Reputation Has a Scar Effect
Why brand damage from a credibility violation doesn't heal the way most recovery plans assume
Reputational damage does not just lead to a lower score; it can modify the system that produces future reputation and, as a result, the requirements for recovery change.
In August 2026, a study followed nearly 290,000 business entries drawn from Yelp, foot-traffic data and credit card transactions. The research team, headed by Yi Cao from George Mason University, investigated what actually takes place to businesses when Yelp issues a consumer warning about suspected fake reviews. The warning remains in place for about 90 days before it disappears. Generally speaking, most businesses and most reputation management approaches regard the deletion of the warning as the conclusion of the issue.
The data showed a disagreement. Immediately after the alert appeared, foot traffic fell by about 8 percent — that was a significant drop in itself. The more important result, however, was what took place afterward: eighteen months after the alert, businesses were still operating at levels below their pre-alert demand. There was no sign that recovery would occur for the majority of the businesses identified in the dataset.
The reason this finding is more than just a warning about fake reviews is the mechanism the researchers have identified. The alert didn't merely affect the business's rating; it harmed the information environment associated with the business. The number of reviews decreased; fewer people decided to leave reviews at all, and the ones that did were more negative. This led to a worsening situation: the positive signals disappeared. In contrast, the negative ones remained, making it all the more difficult, structurally, to regain credibility through the same channel that had been compromised.
The mistake did not affect the score; it only changed the system that would produce future scores.
The Bank Account Model Is Wrong
The way most reputation management works can be described as the bank account model: reputation functions like a running balance, with positive actions serving as deposits and negative events as withdrawals, so that if enough deposits are made over time, the balance will eventually be restored after being depleted. Although this is an intuitive model, it is in fact mostly wrong, or at least wrong enough to cause recovery strategies to perform poorly.
The bank account model suggests that there is symmetry involved: anything which has lowered a reputation can, in turn, restore it. Since negative reviews are damaging, positive reviews should help bring about recovery; if a breach of credibility has created a gap, credibility-building communication should be able to close it. The model therefore states that the way to recover is the reverse of how the decline occurred.
The Yelp data indicate that trust does not build up and wear away symmetrically. Instead, it builds up bit by bit through a series of small confirmations over time—each visit, each review, and each interaction that fulfills the promise. However, it collapses suddenly at the point of a single violation, and this collapse does not diminish in proportion to the amount of effort put into recovery. In the cases the researchers examined, the damage involved a structural change to the information channel. Once a business has lost review momentum after a consumer complaint, it cannot simply ask for more positive reviews to make up for it, since the number of people willing to review that business has already decreased.
Trust builds up little by little and suddenly breaks down; the way back is not the opposite of the way down, since the ground itself has changed.
What Scar Tissue Actually Is
The analogy concerning scar tissue should be taken literally, not merely as a metaphor. Scar tissue does not indicate a lack of healing; it is a different type of tissue with reduced elasticity, diminished function, and a tendency to pull on adjacent structures in ways the original tissue never did. A scar is the body's response to damage, and it is genuine, even though it does not fully restore what was originally there.
Reputational damage works in the same way. A business that has been flagged for review manipulation doesn't just have a lower rating; it has a different relationship with the platform, a different relationship with the consumers who saw the alert, and, according to the research, an altered review ecosystem in which the usual mechanisms for building social proof are working at a reduced level. The brand is not gone; it is there but in a form that is less responsive to the kinds of inputs that normally build up a reputation.
It is important from a strategic point of view since it changes the picture of what recovery really entails. In the case where reputation damage was just a matter of subtraction, recovery would then be an addition exercise: launch a campaign, get positive press, obtain more reviews, and the balance would eventually be restored. But if the damage has altered the underlying system—the rate at which reviews come in, the remaining reviewers' attitudes, and the credibility rules consumers use when assessing the business—then the recovery approach must focus on the system, not just the score.
