AI Visibility & GEO Intelligence
Does AI recommend you? Does it get you right?
We analyse how visible you are in ChatGPT and Google's AI answers, where you stand against competitors, and whether what AI says about you is accurate. We show you what to improve, with evidence, and track what changes.
ChatGPT is measured through OpenAI's developer API (web search enabled); Google AI Mode and AI Overviews from Google's results page. We don't guarantee rankings or visibility.
Live example: pick a sector
Example scenario. Asked ChatGPT: Is Example Clinic’s Kadıköy branch open on Sunday? Example Clinic’s Kadıköy branch is open on Sundays from 10:00 to 16:00. Wrong: open on Sundays from 10:00 to 16:00. Correct: The Kadıköy branch is closed on Sundays. Source (likely): Outdated opening hours on the map listing. To do: Set Sunday to “closed” on the map listing. Fixed ✓ (re-measured): Example Clinic’s Kadıköy branch is closed on Sundays.
Example scenario; not a real business.
Pilot. Truvis currently runs as a managed measurement service: our team runs the measurements and prepares the reports, which arrive by email. The self-service platform is in development.
Evidence-based answers to three questions
AI Visibility
How visible are you in AI answers?
We regularly ask real questions about your business and show your mention rate, engine by engine, with its range.
- Mention rate by engine, with a 95% rangePilot
- Weekly tracking and history, in your reportPilot
- Trend charts in the dashboardComing soon
Competitive Intelligence
Where are competitors ahead?
We show which competitors AI recommends, your share of voice, and the questions where a competitor is ahead.
- Competitor discovery from AI answersPilot
- Share of VoicePilot
- Questions where a competitor is aheadPilot
- Sources that mention competitors but not youPilot
- Competitor management in the dashboardComing soon
Accuracy Intelligence
Is what AI says about you correct?
We check claims in AI answers against the facts you approve: evidence, likely source, to-do list. When you fix something, we re-measure.
- Checks against approved facts, with the evidence sentencePilot
- Branch × service × period checksPilot
- Fixed → re-measured → time to correctionPilot
Pilot: works today; our team runs the measurements and the report arrives by email. Coming soon: not yet available.
Why not just ask ChatGPT yourself?
You can, and it's a good start. But a single answer doesn't tell you much.
| Topic | Asking yourself | With Truvis |
|---|---|---|
| Answers | One answer; the next one may differ | Many answers every week, with a range |
| Time | One day, one moment | Weekly tracking; you see the change |
| Is it right? | You compare it yourself | Automatically compared with your approved facts |
| Branches | No distinction | Branch × service × period |
| Why is it wrong? | Unknown | Likely source and a to-do list |
| Is it fixed? | You keep asking | Re-measured after your fix, with time to fix |
How it works
1. Measure
We regularly ask ChatGPT, Google AI Mode and AI Overviews real questions about you: Google's surfaces every week, ChatGPT on a rotating schedule.
2. Find it, with evidence
We compare each claim with the facts you have approved. Every finding comes with the AI's exact sentence and the engine where it appeared.
3. Source and to-do list
We show where the wrong information most likely came from (your website, your map listing, another site) and give you a prioritised list of what to fix.
4. Re-measure after you fix it
When you mark an issue as fixed, we ask the same questions again and show whether the AI now gets it right.
What we check
Services, branch and period
“Which branch offers swimming, and in which months?”
Branch and address
“Is this branch still open? Is the address right?”
Phone
“Does a caller reach the right number?”
Opening hours
“Is it open on Sunday?”
Age and audience
“What's the minimum age?”
Product information and other kinds of information
Verification for product information (price, features, warranty) and other kinds of information is coming soon.
This is what your report looks like.
Example Business; numbers are examples.
Example Business
ExampleGEO tools that support visibility
- Technical site health: AI crawler access, sitemap, structured dataPilot
- Source analysis: the sites AI citesPilot
- Keyword visibilityPilot
- Evidence-classed recommendations: Verified Issue, Evidence-Based Opportunity, Hypothesis to TestComing soon
- PDF and CSV exportComing soon
- Self-service dashboardComing soon
Why it's different
- Accuracy first, not just visibility. Being mentioned isn't enough; what's said has to be right. We measure how accurately AI describes you against the facts you have approved.
- Branch × service. We check “is this service offered at this branch?” branch by branch, and catch answers that assign one branch's service to another.
- Services that aren't always available. A service offered only in certain periods or on demand is not marked as “not offered”; if the AI mentions it without its period, we flag that separately.
- Honest measurement. AI answers can change every time. We state how many answers we looked at, what we could not measure, and which findings are confirmed versus likely.
- No manipulative tactics. No hidden text, fake reviews or tricks aimed at AI systems; we only recommend publishing correct information, consistently, in the right places.
Who it's for
Anyone who cares whether AI talks about them accurately: independent professionals, product brands, exporters, B2B service companies, local businesses and chains, clubs and associations, public institutions, opinion leaders and content creators; and the agencies that track this for their clients.
Professionals
Doctors, lawyers, consultants: are your address, hours and services right?
Restaurants, cafés, hotels
Menu, hours and branch information.
Clinics
Which service at which branch, and when? Information accuracy only.
Schools and courses
Age, term and branch information.
Chains and e-commerce
Consistency across branches.
Exporters
In every country, in that country's language.
Public institutions and associations
Service and contact information.
Artists and creators
The basic facts said about you.
Agencies
Track it for your clients.
For agencies
Countries and languages. Designed to measure each market in its own language and its own city. We started in Turkey; write to us about another country or language.
Start small; grow as you add branches and markets.
Starter
One branch, one market
$29/mo
Annual $290 (≈ $24.17/mo); 2 months free
1 unit
Extra unit $19/mo
Write to us for early accessGrowth
A few branches or markets
$99/mo
Annual $990 (≈ $82.50/mo); 2 months free
Up to 5 units
Extra unit $19/mo
Write to us for early accessAgency
Agencies tracking clients
from $249/mo
Annual $2,490 (≈ $207.50/mo); 2 months free
15+ units
Extra unit $12/mo
Write to us for early access
FAQ
What is Truvis?
Truvis compares what AI says about you with the facts you've approved, finds what's wrong with evidence, shows where it likely came from, and re-measures after you fix it.
Which AI systems do you measure?
Today: ChatGPT (through OpenAI's developer API, web search enabled), Google AI Mode and AI Overviews (from Google's results page). Gemini measurement is being prepared; we'll list it here once it's live. Method and status are on the Methodology page.
What do I need to start?
Your website and your branches. We extract your information from your site and send it to you for approval; we only measure against what you approve.
Do you guarantee results?
No. Nobody can guarantee what AI will say. We measure what's wrong, where it likely came from, and whether it gets fixed.
Do you use tricks aimed at AI systems?
No. We only recommend publishing correct information, consistently, in the right places.
Where is my data?
During the pilot, measurements run on our own system and are backed up encrypted. We are moving hosting to the EU (Frankfurt). Details on the Methodology page.
Founder's note
Truvis started with a problem I ran into at my own sports club. When I asked AI assistants about our branches, I saw some of our services listed at the wrong branch and some information missing or out of date. There was no easy way to notice this, let alone fix it. This isn't just a sports-club problem; it's a problem for anyone AI talks about.
Request a free AI visibility audit.
Just send your name or business name, your website and (if any) your branches. We'll prepare a short report showing, with evidence, what AI gets wrong about you.
If what you want us to check isn't on our list, write to us anyway; we'll tell you plainly what we can measure.
hello@truvisapp.com