AI Visibility · 19 July 2026

By Eddie Eastwood, Founder

Google Reviews or AI Mentions: Which One Do Patients Actually Trust?

A patient who has been hiding their smile in photographs for years finally decides to fix it. They do not flick through a directory and they do not scroll a page of blue links. They open an assistant and type, "who is a good dentist near me for veneers?" Seconds later they have a name, a sentence of praise, and a decision half made. The practice in that answer wins a case that, on a typical course of veneers, runs well into four figures. The practices that are not in the answer never find out the patient existed.

That is the money sitting underneath this whole question, so it is worth naming the stakes plainly before we get to the psychology. To put illustrative numbers on it (these are typical UK ranges, not researched figures): a single dental implant commonly sits somewhere in the region of £2,000 to £2,500 per tooth, a course of clear aligners often lands between £2,500 and £4,500, and a smile makeover in veneers can reach several thousand pounds.

A patient who simply joins your list for routine care is worth years of check-ups, hygiene visits and the occasional larger treatment: a relationship measured in thousands of pounds across a decade. These are exactly the decisions a nervous patient researches carefully, and increasingly they research them by asking an assistant and by reading reviews, often in the same sitting.

So the real question owners ask us is not academic. It is, "where do I build my credibility, and which one do patients believe?" This article answers that. It is written for UK practices, it stays inside General Dental Council rules throughout, and each section answers a question an owner has actually put to us, so you can read the part that fits your situation and act on it.

Google reviews or AI mentions: which one builds more trust for a dental practice?

Google reviews build the deeper, more durable trust, because a real, named patient staked their own experience and a prospective patient can read it and judge it for themselves. An AI mention carries a different kind of weight: it sounds authoritative, but the patient cannot see who vouched for you or why, and the assistant is largely repeating what your reviews and listings already say.

So this is not a fair fight between two rival trust sources. Reviews are the foundation of trust; an AI mention is an amplifier of that trust, one you do not fully control, and one your reviews mostly earn in the first place.

Hold onto that hierarchy, because it resolves the false choice the rest of the field keeps setting up. The two things are not interchangeable and they are not opponents. They build trust in opposite directions: a review is bottom-up proof from a peer, while an AI mention is a top-down claim from an authority. A patient weighs the two very differently, and so should you when you decide where to put your limited time.

Both, meanwhile, are now mainstream at once, which is why ignoring either is expensive. Online reviews remain, in the words of the largest annual survey of the field, "one of the most powerful drivers of trust and decision-making," even as the way people find them shifts ^[1].

In the same study, assistants like ChatGPT have surged into third place as a source people use for local-business recommendations, up from 6% of people a year earlier to 45%. Patients bring health questions to these tools in particular: roughly a third (32%) of US adults now use AI for health information or advice ^[2].

And inside ordinary Google, 58% of users saw at least one AI summary in their searches in a single recent month, and they rarely clicked the sources underneath it ^[3]. Read those together and the job is clear: reviews are how patients still decide, AI answers are how a growing share first hear of you, and the two are wired together underneath.

What kind of trust does a Google review actually build?

A Google review builds attributable peer trust: a named human has put their real experience on record, in public, where a prospective patient can read it, weigh it, and check whether other people say the same thing. That last part is the quiet engine of it.

A patient does not trust a single glowing entry; they trust a pattern of them. When the largest review survey asked what mattered most when judging reviews, the top answer was that the review is backed up by other reviews with similar sentiment, chosen ahead of raw volume ^[1]. Trust here is consensus a stranger can verify with their own eyes.

That is why a few specific levers matter far more than the headline count. Recency is one: patients increasingly discount old praise, and a slow or generic reply to a recent review now reads as a warning sign rather than a neutral. Star quality is another, with a sharp rise in people who only use a business rated 4.5 stars or higher.

Specificity is the third, and it is the one most practices underuse. A review that says "they talked my anxious son through every step and told me the cost before they started" earns trust in a way "lovely practice, five stars" never can, because the detail is checkable and human. None of this is something you can manufacture. You earn it patient by patient, and that is precisely what makes it credible.

The most important thing to understand about review trust is that the patient still clicks in to read it. They want to see the words, the dates, the pattern, and your replies. That act of reading and judging is irreplaceable, and no machine summary substitutes for it. A review is slow trust, built in public, owned by the people who gave it. That is its strength and, as we will see, the reason an AI mention cannot stand without it.

What kind of trust does an AI mention build, and why is it different?

