Medical Providers Were the Largest Client Category in Korea’s Fake-Review Scheme

Busan police said businesses in the medical sector accounted for 34.5 percent of 707 clients in a nationwide review-manipulation operation. The authorities have not released the institution types, specialties, locations or campaign sizes needed to explain where that demand came from.

Businesses in the medical sector formed the largest client category in a nationwide fake-review operation uncovered by Busan police, accounting for 34.5 percent of the 707 businesses identified in the investigation. The network allegedly posted 28,886 fabricated positive reviews between October 2020 and January 2026 and received about 1.9 billion won from clients seeking greater visibility on an online platform. Police arrested the lead operator and referred 23 other people to prosecutors without detention.

Investigators said the operation used compromised accounts, receipts obtained from actual customers, nominal transactions created for review purposes and automated tools that generated searches or changed internet protocol addresses. Assignments moved from client businesses to marketing agencies and then to subcontractors responsible for creating and posting reviews. The operation was built to reproduce several signals associated with ordinary consumer activity rather than simply place favorable sentences beneath a business listing.

The medical category requires more precise language than the police summary and much of the Korean coverage provided. Reports commonly described the clients as byeongwon, an everyday Korean term often used loosely for places where patients receive treatment, but the released information did not distinguish hospital-level institutions from outpatient medical clinics, dental clinics or Korean medicine clinics. Korean law treats clinic-level and hospital-level medical institutions as separate categories, with the former serving primarily outpatients and the latter primarily inpatients. The evidence therefore supports describing the group as medical providers or medical institutions, not hospitals alone.

That classification gap is central to interpreting the case. Medical providers operate under markedly different revenue structures, and the commercial value of attracting one additional patient varies according to the treatment offered, whether the price is set through National Health Insurance and whether care is delivered once or over repeated visits. The police findings establish that medical businesses generated the largest number of identified clients; they do not establish that those clients spent more in total than other industries or that a particular specialty dominated the demand.

The Review Network and the Missing Client Data

Police described a layered commercial operation rather than a group of individuals independently selling favorable comments. Client businesses purchased marketing services, agencies transferred assignments downstream, developers supplied automated tools, and review-posting contractors used accounts and transaction records to make their activity resemble ordinary customer behavior. Some businesses allegedly provided customer receipts or photographs that could be repurposed as evidence of visits, while nominal payments of only a few thousand won were reportedly used in other cases to create a transaction linked to an account that had never used the advertised service.

Most client businesses reportedly paid about 200,000 to 300,000 won a month, although larger campaigns cost several million won. A medical provider made the largest reported monthly payment, exceeding 10 million won. That case shows that at least one institution regarded manipulated reviews and online exposure as a significant marketing expense, but it cannot establish how representative the contract was of the wider medical group.

Police have not disclosed the institution type, displayed specialty, location, contract value or review count of the medical clients. Without those fields, the 34.5 percent share cannot be compared with the underlying number of providers in each medical market. A specialty containing many institutions and many small purchasers would present a different pattern from a smaller specialty in which a high proportion of providers bought extensive campaigns.

Campaign intensity matters alongside client count. One medical field could account for numerous businesses that ordered a small number of reviews, while another could contain fewer clients responsible for a larger share of the fabricated content and the network’s revenue. The public record does not show whether medical institutions purchased reviews continuously, concentrated orders around particular procedures or used multiple agencies over the five-year period.

The clients’ knowledge also remains unresolved. Police reports indicate that some business owners supplied receipts or other materials and that four client businesses were among those referred to prosecutors, but the entire list of 707 businesses should not be treated as a list of convicted or even formally accused advertisers. Investigators would need contracts, messages, payment records and campaign reports to determine which providers directly requested fabricated patient experiences, which understood that reviews were being manufactured and which purchased broader marketing packages without documented knowledge of the subcontractors’ methods.

Outsourcing does not automatically remove responsibility. A provider purchasing guaranteed increases in reviews, searches or listing activity has reason to examine whether the promised results can be connected to actual patients. Legal responsibility, however, depends on evidence about individual conduct, and the actions of a review contractor cannot be attributed uniformly to every business found in an agency’s client records.

