Every category I have covered has had what you might call ‘a golden age’. A period when buyers could truly judge quality. In every case, I published a story about how the factors we could once judge were somehow removed or concealed from consumers.

Supplements never had that golden age. No one has ever been able to judge the quality of a capsule by inspecting it. Even so, the category has grown even more challenging for consumers to navigate over time.

Our regulations are where this story starts. Since 1994, no supplement maker has been required to demonstrate to the FDA that its claims about its products are true, or even that they contain what the label claims, before selling them. US laws place the burden of proof entirely on the government itself. The statute reads: “the United States shall bear the burden of proof on each element to show that a dietary supplement is adulterated”.

To pull a product, the government must test it itself. That is not possible in today’s market.

In 1994, when Congress wrote the rule, the supplement industry sold roughly 4,000 products and did about $4 billion that year. The FDA now says there may be as many as 100,000 supplement products on the market. Americans spend nearly $73 billion a year, and three in five adults take a supplement in any given month.

The FDA is unable to provide firmer figures because it is hamstrung. Cara Welch, who heads the agency’s Office of Dietary Supplement Programs, explained its predicament plainly: “Under our current authorities, FDA has no systematic way of knowing what dietary supplements are on the market, when new products are introduced, or what they contain.”

The agency responsible for policing this entire industry has no mechanism for even listing the products within it.

This essay is about how that happened, and what you can do to ensure you are getting the supplements you paid for.

Walk into a CVS or Walgreens vitamin aisle and you will be confronted by a sea of gummies. Hundreds of jars covering dozens of independent health claims. Sleep gummies, energy gummies, stress gummies, focus gummies, and so on… Many of the brands you see today did not exist ten years ago. A large share of them are manufactured in the same buildings.

Catalent tells prospective clients it serves "21 of the top 25 self-care companies."

Its competitor, Vitaquest, boasts that it manufactures for "more than 500 brands around the world", launched "over 800 new products in the past year," and proudly lists where its products end up: Amazon, Costco, CVS, GNC, Sam's Club, Target, Vitamin Shoppe, Walmart, Walgreens and Whole Foods.

Robinson Pharma runs a plant producing 20 billion softgels and 7 billion gummies a year.

International Vitamin Corporation calls itself "the leading private label vitamin and supplement manufacturer in the US," and says it is "trusted by thousands of retailers and brands worldwide."

Not one of them will tell you which specific brands they manufacture for.

The prevalence of contract manufacturing practices only becomes visible to the public when something goes wrong.

In January 2020 the FDA published a recall covering every dietary supplement one company had made between January 2013 and November 2019. Seven years of production, pulled in its entirety, at once.

The company in question was ABH Nature’s Products out of Edgewood, New York. Inspectors found “significant violations of manufacturing practice”.

It happens on smaller scales too. In November 2021 a single manufacturer in Norcross, Georgia recalled 77 products in one go, sold under more than seventy different brand names across seventeen states. Alpha, Chemix, Muscle Rage, Wycked Naturals and dozens more you have never heard of. The labels had failed to declare that the products contained milk.

All of this raises the obvious question of how a market in goods so intimately tied to our health can allow so many new participants, with such little oversight.

The supplement game used to be difficult to get into. Making tablets and capsules at scale required something akin to pharmaceutical grade production processes. This created steep barriers to entry.

Making gummies, on the other hand, requires a manufacturing process akin to making Twizzlers.

Once the format transitioned, starting a supplement brand moved from being a manufacturing problem to a marketing one. You do not even need a formula to get started. The same Catalent About page offers brands "250+ Formulations available for Self-Care brands to get a head start."

The cost of launching a supplement brand fell to a logo, an Amazon listing and one production run. Triton Nutra Group in Chandler, Arizona, states that its "minimum order quantity for private label is 1,000 units per SKU". Nutricraft quotes minimum order sizes for gummies at 1,000 bottles and says a first run starts at $8,000.

That explains how a market goes from 4,000 products to over 100,000 without any new brand having to build anything. The plants still had to be built. Somebody else was buying them.

What we are left with is a market that has grown some 20x over the past 30 years, while production concentrated into fewer hands. Following the money reveals some important facts for consumers.