An AI mention builds machine-asserted authority: the assistant tells the patient "this practice is well regarded," and because the answer sounds confident and arrives first, the patient tends to take it at face value. The catch is in what the patient cannot see.

The mention is unattributed (no named person is vouching), unverifiable in the moment (the patient rarely clicks through to check the evidence), and probabilistic rather than fixed (ask the same assistant twice and the wording, even the names, can shift). It is fast trust, borrowed trust, and it is fragile in a way a wall of real reviews is not.

This is not an argument that AI mentions do not matter. They matter enormously, because for a rising share of patients the assistant's answer is the first and sometimes the only impression they form. With most people not clicking the sources behind an AI summary, the assistant's one-line verdict on you is doing the work that a page of reviews used to do. That is real influence over a real decision worth thousands of pounds.

But influence is not the same as a trust source you own. The assistant is not inventing its good opinion of you out of thin air. It is synthesising what it can find: your reviews, your listings, the mentions of you scattered across the web. Strip those away and the mention either disappears or becomes a hollow claim that collapses the moment a careful patient looks closer.

So the plain framing for an owner is this. An AI mention is the trust your reviews and your reputation throw onto a bigger screen. It reaches patients earlier and wider than reviews alone ever could, but it is reflected credibility, not original credibility. Treat it as the amplifier, never the source.

Google reviews vs AI mentions: the plain side-by-side

Here is the contrast drawn cleanly, because no competing dental page does it, and it is the quickest way to see why the two are not rivals. Read the columns against each other and the relationship gives itself away.

Look at the row marked "what builds it." Reviews appear in the right-hand column as one of the things that build an AI mention. The reverse is nowhere to be found: no amount of AI visibility writes you a single patient review. The influence runs one way only. Your reviews help create the mention; the mention creates nothing back. That asymmetry is the whole argument for the order of work we set out later, and it is the part almost every owner gets back to front.

The part most practices miss: your reviews are what earn the AI mention

Reviews and AI mentions are not opponents, because reviews are one of the strongest ingredients that produce a mention in the first place. The clearest statement of that comes from inside the review industry.

Describing what reviews have become, BrightLocal's chief executive says they are "an essential piece of evidence that your business is active, reliable," and "worthy of prominent visibility and citation within traditional Google search and LLMs like ChatGPT, and AI search" ^[1]. The word to notice is citation. The people who measure reviews for a living are telling you that a strong review presence is now part of how the assistants decide who to put forward.

The mechanism is more specific than "good reviews help," and understanding it changes how you ask for reviews. When a patient asks an assistant "which dentist near me is good with nervous patients for implants?", the assistant is hunting for evidence it can point to. A review that describes the team calmly walking an anxious patient through a sedation implant case and explaining the costs upfront hands the model something concrete to repeat. A generic "5 stars, great practice" gives it nothing to say. Compare the two:

• Generic: "Brilliant dentist, highly recommend." Trustworthy to a human skimming, but it gives an assistant nothing to quote about any particular treatment or strength.
• Specific: "I was terrified of implants after a bad experience years ago. They explained every stage, showed me the cost before starting, and I felt no pain on the day." This becomes the substance of a confident, evidence-backed mention about implants and nervous patients.

So your goal is not simply more reviews or fresher reviews, useful as both are. It is reviews specific enough to be repeated, because a quotable review is a mention waiting to happen. This is the bridge that dissolves the false versus. The patient-authored trust on your review profile is the raw material the machine-asserted trust is built from.

A practice with deep, specific, recent reviews is feeding the assistant exactly what it needs to name you with confidence. A practice with a thin or generic profile is asking the assistant to vouch for it on faith, and assistants, like careful patients, do not.

So which signal should a dentist build first?

Build the reviews first, every time, because they are the only investment that pays back in both human trust and AI visibility at once. The order matters precisely because the influence runs one way. Pour effort into being mentioned by assistants while your reviews stay shallow, and you have built an amplifier with nothing worth amplifying. Build the reviews first, and everything downstream has something real to work with.

The practical sequence follows from that. Stage one is your review foundation: a steady, genuine flow of recent reviews, with patients gently encouraged to describe what they came in for and how it went, and a presence on more than just Google, since careful patients cross-check several sites before committing to costly elective work.

Stage two is consistency, the unglamorous step that lets the trust attach to the right practice: your name, address, phone, opening hours and service list reading identically wherever they appear, so an assistant can connect your reviews to a single, verifiable practice rather than hedging between two half-matching records (we cover this in depth in the companion piece on keeping your practice details consistent across the web). Stage three is the citable content: clear, answer-first pages for your high-value treatments, written the way patients actually ask, with realistic price ranges and the facts laid out so a machine reads them without guessing.

