Duplicate listings split a practice's reviews. Patients and AI tools both look at the count.

Who this is for

This is for the physician or medical professional who owns an aesthetic practice, and/or the operations lead who keeps the practice's online listings current. The owner may need to sign off on a merge or a change of specialty.

Google allows one Business Profile per business. Its help pages say multiple profiles for the same business may mislead customers and are against its policies, and that a profile Google treats as a duplicate does not show on Search or Maps.¹

A duplicate Google has not caught stays live, with its own reviews. A review left there does not count toward the profile a new patient opens.

The same split happens on directories such as Healthgrades and RealSelf, which AI engines also read when they compare practices.²

How much do reviews decide which practice a patient picks?

Most patients read reviews before they book, in the physician and dental data points the professional associations publish.

The American Medical Association's Board of Trustees, in a 2022 report on inflammatory and untruthful online ratings, cited data that 37% of patients use online reviews as their first step in searching for a new physician and 60% have selected a physician based on positive reviews. It cited other research putting the share who use reviews in selecting a physician at 70%, and said 83% of patients trust online ratings and reviews of physicians.³

The American Dental Association says 84% of the public trust online reviews to help them make decisions. It also cites an ADA Health Policy Institute survey in which 88% of surveyed dentists reported receiving online patient reviews.⁴

The Medical Group Management Association relays a survey by Binary Fountain, a vendor of healthcare review platforms, in which 75% of healthcare consumers said ratings and reviews influence their choice of physician or provider. From a 2016 Software Advice survey it relays that 82% of patients used reviews to evaluate physicians and 77% searched for a provider online before making an appointment.⁵

One peer-reviewed experiment measured the effect on choice directly. In the Journal of Medical Internet Research in 2018, Yaraghi and colleagues asked 1,000 Amazon Mechanical Turk users, in October 2016, to pick between pairs of primary care physicians with different ratings.⁶

Moving a commercial website's rating of nonclinical care from 2 to 4 stars raised the log odds of a physician being chosen (a standard scale for comparing choices) by 1.32. Moving a government clinical rating the same distance raised it by 1.31. The authors conclude that people weigh the commercial ratings as heavily as the government's clinical ones.⁶

The AMA report gives no sample sizes and does not name its original studies, and the MGMA data points come from vendor surveys. The same AMA passage that reports 83% trust also cites research showing online ratings of physicians do not predict objective measures of quality of care or clinical performance.^3,5

How does a practice end up with more than one listing?

Google's help pages describe the usual causes. Google tells a practice that relocates to keep its existing profile and change the address on it. It tells a practice offering several services to list them all on one profile.¹

Google also describes finding a business already on Maps with no verified owner. The owner is told to claim and verify it, and only after verification does the practice own the profile and manage its information.¹

Individual practitioners at a shared location are eligible for their own profiles under Google's rules. Each of those profiles carries its own reviews.¹

In our audit of 16 aesthetic practices, 9 had directory listings that could be assessed. Of those 9, 7 had stale second listings, wrong specialties or old practice names. Each one spread a practice's reputation across more than one version of it.⁷

The same audit looked at the question-and-answer section on each Google Business Profile. All 11 practices with a profile we could assess had no questions answered by the practice, and the gap held on every one of the 11.⁷

What happens to the reviews on a second listing?

A practice whose reviews sit across two listings shows only part of them on whichever listing a patient opens.

Patients look at the count. BrightLocal, a local search software vendor, surveyed 1,141 US consumers in 2024 and found 59% expected a business to have between 20 and 99 reviews before they would trust its average star rating. The survey covers local businesses in general.⁸

Recency matters too. Tebra, a practice software vendor, cites its Patient Perspectives research, in which 43% of people said recency mattered most and 38% pointed to the average star rating.⁹ In BrightLocal's 2024 survey, 27% said reviews from the past two weeks affect their decisions.⁸

If the recent reviews land on one listing and a patient opens the other, the listing the patient reads shows older reviews than the practice is getting.

Health care reviews split close to evenly. A 2025 study in JAMA Network Open of 1,099,901 Yelp reviews of 138,605 US health care facilities, posted from 2017 to 2023, found 46.3% negative (1 or 2 stars) and 50.1% positive. The themes most tied to negative ratings were payment issues and poor treatment.¹⁰ On a profile no one at the practice reads, those negative reviews go unanswered.

