Doctor Profile Pages Study: What 287 Hospital Pages Show
Data collected on by Gaurav Phogat. Method, coding rules and dataset are published below.
I checked 287 doctor profile pages on 31 Indian hospital websites. Almost every page has a photo, a booking button and a list of degrees. The details that help a patient decide, or check, are rare: only 7% show a medical registration number, 7% show a consultation fee and 24% list the languages the doctor speaks. Just under half carry Physician schema, and that half sits mostly on national chain sites.
New to the topic? Start here: Healthcare digital growth sets out the reading order.
Last updated: 7 November 2026.
When a patient or an attendant shortlists a doctor, the doctor profile page is where the decision is made. It is also the page a search engine or an AI assistant reads when someone asks who to see for a knee or a heart problem. I wanted a measured baseline of what these pages actually contain, so I sampled them across chains and standalone hospitals in 12 cities and coded each one against the same checklist.
Key findings
- 20 of 287 profiles (7%) show a medical registration number. They come from just 4 of the 31 sites. None of the 81 national chain profiles I checked showed one.
- 19 of 287 (7%) show a consultation fee, and all 19 come from 2 national chain sites. No regional chain, standalone or teaching hospital profile in the sample showed a fee.
- 70 of 287 (24%) list the languages a doctor speaks. Regional chains lead at 49% of their profiles; standalone hospitals are at 8%.
- 138 of 287 (48%) carry Physician schema markup. National chains are at 86%, regional chains at 39% and standalone hospitals at 31%.
- 255 of 287 (89%) list qualifications and 200 (70%) state years of experience. Credentials are the one area most hospitals cover.
- 195 of 287 (68%) show none of the three decision details I looked for: registration number, languages and fee.
- No profile showed a visible “last updated” date. 89 (31%), from 9 sites, carried a machine-readable modified date in their structured data.
How I did this
Sample. I started from search results for “multispeciality hospital [city] find a doctor” across 12 cities: Delhi NCR, Mumbai, Bengaluru, Chennai, Hyderabad, Kolkata, Pune and Ahmedabad (metros), plus Jaipur, Lucknow, Kochi and Indore (tier 2). Results were heavy on large chains, so I added well-known standalone and regional hospitals in the same cities to balance the mix. I skipped a short list of brands for independence reasons. I tried 45 hospital websites. 14 could not be used: 4 did not load, 6 had no crawlable profile links (the doctor directory was a search box, an app or a separate portal), 3 had no doctor directory I could find from the homepage and 1 I could not read reliably. That leaves 31 sites: 13 standalone, 9 national chain, 8 regional chain and 1 teaching hospital.
Profiles. On each site I opened the main doctor directory (or one city or specialty listing where the directory was split) and took up to 10 profile links spread evenly across the first listing page. Some listings showed fewer, so the per-site count ranges from 5 to 10. The total is 287 profile pages. Chains are counted once per website, not once per branch, because the profile template is shared.
What I recorded. Each profile was loaded in a desktop Chrome browser on 7 November 2026, rendered fully, and checked for nine things: a medical registration number with digits, a photo next to the doctor’s name, qualifications (MBBS, MD, MS, DNB and similar), languages spoken, a consultation fee in rupees, a booking link or button anywhere on the page, Physician schema in the structured data, years of experience, and a visible “last updated” date. I also recorded any schema modified date. Where a site blocked rendering in a frame, I read the raw page instead and flagged that in the data.
Scoring. Seven of the checks make up a completeness score out of 7: registration, photo, qualifications, languages, fee, booking and Physician schema. Experience and dates are reported separately.
Limits. The checks are automated text and markup tests, spot-checked by hand on several sites, so a registration number shown only inside an image, or a fee shown only after picking a slot, would be missed. The photo test confirms an image sits beside the doctor’s name; it cannot always tell a real portrait from a generic one. The booking check counts a site-wide header button too, which is why it reads 100%. Sampling the first listing page favours doctors a hospital chooses to show first. One teaching hospital (7 profiles) is too small to compare on its own. This is a snapshot of one day, not a ranking, and every hospital and doctor is anonymised.
The trust gap: credentials yes, registration no
Hospitals have clearly settled on a template: a portrait, a degree line, a specialty and a big booking button. 89% of profiles list qualifications and 70% state years of experience. That is the part of the page that sells the doctor.
The part that lets a patient check the doctor is almost absent. A state or national medical council registration number is what a careful patient, an insurer or a referring doctor uses to confirm that a doctor is who the page says. Only 20 of 287 profiles showed one, and they came from 4 sites: two standalone hospitals and two regional chains. On one regional chain every sampled profile carried it. On the 81 national chain profiles I checked, the number did not appear once.
This matters more each year. As more patients ask AI assistants which doctor to see, pages with clear, checkable facts are easier for those tools to quote. A registration number is the simplest verifiable fact a hospital can publish about a doctor, and it costs nothing to add.
Schema follows scale
Physician schema is the structured data that tells search engines a page describes a doctor, with fields for name, specialty, hospital and more. Nearly half the profiles had it. The split by hospital type is sharp: 86% of national chain profiles against 31% at standalone hospitals. Eight of the nine national chain sites used it, compared with 4 of 13 standalone sites.
