Testimonials can be honest and still prove nothing, and federal law agrees

Key takeaways

• In an FTC-commissioned test, 55% of shoppers who saw testimonials of 48- to 72-pound weight losses with a “results not typical” disclaimer still took away that new users would lose at least 48 pounds.
• Federal advertising guides say consumer endorsements are not scientific evidence, and an ad must show its stories are typical or clearly state what buyers can generally expect.
• Since October 2024, an FTC rule has banned fake testimonials, including AI-generated ones; as of December 2025, the agency put the maximum civil penalty at $53,088 per violation.
• Personal reports do real work as warning signals: doctors’ reports of seven liver-failure patients in Hawaii in 2013 started a state and federal investigation of a supplement.
• Treat a testimonial as a lead, ask how many people tried the product and didn’t post, and report fake or harmful claims to the FTC or FDA.

In late 2002 and early 2003, researchers stopped shoppers in malls in 12 U.S. markets and showed them ads for products that didn’t exist¹. One was a weight-loss program called WeightGuard. Its ad featured five happy customers¹.

In one version, those customers said they had lost 48 to 72 pounds. With no disclaimer, 69% of viewers took away the message that a new user would lose at least 48 pounds¹.

Then the researchers added the words “results not typical.” The share barely moved: 55% still took away the 48-pound promise¹. What worked was a plain number. When the ad said the average user loses about 10 pounds in three months, the share fell to 15%¹.

The study, done for the Federal Trade Commission (FTC), helps explain a rule change. In 2009 the agency stopped treating “results not typical” as a safe harbor, a phrase that had shielded advertisers who used it².

If you value lived experience, testimonials can feel like the most honest evidence there is. Often they are honest. But an honest story and proof that a product works are two different things. The research, the law and the record of real cases all show why.

A disclaimer under a testimonial barely changes what you believe

The weight-loss test wasn’t a one-off. In a 2003 study for the FTC, the same researchers, Manoj Hastak and Michael Mazis, showed 200 supplement users a booklet of 18 glowing testimonials³. Four in five were 60 or older, and all had breathing problems, low energy or chronic pain³.

Some booklets carried a boxed notice. One said the testimonials “do not prove our product works” and that readers should not expect similar results³.

Without a notice, 67.6% believed the product would ease pain for at least half of users. With the “doesn’t prove” notice, 55.8% still did³. None of the four notices the researchers tested made a statistically significant dent in those beliefs³.

An ad for a made-up cholesterol supplement told the same story. Adding “results not typical” did not significantly change what people thought other buyers could expect¹.

The lesson is that your mind treats a story as a sample of what usually happens, even when the small print says it isn’t. A real average was the only fix that worked well. Even then, more than 40% of viewers who saw the 48- to 72-pound ad still believed a new user would lose at least 20 pounds, twice the stated average¹.

Federal law says customer stories can’t stand in for studies

The FTC’s Endorsement Guides set the rules for testimonials in ads, and they were last revised in 2023⁴. On this point the wording is blunt: “Consumer endorsements themselves are not competent and reliable scientific evidence”⁴.

That sentence has teeth. If a customer in an ad says a product eased her joint pain, the FTC reads that as the company’s own claim⁴. The company needs the same proof it would need to say so directly⁴.

The guides add a second test, about typical results. A testimonial about a product’s main benefit is likely to be read as what most buyers will get⁴. If the company can’t back that up, it should clearly disclose the results people can generally expect, and that figure needs proof, too⁴.

The guides’ examples read like a catalog of how testimonials mislead⁴:

• A page made up only of testimonials implies the results are typical.
• Results from people who stuck with a program may not reflect everyone who started it.
• When a few extreme results skew an average, the median, or middle result, can be more honest.
• Passing along only the favorable reviews misleads.

The guides are the FTC’s reading of the law, not a law by themselves. In a case, the agency still has to prove a violation of the FTC Act².

The stakes rose in April 2023. The FTC sent notices to about 670 companies that sell supplements, over-the-counter drugs, homeopathic products and functional foods⁵.

The notices said claims to treat a serious disease need at least one well-controlled human clinical trial⁵. They came with a second notice on endorsements, which covers passing off an endorser’s experience as typical. At 2023 rates, each violation could bring a civil penalty of up to $50,120⁵.

Fake testimonials now carry federal fines

Some testimonials aren’t honest at all. In August 2024, the FTC voted 5-0 to approve a rule banning fake reviews and testimonials⁶. It took full effect on October 21, 2024⁷.

The rule bars testimonials from people who don’t exist, including ones generated by artificial intelligence, and from people who never used the product⁶. It also bans rewards that depend on what a review says, positive or negative. Testimonials from company managers or staff must disclose the connection⁶.

The agency has begun using it. On December 22, 2025, FTC staff warned 10 companies about possible violations⁷. The letters said a court could impose civil penalties of up to $53,088 per violation, though a warning letter is not a finding that anyone broke the rule⁷. Our article on online reviews covers how to spot fakes.

How real companies manufactured their proof

Federal cases show what testimonial abuse looks like in health marketing.