The investigation has described the production side in considerable detail: how accounts were obtained, how false visits were constructed, how automated activity was generated and how assignments moved through contractors. The public still lacks an equivalent account of medical demand. Until that information is released, the largest client category remains a broad industry label rather than an identifiable segment of South Korea’s health-care market.

How Treatment Value Shapes Marketing Spending

Official data on medical services outside the National Health Insurance benefit schedule provide a useful framework for examining why online patient acquisition can carry different economic values across medical fields. Medical institutions reported 2.1019 trillion won in fees for 1,251 monitored non-covered services provided in March 2025, an increase of 11.4 percent from the previous March. Clinic-level medical institutions accounted for 1.4155 trillion won, or 67.3 percent of the reported total.

Those figures measure reported fees for monitored services, not household net spending, medical-provider profit or the entire non-covered-care market. They also should not be presented as the size of South Korea’s cosmetic-medicine industry. The monitored items include dental treatment, pain management, imaging, medicines, private rooms and other services with widely different medical purposes.

Dental clinics reported 771.2 billion won, the largest total among institution types. Dental implant treatment accounted for 361 billion won, dental crowns for 246.9 billion won and orthodontic treatment for 84.7 billion won. Together, the three categories represented 82.6 percent of reported non-covered dental fees.

A patient considering an implant may compare several clinics before beginning diagnosis, surgery, prosthetic work and follow-up. The decision remains medical, but the patient has time to evaluate providers, request prices and select where to undergo treatment. Because one completed course can generate considerably more revenue than a routine insured consultation, a relatively small number of additional bookings may cover a meaningful portion of a monthly marketing contract.

Medical clinics reported another 500.6 billion won in monitored non-covered fees. Manual therapy accounted for 121.3 billion won and extracorporeal shock wave therapy for 75.3 billion won. A broader group of services used for musculoskeletal pain reduction and functional recovery generated 241.9 billion won, or 21.9 percent of the reported fees in the medical field.

Repeated care changes the commercial calculation. A new booking may produce one consultation, but it may also lead to several treatment sessions and follow-up visits. Providers evaluating a marketing campaign can therefore compare its cost with the expected revenue associated with a patient’s entire course of care rather than the first encounter alone.

Directly priced cosmetic and aesthetic procedures create another model. Providers can set package prices, manage appointment capacity and advertise individual services to prospective patients already comparing treatment options. Calls, online inquiries, consultation bookings and completed procedures can be traced through a relatively continuous marketing process, allowing agencies to estimate the cost of generating each new booking.

None of those examples identifies the medical clients in the police investigation. The authorities have not shown whether the list was concentrated in dentistry, pain treatment, aesthetic medicine or other areas where patients commonly compare providers and pay substantial fees outside the public benefit schedule. The non-covered-care statistics explain the economic conditions under which search exposure may be particularly valuable; they do not prove that those conditions produced the medical share in this case.

Treatment price is also an incomplete measure of marketing value. Materials, equipment, staff, rent, unsuccessful consultations, cancellations and follow-up obligations reduce the revenue retained by a provider. A patient may encounter several advertisements, receive a personal referral and consult multiple clinics before choosing treatment, making it difficult to assign the resulting revenue to one search result or review.

Digital campaigns nevertheless give providers a chain of measurable events. Search impressions can lead to listing views, listing views to calls or reservations, consultations to treatment and treatment to repeat visits. Reviews sit near the point where a potential patient has already found a provider but has not yet accepted the price, uncertainty or medical risk involved.

A modest improvement in that conversion rate may justify continued marketing when several new patients purchase high-value or repeated care. The same commercial incentive supports lawful advertising, accurate information and genuine patient feedback. It can also create demand for agencies offering a faster or cheaper way to manufacture the signals associated with patient approval.

Police could test that relationship by releasing anonymized data. The number of clients in each field should be compared with the total number of comparable institutions, while contract amounts and review volumes would show where the most intensive campaigns occurred. A raw percentage for the entire medical sector cannot answer either question.

