Marketing reputation

Being found in AI search answers

What AI search features read, how dental practices and private clinics can be cited in AI answers, and an audit you can run in an afternoon.

Ava research teamPublished 7 min read

On this page
  1. Patients are asking machines first
  2. What AI search actually reads
  3. Myths and what to do instead
  4. Make your pages easy to quote
  5. Pages worth having before anything else
  6. Be the same practice everywhere
  7. Reviews and ratings will matter more
  8. Accuracy is the real advantage in health
  9. An afternoon audit
  10. Example: correcting a stubborn error
  11. Frequently asked questions
  12. References

Key takeaways

  • AI search answers draw on ordinary web pages; Google says there are no extra requirements beyond being indexed and eligible for a snippet[1].
  • 40% of UK adults now use AI chatbots for personal purposes at least monthly, so patients will ask them about local clinics[2].
  • Clear text answers, consistent listings and accurate structured data make a clinic easier to cite correctly.
  • Health is the area where the public most often wants AI handled carefully, so accuracy and dates matter more than volume[3].

Patients are asking machines first

A patient looking for a dentist who sees nervous adults, or a clinic that treats a skin condition, may no longer type a few words and scan ten blue links. Increasingly they ask a full question and read a summarised answer, either in a search engine's AI feature or in a chatbot. In the government's public attitudes tracker, 40% of UK adults said they use AI chatbots for personal purposes at least once a month, up from 34% in the previous wave[2].

UK adults using AI chatbots for personal purposes at least monthlyDSIT public attitudes tracker
UK adults using AI chatbots for personal purposes at least monthlyComparison. Previous tracker wave: 34%; Wave 4: 40%.34%Previous tracker wavePrevious tracker wave: 34%40%Wave 4Wave 4: 40%UK adults using AI chatbots for personal purposes at least monthlyComparison. Previous tracker wave: 34%; Wave 4: 40%.34%Previous tracker wavePrevious tracker wave: 34%40%Wave 4Wave 4: 40%

Source: GOV.UK (Department for Science, Innovation and Technology), 2024[2]

Show the data
ItemValue
Previous tracker wave34%
Wave 440%

For a clinic, the question is not whether to "do AI marketing". It is whether the information these systems find about you is accurate, current and easy to quote. When it is not, an AI answer may describe old opening hours, a service you no longer offer, or a competitor instead.

What AI search actually reads

Google's own guidance for site owners is useful because it is specific. It says there are no additional requirements to appear in AI Overviews or AI Mode, and no special schema.org structured data you need to add[1]. To be shown as a supporting link, a page must be indexed and eligible to be shown in Google Search with a snippet[1]. The best practices it lists are familiar: make important content available in text form, provide a good page experience, let crawlers reach the page, use internal links, and make sure structured data matches the visible text[1].

Other AI assistants work in different ways, and not all of them explain their sources. The safe assumption is that they reward the same things: plain text that answers a question directly, consistent facts across the web, and pages that are easy to crawl.

If a person could not find the answer on your page in ten seconds, an AI system is unlikely to quote it accurately either.

Myths and what to do instead

Common beliefs about AI search compared with what clinics can control
MythRealityWhat to do
You need special AI code on the siteGoogle says no special structured data is required[1]Fix basic crawlability and accurate structured data
More pages means more mentionsThin, repeated pages add littleFewer, clearer pages that answer specific questions
AI answers ignore reviewsRatings and patient experience feed many local resultsAsk for honest reviews and respond professionally
You can pay to be in AI answersOrganic AI answers are not sold as placementsTreat paid ads as a separate channel
Once listed, you stay listedAnswers change as pages and listings changeReview key facts and answers every quarter

Make your pages easy to quote

AI summaries lift short, self-contained answers. Pages that bury key facts in images, PDFs or long introductions are harder to use. For each service and for your practical information, write the answer first and the detail second.

  • Put opening hours, address, parking, step-free access and phone number on a page in text, not only in an image or a map embed.
  • On treatment pages, start with one or two sentences that say what the treatment is, who it suits and whether you offer it privately, on the NHS or both.
  • Use question-style headings that match how people ask: "Do you take new NHS patients?", "How much is a hygiene appointment?".
  • Name the clinicians and their qualifications in text, with registration details where your regulator expects them.
  • Show a last-updated date. Dental practice websites must already show when the site was last updated[4], and a visible date helps any reader judge freshness.
  • Keep structured data, such as opening hours, consistent with what the page says. Google asks for structured data to match the visible text[1].

Pages worth having before anything else

Most clinics do not need more content; they need a handful of pages that answer the questions AI tools are most often asked about local healthcare. Check you have each of these, written in plain text and kept current:

  • Practical information: hours, bank holiday arrangements, address, parking, public transport, step-free access, and how to get urgent help when you are closed.
  • New patients: whether you are accepting new NHS or private patients, how to register or book, and what to bring to a first appointment.
  • Fees: an up-to-date fee guide, what is included and how payment or finance works.
  • One page per main treatment, each answering who it suits, what happens, who provides it and what it costs.
  • The team: names, roles, qualifications and registration details for each clinician.
  • Accessibility and languages: interpreters, hearing loops, longer appointments for people who need them.