I only checked whether Physician schema was present, not how completely each field was filled, so the real gap is likely wider. Standalone and regional hospitals that run a single profile template can close most of it in one release.
Freshness is the other blind spot. Not one profile showed a visible “last updated” line. 89 profiles from 9 sites carried a modified date inside their structured data, and 66 of those were dated 2026. For the other 22 sites there is no signal at all, for a patient or a crawler, of whether the timings, fee or OPD days on the page are still current.
Languages and fees: the decision details
Two details decide whether a patient books. Can I talk to this doctor in my language, and what will the first visit cost? Both are rare.
Languages spoken appeared on 70 profiles (24%) across 9 sites. Regional chains did best, with 36 of 74 profiles (49%), which makes sense for groups that serve one or two states. National chains showed languages on 28% of profiles and standalone hospitals on 8%. In a country where a patient in Kochi, Hyderabad or Pune may prefer Malayalam, Telugu or Marathi, and where attendants often travel from another state, a line saying which languages the doctor speaks is a small addition with a direct effect on bookings.
Consultation fees were the rarest item: 19 profiles (7%), all from 2 national chain sites. Many sites ask a visitor to pick a slot before showing any price, which my check would not see. Even so, on the profile itself, where most patients decide, the fee is almost never stated. My earlier study of OPD booking flows looks at what happens after the booking button.
How complete is a typical profile?
The median profile scores 4 out of 7. In practice that is photo, degrees, booking button and either schema or languages. Only 6 profiles scored 6, and none scored 7. 195 profiles (68%) had none of the three decision details: registration number, languages or fee.
Scores cluster by site, not by doctor. Profiles on the same site look almost identical, because they come from one template. That is good news for marketers: the fix is a template change, made once, not 300 page edits.
What this means for hospital marketers
These are digital and brand fixes, not clinical advice. Each one adds facts rather than claims, which keeps profiles on the safe side of the NMC and ASCI rules on hospital advertising.
- Add the registration number to the template. One field, the issuing council and the number, shown near the name. It lets patients and insurers check a doctor, and it gives AI assistants a hard fact to cite.
- Show languages spoken. Collect it once from each doctor and display it as a short line. For hospitals that take patients from other states or abroad, this is often more useful than another award.
- Decide your fee policy and state it. If you publish fees, show the first consultation fee on the profile. If you do not, say where the fee will be shown. Hiding it until the last step costs trust.
- Fill the schema you already have. If your profiles carry Physician schema, add the fields that match what is on the page: medical specialty, the hospital location, languages and the registration identifier. If they carry none, start with name, specialty, image and the hospital.
- Show when the page was last checked. A visible “profile last updated” date, backed by the same date in structured data, tells patients and crawlers the OPD timings are current. Then make sure someone owns that review.
- Keep the copy factual. Degrees, experience, specialty, languages, timings and fee. Avoid superlatives such as “best” or “No. 1” and patient testimonials on profiles; they add risk and add little a patient can verify.
If you want to see how these pages fit the wider site, my Indian hospital website benchmark covers speed, accessibility and tracking across 100 sites, and my guide to doctor profile pages that convert covers layout and copy.
Download the data
The full dataset (CSV) has one row per profile page in shuffled order: an anonymised site ID, hospital type, city tier, a yes or no column for each of the nine checks, any schema modified month, the completeness score and the collection date. It contains no hospital names, doctor names, URLs or city names, so pages stay anonymous. Collection date: 7 November 2026. The data is licensed under CC BY 4.0: you can reuse it with credit and a link to this page.
How to cite this study
Phogat, G. (2026). Doctor profile pages on Indian hospital websites: what 287 pages show. gauravphogat.com. https://gauravphogat.com/doctor-profile-pages-study/. Data collected 7 November 2026. Licensed under CC BY 4.0.
Questions people ask
Sources and further reading
- CC BY 4.0 (creativecommons.org)
At minimum: the doctor’s name and photo, qualifications, specialty, years of experience, OPD days and timings, a booking option, languages spoken, the medical registration number and, if the hospital publishes fees, the first consultation fee. In my sample of 287 profiles, most covered the first group but only 7% showed a registration number and 24% listed languages.
Rarely. On 7 November 2026 I found a registration number on 20 of 287 doctor profile pages (7%), from 4 of 31 hospital websites. None of the 81 national chain profiles in the sample showed one.
Physician schema is structured data that tells search engines a page describes a doctor, with fields such as name, specialty and hospital. 138 of the 287 profiles I checked (48%) used it. National chain sites used it on 86% of profiles, standalone hospitals on 31%.
It is a business decision, but stating the fee, or saying where it will be shown, reduces drop off in the booking flow. Only 19 of 287 profiles (7%) in my sample showed a fee, all from 2 national chain sites.
I sampled up to 10 profiles from the doctor directory of each of 31 Indian hospital websites in 12 cities, loaded each page in Chrome on 7 November 2026 and checked it against nine items. Hospitals and doctors are anonymised and the dataset is free to download under CC BY 4.0.
Read my takes first in Google Search
