01 / Today

The search landscape, now.
And how we keep Newmedica ahead of it, together.

A working session, not a presentation. Six short sections. Stop and talk wherever you like, and click on things together as we go. The middle sections are aimed at the local marketing managers in the room.

SessionFuture of search and AI
Prepared forPaul and the wider Newmedica team
DateMay 2026
Use the dots, the arrow keys, or just scroll
02 / The landscape

Search has fragmented.

A patient looking for cataract surgery in 2026 will likely touch five or six different surfaces before they book a consultation. Click each one to see what role it now plays in that journey.

Surface 01
Google Search
Still the starting line for most patients, but no longer the destination.
Ten blue links increasingly sit beneath an AI Overview, a map pack, and a "people also ask" block. The traditional first organic result now sits much further down the page. Patients click less, scan more.
Expand
Surface 02
Google AI Overview
An answer at the top of the page, written by Google, citing a handful of sources.
Appears on a growing share of eye-care queries, particularly informational ones. Conservative on YMYL topics like surgery, but expanding. Whoever gets cited here owns the moment.
Expand
Surface 03
Google AI Mode
The conversational tab. Patients ask, refine, ask again.
Where the patient asks "what should I look for in a cataract clinic" and then follows up with "which is good in Birmingham" and then "what will it cost". Each turn pulls different sources, often more than ten at a time.
Expand
Surface 04
ChatGPT
Sits entirely outside Google. Different sources, different rules.
Tends to lean on editorial sources, named hospitals (Moorfields appears constantly), forum threads, and individual surgeons. The patient never sees a SERP. They see a paragraph and a map.
Expand
Surface 05
Gemini
Google's assistant. Heavy on official sources, schema, and the Knowledge Graph.
Rewards structured, credentialed, regulated content. Clinic websites with strong schema and verified credentials do well here. Less reliant on reviews and forum content than ChatGPT or AI Mode.
Expand
Surface 06
Google Maps and reviews
Where local intent finally converts. Often the final step.
After the AI has answered, the patient still wants to see who is near them, what the reviews say, and what the phone number is. Google Business Profile is where intent becomes a call.
Expand
The big idea

There is no single page one anymore. There are five or six. Each one rewards slightly different things. The work is to win across all of them.

03 / Real searches, real sources

Four real eye-care searches, four different answers.

These are real searches run on each AI surface in May 2026. The queries differ because patients use each surface differently. AI Mode is a back-and-forth conversation. ChatGPT is a longer, more open-ended prompt. AI Overview is a quick one-line lookup. Click each tab to see what came back and where it pulled from.

