I Hired a Lab to Counterfeit-Test a Dozen Suspicious Beauty Products I Bought Online. Every Single One Had a Problem.

U.S. Customs and Border Protection seized 500,716 counterfeit personal-care items — a category that includes cosmetics, skin care, soap, and toothpaste — during the 2024 fiscal year. That’s peanuts considering how many fake handbags and wallets (5,132,402), fake pharmaceuticals (3,743,413), fake clothing items (1,043,853), fake pairs of sunglasses (771,533), and fake pieces of jewelry (713,036) the agency intercepted over that same time frame.

That figure is even more minuscule compared to the number of bootleg beauty products that aren’t confiscated. Authorities capture less than 2.3% of all counterfeit goods sent to the United States, according to a July 2020 report from the National Association of Manufacturers.

If those 500,716 counterfeit personal-care items represented 2.3% of the fakes arriving stateside, that means another 21,269,544 snuck across the border that same year.

Then there are domestically manufactured fakes, like the $700,000 worth of illegitimate Kylie Cosmetics, MAC, NARS, and Urban Decay products seized by Los Angeles police in a 2018 raid. Authorities were reportedly tipped off by buyers who, not realizing they’d purchased counterfeits, complained to the actual brands about developing bumps and rashes after using “their” products. The fakes were ultimately “found to contain bacteria and human waste,” as LAPD deputy chief Marc Reina later tweeted.

And that’s why counterfeit beauty, even though it represents a small fraction of the fakery out there, is one of the most critical categories to avoid. These are products that go on or in your body, and that can be risky. Over the past two decades, fake personal-care products have been found to contain ingredients like antifreeze (in counterfeit toothpaste as a sweetener, I’m sorry to tell you) and urine (to give a fake fragrance a golden hue, I’m even more sorry to say).

“People go to the hospital because of these products,” Bob Barchiesi, president of the International AntiCounterfeiting Coalition, said in a video interview. “They end up in the emergency room because of them.”

To see how easily someone might unwittingly buy bogus beauty items online, I shopped for skin care and makeup just as millions of people do every day: on my phone and laptop, browsing several of the world’s most popular e-retailers.

A big difference, of course, was that I kind of wanted to get scammed. For journalism.

To mitigate that inclination, I looked for third-party storefronts with lots of positive customer reviews and lots of beauty items for sale. I didn’t want to stack the deck by engaging with an obvious swindler or fly-by-night seller (although it seems that one of my eBay sellers has since vanished). I also aimed to approximate the reassurance of finding a seller who’s been doing business for a while and seemingly has plenty of testimonials from happy customers.

I focused exclusively on third-party sellers who listed through big online marketplaces. Defined broadly by Amazon as “independent sellers who offer a variety of new, used, and refurbished items,” third-party sellers are legion. They’re responsible for 17.4% of sales on Shein, more than 60% of Amazon’s sales, 95% of all products listed on Walmart.com, and virtually all of eBay’s sales.

They’re also frequently called “unauthorized sellers,” because in many cases they haven’t been officially approved by a brand to sell its merchandise, according to the retail-intelligence platform Wiser. For example, I easily purchased that puffy tube of Rhode lip gloss off Amazon even though Rhode does not list Amazon as an authorized sales channel on its FAQ page. When you purchase Rhode-labeled products via a third party, the company says, it “cannot guarantee that you’re buying genuine products or vouch for their authenticity, quality, or condition.”

Even when platforms such as Amazon or Walmart have an authorized partnership with a brand, that brand’s products might still be available on those same sites via unauthorized, third-party sellers, which can be confusing even for knowledgeable shoppers. For example, if you’re looking to buy CeraVe products on Amazon, you can do so via the brand’s official Amazon store — but if you search Amazon for a specific CeraVe product and filter your search results by “Seller,” you may surface some third-party options that do not meet CeraVe’s standards for authenticity, as noted on the company’s FAQ page.

