Why Do Patient Testimonials and Before and After Pages Matter for Dental SEO?
This is the strongest evidence a dental practice has and the easiest to get wrong. The rules are specific rather than prohibitive. The practices avoiding this material entirely are throwing away their best asset out of vague worry.
Why This Is The Strongest Evidence You Have
A patient cannot assess clinical skill. So they assess the next best thing available to them, which is other patients' experiences. That makes this material more persuasive than anything a practice can say about itself.
Why self description fails. Everybody says the same thing.
Caring, experienced, modern, patient focused. A reader has seen those words on every practice website they have opened and they now carry no information whatsoever.
What a patient account carries instead. Something checkable.
A named person describing a specific experience is a claim somebody could in principle verify. That is a different category of statement from an adjective the practice chose about itself.
The evidence that patients seek it out. By practice name.
Our own keyword research in August 2026 found around 14 review related terms carrying roughly 3,910 searches a month between them. Twelve of those fourteen were searches for a named practice plus the word reviews, carrying roughly 2,650 a month between them.
What that behaviour tells you. This is a verification search.
Somebody typing a practice name and the word reviews has already found you. They are checking before they ring, which is the same behaviour recorded on the registration searches. It means this material is consulted at the decision point rather than browsed idly.
The generic demand too. Cheap and unserved.
The same research found dentist reviews at around 1,000 searches a month at a competitive difficulty score of about 16 and review for dentist at roughly 320 near only 9.
What follows. The rules, which are narrower than most practices assume.
A great deal of the caution around this material comes from nobody having established what is actually restricted. Block two sets that out. The answer is smaller than the worry.
The Rules Are Specific, Not Prohibitive
Before and after imagery is restricted rather than prohibited. Most practices believe otherwise and publish nothing, which throws away the most persuasive material a cosmetic practice has on the strength of a misunderstanding.
What we will not do on this page. State the requirements for you.
We describe how a practice is found and what it publishes. We do not set out what any regulator or advertising authority currently requires, which is not ours to summarise and which changes.
The four conditions. Wherever you publish a position.
Name the regulator or advertising authority it comes from. Give the date. Verify it against that body's own current wording immediately before the page goes live, then at every review. And carry a line stating this is general information rather than regulatory advice.
The four practical requirements. Applied to every image.
Genuine, meaning your own patient and your own work. Representative of a typical result rather than the best one ever achieved. Unretouched, which includes lighting, angle and whether the person is smiling the same way in both images. And carrying context, meaning the treatment stated and the timeframe given.
Why avoiding it entirely is the wrong answer. It costs conversions.
A cosmetic patient is buying an outcome they cannot otherwise assess. A practice showing properly handled results outperforms one showing none. The requirements above are achievable with a phone and a consistent process.
The practical fix. A fixed setup.
Same position, same lighting, same camera, every time. That single decision makes every future case usable and removes most of the difficulty at a stroke.
Consent Properly Obtained
Written, specific and capable of being withdrawn. That is the shape of it. The failures are almost always in the specifics rather than in whether consent was obtained at all.
What it has to cover. Four things.
What exactly is being used, whether images, words or both. Where it may appear, which should name the places rather than saying generally. How long it may be used for. And that it can be withdrawn.
Why written matters. A conversation is not a record.
Verbal agreement at the end of an appointment is not specific, is not dated and cannot be produced later. It also puts a patient on the spot at a moment when they are unlikely to say no.
Where practices go wrong. Consent for one thing, used for another.
An image agreed for the practice website appearing on social platforms. Or in advertising. Or years after the treatment. Each of those is a use the patient did not agree to. The fact that they agreed to something is not a defence.
What withdrawal means practically. A process, not a promise.
Somebody has to be able to act on it. That means knowing where every image is published and being able to remove it, which is an argument for keeping a simple record of what is used where.
The timing point. Ask later rather than immediately.
A patient asked weeks after treatment, when they are pleased and under no pressure, gives more considered consent than one asked in the chair.
Verify the current position before publishing. Per block two.
When A Patient Makes A Clinical Claim
A testimonial saying a treatment was completely painless is a clinical claim. So is one saying it cured something. The patient may believe it sincerely. By publishing it on your own website you have made the claim. You are held to it as though you had written it yourself.
Why this catches people out. It feels like somebody else's words.
Practices assume that quoting a patient transfers the statement to them. It does not. A practice is responsible for what appears on its own site regardless of who composed the sentence.
What to look for. Three patterns.
Absolute claims about experience, such as painless or completely comfortable. Claims about effect, such as cured or fixed. And comparative claims, such as better than the last practice, which brings its own problem entirely.
What to do about it. Two legitimate options.
Edit it with the patient's permission so their account of their own experience survives without the general assertion. Or do not publish that one. What is not available is publishing it unchanged on the basis that somebody else said it.
How to edit without distorting. Cut rather than rewrite.
Removing a sentence is defensible. Rewriting somebody's words into something they did not say is a different problem, so cut and show the patient what remains before it goes live.
The comparative one deserves particular care. It is about another practice.
A patient saying they had a worse experience elsewhere has made a claim about somebody else's clinical standards, which you adopt by publishing it. That one is rarely worth the risk.
The test. Would you have written this sentence yourself?
What Makes A Testimonial Credible
A first name and a star rating proves nothing. Readers discount the generic automatically, which means a page of them is worse than three good ones because it demonstrates that the practice collects without curating.
What credible looks like. Four elements.
A named person rather than an initial. The treatment they had. Roughly when. And something specific enough that it could not have been written about any other practice.
Why the specific detail does the work. It cannot be fabricated generically.
Somebody mentioning that they were nervous about a particular thing and how it was handled is describing something that happened. Somebody writing that the staff were lovely is producing a sentence available to anybody.
