15 questions people ask AI before booking an ayurvedic treatment, and what answers them
Before a patient ever calls your clinic, they are asking an AI assistant these 15 questions. I checked whether 18 real ayurveda sites give the AI a clean answer to each one.
A prospective patient researching ayurveda treatment increasingly starts with an AI assistant, not a search bar. They ask ChatGPT or Perplexity a direct question, and the AI answers from whatever pages it can read cleanly, then names a clinic or two if the evidence is strong enough.
Whether your clinic gets named comes down to whether your site answers these questions in a form the AI can actually lift: a structured block, a clear FAQ, a single unambiguous headline, not just a well written paragraph a human would understand fine. I scanned 18 real ayurveda and panchakarma sites to see how close they come.
The 15 questions, and the fix for each
List the specific conditions you treat in plain text on the page, and back it with FAQ schema so the AI can match a patient's condition directly to your answer.
Name the conditions, do not just say holistic.
State each doctor's qualification (BAMS, MD Ayurveda) on an about or team page, and mark it up with structured data so it reads as a verifiable fact, not a claim.
Credentials in text, not just a photo caption.
Describe your treatment lineage and methodology plainly, and name the specific therapies (abhyanga, shirodhara, basti) rather than generic wellness language.
Specific therapy names beat vague wellness talk.
Publish a short breakdown of which panchakarma therapy addresses which condition, ideally as an FAQ block the AI can extract directly.
Match therapy to condition on the page itself.
State typical program length in days explicitly on the treatment page instead of leaving it to a phone call.
Say the length in days, not just contact us.
Publish at least a starting price or range. A page with zero pricing gives the AI nothing to answer with, so it skips you for a competitor who does.
A price range beats no price at all.
Say plainly whether patients stay onsite or visit daily, near the top of the page, not buried in a brochure PDF.
Answer this in the first paragraph, not a PDF.
Describe the diet plan, sattvic, personalised per dosha, whatever applies, in plain text on the page so it can be quoted directly.
Describe the diet, do not just say customized.
Address safety and side effects honestly on the page or in an FAQ block. Silence here reads as evasive to both patients and AI engines.
Answer the safety question, do not dodge it.
Walk through a typical first day in plain language, ideally as a numbered or FAQ style block the AI can lift wholesale.
Describe day one step by step.
State the booking process and whether the first consultation is free directly on the page, with a clear structured contact block.
Make the booking steps explicit, not implied.
If insurance is not accepted, say so plainly rather than leaving the question unanswered. A clear no is more useful to the AI than silence.
A clear no beats no answer at all.
Publish full address and directions as structured data, this is one of the checks I score for, and it is the fact an AI is most confident answering when present.
Put the full address in structured data.
Name the specific formulations or oils used where relevant, rather than generic terms like natural herbs.
Name specific oils and formulations.
Describe the aftercare plan, follow-up consultation, diet guidance, on the page, since patients weigh this before choosing a program.
State what happens after the program ends.
What I found scanning 18 ayurveda and panchakarma sites
Scanning the 18 sites against these 15 questions confirms the same pattern from the wider study. Structured data is nearly solved: 16 of 18 sites publish JSON-LD. But FAQ schema, the format built for exactly this kind of question and answer, exists on only 2 of 18.
The median overall score was 87, and a single clear H1, the signal that tells an AI what the page is even about, appears on only 4 of 18 sites. Most sites can tell an AI what kind of business they are, but not necessarily which specific question they answer.
The result: a patient's question is often answered somewhere on the page in prose, but not in a form the AI can lift cleanly and attribute to you.
The pattern under all 15
Every one of these 15 questions has the same underlying problem. The trust signal usually exists on the site: the doctor's qualification is in a bio, the price is in a brochure, the address is in the footer. The information is there for a human to find.
What is missing is the machine readable version: FAQ schema pairing the question with the answer, structured data confirming the fact, one clear headline stating what the page is about. An AI answer engine does not read the way a human does, it extracts. If the answer is not in a block it can extract, it will name the clinic down the road that made it easy.
Frequently asked questions
Do AI assistants really influence where patients book ayurveda treatment?
Increasingly, yes. Patients researching treatment now ask ChatGPT or Perplexity direct questions instead of only searching Google, and the AI names the clinics whose sites answer clearly. If your site does not give it a clean answer, it will name someone else's.
Which of these 15 questions matters most to fix first?
Structured data first, since it is the foundation an AI reads before anything else, then FAQ schema, since that is the exact format built to answer a direct question. Both are cheap fixes and neither requires touching your design.
My site already has the answers in normal text. Isn't that enough?
It helps, but it is not the same as a machine readable block. An AI extracting an answer from a paragraph has to work harder and may skip it, whereas FAQ schema or structured data hands the answer over directly. Text plus markup beats text alone.
How do I find out which of these gaps my own site has?
Run my free Report Card tool on your homepage. It checks the same signals used in this study, structured data, FAQ schema, Open Graph tags, and H1 clarity, and shows you exactly what an AI engine sees when it reads your page.