As common as third-party sales are, they’re unfortunately the method through which much counterfeit merchandise is moved online. About 43% of goods purchased from third-party sellers on popular e-commerce platforms, including 13 out of 13 cosmetics products, were found to be counterfeit in a 2018 report from the United States Government Accountability Office.

In the beauty industry, first-party sellers and authorized retailers include brick-and-mortar Sephora stores, Ulta locations, and department stores, as well as those retailers’ digital storefronts. (Ulta now invites select brands to drop-ship through its Ulta Beauty Marketplace, but it isn’t open to third-party sellers, per a company statement.) Such retailers provide “well-designed, well-thought-out supply chains,” Zac Rogers, an associate professor of supply chain management at Colorado State University, said in a video interview. The supply chains typically involve moving goods straight from Point A to Point B, so their custody and provenance are relatively guaranteed.

A third-party chain, however, might move inventory from questionable origins or include who-knows-how-many intermediaries en route to the final customer, remaining open to vulnerabilities. “If I’m seven steps removed from the manufacturer, I don’t know if something was stolen,” Rogers said. “I also have to find out on my own if there has been a recall, because everything is so ad hoc. There’s all sorts of weird things that can happen.”

Overall, I purchased 12 beauty products from third-party sellers. That includes three products for which I ordered two suspicious versions, simply because more than one listing caught my eye.

I also bought those same products from the brands’ own websites (which, by default, are authorized, first-party sellers). These items served as my control specimens, against which we compared our third-party samples.

I sent all of my purchases to Johnson, the cosmetic chemist in New Jersey, who conducted a series of tests to compare our controls with our more suspicious samples.

To assess chemical similarities and differences between the products, Johnson relied on a Fourier Transform Infrared (FTIR) spectrometer. Frequently used for quality control or material authentication, an FTIR shoots a laser at a substance and provides a breakdown of what’s inside. It can’t spit out a word-for-word ingredient list, but it can determine the number of ingredients in a substance and (depending on the complexity of the mixture) the concentration of each as a percentage of the overall mixture.

It can also compare results from two separate substances and produce a percentage match. The lower that percentage, the more likely that the two substances are different — or, for our purposes, that one is a spurious version of the other. Analyzing with FTIR involves some margin of error, so Johnson told me that she would want to see a 95% match or higher between an authorized product and its unauthorized counterpart to call the latter “real.”

Only three of our 12 samples — two bottles of Drunk Elephant glycolic serums and a Dior concealer — surpassed that 95% benchmark.

Because there are no precise rules for exactly how fake a product should appear to be before it can be deemed definitely fake, I cross-checked our results with Andrew Koenig, a technical and formulation consultant at CosChemist Consulting in California. He said that a sample could receive as low as an 80% to 90% FTIR match against its real counterpart and still wind up being a legitimate product. (Johnson had mentioned that other chemists might accept an 80% match as authentic. Again, there’s no industry standard.)

By that 80% minimum, five of our 12 samples would be definitively counterfeit. This group included a supposed Nars concealer that was only a 2% match to its authentic counterpart and two alleged bottles of Charlotte Tilbury Airbrush Flawless Setting Spray that got FTIR readings of 21% and 25% compared with the real thing. Four other samples’ FTIR results fell into the 80% to 90% range.

Johnson also employed a 14-point sensory analysis that she uses when formulating “dupes,” which is when a legitimate brand intentionally produces a product similar to one that’s already on the market. For the sensory analysis, Johnson graded each product’s opacity, shine, tackiness, absorption, and similar qualities on a scale from 0 to 10 and then compared the two outcomes.

None of our suspect samples, including those with high FTIR percentage matches, matched its authentic counterpart on all 14 points. In combination with the spectrometer testing and packaging-discrepancy observations, this analysis helped solidify Johnson’s conclusion that all of my third-party products were counterfeit.