The Asymmetry in Practice
There is also an asymmetry in how audiences respond to positive and negative reputation signals. Negative information about a brand has more cognitive impact than equivalent positive information—not just because consumers are cynical—but because the potential risk of making a wrong choice is more striking than the benefit of making a correct one. A company with 100 positive reviews and one credibility scandal does not seem to be regarded as having received a 99-to-1 vote in its favor; the scandal serves as a disqualifying signal, so the positive reviews cannot simply outweigh it.
The study illustrates this point: businesses with a high reputation in the dataset—those that mostly had four- and five-star reviews—saw about 2.9 percent more foot traffic than their lower-rated competitors. This advantage is genuine and measurable. However, it was also this advantage that led some business owners to attempt to manipulate the system, and the data make it clear that the penalty for being caught was many times greater than the value of the benefit they sought to secure. The businesses that lost 8 percent of their foot traffic and continued to decline for 18 months were paying a price several times larger than the marginal benefit they had been attempting to create.
A business with a 100 positive reviews but involved in a credibility scandal is not seen as having 99-to-1 vote; the scandal serves as a disqualifying signal, so the positive reviews cannot outweight it.
What Recovery Actually Requires
If the bank account model is incorrect, the recovery strategy based on it must also be incorrect. Simply inundating a damaged reputation with positive messages does not tackle the underlying mechanism, and in certain cases, a recovery approach that relies heavily on communications can make the situation worse by conveying a defensive attitude rather than taking responsibility.
The research identifies three clear recovery needs that most conventional methods fail to address.
1
The first aspect is behavioral rather than communicative. Trust is lost when a particular action breaches an expectation, so restoring it requires a specific behavioral change that can be observed and verified—not a statement that the issue has been dealt with, but actual evidence that it has been. It is important to make this distinction because consumers who have witnessed a breach of credibility are particularly sensitive to whether the response is substantial or merely performative. A statement expressing a commitment to authenticity coming from the same company that was caught manipulating reviews does not constitute a trust signal; it is just noise.
2
The second point is about having patience in line with the real timeline. It is valuable that the finding after eighteen months counters the tendency to consider recovery complete as soon as the immediate crisis is over. A ninety-day alert period should not be taken to mean that recovery has taken place. Companies that thought the removal of the flag signified the end of the problem were working from a timeline unsupported by the data. Any recovery plan that fails to account for the long-term effects of reputational damage will always underestimate the resources required.
3
The third measure is to diversify your channels. If the main reputation channel—take Yelp reviews, for example—has suffered structural damage, the recovery plan cannot rely solely on that channel to restore the signal. Instead, it must build credibility through other channels that remain unaffected: direct relationships with customers, third-party endorsements, press coverage, and a presence in the community. These channels do not replace the compromised one, but they can rebuild the fundamental trust that ultimately enables the primary channel to work again.
The Regulatory Layer
This situation also has a compliance aspect that goes beyond the strategic point. The FTC now regards fake reviews as an unfair method of competition under Section 5 of the FTC Act, which provides for per-violation fines. The amount paid in the Amazon Prime settlement — $2.5 billion, with the settlement finalized in September 2025 — shows the level of enforcement the agency is willing to pursue against manipulative engagement mechanisms. Although the FTC's Click-to-Cancel rule was overturned on procedural grounds, enforcement actions under the agency's current authority have not decreased.
This indicates that for brand leaders, the assessment of risk has altered. Manipulation of online reviews is no longer just a reputational risk; it has become a regulatory issue with serious financial implications. Moreover, according to Yelp's research, the reputational damage alone, in terms of losses in foot traffic and transactions over more than 18 months, had already exceeded the marginal benefit for the majority of businesses in the dataset. A regulatory penalty has now been added to a reputational penalty that was already expected to be negative.
The key conclusion is far from complicated: the systems that most efficiently build trust—through consistent service, real customer relationships, and operational integrity—are also the ones that create no regulatory exposure and no scar tissue when functioning properly. The shortcut is more expensive than taking the longer route, even before any fines are taken into account.
Reputation isn't like a bank account; it's a living system, and if you damage that system, you don't just end up with less, you're working with different equipment from the start.
Some ideas are worth discussing in the context of your organization.