What readers actually discount. More than practices realise.
Five word entries, uniform enthusiasm, no dates and no treatments named. A reader scanning that concludes it was collected rather than earned, whether or not it was.
The counterintuitive one. Mixed content is more credible.
An account mentioning something that was difficult, alongside how it was handled, reads as real. A page where every patient was uniformly delighted reads as filtered, because it is.
How to collect better ones. Ask a question, not for a review.
Asking somebody what they had been worried about beforehand produces specific material. Asking for a review produces a sentence about lovely staff.
How many you need. Fewer than you think.
Three specific accounts on a treatment page outperform forty generic ones on a dedicated page. Volume is not the variable and treating it as one is why most practices have a long list nobody reads.
Where This Material Should Live
Attached to the treatment pages rather than corralled into a testimonials section nobody visits. That single structural decision does more than doubling the amount of material ever would.
Why the dedicated page underperforms. Nobody goes there.
A testimonials page is visited by people already convinced and by nobody else. The reader who needed persuading was on the treatment page and left without seeing anything.
The evidence for treatment level placement. The searches are treatment specific.
Our own keyword research in August 2026 found the before and after demand attached to particular treatments rather than to practices generally, with dentist teeth whitening before and after at around 140 searches a month at a competitive difficulty score of only 13 and before and after dentist cleaning at roughly 210 near 6.
What that means. The demand is organised the way the site should be.
People are not searching for a practice's results in general. They are searching for results for the treatment they are considering, which is exactly where the material should sit.
How to arrange it. Two or three per treatment page.
Placed alongside the description of that treatment, with the expectation setting immediately adjacent rather than three screens below. Per the treatment guides, an image without the expectations beside it does the wrong work.
What about the dedicated page. Keep it. Do not rely on it.
It has some value for people arriving already interested. It should not be where the material lives.
The measure. Whether it appears where somebody is deciding.
Reviews As Distinct From Testimonials
The mechanics of collecting and responding to reviews belong elsewhere. What matters here is a distinction practices consistently blur, which is that the responsibility position is not the same.
The difference in one sentence. Whose site it is on.
A testimonial sits on your website and you published it. A review sits on somebody else's platform, you did not publish it and you cannot edit it. That changes the position materially.
Why the blurring is costly. Practices apply the wrong caution.
Having correctly understood that a testimonial making a clinical claim is a problem, a practice concludes that reviews are dangerous too and stops encouraging them altogether. That is a substantial loss for no reason.
What the search behaviour shows. They are checked by name.
Per block one, twelve of fourteen review terms in our own keyword research in August 2026 were searches for a named practice plus reviews. That demand exists whether a practice participates or not.
Where care is still needed. What you do with them.
Quoting a review onto your own website turns it into a testimonial, at which point everything in block four applies. Reproducing a review is a publishing decision.
Responding to reviews. One point worth stating.
A reply confirming or discussing a patient's treatment discloses that they are a patient. That is a matter to handle carefully and it is why replies should stay general.
Where the mechanics sit. The local search material rather than here.
Video And Its Advantages
A patient speaking is more credible than a patient quoted, for a simple reason. Text can be written by anybody. A person on camera saying something in their own words is considerably harder to manufacture and readers know it.
Why it converts. The hesitations are the point.
Somebody pausing, correcting themselves or laughing is demonstrably real. A polished script read to camera loses most of the advantage, which means the production value should be deliberately modest.
What to ask. The same question as block five.
What were you worried about before you came, then what actually happened. That produces material nothing else does.
The practical requirements. Lower than practices assume.
A phone, a quiet room, decent light and two minutes. Anything more elaborate begins to look produced, which works against the whole purpose.
The consent position is the same. With one addition.
Everything in block three applies. Video also needs an explicit understanding about where it may be posted, since it travels further than a photograph and is harder to retrieve.
The clinical claim rule applies equally. And it is harder to fix.
A patient saying on camera that it was painless has made the same claim as one writing it. Editing video without distorting somebody's meaning is considerably more difficult. Better to explain the constraint before recording than to salvage it afterwards.
Where it belongs. On the treatment page, per block six.
What Never Goes On The Site
Five things. Each appears on dental websites regularly and each is straightforward to avoid once somebody has decided the rule in advance rather than case by case.
Retouched images. Including the subtle version.
Not only editing teeth. Adjusting lighting, exposure or colour between two images produces a difference the treatment did not, which makes the pair misleading however innocent the intention.
Exceptional results presented as typical. The commonest failure.
Publishing only your most striking cases creates an expectation somebody then has to manage in the chair. It is also the thing block two most explicitly rules out.
Images of anybody who has not consented. Including old ones.
Material published years ago, under arrangements nobody recorded, is still on the site and still your responsibility. An audit of what is currently published is worth doing before anything new goes up.
Any claim you would not make yourself. Per block four.
If you would not write the sentence in your own copy, publishing it in quotation marks does not change what it is.
Stock photography presented as patients. The one that ends trust instantly.
Models used as though they were people you treated. A reader who recognises a stock image discounts everything else on the site. They are recognised more often than practices expect.
What to do first. Look at what is already published.
Most practices have something on this list live right now. That is set out further in GDC registration and dental SEO. Our approach is on the dental SEO page and across our SEO guides for dentists.
Your best evidence,
used properly.
Results published within the requirements rather than avoided entirely, patient accounts specific enough to be believed, everything placed on the treatment pages where people are deciding, plus an audit of what is already live.
What is included every month:
One monthly rate covering everything listed above. No setup fee. Nothing billed separately.
Every guide.
One profession.
Marketing compliance, regulated healthcare search, the access question, emergency and nervous patients, implants, aligners, bonding, whitening, veneers, smile makeovers, children's dentistry, credentials and the corporate groups.