Google's remedy for two profiles of the same business is a merge, requested by reporting a suggestion on Google Maps and subject to Google's review. When two profiles merge, their reviews are combined, though replies to those reviews may be lost.¹

Removing a profile works differently. Google says that when a duplicate profile is removed, all of its content and linked managers are removed with it, which would take its reviews too.¹ Google's two remedies treat reviews in opposite ways: a merge keeps and combines them, and a removal deletes them along with everything else on the profile.

Do AI engines name a practice on its reviews?

Google is one of several places a patient, or an AI engine, reads about a practice. In our July 2026 measurement of AI answers about aesthetic practices, the sources the engines leaned on when comparing practices were review platforms and directories, with Yelp, RealSelf and Healthgrades the most frequent.²

So a stale second profile or a wrong specialty on one of those sites reaches the engines as well as the patients who land there.

The evidence on how engines choose comes from vendors and from our own measurement. No published controlled study tests it.

Uberall, which sells listings and review management, says 47% of patients now use AI tools such as ChatGPT, Gemini or Perplexity to find a healthcare provider.¹¹ RepuGen, which sells healthcare reputation management software, puts the share at 39.7% in its 2025 AI and Healthcare Search Study.¹² Neither page gives a sample size.

BrightLocal's 2025 survey of 1,026 US adults found 6% used ChatGPT or another AI tool to find reviews, down from 9% the year before. That measures finding reviews of local businesses, a narrower question than finding a provider, so the three data points measure different things.¹³

RepuGen also reports a 2026 study in which 56.8% of respondents chose a provider rated 4.5 stars with 1,050 reviews over one rated 4.9 stars with 25. That volume effect was larger than we expected. RepuGen says the study does not prove AI uses the same decision process.¹²

Uberall says AI recommends the practices whose listings, reviews and photos agree with each other.¹¹ Both vendors sell the fix for what they describe.

In our own measurement, the engines wrote their reasons for naming a practice into their answers, and the reason they gave was its reviews. That comes from what the engines said, and no controlled experiment has tested it.²

Our reading, which the sources do not state: if review count and listing consistency both feed what an engine names, a duplicate listing hurts on both counts. It splits the reviews, and the second profile is itself an inconsistency.

How can a practice rebuild a split review count within the fake-review rules?

A split count is repaired by merging listings and by more reviews from real patients. Buying reviews carries legal risk in the US and the UK.

The Federal Trade Commission's rule on consumer reviews and testimonials, 16 CFR Part 465, took effect on October 21, 2024. It prohibits creating, selling or buying fake or false reviews, and buying reviews in exchange for, or conditioned on, a particular sentiment. The FTC says the rule lets it seek civil penalties against violators.¹⁴

The UK Competition and Markets Authority's fake reviews guidance of 4 April 2025, issued under the Digital Markets, Competition and Consumers Act 2024, bans fake reviews and concealed incentivised reviews. It defines a fake review as one that purports to be, but is not, based on a person's genuine experience. Any trader that publishes consumer reviews must take reasonable and proportionate steps to prevent and remove them.¹⁵

UK government guidance for businesses says misleading practices around reviews can lead to a fine of up to 10% of global turnover.¹⁶ For a US practice, the FTC rule is the federal one that applies.

The practice's own listings, on Google and on the directories, are the place to start, since a split count is made of reviews the practice already earned.

One claimed, correct profile per practice is the goal, and a duplicate that holds reviews is a candidate for a merge before a removal.

The count grows from real patients over time, and never from paid reviews.

About this review

We read patient-use data points from three professional associations (AMA, ADA, MGMA), two peer-reviewed studies (Journal of Medical Internet Research 2018, JAMA Network Open 2025), Google's Business Profile help pages, and vendor material from Uberall, RepuGen, Tebra and BrightLocal, each labeled by who produced it. The fake-review rules come from the FTC, the UK CMA and UK government guidance. Sources were read in October 2026.

The first-party data points are anonymous totals from our 2026 audit of 16 aesthetic practices, and findings from our July 2026 measurement of AI answers about aesthetic practices. Practice names are withheld.

The association reports give no sample sizes and do not name their original studies. The JMIR experiment used online survey takers choosing between physicians described to them. Uberall and RepuGen give no sample sizes for their AI data points, and BrightLocal surveys local businesses in general. The link between reviews and being named by an AI engine rests on what the engines say in their answers. Our audit is a small sample and was not drawn at random.

Fractional Aesthetics is a growth strategy practice for aesthetic medical businesses.