89% of the sites scanned scored an A or a B. This is not a story about widespread invisibility, and anyone telling you it is has not counted. The real finding is narrower and more useful: 4 of 35 sites (11%) publish no structured data at all — and they are exactly the sites at the bottom. Structured data isn't the only thing that matters, but here it separates the tail from everyone else more cleanly than any other single signal.
$The distribution
Every site scanned, by grade. The shape matters more than any single row: a healthy curve with a thin tail, not a crisis.
$The sites below 85
The 16 sites scoring 85 or under, lowest first. Open any row to see what was actually found, what it costs, and where the site sits against others in its own city. Every finding was re-checked against raw HTML before publishing.
| Clinic ↕ | City ↕ | Speciality ↕ | Score ↑ | Structured data ↕ | |
|---|---|---|---|---|---|
| Chisel Dentalchiseldental.co.in | Bangalore | Dental / Implants | 42D | none | |
|
10 clinics in Bangalore scored higher than this one — 7th of 7 dental / implants sites scanned in the city.
What an AI finds when it looks for structured facts about this clinic
// nothing — no structured data on the page Verified against the raw HTML: zero JSON-LD blocks.
What a visitor sees
A working dental clinic site — 1,377 words of real content, photos, treatments, contact details.
The information exists. It just isn't in a form a machine can read.
No structured data at all When someone asks ChatGPT or Google's AI for a dental clinic in Bangalore, there is no machine-readable record saying this practice exists, where it is, or what it treats. The AI has to infer all of it from prose — or name someone else. No Open Graph tags Every time a patient forwards this link on WhatsApp, it arrives as a bare URL — no name, no photo, nothing to react to. That is the moment a recommendation either lands or dies. No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. Title is 77 characters Google shows about 60. The tail is invisible to a human and reads as padding to a search engine. |
|||||
| Dr Madhu's Hair Transplant Centrehairhospi.com | Hyderabad | Hair Transplant | 50C | none | |
|
3 clinics in Hyderabad scored higher than this one — 4th of 4 hair transplant sites scanned in the city.
What an AI finds when it looks for structured facts about this clinic
// nothing — no structured data on the page Verified against the raw HTML: zero JSON-LD blocks.
What a visitor sees
A working hair transplant clinic site — 292 words of real content, photos, treatments, contact details.
The information exists. It just isn't in a form a machine can read.
No structured data at all When someone asks ChatGPT or Google's AI for a hair transplant clinic in Hyderabad, there is no machine-readable record saying this practice exists, where it is, or what it treats. The AI has to infer all of it from prose — or name someone else. No Open Graph tags Every time a patient forwards this link on WhatsApp, it arrives as a bare URL — no name, no photo, nothing to react to. That is the moment a recommendation either lands or dies. No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. |
|||||
| Dr Kheur Dentaldrkheurdental.com | Pune | Dental / Implants | 58C | none | |
|
6 clinics in Pune scored higher than this one — 7th of 7 dental / implants sites scanned in the city.
What an AI finds when it looks for structured facts about this clinic
// nothing — no structured data on the page Verified against the raw HTML: zero JSON-LD blocks.
What a visitor sees
A working dental clinic site — 1,139 words of real content, photos, treatments, contact details.
The information exists. It just isn't in a form a machine can read.
No structured data at all When someone asks ChatGPT or Google's AI for a dental clinic in Pune, there is no machine-readable record saying this practice exists, where it is, or what it treats. The AI has to infer all of it from prose — or name someone else. No Open Graph tags Every time a patient forwards this link on WhatsApp, it arrives as a bare URL — no name, no photo, nothing to react to. That is the moment a recommendation either lands or dies. No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. |
|||||
| WEA Clinicweaclinic.com | Chennai | Derm / Skin | 69C | yes | |
|
1 clinic in Chennai scored higher than this one — 2nd of 2 derm / skin sites scanned in the city.
No Open Graph tags Every time a patient forwards this link on WhatsApp, it arrives as a bare URL — no name, no photo, nothing to react to. That is the moment a recommendation either lands or dies. No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. |
|||||
| World IVF Centreworldivfcentre.com | Delhi NCR | Fertility / IVF | 70B | none | |
|
3 clinics in Delhi NCR scored higher than this one — 4th of 4 fertility / ivf sites scanned in the city.
What an AI finds when it looks for structured facts about this clinic
// nothing — no structured data on the page Verified against the raw HTML: zero JSON-LD blocks.
What a visitor sees
A working fertility clinic site — 1,690 words of real content, photos, treatments, contact details.
The information exists. It just isn't in a form a machine can read.
No structured data at all When someone asks ChatGPT or Google's AI for a fertility clinic in Delhi NCR, there is no machine-readable record saying this practice exists, where it is, or what it treats. The AI has to infer all of it from prose — or name someone else. No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. |
|||||
| Dr Parag Telangdrparagtelang.com | Mumbai | Cosmetic / Plastic | 76B | yes | |
|
6 clinics in Mumbai scored higher than this one — 7th of 7 cosmetic / plastic sites scanned in the city.
No Open Graph tags Every time a patient forwards this link on WhatsApp, it arrives as a bare URL — no name, no photo, nothing to react to. That is the moment a recommendation either lands or dies. No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. Title is 82 characters Google shows about 60. The tail is invisible to a human and reads as padding to a search engine. |
|||||
| Caree Fertility Centrecareefertilitycentre.com | Bangalore | Fertility / IVF | 77B | yes | |
|
7 clinics in Bangalore scored higher than this one — 4th of 4 fertility / ivf sites scanned in the city.