How Paid Promotion Was Made to Look Like a Patient Review

A patient searching for a provider on Naver may encounter labeled advertising, clinic-operated websites, promotional blog content, a Naver Place listing, business hours, photographs, prices, reviews and reservation tools within the same search process. Each item has a different source and commercial relationship. The provider openly controls its advertisement and website, while a review appears to describe the experience of someone who has already visited.

That apparent independence gives reviews a function that conventional advertising cannot fully reproduce. A provider can describe its doctors, equipment and services, but readers recognize those statements as promotional claims. A favorable review appears to confirm them from outside the business, even when the writer can evaluate courtesy and waiting time more readily than diagnostic accuracy, clinical necessity or the durability of a result.

Naver’s published rules identify several methods as review manipulation, including fake or altered receipts, false reservations created for review purposes, reviews from people who did not visit or experience the business, the use of another person’s receipt and reviews produced through marketing agencies or reward services. The company says a confirmed instance of manipulated visit evidence or a fabricated review can result in all reviews for a business becoming hidden and the listing being placed under continued monitoring.

The operation investigated by Busan police allegedly exploited the evidentiary layer behind Naver’s Visitor Reviews. Compromised accounts were paired with customer receipts, photographs or nominal transactions so that the resulting posts could appear connected to genuine visits. The scheme’s commercial product was therefore broader than favorable wording: it supplied an account, apparent proof of interaction and supporting search activity.

Visit verification has a limited function even when it operates properly. A receipt, payment record or reservation can support a conclusion that an account interacted with a business, but it cannot verify every medical statement in the review. It does not establish that the writer underwent the treatment described, paid the usual price, received an appropriate recommendation or experienced the reported outcome.

Several forms of medical review must also remain distinct. An independent patient may post an unsolicited account. A patient may receive a benefit and clearly disclose it. A provider may select or edit testimonials for advertising. A real patient may receive discounted treatment while the financial relationship remains hidden. A person who never received care may invent the entire experience.

The Fair Trade Commission addressed the fourth category in July 2026 when it sanctioned three plastic surgery clinics over favorable reviews posted by promotional models who received discounted procedures without adequate disclosure of their financial relationship with the clinics. Those reviewers had received treatment, making the conduct different from the fabricated visits alleged in the Busan police case. Both practices, however, depend on readers perceiving commercially influenced material as independent patient feedback.

Marketing subcontracting can obscure that relationship. Medical providers routinely hire agencies to manage search advertising, websites, content and business listings, and those agencies may divide work among several contractors. Review manipulation can be inserted into an otherwise familiar campaign and described through broad terms such as optimization, exposure or reputation management.

The investigation therefore needs to reach beyond the people operating the accounts. Medical providers should be assessed according to what they ordered, what results were promised, what campaign reports they received and whether the volume of reviews could plausibly be connected to actual patients. Agencies should be examined for the subcontractors they hired and the instructions they transmitted.

Naver holds another category of evidence. The platform can compare account histories, transaction patterns, devices, internet addresses, review timing and links among businesses served by common contractors. Reports on the investigation said police supplied information about compromised accounts and that the platform took protective measures and hid fraudulent reviews, but the number detected before the police inquiry and the duration of their public exposure have not been disclosed.

Removing the posts identified by investigators addresses only the confirmed output of the network. A broader review would need to examine other accounts, businesses and campaigns connected through the same agencies, tools or transaction patterns. Otherwise, one set of reviews may disappear while related activity remains outside the case file.

Legitimate providers also have an interest in stronger detection. Clinics collecting genuine feedback compete with businesses able to purchase artificial review volume, and visible success can pressure other providers to spend more on opaque reputation-management services. Manipulation therefore affects competition within the medical market as well as the patients who read the fabricated accounts.

Patients cannot perform the same network analysis. They may compare reviews, verify professional credentials and seek another medical opinion, but they cannot determine whether several accounts were operated by one contractor or whether a receipt was transferred privately to someone who never visited. Organized review manipulation requires investigation at the agency and platform level rather than relying primarily on individual consumer vigilance.

The police findings do not show how many consultations, treatments or payments resulted from the reviews, nor do they establish the extent to which search exposure changed. They show that businesses paid for content designed to look like independent customer experience and that the medical sector formed the largest client category identified in the operation.