For example: Our Amazon-purchased sample of Rhode Peptide Lip Treatment was an 80% FTIR match to its real counterpart. But in addition to all of the peculiarities we saw on the packaging, as well as the potential bacteria lurking inside, we could easily tell just by seeing what came out of each tube that something was different. The real Rhode balm looked opaque, while the other substance looked translucent.

Another example: A tub of supposed Drunk Elephant Protini Polypeptide Cream from eBay achieved an 83% match. But that sample was noticeably more yellow in tone, had a glassier shine (almost as if the product was separating a bit), and did not form peaks as stiff as what we got from the Protini Polypeptide Cream I’d purchased on Drunk Elephant’s website.

Even when a suspected fake earned an outstanding FTIR match percentage, there were reasons for skepticism.

I bought two bottles of another Drunk Elephant product, T.L.C. Framboos Glycolic Night Serum, from a Walmart third-party seller and an eBay third-party seller. Both samples were a 97% FTIR match to the real thing.

Johnson, however, said that glycolic acid’s molecular structure is rather simple, so even something with a 97% FTIR match might be an inauthentic copycat. (Koenig agreed about the molecular structure but still said that a 97% match was likely not counterfeit.) Additionally, Johnson detected differences in the packaging that led her to believe that the ones I’d purchased via third party might be illegitimate: On the boxes that these serums came in, for example, the lot codes had been stamped in slightly different locations with what appeared to be different marking technologies. (I asked Drunk Elephant if it could confirm any changes in how it printed lot codes but did not receive a response before publication.)

Because packaging differences were a consistent tell across many of my purchases, I also asked Elizabeth Carey Smith, a branding and typography designer in New York, to examine photos we’d taken of our real and maybe-not-real packages and weigh in. Although Smith said she was “very surprised at how close some of these things are,” she pointed out several inconsistencies, including some bits of heavier typeface on the T.L.C. Framboos Glycolic Night Serum boxes, which could indicate that they were “probably using an image [of the legitimate packaging] or scanning it.”

Other smoking guns may have been hiding online. The eBay seller who sent me one of those Drunk Elephant serums has seemingly disappeared in the four-ish months since I placed my order, which certainly doesn’t instill confidence. Meanwhile, the Walmart seller who sent me the other has received at least 10 one-star customer reviews for a variety of items since December 1, 2025, peppered with claims like “This is not an authentic product,” “This seller is selling knockoff perfumes,” “The description and the pictures do not match the product,” and “I see you did [a] bait and switch.”

My most confounding specimen was a package of SkinCeuticals C E Ferulic serum. The sample I bought from a Walmart third-party seller had an “e-mark,” a quality-assurance emblem from the European Union, which made me wonder if we were dealing with someone who had traveled to Paris, loaded up a suitcase with serums, and come home to turn a quick buck. (Then again, the e-mark could just be fake, perhaps added to the packaging to make me think the item had come from Europe.)

Compared with an identical serum I purchased from the SkinCeuticals site, the one from Walmart registered as a 90% FTIR match — right on the cusp of what may be considered legitimate, according to Koenig (but not Johnson). But the colors of the two serums were significantly different: The one from the SkinCeuticals site was pale yellow, while the e-mark sample looked orange. The latter shade, as experienced vitamin C serum users may know, typically indicates that the product has oxidized, which renders it less potent.

When I contacted L’Oreal, the parent company of SkinCeuticals, for comment, it offered to retest my third-party-purchased sample. I hand-delivered it to the company in New York. L’Oreal claimed it was legit, telling me that the European version is darker in color even when it’s fresh. For what it’s worth, both Johnson and Koenig were skeptical of that claim: As Koenig put it, “No one’s gonna say that their product is oxidizing.”

According to other experts I spoke to, no one’s gonna say that their product is counterfeit, either.