Run in that order and the spend compounds. The specific reviews you gather in stage one become quotable evidence once stages two and three make you a clean, verifiable entity. Reverse the order and you build a tidy, citable shell with no compelling trust inside it. The sequence is the strategy, and it starts with the one signal patients believe most.

Can you trust an AI mention that has no real reviews behind it?

No, and neither can your patients, which is the whole risk of treating the mention as the goal. An assistant can occasionally name a practice on thin evidence, and for a short while a burst of fake or AI-written reviews can inflate the signal that earns a mention. Both routes fail the moment a real patient applies real scrutiny.

The patient clicks through, finds a sparse or stale review profile that does not match the confident answer, and the borrowed trust evaporates. Worse, fabricated or incentivised reviews are not a clever shortcut; they breach the review platforms' policies and they breach the standards a dental practice is held to, and they are exactly the kind of thing that destroys trust permanently when it is found out.

The deeper point is that an AI mention is only ever as trustworthy as the evidence underneath it. That is good news, because it means you cannot be locked out by a competitor with a bigger budget, and it means the durable path is also the straight one. A real, specific, well-tended review base is both the trust patients believe and the evidence the assistant needs. Chase the mention directly and you are building on sand. Build the reviews and the mention becomes the natural, defensible result.

How do you build both signals the right way? The dental playbook

This is the practical part, grouped into the moves that build patient trust and machine-readable trust together. None of it needs a large budget, and you can verify all of it yourself.

Keep a steady, authentic review drip, never gated or incentivised

Because patients increasingly value only the most recent reviews, a one-off campaign decays fast and a steady habit wins. Ask satisfied patients at the natural moment, after a completed course of treatment rather than mid-appointment, and make leaving a review a single tap from a text or email. Crucially, never filter so that only happy patients are invited, and never offer anything in return. Genuine, unfiltered feedback is both the ethical standard and the only kind that builds trust a patient or an assistant will rely on.

Ask for specifics, so the trust is checkable and quotable

A review that names the treatment and the experience is worth several generic raves, for humans and for assistants alike. You can prompt this without scripting anyone: simply invite patients to mention what they came in for and how it felt. "They explained every cost before my implant and the whole thing was painless" is the kind of sentence a careful patient believes and an assistant can repeat. Specificity is where patient trust and machine trust meet.

Respond to every review, because silence now reads as a red flag

Reply to reviews, good and bad, as routine. It signals an attentive practice to patients and to the systems weighing your reputation, and slow or generic responses are increasingly taken as a warning. One firm rule for dentistry: never put a patient's clinical details in a public reply. Thank them, address the general point, and take anything specific to a private channel. That single discipline protects confidentiality and reads well to every reader, human or machine.

Spread beyond Google, so the trust is corroborated

Google is necessary but not the finish line, because careful patients and assistants both cross-check across several sources before trusting an expensive decision. A presence on reputable healthcare and local platforms gives patients more places to confirm the pattern and gives an assistant more independent evidence to agree on. Claim and complete the listings that matter for a UK dental practice, and keep every one accurate.

Keep your details identical, so the trust attaches to the right practice

Inconsistent information is the most common reason a practice that deserves to be named is passed over. Audit your name, address, phone, hours and services everywhere they appear and bring every version into exact agreement, then make those facts legible to a machine on your own site. This is the step that converts a strong reputation into something an assistant can confidently connect to you rather than hedge around.

Is any of this against GDC rules?

No, done as described here, and that is deliberate, because the compliant path and the effective path are the same path. The General Dental Council's Standards for the Dental Team require that any advertising, promotional material or other information you produce is accurate and not misleading ^[4]. That covers your website, your listings and how you handle reviews.

In practice that means a few clear lines. You must not gate reviews so that only happy patients can leave them, and you must not offer incentives, because filtered or paid-for praise is neither straight nor trusted. You must not fabricate reviews or use AI to write them, which is a breach on every front and a fast way to lose the trust you are trying to build.

Keep your claims specific and provable rather than superlative, because "the best implant dentist in town" is exactly the unverifiable boast the rule is built to catch, and it gives an assistant nothing it can responsibly repeat either.

State realistic price ranges and real qualifications, keep your "accepting new patients" status current, and you satisfy the regulator and the assistants at the same time. The trust bar is high for health topics because the decisions matter, and clearing it is what makes you both recommendable to patients and safe for an assistant to name.

How do you check where your practice stands on both?

