Testimonials, guarantees and "pain-free": an advertising compliance screen of 618 Australian clinic websites

Written by Sara Smith, Head of SEO, and reviewed by Lucas Durante, Founder & CEO, Odin Digital.

Headline finding

More than half of Australian clinic websites carry at least one phrase that AHPRA's advertising guidelines identify as a likely breach: a testimonial, a guarantee, a "pain-free" promise, or a "best" or "award-winning" claim. Dentists are the worst at three in four. Psychiatrists and psychologists are the best at under a third. Fewer than a quarter of clinics show a practitioner registration number anywhere on the pages we read, and the clinics with the cleanest copy are the least likely to show one.

Key findings

• 57.3 percent of clinic sites (354 of 618) contain at least one phrase matching a prohibited-advertising rule under section 133 of the Health Practitioner Regulation National Law as AHPRA's guidelines describe it. 32.0 percent (198) match two or more distinct rules. Excluding the "pain-free" rule, the most arguable one, the figure is still 51.0 percent. Excluding superlatives as well, it is 40.6 percent.
• 31.1 percent show testimonial or review content on their own site, the single clearest prohibition in the guidelines. 69 sites do it with visible text ("what our patients say", a Google rating), 69 through an embedded review widget or star-rating markup, and 54 do both.
• 26.7 percent make a superlative or comparative claim: "world-class" (35 sites), "best dentist" (32), "most experienced" (16), "award-winning" (26 across two spellings).
• 25.2 percent promise "pain-free", "painless", "no pain" or "no side effects". Dental sites drive this: 46.9 percent of dentists and 40.0 percent of physiotherapists.
• 25.6 percent offer an inducement: "complimentary" (60 sites), "free consultation" (37), a percentage off (27), "no gap" (22). In only 17 of the 158 cases did the same passage carry any terms, eligibility or expiry language.
• 13.4 percent make a guarantee or outcome promise, including "guaranteed", "money back", "instant results" and "no downtime".
• Only 22.8 percent display an AHPRA registration number or practitioner registration anywhere on the pages we read. Plastic surgeons are the exception at 77.6 percent. Cardiologists and orthopaedic surgeons are at zero. 6.1 percent of all clinics carry a "results may vary" disclaimer.
• In cosmetic and skin practices (n = 144), 39.6 percent reference before-and-after imagery, 34.0 percent use trivialising or finance framing the 2023 cosmetic rules address ("Afterpay" on 37 sites, "interest-free" on 42, "payment plan" on 27, "medical-grade" on 42), 6.9 percent name a prescription-only medicine brand, and 5.6 percent mention a cooling-off period.
• The breach is on the homepage for 41.9 percent of all clinics. Only 15.4 percent have flagged copy solely on internal pages. This is not buried in an old blog post; it is the first thing a patient and a regulator see.
• The rate barely varies by city: Hobart 52.2 percent, Canberra 65.0 percent, and the rest between 54 and 61 percent. It is a national pattern, not a local one, and it is the same whether the clinic ranks in search or not.
• Platform matters. Wix clinic sites are flagged 38.2 percent of the time, Squarespace 45.1 percent, WordPress 61.0 percent. The difference is almost entirely review widgets: 23.9 percent of WordPress clinic sites embed one, against 2.9 percent on Wix.

Why this matters in Australia

Australia regulates health advertising more tightly than almost any comparable country. Section 133 of the National Law prohibits advertising a regulated health service in a way that is false or misleading, offers an inducement without terms, uses testimonials, creates unreasonable expectations of benefit, or encourages indiscriminate use. AHPRA's guidelines, updated in 2020 and again for cosmetic procedures in 2023, spell out what that means: no patient testimonials on your own site, no "pain-free", no "best", no guarantees, no prescription-only brand names, no financing offered as an inducement for cosmetic surgery.

Every one of these rules applies to the practitioner's website, and to the marketing agency that builds it. The penalties are per advertisement: up to $60,000 for a body corporate and $30,000 for an individual. And the rules are, on the evidence of this study, being broken by most practices most of the time, not through malice but through templates, plugins and copy conventions imported from industries where "5-star rated" is normal and "Melbourne's best" is just how a headline is written.

This paper is for practice owners, practice managers and the agencies that serve them. It is not legal advice. It is a measurement of how far the average Australian clinic website sits from the published rules, so that the fix can be sized.

Methodology

Sample. The 736 healthcare domains from the State of Australian Small Business Websites sample (Odin Digital Research, September 2026), which was built from Bing search results and OpenStreetMap business points across eight capital cities. Thirteen regulated professions are in scope: dentists, physiotherapists, psychiatrists and psychologists, chiropractors, dermatologists, plastic and cosmetic surgeons, cosmetic clinics, cardiologists, orthopaedic surgeons, audiologists, fertility clinics and vein clinics. 618 sites could be read; 118 blocked automated access, timed out, or turned out not to be a clinic. 364 of the 618 came from the map and 209 from search results; the remainder from both.