Although brands may work diligently behind the scenes to thwart bad actors who are making or selling fakes based on their intellectual property, admitting publicly that fakes are floating around out there is something that’s rarely done. I learned this firsthand when I contacted the eight beauty brands whose products my testing encompassed. None would grant me an interview (including L’Oreal, which shared those serum-testing results through a spokesperson).

“The bottom line is, [companies] don’t want their brand affiliated with counterfeit goods,” Barchiesi, of the International AntiCounterfeiting Coalition, said. “They don’t want to scare the consumers.” Harley Lewin, an intellectual-property lawyer renowned for his raids on fake-handbag operations, similarly told me, “It’s very easy to get drawn into a debate that you can’t win. That debate being that all of your stuff out there on the market, legitimately or not … is fake.”

In the end, if there really was nothing wrong with that third-party SkinCeuticals serum, all I could think was “How ironic.” Because if you’re a regular user of the North American version of the serum, and you decide to roll the dice on a third-party purchase to save (in my case) about $60, you might look at what you get in the mail and, if it’s like my sample, you may quickly surmise that it’s fake, expired, or otherwise suspicious — when, apparently, you actually won the third-party-seller lottery and found a true bargain, a legit product that you could use with confidence.

But you’d probably just throw it away.

When I reached out for comment on my findings, Amazon, eBay, Shein, and Walmart all expressed a zero-tolerance policy with regard to counterfeit products.

An Amazon spokesperson called out “proactive measures” that are meant to prevent counterfeit products from being listed.

An eBay spokesperson said that it uses “seller compliance audits, block filter algorithms, and AI-supported monitoring by our team of in-house specialists” to proactively remove potentially counterfeit and prohibited items.

A Shein spokesperson cited its Anti-Counterfeiting and Intellectual Property Infringement Policy, which prohibits the sale of counterfeit products.

And Walmart said it’s “investing in tools and technology to help ensure only trusted, legitimate items are available on our site.”

But none of that seems to be enough. Even if these platforms are trying to stymie bad actors, it’s still distressingly easy to buy fake products inadvertently, and even smart and savvy consumers regularly fall prey.

In a 2023 survey from Michigan State University’s Center for Anti-Counterfeiting and Product Protection, 68% of participants said they “were deceived into buying counterfeits at least once in the past year.” Overwhelmingly, those deceptions occurred online, as social media and e-commerce platforms each accounted for 39% of counterfeit purchases.

The survey, which polled 13,053 people across 17 countries, also found that more-frequent online shoppers were 6% likelier to unknowingly “engage in deceptive counterfeit purchase,” and for each level of education a person had completed, that person was more likely to unknowingly buy fakes by another 7%. In addition, male participants were 39% likelier to have unknowingly bought counterfeit goods than female participants.

Some of those statistics are explainable through correlation. If you shop online more frequently than most people, and online is how most counterfeits are sold — in the 2024 fiscal year, 97% of U.S. Customs and Border Protection’s intellectual-property rights seizures occurred on small-package, “de minimis” deliveries rather than large-scale, retail cargo shipments — you’re just shopping more often in a space where the risk is inherently higher.

There’s a similar explanation for the education-related statistic, which is that “those who are more educated [tend to] have more access to technology,” Saleem Alhabash, professor of advertising and public relations at Michigan State and associate director of research at MSU’s Center for Anti-Counterfeiting and Product Protection, told me.

But perhaps some inferences into the shopping behaviors of different demographics can be made as well. “The consumer research that I teach in my consumer behavior class shows that heteronormative men generally shop differently than women,” Alhabash said. “They’re much more task-oriented. They don’t do a lot of scouting around of different platforms.… Generally, men could be more comfortable with taking risks, with pressing the button and buying a product even though it is counterfeit.”

The survey finding that Alhabash is most boggled by, however, is that “perceived threat susceptibility,” namely a person’s awareness of the risks of buying counterfeit, “increased deceptive counterfeit purchases by 32%.” In other words, people who acknowledged being aware of counterfeit risks were more likely to buy counterfeit products than others.