What Busan’s Clinic Map Can Show

Busan’s medical geography offers a way to test the client data if police eventually release it. A 2026 study examining medical institutions and personnel from 2013 to 2023 found that primary-care institutions expanded overall, while several departments showed substantial geographic concentration and uneven access across districts. The researchers cautioned that changes in institution counts did not translate automatically into an even or adequate supply of medical services.

The number of clinics displaying internal medicine rose from 336 to 447 during the period. Clinics displaying general surgery declined from 73 to 64, obstetrics and gynecology from 82 to 69, and pediatrics from 157 to 137. The study found that internal medicine remained relatively dispersed across residential areas, while general surgery and obstetrics and gynecology were more concentrated in particular districts.

Displayed specialty is an imperfect measure of clinical capacity. It does not always establish the physician’s board certification, the principal source of the clinic’s revenue, its operating hours or whether it provides inpatient or emergency services. A clinic displaying obstetrics and gynecology may not operate a delivery room, while a pediatric clinic may provide daytime outpatient appointments without accepting seriously ill children at night.

The figures therefore cannot prove that physicians moved from pediatrics, obstetrics or general surgery into aesthetic medicine. That claim would require individual employment histories, specialist qualifications and current service data. The available study shows that different parts of Busan’s medical supply moved in different directions and that some services became geographically concentrated; it does not identify a direct transfer of doctors between fields.

Scheduled treatment and continuously available care also have different geographic requirements. Patients can travel to a commercial district for a planned implant, eye procedure or aesthetic consultation after comparing providers online. A child requiring urgent assessment, a patient approaching delivery or someone needing emergency surgery depends more heavily on nearby staff, equipment, operating hours and an institution able to accept the case.

The distinction provides context for the review investigation without proving a causal connection. Medical services that draw scheduled patients from a wide area may compete more intensively for procedure-specific searches and online inquiries. Services whose primary function is local readiness operate under other constraints, including staffing, night coverage and the ability to maintain teams even when patient volume is unpredictable.

Police data could turn the geographic comparison into an empirical test. An anonymized list of client institution type, displayed specialty, district, contract range and review volume would show whether participation reflected the general size of Busan’s medical sector or was unusually concentrated in particular markets. Contract values would also reveal whether some specialties financed larger campaigns even when they produced fewer individual clients.

Busan’s clinic map supplies context rather than an answer. It identifies a city in which medical services are distributed unevenly and in which some forms of scheduled care compete within concentrated commercial districts. It cannot identify the providers that purchased fabricated reviews until the client records are classified and released.

The Data Police Have Yet to Release

The Busan investigation has established a finding that warrants further disclosure: medical providers were the largest client category in a five-year market for fabricated reviews. The network used compromised accounts, customer records, nominal transactions and automated activity to make paid marketing resemble ordinary consumer behavior. At least one medical client reportedly paid more than 10 million won a month, and some client businesses allegedly supplied materials used to construct false visits.

The evidence does not support describing all 707 clients as proven offenders, treating every medical provider as aware of the methods used or claiming that one specialty dominated the scheme. It also does not show that fake reviews caused broader changes in Busan’s medical supply. Those conclusions require information that has not been released.

Police can provide much of the missing picture without naming institutions before their legal status is resolved. An anonymized breakdown should include institution level, displayed specialty, district, campaign period, contract range, review count and investigative status. Providers suspected of directing or facilitating fabricated reviews should be separated from businesses whose knowledge remains unproven.

Naver can disclose how many reviews it detected before police intervention, how long the fabricated content remained visible, whether related accounts and listings were examined as a network and what action was taken against businesses that benefited from the manipulation. The company should also explain clearly what its visit-verification process establishes and what users should not infer from it.

The scheme did not create the economic differences within South Korea’s medical market. It revealed that some medical businesses were willing to purchase review volume and search activity through a supply chain capable of manufacturing the appearance of patient approval.

Police have disclosed the medical sector’s share of the client list. The institution types, specialties and campaign records behind that percentage remain the most important unanswered part of the case.