A weight-loss telehealth firm. In July 2025, the FTC charged NextMed, a New York company selling memberships in weight-loss programs built around GLP-1 drugs⁸. According to the complaint, its testimonials came from people the company hired, from its own staff and from their relatives. None had used its programs⁸.

The agency also said the before-and-after pictures showed strangers, not clients, and that a promised average loss of 53 pounds had no support⁸. Unhappy customers, it alleged, were offered Amazon gift cards to soften or delete their reviews⁸.

NextMed settled and agreed to pay $150,000. The FTC made the order final in December 2025, and it requires solid evidence behind any claim about typical results⁸.

A brain-training app. The FTC said Lumosity’s ads implied its games could protect users against dementia and Alzheimer’s disease. Lumos Labs, the maker, settled those charges in January 2016⁹.

Part of the case involved its website’s success stories. The company had gathered some of them by running contests, with an iPad, a lifetime subscription and a San Francisco trip as prizes, and never told readers⁹. Lumos Labs paid $2 million; the rest of a $50 million judgment was suspended because of its finances⁹.

Network sellers. In April 2020, the FTC warned 10 multilevel marketing companies about health and income claims, including distributor posts suggesting products could prevent or treat COVID-19¹⁰. The agency’s consumer protection director said such companies are responsible for the claims their distributors make¹⁰. Our article on wellness MLMs shows how those posts blend a personal story with a sales pitch.

Why one result can’t answer your real question

When you read a testimonial, the question in your mind is “Will this work for me?” One story can’t answer it, for a simple reason: it has no denominator.

A denominator is the bottom number of a fraction. “It worked for my neighbor” is a top number with nothing under it. You don’t know whether 10 people tried the product or 10,000. You don’t know how many improved, how many didn’t, and how many would have improved anyway.

Testimonials are also chosen. A company picks its best stories, and the FTC’s own examples warn that people who stick with a program differ from people who start it⁴. The ones who quit, got worse or saw nothing rarely get a page.

Our article on healing stories covers the other traps, such as illnesses that fade on their own and extreme symptoms that drift back toward normal. Taken together, one honest success story fits a product that works, one that does nothing, or one that helps fewer people than it harms.

Where personal reports earn real weight

None of this makes personal reports worthless. In two places, they are essential.

Warning signals. The FDA collects reports of bad reactions to foods, supplements and cosmetics, and makers and distributors must pass along the serious reports they receive¹¹. Like testimonials, these reports can’t prove a product caused the problem, and the FDA says so plainly¹¹.

But they can sound an alarm. In September 2013, doctors in Hawaii flagged a cluster no one could explain: seven people whose livers were failing, with no clear cause¹².

The common thread was a weight-loss and muscle-building supplement, OxyELITE Pro. Every one of the seven had taken it. The state health department heard about the cluster on September 9 and, with the CDC and FDA, opened an investigation¹².

A public alert brought more reports. By October 2013, 29 people met the case definition. Eleven were hospitalized, two needed liver transplants and one died¹². A later review counted 44 cases among 76 doctor reports and called it one of the largest outbreaks of liver injury tied to a supplement¹³.

Notice how those reports were handled. No one declared the product guilty from the first story. Officials investigated, counted and compared. That’s the right way to treat a glowing story, too: as a reason to look closer.

The patient’s voice in trials. Researchers also gather personal reports on purpose. The FDA has guidance on patient-reported outcome measures, questionnaires that collect reports straight from patients in a clinical trial¹⁴. The difference from a testimonial is that everyone in the trial answers, including people who didn’t improve and people who got a placebo. Our article on feeling better versus getting better explains how to read them.

Testimonial or evidence? A side-by-side check

Question	What a testimonial can tell you	What a controlled study can tell you
Did something happen to a real person?	Yes, if the person is real and honest	Yes, for each participant
How many people tried it?	Unknown	Stated
How many got no benefit or got worse?	Usually left out	Counted
Would they have improved anyway?	Can’t tell	Compared with a group that didn’t get it
Who chose which results you see?	Often the seller	Everyone enrolled is supposed to be counted
What result is typical?	Unknown unless the ad states it4	An average or median, with a range
Can it flag a possible harm?	Yes, as a signal to report11	Yes, with rates

Before a testimonial moves your money or your health choices, run it through these six checks:

1. Is the person real? Fake and AI-written testimonials are banned, but bans don’t make them disappear⁶.
2. Is a typical result stated? If an ad shows only stars and stories, expect the usual result to be smaller¹⁴.
3. Was the person rewarded? Free products, payments, contest prizes, a job with the company or a distributor account all count⁹¹⁰.
4. Is it about a disease? The FTC expects at least one well-controlled human trial behind claims to treat a serious disease⁵.
5. Was the product itself tested? A study of one ingredient, in a lab dish or at a different dose, isn’t the same thing.
6. What would being wrong cost you? Wasted money on a skin cream differs from delayed care for chest pain.

