In July 2026 the Federal Trade Commission finalised an order against the supplement brand TruHeight carrying a $4 million judgment, partially suspended on payment of $750,000. Part of what the company was accused of doing will sound uncomfortably familiar to a lot of marketing teams: several thousand five-star reviews on its own site written by employees, and free or discounted product handed out in exchange for five-star ratings elsewhere. Social proof stopped being a growth tactic that quarter. It became a regulated claim.
Key takeaways
- The FTC rule took effect 21 October 2024; first warning letters went out in December 2025.
- An EU screening of 223 websites found 55% potentially in breach on review reliability.
- Purchase likelihood peaks between roughly 4.0 and 4.7 stars, not at 5.0.
- Displaying five reviews raised purchase likelihood by 270% in the Spiegel research.
Enforcement caught up in 2026
The rule that matters most for US-facing brands is the FTC’s Rule on the Use of Consumer Reviews and Testimonials, effective 21 October 2024. It bans buying or selling fake reviews, prohibits paying for either positive or negative reviews, sets out how insider reviews must be disclosed, and lets the Commission seek civil penalties from knowing violators. Because each fake review is generally treated as its own violation, the arithmetic on a batch of a few hundred gets alarming quickly.
For a year it looked like a rule without a body count. That ended in December 2025, when the FTC issued warning letters to ten companies, later published in April 2026: six property management firms, three personal injury law firms and one accounting firm, flagged for offering lease discounts, gift cards and similar incentives in exchange for positive reviews. The TruHeight action followed in April 2026 and settled in July.
Europe got there earlier and by a different route. The Omnibus Directive amended the Unfair Commercial Practices Directive with rules applying from 28 May 2022: traders who give access to consumer reviews must state whether and how they check that those reviews come from people who actually bought or used the product, and taking no steps at all while implying otherwise is a misleading practice. A coordinated screening of 223 major websites, announced by the European Commission in January 2022, found 55% potentially in breach on this point, and in 144 of them authorities could not confirm the trader was doing enough to ensure reviews were authentic.
What the ratings evidence actually shows
The most useful body of work here remains the Spiegel Research Center study at Northwestern University, published in June 2017 with PowerReviews. Three findings from it have held up well enough to plan around.
First, the jump comes early. Purchase likelihood for a product with five reviews was 270% higher than for the same product with none, and the marginal value of each additional review falls away quickly after that. Second, the effect scales with price: displaying reviews lifted conversion 190% for a lower-priced product and 380% for a higher-priced one. Third, and least intuitive, purchase likelihood peaks in the 4.0 to 4.7 star band and declines as ratings approach 5.0. A perfect score reads as curation rather than consensus.
That last finding is worth sitting with, because it inverts the instinct that drives most review programmes. Chasing a flawless average is not just legally risky if you do it by filtering, it appears to be commercially self-defeating.
Reviews and testimonials do different work
Reviews are a distribution of opinion
A review’s value comes from the fact that you did not choose it. Volume, recency and spread are what make the average legible, which is why a page with sixty reviews and a 4.4 average persuades more than one with four glowing entries. Reviews answer the question “is this what it claims to be.” They are also the format regulators watch most closely, because they carry an implicit promise of independence.
Testimonials are a curated argument
A testimonial is selected, and everyone knows it. That makes it a weaker signal of consensus and a stronger vehicle for specificity: the customer who names the problem they had, the number they moved and the thing that nearly stopped them buying. Testimonials answer “is this for someone like me.” The compliance line here is narrow and worth stating plainly: a testimonial describing an atypical result needs enough context that a reader does not take it as the normal outcome, and any material connection to the brand, employment, equity, free product, has to be disclosed.
Collecting reviews without walking into the rule
The pattern that gets companies into trouble is almost never sophisticated fraud. It is a well-meant incentive attached to the wrong condition. Five habits keep a programme on the right side of both frameworks.
- Ask everyone, not just the happy ones. Sending the request only to customers who scored well in a satisfaction survey is review gating, and it manufactures exactly the inflated average the data says underperforms.
- Never condition anything on sentiment. If an incentive exists, it is for leaving a review, full stop, and that has to be true in the wording as well as in the intent.
- Disclose insider reviews. Employees, contractors, agencies and anyone with equity must say so clearly, in the review itself rather than in a policy page.
- Say how you verify. In the EU this is an obligation. Everywhere else it is free credibility: a short line explaining that reviews come from confirmed orders does real work, and verified-buyer badges were associated with a 15% improvement in purchase odds in the Spiegel research.
- Keep the audit trail. Order reference, timestamp, unedited text. If you cannot reconstruct where a review came from two years later, you cannot defend it.
Negative reviews are an asset you are allowed to keep
Suppressing unflattering reviews is both the most tempting shortcut and the one with the worst risk profile, since it goes to the heart of what the FTC rule and the UCPD are aimed at. It also removes the material that makes the rest believable. The 4.0 to 4.7 finding is, read another way, a finding about the value of visible dissent.
A published complaint with a specific, unembarrassed reply from the company does more for trust than a tenth compliment. It demonstrates something no testimonial can: what happens when things go wrong. Readers looking for that signal will find it whether or not you provide it, and the version they find on a forum will be less generous than yours.
Questions we get from clients
Can we still offer an incentive for leaving a review?
Generally yes, provided it is offered regardless of what the review says and is disclosed. The December 2025 warning letters targeted incentives tied to positive reviews specifically, which is the distinction to hold on to.
Do these rules apply if we only sell in one market?
They apply where your customers are, not where your office is. A brand shipping into the EU is subject to the disclosure obligation, and one selling into the US is within reach of the FTC rule regardless of where the reviews were collected.
How many reviews do we need before displaying them?
Fewer than most teams assume. The Spiegel data puts most of the lift inside the first handful, so publishing five honest reviews beats waiting for fifty.
The framing that has aged best, in our view, is to stop treating social proof as persuasion and start treating it as evidence. Evidence you can source, date and hand to a regulator is also, conveniently, the kind a sceptical buyer believes. The same instinct is reshaping how brands talk about their supply chains and their claims, which we looked at in our piece on marketing in the age of ethical consumerism.
Proof earns the click, the button has to deserve it
Trust built by reviews gets spent at the moment you ask the reader to act.
Sources: Federal Trade Commission, Rule on the Use of Consumer Reviews and Testimonials, effective 21 October 2024, which authorises civil penalties against knowing violators; FTC warning letters issued December 2025 to ten companies, published April 2026, covering six property management companies, three personal injury law firms and one accounting firm; FTC action against TruHeight (Vanilla Chip LLC) announced April 2026, final order approved July 2026 with a $4 million judgment partially suspended on payment of $750,000. Directive (EU) 2019/2161 amending the Unfair Commercial Practices Directive, applicable since 28 May 2022, on disclosure of review verification. European Commission, coordinated screening announced January 2022 covering 223 websites in 26 member states plus Iceland and Norway, 55% potentially in breach, authenticity unconfirmed on 144 sites. Spiegel Research Center, Medill, Northwestern University, with PowerReviews, June 2017, for the 270% figure at five reviews, the 190% and 380% conversion lifts by price point, the 4.0 to 4.7 rating band and the 15% verified-buyer effect; these are observational e-commerce findings and do not guarantee a result in any specific programme. This article is general information, not legal advice. Updated August 2026.