You can run a useful self-audit today, free, in about fifteen minutes, and it tells you which signal is your weak point. Start with your reviews, and read them the way a nervous patient would. Are they recent? Are they specific about treatments, or vague? Is there a consistent pattern of sentiment, or a few old highlights and then silence?

Have you replied? Then do the same quick read for two or three nearby practices. If your reviews are fewer, older, generic or unanswered while a rival's are fresh, specific and tended, you have found a trust gap that is costing you at the decision moment.

Then test the amplifier. Open ChatGPT, Perplexity and Google's AI answers and ask what a real patient asks: "is [your practice] a good dentist in [your town]?", "best dentist for Invisalign in [your town]", "dentist for nervous patients near me." Note three things: whether you are named, what the assistant actually says about you, and which practices appear when you are absent.

Run each prompt a few times, because answers vary, and it is the recurring pattern that counts. Finally, check your own details for drift across your website, Google profile and any NHS listing. Strong reviews that the assistants are not repeating usually points straight at a consistency problem; absence from the answers plus a thin review base points at the foundation. Either way, you now know where the trust is leaking.

(For the related questions of how many reviews you need before AI results respond, and how reviews and AI citations drive discovery rather than trust, see the companion guides on review volume for AI results and on reviews versus AI citations.)

How QBiz helps you build trust patients and assistants both believe

Most providers in this space sell one half of the answer: they tidy your website and call it done. QBiz is built around the fact that durable trust has two halves that feed each other, so the work runs on two fronts at once.

On the first, we get your own house in order for the assistants: we make your practice details consistent and machine-readable everywhere they appear, shape your reviews and ratings so an assistant can connect and repeat them, and build answer-first pages for your high-value treatments so there is real, citable substance behind any mention.

On the second front, we do the distribution work, getting your practice referenced and corroborated across the wider web that the models actually read, so the trust you have earned is amplified rather than left to chance. Putting your own site in order is the groundwork; getting your practice out into the wider web is what turns that groundwork into mentions and enquiries. You need both, and we do both.

It begins with a QBiz AI Visibility audit. We ask the assistants the questions your prospective patients ask, show whether you are named and what is said about you, measure your review depth, recency and specificity against nearby rivals, and flag where your details disagree across the web. You get back a prioritised, GDC-safe plan covering the on-site fixes and the distribution work, highest-value cases first. It is the no-cost first step, and it tells you exactly which signal is costing you the patients you should be winning.

Frequently asked questions

Do patients trust Google reviews or AI recommendations more?

Patients place deeper, more durable trust in Google reviews, because they are written by real, named people whose experience a prospective patient can read and judge. An AI recommendation carries authority and often reaches the patient first, but it is unattributed and the patient rarely checks the evidence behind it. The practical reading is that reviews are the trust patients believe most, while an AI mention is an amplifier of that trust that your reviews largely earn.

If an AI mentions my practice, do I still need reviews?

Yes, more than ever. An AI mention is only as trustworthy as the evidence underneath it, and your reviews are most of that evidence. A patient who hears your name from an assistant very often goes to read your reviews before booking, so a thin or stale review profile undermines the very mention you were pleased to get. The mention opens the door; the reviews are what convince the patient to walk through it.

Do my Google reviews help AI recommend me?

Yes, and increasingly directly. The review industry now describes reviews as evidence worthy of citation within tools like ChatGPT, and a specific review that names a treatment and describes the experience gives an assistant something concrete to repeat. Generic five-star reviews help your human trust but give a model little to work with, so encourage patients to mention what they came in for. This link is covered more fully in the companion guide on reviews and AI citations.

Can I pay to be mentioned by ChatGPT?

No. There is no paid placement that buys you into an AI answer the way an advert buys a slot. A mention is earned by being a practice the assistant can verify from several independent directions and describe with confidence, and reviews are one of the strongest things that earn it. Anyone selling guaranteed AI mentions is selling something that does not exist.

How many reviews do I need before AI trusts my practice?

There is no fixed number that switches AI trust on. Volume helps your human credibility, but for whether an assistant will name you, the recency, specificity and consistency of your reviews and the agreement of your practice details across the web do more work than any single count. We cover this directly in the companion guide on how many reviews a dentist needs for AI results.

Are AI-written reviews a fast way to build trust?

No, they are the fastest way to destroy it. Fabricated or AI-generated reviews breach the review platforms' policies and the standards a dental practice is held to, they collapse the moment a real patient scrutinises them, and they put your registration at risk. The only trust worth building is the genuine kind, earned one real patient at a time.