Alhabash called this figure “incredibly alarming” and attributed it to the sense of fatalism shoppers sometimes feel when wading into the quagmire of e-commerce. “There are certain instances where we sometimes feel, ‘Oh, you know, how’s one product going to make a difference? I am too small of a unit to fight this big problem,’” he said. “Because consumers feel that way, they have this ingrained assessment that, yeah, when they are online, there’s going to be counterfeit products. They feel a sense of fatalism, and that drives their justification or even engagement in risky behavior.”

I agree. Working on this article taught me that it can be much trickier than you may think to guarantee that you aren’t adding a counterfeit to your cart. But as Alhabash noted, “It should not prevent you from questioning, from stopping and thinking, and from looking at red flags and cues.”

Here are several strategies I recommend to help you stay on the safe side when shopping online.

Before you click the buy button:

• Shop directly from brands and first-party retailers as much as you can. If you’re shopping on a brand’s own direct-to-consumer site, I believe you can consider yourself safe from counterfeits, as orders from such sites are processed and shipped by the brand itself. The URL of a brand’s direct-to-consumer site is typically an exact match for the brand’s name (as with drunkelephant.com and skinceuticals.com), but sometimes there’s a slight variation. (Rhode’s site, for example, is rhodeskin.com; Nars’s is narscosmetics.com.)

One important caveat: Beware of potential scam sites with sneaky typos in the URL. In 2025, I encountered a slew of scam sites for Jones Road cosmetics with URLs like jonesroedbeauty.com and jonesraodbeauty.us.com; they were convincing enough to fool at least a couple dozen makeup enthusiasts on Reddit.

Another safe bet is a first-party supplier such as Dermstore, Sephora, or Ulta. These sites sell only brands that they are directly contracted with to sell.

• On giant e-commerce platforms that host third-party sellers, confirm who’s actually doing the selling. Amazon, Walmart, and other big online retailers are sometimes authorized, first-party sellers of a particular brand’s products but sometimes not. (Or, confusingly, sometimes both.)

To make sure you’re getting a product through first-party channels, look on the product page for phrases like “Ships from,” “Sold by,” “Sold and shipped by,” or “Shipper/seller.” What follows should either be the name of the brand (CeraVe, Too Faced) and/or the name of the platform (Amazon, Walmart), either of which would indicate an authorized, first-party transaction.

On Amazon and Walmart, I recommend cross-checking that info by looking at the brand byline, which is located either directly above or below the name of the product at the top of the page. If the brand byline reads “Visit the CeraVe store,” for example, or appears as the brand’s logo, that typically indicates a first-party sale. If the brand byline simply reads something like “Brand: Rhode” or “Brand: SkinCeuticals,” take that as unofficial code for “unauthorized seller.”

• Dismiss any other words of assurance about the product or seller. A product page from a third-party seller may be littered with phrases like “Pro Seller” or “Amazon’s Choice,” or it may boast what looks like a high customer-review rating — all of which might lead you to believe that you’re in professional hands. Don’t put stock in these little participation trophies.

A Charlotte Tilbury Airbrush Flawless Setting Spray I bought from a third-party seller on Shein was “100% authentic,” as stated its product page, but according to our FTIR analysis, it wasn’t even 30% authentic. Meanwhile, the supposed Nars concealer, which was only a 2% match to our real Nars concealer, was sold by an Amazon third-party seller whose customer ratings are currently “99% positive in the past 12 months,” according to its seller page.

• Note if the price seems too good to be true. A “Fake Cosmetics and Their Health Risks” consumer alert that I found from the New York State Department of State notes that “prices that are discounted by a third or more are highly suspicious.” In other words, if an asking price seems too good to be true, it probably is. That was certainly the case for a few of my purchases, such as the Drunk Elephant glycolic serum I bought on eBay, which was listed for $20 even though its street price is $90.