What each AI surface actually returned

Click each tab
how do i choose a private clinic for cataract surgery
What the AI Overview said
A bulleted guide covering surgeon selection (research consultant ophthalmologists with high success rates), credential verification (GMC specialist register), lens options, transparent package pricing, and the value of multiple consultations. Each bullet linked to a named source.
Sources cited in the answer
moorfields.nhs.uk practiceplusgroup.com mr zaid shalchi (surgeon)
Side panel also surfaced
practiceplusgroup.com practiceplusgroup.com moorfields.nhs.uk
Newmedica is not cited here. The sources are surgeon-led content and named consultants from competitors. AI Overview is rewarding pages where a real, named expert is talking about a single specific topic. The lesson: when a named, credentialed surgeon writes about cataract surgery on a clinic's site, AI tends to pick them as the authority.
Where the answer pulled from
Newmedica
0%
Competitor clinics
~65%
Named surgeons
~20%
NHS / regulators
~15%
Editorial / news
0%
Reviews / forums
0%
Breakdown from this single example. Patterns vary by query, but surgeon-led content keeps appearing.
how do i choose a cataract clinic → which is good in birmingham → what will it cost
What the AI Mode answer said
A full cost breakdown by lens type (£2,295 to £4,000 per eye), then a Birmingham-specific package comparison naming Newmedica Aston Cross, The Priory Hospital, and Spire Parkway Hospital with starting prices for each. Eleven sites cited in total, with Newmedica appearing more than once.
Sources cited in the answer
newmedica.co.uk circlehealthgroup.co.uk spirehealthcare.com + 8 more
Newmedica is cited here. The Birmingham clinic page, the lens options content, and the pricing pages all pulled through. AI Mode rewards depth: detailed pages that anticipate the follow-up questions. A page with itemised pricing, lens options, and named consultants gets pulled far more often than a generic "we offer cataract surgery" page.
Where the answer pulled from
Newmedica
~25%
Competitor clinics
~55%
Named surgeons
~5%
NHS / regulators
~10%
Editorial / news
0%
Reviews / forums
~5%
Eleven sites cited across the conversation. Newmedica appeared on the cost question and on the Birmingham clinic comparison.
help me pick a private cataract clinic, what should i look for and who is good
What ChatGPT returned
An embedded map highlighting Moorfields Private Eye Hospital, followed by a guide centred on surgeon experience and subspecialty. The advice was to ask about annual cataract volume, handling of complex cataracts, and whether the surgeon is also an NHS consultant at a major eye hospital. Recommended researching individual consultants over choosing on clinic brand.
Sources and entities mentioned
moorfields private eye hospital gmc specialist register major teaching hospitals
Newmedica is not visible here. ChatGPT is pushing patients toward named surgeons and teaching hospitals, with Moorfields as the dominant entity. ChatGPT leans heavily on editorial sources and third-party mentions rather than clinic websites. A brand becomes visible here when it gets talked about in trusted external publications, and when named consultants are discoverable as individual experts in their own right.
Where the answer pulled from
Newmedica
0%
Competitor clinics
~50%
Named surgeons
~25%
NHS / regulators
~15%
Editorial / news
~5%
Reviews / forums
~5%
Heavy bias toward named hospitals and individual surgeons. Forum content (Reddit) does show up on broader ChatGPT queries.
how do i choose a private clinic for cataract surgery
What Gemini returned
A five-section guide covering surgeon credentials (Consultant Ophthalmologist, GMC-registered, surgical volume of 500+ cases per year, PCR rates), lens options (Monofocal, Multifocal/Trifocal, Toric, named brands like ZEISS and Alcon), technology and diagnostics (biometry, laser-assisted surgery), full cost (£2,500 to £5,000 per eye), and clinic standards (CQC rating, continuity of care). Personalised to the user's location (Cheltenham).
Sources cited in the answer
bluefinvision.com sussexeyelaserclinic.co.uk centreforsight.com sharminakhan.co.uk hertseyesurgeon.co.uk lasereyesurgeryhub.co.uk thebestofhealth.co.uk
Newmedica is not cited here. Neither are the bigger competitor clinics (Moorfields, Practice Plus, Optegra, SpaMedica). Gemini pulled instead from smaller specialist providers and a surgeon's personal site. The lesson: Gemini seems to favour content that reads like one-on-one expert advice from a named clinician, over brochure-style corporate clinic content. Tone of voice and authorship matter.
Where the answer pulled from
Newmedica
0%
Smaller / specialist clinics
~55%
Named surgeons
~25%
NHS / regulators
0%
Editorial / news
~20%
Reviews / forums
0%
Seven sources cited. None of them the major UK eye-care chains. Notably surgeon-led and specialist-led.
04 / Zooming out

Where AI is pulling from.

Industry research in 2025 and 2026 has tracked over a billion AI search citations. The pattern is consistent across all four surfaces: AI responses are built from a mix of source types. Brand sites like Newmedica's play a real role, alongside community discussion, video, editorial coverage and encyclopedic sources.

