Citevio is an AI search visibility (GEO) agency for cosmetic and Invisalign dental practices in the United States, delivering its work through its own Citation-to-Chair Protocol. The protocol's discovery layer identifies the buying questions patients actually ask, and the question set is agreed with the practice. Citevio then analyzes daily how ChatGPT, Perplexity, Gemini and Google AI Overviews answer those questions, and the results set the direction of the next round of work. Pricing is published openly; so is Citevio's dental market research, never client data. Unlike a typical agency retainer, there is no fixed-term contract and no minimum commitment — a clinic can end the engagement at any time.
This page exists because the place we usually send people, citevio.com/data, answers "what did you measure" but not "who are you and can I quote you." Below: the three datasets you can cite today with the CSV attached, the six city reports, how we ask to be credited, what we will not tell you, and who to contact.
Data you can use
Three datasets, each aggregated and free to download as CSV under CC BY 4.0. Every figure below states its sample and its scan date — no live feed, no rolling average.
| Dataset | Sample | Scan date | What it found |
|---|---|---|---|
| Dental practice web anatomy | 527 dental practice websites, 7 US metros | June 2026 | 53.9% of the 527 carry no LocalBusiness or Dentist schema at all. Separately, of the 523 with a Google rating, 96.7% already sit at or above the roughly 4.3-star average AI-recommended practices tend to carry. |
| AI crawler access in robots.txt | 6,497 readable domains, of 7,632 requested | 22 July 2026 | 7.6% block at least one crawler run by OpenAI, Anthropic or Perplexity; widen to all 13 crawlers tested and it's 11.4%. |
| Live crawler access | 499 sites whose robots.txt already allows AI crawlers | 22 July 2026 | 14.4% were refused by their own server anyway, while Googlebot was served normally in 12.6% of that same sample. |
The 7.6%/11.4% and 14.4% figures are not the same measurement: the first two read a policy file (robots.txt) across 6,497 domains, the third tests what 499 live servers actually do when asked. Do not quote them as one rate. Full method, limits and what we chose not to publish: citevio.com/data.
City reports
Six of the seven metros scanned in June 2026 have a published city report: how many practices were scanned, how many ChatGPT or Perplexity actually named, and what the invisible ones had in common.
| Metro | Practices scanned | Headline finding | Report |
|---|---|---|---|
| Charlotte, NC | 154 | ChatGPT named 2 (1.3%) | Read report |
| Austin, TX | 111 | Perplexity named 7 (6.3%) — our strongest single-engine result | Read report |
| Raleigh, NC | 64 | ChatGPT named 0 | Read report |
| Nashville, TN | 59 | ChatGPT named 0; highest AI-crawler blocking rate of the seven metros scanned (8.5%) | Read report |
| Columbus, OH | 58 | ChatGPT named 0; Perplexity named 3 of the 21 it could read | Read report |
| Tampa, FL | 42 | ChatGPT named 0; 4 of the 42 sites had no robots.txt file at all | Read report |
Perplexity could not be scored at all in Raleigh (an identification error affected 63 of 64 responses), Nashville or Tampa, and was only partly readable in Columbus (21 of the responses came back). ChatGPT itself was rate-limited on 90 of Austin's 111 attempts, leaving 21 to read. Each city's own report has the full per-engine detail. Full seven-metro comparison: AI visibility reports by city.
How to cite us
Every dataset on this page is released under CC BY 4.0. You do not need to ask permission first — credit Citevio, link to citevio.com/data, and say which scan date you used, because these are dated snapshots, not a live feed.
Ready-made citation lines, one per dataset:
Citevio, AI crawler access in dental robots.txt files, 22 July 2026 (n = 6,497). citevio.com/data
Citevio, live crawler access on dental homepages, 22 July 2026 (n = 499). citevio.com/data
Want a cut for your own readers — a single state, a single crawler, or one of the six cities above re-run closer to your publish date? Email contact@citevio.com and we'll run it, within the limits below.
What we will not say
The honest limits of this data matter more than the headline numbers. Seven things we will not claim, so a reporter doesn't have to find out the hard way.
- No trend. Every figure on this page is a single dated scan — June 2026 or 22 July 2026 — not a repeated series or a live feed. We do not claim a metro or a metric is improving or getting worse.
- No causation. Missing schema and blocked crawlers correlate with being skipped by AI assistants, but nothing here tests cause. A site could check every box and still not be named, or the reverse.
- No name-by-name ranking. The city reports above count how many practices were named, not which ones. We withhold the separate study that would extract and rank practices by name, because its name-extraction step does not yet separate practice names from generic phrases precisely enough to publish.
- No Gemini in this scan. Google Gemini was not tested for practice mentions in the scans behind the city reports and the June dataset — only ChatGPT and Perplexity were.
- No Bing figure by city. Bing index results are published for the full 527-practice sample only, not broken out by metro — 236 of those checks failed on our side, leaving too thin a base to publish per city.
- No cut below our own threshold. State and category breakdowns need at least 300 sites; metro breakdowns need at least 20 practices. Below that, a rate moves with a handful of websites, so we don't publish it and won't manufacture one on request.
- No outcome guarantees, and no borrowed proof. We don't guarantee a ranking, a citation, a patient count or a revenue figure to anyone, and we don't publish a number we did not measure ourselves. The one guarantee we do publish is financial: the 45-day citation guarantee refunds a client's setup fee if they have not been named by at least two of the four engines in an agreed question within that window.
Who to quote
"Every figure on this page comes from a scan we ran ourselves, with the sample size and the date attached. If something here is wrong, email me and the correction goes on the page, in public."
— Muhammed Veysel Erin, Founder, Citevio
Contact
contact@citevio.com — replies within one business day. Full company registration, address and phone for verification: see About Citevio.
Want to see how this scoring works on one specific practice? Free AI visibility checker for dentists.