Pages read. The homepage plus up to six internal pages whose URL or link text suggested about, team, services, treatments, reviews, testimonials, results, gallery, pricing or offers. Seven pages were read for 334 sites; the median is seven. All were fetched on 3 September 2026 with no JavaScript execution, so content injected by scripts, including some review widgets, is under-counted.

Rules. Thirteen pattern rules written from AHPRA's Guidelines for advertising a regulated health service and the 2023 cosmetic surgery advertising guidelines. Each rule is a fixed list of phrases. Seven are treated as likely breaches: testimonials and review content in text; embedded review widgets or self-declared star-rating markup; guarantees and outcome promises; superlative and comparative claims; pain and safety absolutes; urgency and scarcity. Six are treated as risks that need context: inducements, before-and-after references, prescription-medicine brand names, cosmetic trivialising or finance framing, unsubstantiated clinical claims, and price framing. Guarantee matches preceded by a negation ("no procedure can guarantee") and price-match guarantees are excluded. Clinical-scope terms (ADHD, autism, colic) are reported only for professions where they indicate a scope-of-practice concern. The full phrase list is published with the dataset.

Precision. Pattern matching on medical copy produces false positives. In a hand review of 54 randomly selected breach-rule matches across two runs, 45 were judged correctly flagged and 9 were not, a precision of 83 percent. The false positives cluster in the guarantee rule ("guaranteed a good laugh", a Medicare "money back" answer) and the scarcity rule ("only a few implants"). Testimonial, superlative and pain-free matches were almost always correct. Read the headline rate as "carries at least one phrase the rules flag", not as a legal finding, and read the 31.1 percent testimonial rate, the 32.0 percent two-or-more-rules rate and the 40.6 percent rate excluding the two most arguable rules as the conservative figures.

What we did not do. We did not assess images, videos, social media, Google Business Profiles or paid ads. We did not verify whether a practitioner described as a specialist holds specialist registration. We did not contact any practice. No practice is named in this paper.

Finding 1: testimonials are everywhere

The guidelines are explicit that a practitioner may not use testimonials about clinical aspects of a service in their own advertising, and AHPRA treats an embedded Google review widget on the practitioner's site as the practitioner using them. A third of clinics do it anyway. Dental (45.6 percent) and cosmetic clinics (52.0 percent) are the heaviest users, and both are the sectors where review widgets are sold as standard by website builders.

The widgets are identifiable. Across the 123 sites with embedded review evidence we found self-declared aggregateRating markup on 130 pages, Trustindex on 110, Podium on 54, Birdeye on 50, a generic Google reviews embed on 50, and Elfsight on 43. These are products sold to clinics by their web and marketing suppliers, often as part of a "reputation" package. The most common visible-text triggers are "Google reviews" (51 sites), "testimonials" (42), "Google rating" (16), "patient stories" (13) and "what our patients say" (24 across two phrasings).

Finding 2: the words the rules name

The inducement figure needs its context. Offering a free consultation is legal. Offering it without stating the terms in the same advertisement is not. In 141 of the 158 sites with an inducement phrase, the passage around it carried no terms, eligibility or expiry language at all. "Complimentary consultation" as a menu item, with nothing attached, is the standard form.

The "clinically proven" figure is reported separately from the scope-of-practice terms because they mean different things. A psychologist who mentions ADHD is describing their work; 62.4 percent of psychiatry and psychology sites do. A chiropractor who mentions ADHD, autism or colic is describing a claim AHPRA has named as an enforcement priority; 10.5 percent of chiropractic sites and 12.9 percent of physiotherapy sites do.

Finding 3: by profession

The pattern is consumer-facing versus referral-based. Professions patients choose directly from a search result (dentistry, cosmetic, physio, chiro, audiology) market like retailers and break the rules like retailers. Professions reached mostly by referral (cardiology, psychiatry, dermatology, orthopaedics) have less marketing copy and fewer breaches.

Dentistry deserves its own paragraph. Three in four dental sites carry a flagged phrase, half carry two or more distinct rules, and the profession is the heaviest user of every rule except inducements. "Pain-free dentistry" is so standard a phrase that 46.9 percent of dental sites use it, and 39.4 percent offer a "no gap" check-up or a free consultation with no terms attached. Excluding the pain-free rule entirely, 65.6 percent of dental sites are still flagged.

Plastic surgery is the counter-example worth studying. It has the highest rate of registration display in the study, 77.6 percent, and the highest rate of "results may vary" disclaimers at 36.7 percent, which reflects a profession whose college has drilled the cosmetic rules into its members. It also has the highest before-and-after rate at 61.2 percent. The compliance culture is real, and it stops at the gallery.

Finding 4: the cosmetic rules, two years on

Among 144 sites in cosmetic surgery, plastic surgery, dermatology, cosmetic clinics and vein clinics:

Two years after the cosmetic rules were tightened, one in three cosmetic sites still uses the language the rules single out, and one in twenty mentions the cooling-off period the rules require for surgery. The finance framing is the most common: "Afterpay", "Zip Pay", "interest-free" and "payment plan" appear on 116 sites across the whole sample, and the 2023 guidelines are explicit that promoting finance for cosmetic surgery is an inducement. "Medical-grade", which the guidelines treat as a meaningless comparative, appears on 42.