But also, if a price seems like the norm, what you’re buying at that price could still be abnormal. My problematic Rhode lip gloss cost $18, only two bucks less than its $20 street price.

On at least two other occasions, I even saw a third-party seller list a brand-name product at a price above what the actual brand was asking (which seems especially shady). Even if the seller indicates that they’re running a promotion that brings the asking price below the street price, I’d stay away.

• Compare language between the seller’s listing and the brand’s website. One of the suspicious listings I encountered said that the product was “unflavored,” even though the brand referred to it as “unscented.” Another listed a mascara’s ingredients as “gel.” (That’s it. “Gel.”) Such shortcomings in the description don’t necessarily mean that you’ll receive a counterfeit, but if you’re hoping for some reassurance of professionalism, they certainly don’t help.
• Dig deeper for a seller’s one-star reviews. That’s where customers’ claims of counterfeits may be lurking. I never found these negative reviews front-and-center on a product page; I always had to click through once or twice to reach them. As some experts pointed out to me, these reviews could be bogus as well, perhaps posted by rival sellers looking to tank their competition.

Or, you may not find such customer feedback at all. I was told by two Wirecutter readers — both of whom purchased what they believed to be fake moisturizers on Amazon — that the company would not publish their reviews of the third-party sellers who sold them those products because their allegations of counterfeiting were, as one of those readers recalled, “derogatory.” (In response, Amazon told me that it doesn’t prohibit such reviews.) But again, you’re just trying to get an overall assessment of what you may be dealing with, and what you read in the one-star reviews might be enough to give you pause.

After you receive your order:

• Look up its lot code. Also known as a lot number, batch number, or batch code, this is a string of letters and numbers that a manufacturer prints on an item to identify it if necessary during a recall. These codes can also indicate when a product was manufactured. CheckFresh and Cosmetic Calculator allow anyone to verify a beauty item’s lot code, although each site’s code database lacks certain brand names. (On Cosmetic Calculator, I couldn’t verify CeraVe, Drunk Elephant, or Charlotte Tilbury lot codes, but they were all in CheckFresh’s database.)

You might find that an item is expired or at least past its recommended use-by date. You might also discover that your item’s lot code doesn’t exist, an obvious red flag that it may be fake.

Just remember that the opposite doesn’t necessarily apply. Most of our suspected counterfeits still had legitimate lot codes. Anyone can walk into Sephora and read a few letters and numbers off a box.

• Try to find a real specimen for comparison. If you’re a repeat user of a particular product, you may open your order and instantly sense that what you have on your hands is not what you’re used to. Newer customers have it much harder.

If you’re unfamiliar with the product you purchased, I suggest at least trying to determine if the color and scent seem off, since those are two qualities that are relatively easy to describe and compare. Look for Reddit threads or buyer reviews that corroborate what you see and smell. Or, if you’re really getting the ick, just return it or toss it.

I began working on this article with a question: How can you know if that popular beauty product you bought on the internet is actually the real thing?

Well, here’s my answer: If you can’t verify the source, knowing whether it’s real can be exceedingly tricky. Even when experienced chemists and high-tech lasers are at your disposal, reaching clear-cut conclusions is challenging, perhaps even impossible.

What compounds that problem is how people tend to shop online, in ways that unintentionally put themselves at further risk.

Perhaps someone who’s reluctant to try a miracle makeup product at full price filters their search results to surface the cheapest offers first, unwittingly setting themselves up for a bogus purchase with a price that’s too good to be true.

A busy parent hurriedly knocking out a medicine-cabinet restock on their phone while waiting for their kids at soccer practice can miss the visual cues on a product page that would tip them off to potential counterfeits.

Kids themselves, such as the tweens and teens chasing the newest viral beauty trends, may behave like the adolescents they are and ignore any and all words of caution while scrolling for a deal.

No matter the scenario, it takes diligence to shop smart online. Stay safe out there, my beauties.

This article was edited by Maxine Builder and Ben Frumin.