These pages are also the ones reception spends the most time explaining on the phone. Writing them well reduces calls as well as improving how you are described elsewhere.

Be the same practice everywhere

AI systems cross-check facts across sources. If your website says you open at 8am, your map listing says 8.30am and an old directory lists a number you no longer use, the answer may pick the wrong one or hedge. Online search is already a main route to a new practice: among adults without a dentist seeking urgent NHS care, 30.9% found one through an online search[5]. The same consistency that helps searchers helps AI answers.

  1. List every place your practice appears: website, map listings, NHS service profile, directories, social profiles, insurer and referral directories.
  2. Agree the master details: exact name, address, phone number, hours, services and NHS or private status.
  3. Update each listing, starting with the ones that show up first when you search your name.
  4. Remove or claim duplicate listings and old addresses.
  5. Set a calendar reminder for bank holidays and seasonal hours so listings are updated in advance.

Reviews and ratings will matter more

Patient feedback is becoming part of how people choose. The 10 Year Health Plan says patients will be able to choose providers on the NHS App using quality data including waiting times, patient ratings and clinical outcomes[6]. Private clinics already compete on reviews. Ask every satisfied patient for an honest review through a simple link, never offer incentives that are not disclosed, and reply to criticism calmly without discussing any patient's care in public.

29%

of UK adults pick healthcare as the area where AI most needs careful management[3]

Accuracy is the real advantage in health

People are open to AI in healthcare but cautious. 52% of UK adults expect AI to have a positive impact on healthcare[7], yet healthcare is also the area most often chosen as needing careful management to prevent negative outcomes[3]. A clinic whose pages are accurate, clinically reviewed and dated is a better source than one with dozens of vague articles. Make sure health information is checked by a clinician, cite reliable sources for any claims, and remove pages that are out of date. If a page gives general health advice, say who wrote it, who reviewed it and when, and point readers to NHS 111 or 999 where urgent symptoms are mentioned. That signals care to readers and gives any system quoting the page the context it needs.

An afternoon audit

  1. Ask the questions patients ask

    In two or three AI tools and an ordinary search, ask questions such as "dentist near [town] taking NHS patients" or "private physio [town] Saturday". Note what is said about you.

  2. Check the facts

    Compare each answer with your real hours, services, fees and status. Note every error and where it seems to come from.

  3. Fix the sources

    Correct the website and listings that carry the wrong facts. Add short text answers where the information was missing.

  4. Check crawlability

    Use your search console to confirm key pages are indexed, and that nothing blocks crawlers or removes snippets by accident.

  5. Repeat quarterly

    Run the same questions every three months and log changes. AI answers move as the web changes.

Example: correcting a stubborn error

Example: Harbour Smile Studio (fictional) found that AI answers kept saying it offered NHS treatment, which it had stopped two years earlier. The source was an old directory listing and a blog post celebrating a past NHS contract. The practice updated the directory, added "We are a private practice" to its home page and contact page in text, and added a short note to the old post. Within a few months the answers described the practice correctly, and reception received fewer calls asking for NHS appointments.

Frequently asked questions

Do we need special code to appear in AI Overviews?

No. Google says there are no additional requirements and no special structured data needed; a page must be indexed and eligible for a snippet[1].

Can we stop AI features using our content?

Google points site owners to preview controls such as nosnippet, data-nosnippet, max-snippet and noindex to limit what is shown[1]. Blocking snippets can also reduce how your pages appear in ordinary search.

Should we write lots of short articles for AI search?

Rarely. A smaller number of accurate, specific, clinically reviewed pages that answer real questions is more useful to patients and more likely to be cited correctly.

How do we know if AI answers mention us?

There is no complete report. Ask the same patient-style questions in several tools every quarter, note what they say, and track enquiries that mention finding you through an AI answer.

References

  1. Google Search Central (2026). AI features and your website (opens in a new tab) ↑

  2. GOV.UK (Department for Science, Innovation and Technology) (2024). Public attitudes to data and AI: Tracker survey (Wave 4) report (opens in a new tab) ↑

  3. GOV.UK (Department for Science, Innovation and Technology) (2024). Public attitudes to data and AI: Tracker survey (Wave 4) report (opens in a new tab) ↑

  4. General Dental Council (2026). Guidance on advertising (opens in a new tab) ↑

  5. Office for National Statistics (2026). Experiences of NHS healthcare services in England: September 2026 (opens in a new tab) ↑

  6. GOV.UK (Department of Health and Social Care) (2025). Fit for the future: 10 Year Health Plan for England (opens in a new tab) ↑

  7. GOV.UK (Department for Science, Innovation and Technology) (2024). Public attitudes to data and AI: Tracker survey (Wave 4) report (opens in a new tab) ↑

Book a demo

See Ava handle your calls.

Book a short demo. We’ll walk through your calls and workflow, and agree what Ava should take care of.

Book a demo