The prescription-medicine finding is small but unambiguous: 18 sites use the word "Botox" in their own copy, and a handful name Dysport, Juvederm, Restylane or Ozempic. The Therapeutic Goods Administration prohibits advertising Schedule 4 medicines to the public by name, and AHPRA's guidelines repeat it. Some of the 18 are FAQ answers rather than promotions, which is why this rule is scored as a risk and each match is published for hand review.

Finding 5: where the breach sits

The most common co-occurrences are a superlative with a pain-free claim (60 sites), a review widget with a pain-free claim (60), and testimonial text with a pain-free claim (59). That is the standard dental homepage: "award-winning, pain-free dentistry, rated 4.9 on Google". It is three breaches in one hero section.

The platform split is almost entirely the review-widget ecosystem. WordPress clinics embed Trustindex, Podium, Birdeye or Elfsight at eight times the Wix rate, and those widgets are sold and installed by the marketing suppliers who also write "Melbourne's best". The template is the breach.

Sydney clinics are the most likely to show a registration number, 40.2 percent, and no other city passes 25 percent. Whether a clinic came from search results or the map makes no difference: 52.6 percent versus 58.5 percent flagged.

Finding 6: who you are dealing with

22.8 percent of clinic sites display a registration number or an explicit AHPRA registration reference anywhere on the pages we read. The guidelines do not strictly require it on a website, but it is the single most credible thing a practice can put on a page, it is free, and it is the direct substitute for the testimonials and superlatives the rules prohibit.

The uncomfortable detail is that the clinics with the cleanest copy are the least likely to show it. Among the 143 sites with no flagged phrase at all, 7.7 percent show a registration. Among the 354 flagged sites, 25.4 percent do. Clean copy is, in many cases, thin copy: a brochure site with a phone number and no claims because there is no content. The practice that has invested in its website has invested in the wrong sentences.

What to do about it

For a practice, in order:

1. Remove the review widget and the patient quotes from your own site. Reviews can live on Google. You cannot reproduce them, and you should not solicit them. This one change clears the most common breach in the study, and on WordPress it usually means deactivating one plugin.
2. Search your site for "pain-free", "painless", "guarantee", "best", "leading", "award-winning", "world-class", "most experienced". Replace each with a fact: years practising, registration category, fellowships, the procedure described plainly. "We use topical anaesthetic and discuss pain management with every patient" is compliant. "Pain-free dentistry" is not.
3. Put the registration on the page. Practitioner name, registration category, AHPRA number, on the team page and in the footer. Three-quarters of practices are leaving it off while a third are running review widgets they are not allowed to run.
4. If you offer anything free, discounted or "no gap", put the full terms beside it. Who is eligible, what is included, when it ends. In the same block of copy, not behind a link.
5. Cosmetic practices: audit against the 2023 guidelines specifically. Brand names of prescription medicines off the site, finance promotion off the site, cooling-off period stated, before-and-after images only with consent, matched conditions and a disclaimer, and no "mummy makeover".
6. Ask your agency which of these rules they know. Most of the copy we flagged was written by marketers, not clinicians, and most of the widgets were installed by them.

A free screening tool that runs these rules on any block of copy is at /tools/ahpra-ad-checker. It is not legal advice, and neither is this paper.

Frequently asked questions

Is a Google review widget really a breach? AHPRA's guidance is that a practitioner must not use testimonials in their advertising, and that embedding or reproducing reviews on the practitioner's own site is using them, even though the reviews themselves may legitimately exist on Google. A review widget is the practitioner reproducing them.

What about reviews that only mention the receptionist or the parking? The prohibition covers testimonials about clinical aspects of the service. AHPRA's position is that a practitioner cannot reliably filter reviews to non-clinical content, and the safe course is not to reproduce them.

Does "pain-free" really count? The guidelines list claims that create unreasonable expectations of beneficial treatment, and AHPRA has specifically named "painless" and "pain-free" as examples. Our rule is scored as a likely breach, and we also publish the figure with the rule excluded.

Is this the same for allied health? Yes. The National Law covers every AHPRA-registered profession, including physiotherapists, chiropractors, psychologists and dental practitioners, not only doctors.

Will you tell a practice whether it was in the sample? Yes, on request, with the flagged phrases and their context so they can be checked.

Dataset and citation

• Dataset: ahpra-screen-2026-09.csv, one row per site with each rule's hit count, the pages read, and registration and disclaimer flags. The matched phrases and their context are in the accompanying JSONL so any flag can be re-checked by hand. Domains are included for replication; no practice is discussed individually.
• Rubric: the thirteen rules and every phrase in them, with the guideline section each maps to.
• Suggested citation: Smith, S. (2026). Testimonials, guarantees and "pain-free": an advertising compliance screen of 618 Australian clinic websites. Odin Digital Research. odindigital.com.au/research/ahpra-advertising-compliance-australian-clinics
• Media enquiries: [email protected]

Back to Odin Digital Research