Top source categories cited in AI search

Blended across all four surfaces · 2025-2026 data
Community & forums
Reddit, Quora, Stack Exchange
~25%
Video
YouTube
~17%
Brand & business websites
~15%
News & editorial
Forbes, BBC, Guardian
~13%
Encyclopedic
Wikipedia
~12%
Reviews & aggregators
G2, Yelp, Trustpilot
~9%
Professional networks
LinkedIn
~9%
Approximate share of citations across Google AI Overviews, AI Mode, ChatGPT and Gemini, blended from 2025-2026 industry studies. Specific weighting varies by surface and query type.
FINDING 01

Real voices carry real weight

Reddit is the most-cited domain in AI search overall, ahead of YouTube, LinkedIn and Wikipedia. For an eye care brand, the practical translation is that genuine patient stories, honest reviews and real local press coverage shape AI's picture of what a clinic is like. Strong content on your own site does the same job from the other direction.

Search Engine Land (March 2026); Peec AI 30M source analysis (March 2026)

FINDING 02

Brand sites still earn their share

Brand and business sites account for around 15% of AI citations, and the ones AI rewards are those with specific, useful detail: real pricing, named consultants, procedure information, location specifics. Newmedica's site is well-placed for this. The other 85% comes from sources Newmedica influences alongside its own content, through patient stories, local press and real reviews.

Profound 1B citation study (June 2025); Tinuiti Q1 2026 Report

FINDING 03

Google rankings do not guarantee AI citations

Only 38% of pages cited in Google AI Overviews also rank in the top 10 organic results. Around 90% of ChatGPT citations come from pages ranked position 21 or lower. Being visible in AI is a different game from ranking in Google.

Ahrefs (March 2026); Radyant (March 2026)

References
  • Profound · AI Platform Citation Patterns · One billion citations tracked across ChatGPT, Google AI Overviews and Perplexity, Aug 2024 to June 2025
  • Semrush · The Most-Cited Domains in AI · 3-month study, July to October 2025
  • Ahrefs · AI Overview Citation Analysis · 38% organic ranking overlap, March 2026
  • SE Ranking · 70+ AI Search Stats for 2026 · December 2025
  • Tinuiti · Q1 2026 AI Citation Trends Report · Quarterly citation share tracker
  • Peec AI · Top cited domains analysis, 30M data points, March 2026
  • Search Engine Land · AI citation analysis, March 2026
05 / Where you come in

Six things the local team is closer to than we are.

For the local marketing managers in the room. Each one of these directly addresses the gaps we just saw in AI Overview and ChatGPT. Click any card to see what we'd ideally need and why it works.

For the local marketing managers
ASK 01
Real patient stories, named and photographed
A real patient, a real consultant, a real clinic, a real outcome. With consent, ideally a photo and a quote.

What we would ideally need

  • One or two patient stories per clinic per quarter
  • The patient's first name and city
  • The named consultant who treated them
  • The specific procedure
  • A quote from the patient about the experience
  • A photo where the patient agrees (with the relevant consent form signed)

Why this helps AI

First-person verifiable experience is the single strongest signal AI systems look for in YMYL content like eye care. Pages with named patients, named consultants and specific outcomes get cited far more often than pages without.

What we need + why
ASK 02
Richer consultant material
Surgery volumes. Sub-specialisms. Languages spoken. Memberships. Training history. The detail that makes a consultant a discoverable authority.

What we would ideally need

  • Annual surgery volumes per consultant (cataract, glaucoma, etc.)
  • Sub-specialisms and areas of clinical interest
  • Professional memberships and training history
  • Languages spoken (matters more than people realise for local SEO)
  • A short video introduction where consultants are open to it
  • A current high-quality headshot

Why this helps AI

Look at the AI Overview example. The named surgeon (Mr Zaid Shalchi) got cited alongside competitor hospitals. The richer the consultant page (credentials, volumes, sub-specialisms, languages), the more often this happens.

What we need + why
ASK 03
Local press, partnerships, and PR hooks
Anything that puts a clinic's name in a trusted local source. Press mentions, charity work, partnerships with local opticians or GPs, equipment launches, open days.