No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. |
|||||
| Dr C Jagadeesh Dentaldrcjagadeesh.com | Bangalore | Dental / Implants | 77B | yes | |
|
7 clinics in Bangalore scored higher than this one — 5th of 7 dental / implants sites scanned in the city.
No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. Title is 80 characters Google shows about 60. The tail is invisible to a human and reads as padding to a search engine. |
|||||
| Cosmetic Dental Clinicscosmeticdentalclinics.in | Bangalore | Dental / Implants | 77B | yes | |
|
7 clinics in Bangalore scored higher than this one — 5th of 7 dental / implants sites scanned in the city.
No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. Title is 75 characters Google shows about 60. The tail is invisible to a human and reads as padding to a search engine. |
|||||
| Sapling IVFsaplingivf.com | Delhi NCR | Fertility / IVF | 80B | yes | |
|
2 clinics in Delhi NCR scored higher than this one — 3rd of 4 fertility / ivf sites scanned in the city.
No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. |
|||||
| Burute Dentalburutedental.in | Pune | Dental / Implants | 81B | yes | |
|
5 clinics in Pune scored higher than this one — 6th of 7 dental / implants sites scanned in the city.
No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. |
|||||
| Aesthetic Artaestheticart.in | Mumbai | Cosmetic / Plastic | 82B | yes | |
|
5 clinics in Mumbai scored higher than this one — 6th of 7 cosmetic / plastic sites scanned in the city.
No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. Title is 75 characters Google shows about 60. The tail is invisible to a human and reads as padding to a search engine. |
|||||
| Beyond Smiles Dentalbeyondsmiles.co.in | Bangalore | Dental / Implants | 83B | yes | |
|
6 clinics in Bangalore scored higher than this one — 4th of 7 dental / implants sites scanned in the city.
No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. |
|||||
| Hair Surehairsure.in | Hyderabad | Hair Transplant | 83B | yes | |
|
2 clinics in Hyderabad scored higher than this one — 3rd of 4 hair transplant sites scanned in the city.
No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. Title is 92 characters Google shows about 60. The tail is invisible to a human and reads as padding to a search engine. |
|||||
| Smilekraft Dentistrysmilekraftdentistry.com | Pune | Dental / Implants | 83B | yes | |
|
4 clinics in Pune scored higher than this one — 5th of 7 dental / implants sites scanned in the city.
Title is 71 characters Google shows about 60. The tail is invisible to a human and reads as padding to a search engine. |
|||||
| Dr Mohan Thomas Aestheticsdrmohanthomasaesthetics.com | Mumbai | Cosmetic / Plastic | 84B | yes | |
|
4 clinics in Mumbai scored higher than this one — 5th of 7 cosmetic / plastic sites scanned in the city.
No FAQ markup The answers are on the page as prose, which a machine can already read. FAQ markup only labels which text answers which question, so nothing has to be inferred. It's the smallest item here — the answers themselves are what get quoted, not the JSON wrapped around them. Title is 77 characters Google shows about 60. The tail is invisible to a human and reads as padding to a search engine. |
|||||
$Methodology
- Sourcing. Independent clinics found through public search across 6 cities and 5 specialities. Hospital chains, aggregators and directories (Practo, JustDial and similar) excluded — the point was sites a single practice actually controls.
- Scanning. Each homepage scanned once with the free AEO Checker. It's deterministic: every check is a concrete, verifiable signal and no AI judges anything.
- Verification. Every published finding re-checked against the page's raw HTML source, not a rendered DOM. Where the scanner and the source disagreed, the row was dropped — the tool being wrong is not the clinic's problem.
- Statistics. Every number on this page — median, distribution, percentages, the with/without split, and each row's position against its city — is computed in code from the scan data at build time. None are typed by hand.
- Exclusions. 35 candidates scanned; 35 produced usable data; 16 scored 85 or under and appear above. Dropped: anything unscannable, anything above 85, and any page resembling a shell or mid-migration state where a score would mislead.
- What the number is. A measure of how legible a page is to software. Not a measure of the clinic, the doctors, or the care.
- It's a small sample. 35 sites is enough to see a pattern, not enough to describe every clinic in India.
- It's one day. Scanned 15 July 2026. Some of these numbers are probably already stale.
- It's the homepage only. A site could have good structured data on inner pages and none on the homepage.
- Correlation, not causation. The 34-point gap is real in this sample, but sites missing structured data tend to be under-maintained generally. Adding JSON-LD alone won't move a site from 54 to 88.
- It doesn't measure citations. Being machine-readable makes you easier to quote. It doesn't guarantee any AI recommends you.
$Questions
What is the AI Visibility Index?
A snapshot of how 35 independent Indian clinic websites are built for AI answer engines, scanned on 15 July 2026. It measures whether a page is machine-readable — not the quality of the clinic's medical care.
Does a low score mean the clinic is bad?
No, and this matters. The score describes a website's markup and nothing else. An excellent clinic can score badly because nobody added structured data to its site. This measures code, not care.
Are most clinics doing badly?
No. The median across all 35 scanned was 87/100 and 89% scored an A or a B. Most are fine. The finding is a specific minority: 11% publish no structured data at all, and they hold the 3 lowest scores in the sample.
What is structured data?
A machine-readable label in a page's code stating what a business is, where it is and what it does. In this sample, clinics with it had a median of 88; those without had a median of 54.
How was this measured?
Each homepage was scanned once with the free AEO Checker on vishkul.com, then every published finding was re-checked against the page's raw HTML source. Anything that didn't reproduce was dropped. Nothing is rendered-DOM based; nothing is an AI's opinion.
How do I get my clinic updated or removed?
Email [email protected] and it's updated or removed, no questions asked. You can also re-run the scan yourself — the tool is free and public.