- Score, competitor share-of-voice, and source tracking over time
- Accuracy watch: we flag it when AI starts saying something wrong about you
- Need one sooner? Extra reports on demand, $15 each
A deeper scan and a working session: what is actually keeping you out of AI answers, and the exact order to fix it in.
- Expanded scan at our full 7-run statistical sample
- Full source + accuracy audit, plus each assistant's own explanation of what it would take to rank you first
- 30-minute consult to walk the findings and plan
- Run it once, quarterly, or monthly — your call
We do the work: profiles, listings, review cadence, and website updates. Scope depends on what the deep dive finds and the access you grant — so we quote it, in writing, before any work starts.
- Google Business Profile, Yelp & directory cleanup
- Website content and structure updates where needed
- Review request cadence + response strategy (never fake reviews)
- Measured against your baseline in next month's report