What we would ideally need

  • A heads-up when any clinic is featured in local press
  • Existing relationships with local opticians, GPs, care homes, sports clubs
  • Charity work and community partnerships, current or planned
  • New equipment installations or service launches at a specific clinic
  • Open days, talks, education events
  • Any local awards or recognition

Why this helps AI

Look at the ChatGPT example. The sources aren't clinic websites, they're editorial mentions and named hospital entities. AI systems use external mentions as a confidence check. The more often a clinic gets mentioned in trusted local press, the more often AI will recommend it.

What we need + why
ASK 04
The real questions patients ask
Not what we think patients ask. What they actually ask, at reception, during the consultation, the day after surgery.

What we would ideally need

  • The five questions reception fields most often
  • What patients are anxious about, in their own words
  • Questions that come up in pre-op assessments
  • Questions patients ask in the post-op follow-up
  • Even (especially) the awkward ones nobody wants to write FAQs for

Why this helps AI

AI search is built around answering questions. Pages that mirror the actual phrasing patients use become the source AI cites. We want our FAQ pages to read like a transcript of what happens in the clinic, not a marketing department's guess at what patients are wondering.

What we need + why
ASK 05
Reviews with real detail
When asking for reviews, prompt patients to name the consultant, the procedure, and the clinic. Detail beats stars.

What we would ideally need

  • A consistent prompt when asking for reviews ("if you can mention the consultant and what you had done, that helps others")
  • Encouragement to write more than one sentence
  • Reviews that mention the city or clinic specifically
  • A small system for nudging happy patients to leave a review the same day

Why this helps AI

"Five stars, would recommend" tells AI nothing. "Mr Patel did my cataract surgery at the Norwich clinic in March, the whole team were brilliant" tells it everything. Detailed reviews give AI specific things to cite, and feed Trustpilot, Google and Doctify visibility at the same time.

What we need + why
ASK 06
What makes each clinic different
The specifics. Equipment, accessibility, parking, languages, NHS partnerships, specialist procedures only available at one location.

What we would ideally need

  • Specialist equipment or technology at a specific clinic
  • Procedures only available at certain locations
  • Languages spoken by reception staff
  • Accessibility features (wheelchair access, hearing loops, sensory accommodations)
  • Parking, public transport, drop-off arrangements
  • NHS partnerships or referral routes specific to that clinic
  • Anything that makes that clinic genuinely the right answer for a specific patient's circumstance

Why this helps AI

AI Mode is already pulling from the Birmingham clinic page because it has real detail (Aston Cross location, package pricing). Every clinic should be this rich. Specificity is what wins local AI searches.

What we need + why
06 / Quick quiz

Four questions to check the room.

No marks. Just a quick way to test that we're all looking at AI search the same way before we head into the discussion. Click an answer to see if it lands.

QUESTION 01 of 04
For AI Mode to cite Newmedica's Birmingham page, what helps most?
Why
QUESTION 02 of 04
When AI Overview cites a competitor's surgeon page over a Newmedica page, what is usually different about that surgeon's page?
Why
QUESTION 03 of 04
To close the ChatGPT visibility gap, what helps Newmedica most?
Why
QUESTION 04 of 04
A patient asks AI Mode "what's the best clinic in Norwich for cataract surgery". What gives Newmedica Norwich the best chance of being recommended?
Why
07 / Talk it through

Four questions worth sitting with.

Click any card to flip it. There are no right answers. The point is to talk through them and let what comes out of the conversation shape what we work on next quarter.

Question 01 · click to reveal
Which clinic has the patient story we should tell first?
Question 02 · click to reveal
Which consultant feels most ready to be the visible face of their specialism online?
Question 03 · click to reveal
What question do patients ask in person that nothing on the website currently answers?
Question 04 · click to reveal
What makes your clinic genuinely different from the Newmedica down the road, that nobody outside the clinic knows yet?

Thanks for your time today.

← → to move sections