What the evidence shows

Claim	What the research or record shows	Evidence level
A “results not typical” or “doesn’t prove it works” notice corrects the impression	Two FTC-commissioned studies found these notices mostly failed13	Evidence says no
Stating the average result helps people judge testimonials	It sharply cut misleading takeaways in one FTC study but didn’t erase them1	Promising but early
Customer testimonials show a product works	Federal guides say endorsements aren’t scientific evidence; missing denominators and hand-picked stories can’t be ruled out4	Evidence says no
Reports of harm can reveal a real problem	Doctors’ reports set off the 2013 Hawaii investigation of liver injury tied to a supplement1213	Strong evidence that reports can flag problems, though one report can’t prove cause11

Your experience counts. Make it count more

Natural health has always traveled by word of mouth, from neighbor to neighbor and parent to child. That tradition deserves respect, and so does the honesty behind most personal stories.

Respect also means holding sellers to the standard you’d hold yourself. A company that profits from your story should be able to show what happens to most buyers, not just its happiest ones.

Your own experience can do public good, especially when something goes wrong. If a product makes you sick, tell your clinician and report it to the FDA through MedWatch or at 1-800-FDA-1088¹⁵. Report fake or deceptive health ads to the FTC at ReportFraud.ftc.gov.

And if a testimonial tempts you to stop a prescribed medicine or skip care for a serious condition, talk with your clinician first. The FDA warns that success stories are easy to make up and are no substitute for scientific evidence¹⁵.

Questions readers ask

The bottom line

A testimonial can be completely true and still tell you almost nothing about what a product will do for you. It has no denominator, it’s usually chosen by the seller, and FTC research shows small-print disclaimers barely dent the impression it leaves. Federal law agrees: customer stories are not scientific evidence.

Use personal reports for what they do well. Treat a glowing story as a lead, ask who tried it without results, and look for the typical outcome. Treat a story of harm as a signal worth reporting. Save big decisions for evidence that counts everyone.

Related reading

• How to weigh healing stories, which move us most when the stakes are highest
• Fake reviews became illegal in 2024, yet reviews still mislead you
• Wellness MLMs sell real products and a business few people profit from
• A fake inhaler made asthma patients feel better while their lungs showed no change
• What an influencer is legally required to tell you

Sources

1. Hastak M, Mazis MB. “Effects of Consumer Testimonials in Weight Loss, Dietary Supplement and Business Opportunity Advertisements.” Report submitted to the Federal Trade Commission, 2004. ftc.gov
2. Federal Trade Commission. “FTC Publishes Final Guides Governing Endorsements, Testimonials.” FTC, 2009. ftc.gov
3. Hastak M, Mazis MB. “The Effect of Consumer Testimonials and Disclosures on Ad Communication for a Dietary Supplement.” Report submitted to the Federal Trade Commission, 2003. ftc.gov
4. Federal Trade Commission. “Guides Concerning Use of Endorsements and Testimonials in Advertising: Consumer endorsements,” 16 CFR 255.2, as amended July 26, 2023. eCFR. ecfr.gov
5. Federal Trade Commission. “FTC Warns Almost 700 Marketing Companies That They Could Face Civil Penalties if They Can’t Back Up Their Product Claims.” FTC, 2023. ftc.gov
6. Federal Trade Commission. “Federal Trade Commission Announces Final Rule Banning Fake Reviews and Testimonials.” FTC, 2024. ftc.gov
7. Federal Trade Commission. “FTC Warns 10 Companies About Possible Violations of the Agency’s New Consumer Review Rule,” December 22, 2025, ftc.gov ; and the accompanying warning letter template, search.ftc.gov
8. Federal Trade Commission. “FTC Takes Action Against Telemedicine Firm NextMed Over Charges It Used Misleading Prices, Fake Reviews, and Deceptive Weight Loss Claims to Sell GLP-1 Weight-Loss Programs,” July 14, 2025, ftc.gov ; and “FTC Approves Final Order against Telehealth Provider NextMed,” December 3, 2025, ftc.gov
9. Federal Trade Commission. “Lumosity to Pay $2 Million to Settle FTC Deceptive Advertising Charges for Its ‘Brain Training’ Program.” FTC, 2016. ftc.gov
10. Federal Trade Commission. “FTC Sends Warning Letters to Multi-Level Marketers Regarding Health and Earnings Claims They or Their Participants are Making.” FTC, 2020. ftc.gov
11. U.S. Food and Drug Administration. “Adverse Food Event Overview” (CAERS adverse event reports). openFDA. open.fda.gov
12. Centers for Disease Control and Prevention. “Acute Hepatitis and Liver Failure Following the Use of a Dietary Supplement Intended for Weight Loss or Muscle Building — May–October 2013.” Morbidity and Mortality Weekly Report, 2013. cdc.gov
13. Council of State and Territorial Epidemiologists. Conference abstract on the 2013 Hawaii investigation of hepatitis linked to OxyELITE Pro. CSTE Annual Conference, 2015. cste.confex.com
14. U.S. Food and Drug Administration. “Patient-Reported Outcome Measures: Use in Medical Product Development to Support Labeling Claims.” Guidance for Industry, 2009. fda.gov
15. U.S. Food and Drug Administration. “6 Tip-offs to Rip-offs: Don’t Fall for Health Fraud Scams.” FDA, 2021. fda.gov