In 2017 a private equity firm called Highlander Partners started buying gummy plants. It bought four. Its president, Jeff L. Hull, described the plan: "In 2017, we started with a simple investment thesis of creating manufacturing capabilities around fortified/functional products in consumer-preferred formats." In August 2021 Highlander sold the assembled business to Catalent for $1 billion.

IVC bought Perrigo's entire US vitamins and supplements business in 2016, then bought GNC's factory in 2019 for $176 million.

And Vitaquest, the plant producing supplements for more than 500 brands, is wholly owned by CK Life Sciences, the Hong Kong company in Li Ka-shing's group.

More than four billion dollars in disclosed prices, most of it inside five years, and every dollar of it paid for plants and production lines.

So what was happening to the brand names over the same period?

Church & Dwight paid $650 million in 2012 for Vitafusion and L'il Critters, then the biggest gummy vitamin brands in America. It called the purchase "a new growth platform in one of the fastest-growing segments of the attractive vitamin/mineral/supplement category." In December 2025 it sold both. It did not say for how much, though it did say that the two brands had shrunk to under 5% of its expected sales for the year.

GNC sold its factory in March 2019 and filed for bankruptcy fifteen months later.

Bayer switched off the personalized vitamin service Care/of in July 2024.

The Vitamin Shoppe changed hands twice in six years, with a $2.6 billion buyout and a Chapter 11 in between.

This can’t be explained as a category going out of fashion. The plants were being bought at record prices over the same years. What lost its value was the brand itself.

A supplement brand is now cheap to start, cheap to abandon and cheap to replace. Names like that do not accumulate reputations worth protecting. The value in this market sits in the production plants instead. The plants do not sell to you. They do not appear on the label and do not compete for your trust. They are completely isolated from reputational damage.

The gummy format is the hardest to produce honestly. The people who make them say so in their own trade press.

Pierre Albert Thomas of Rousselot, which supplies the gelatin, says "the real complexity comes with the introduction of active ingredients," and that brands historically had to "reduce the active-ingredient content of gummies to mitigate interactions with the base, or switch to a more conventional—a.k.a. less-exciting—supplement format."

Doug Brown of Sirio Pharma describes a process running on "complex temperature, humidity, and pH conditions that require careful stewardship of ingredients."

Some direct testing has also been done.

Pieter Cohen's team bought 25 melatonin gummy products and found 22 of them inaccurately labeled, running anywhere from 74% to 347% of the dose printed on the jar. One contained no melatonin at all.

NOW Foods, which makes and sells supplements itself, bought 25 vitamin C gummy brands off Amazon and paid an accredited lab to test them. They published the results last week. Twelve had almost no detectable vitamin C. Five claimed 1,000mg or more per gummy and contained under 2mg.

When we buy supplements, we are often not getting the dose we paid for. Sometimes that is frustrating, but benign. In other cases, like the Norcross recall outlined above, the label is wrong in more dangerous ways: those seventy-seven products contained milk and did not say so.

That is not to say there are no supplement makers that hold brand value. Earlier this month, Procter & Gamble agreed to buy Thorne for $3.8 billion in cash. The private equity firm L Catterton had taken Thorne private in October 2023 for about $680 million. They made roughly five and a half times their money in thirty-four months.

What P&G is buying, according to Thorne's own announcement is a vertically integrated plant in South Carolina, in-house doctors and researchers, products certified under NSF Certified for Sport, and more than a hundred professional sports teams as customers. Paul Gama, who runs P&G's health care business, said Thorne "strengthens our position in premium wellness with a trusted, science-backed brand."

Look at what the two ends of this market have in common. The gummy brands that got written off were selling a name attached to a formula from a catalogue. The brand that appreciated is selling verifiability and reputation; the fact that somebody else checks its work. In a market where consumers can verify nothing for themselves, being the brand that employs an independent third-party to check its products turned out to be worth $3.8 billion.

Which makes what happens next incredibly telling.

P&G sells to everybody, through Target and Walmart and other such mass-outlets. Thorne sells premium, through practitioners, professional teams and direct to top-of-market consumers.

The commercially obvious move is to take the brand into the mass channel. NSF certifications screen every production lot rather than certifying a formula once, so the cost of the certification scales with volume. The most profitable way for P&G to sell Thorne at Walmart is to stop paying for the testing.

To be clear, that may not happen. P&G has done nothing untoward yet.

But just so we have them, here are the three things that would tell us the extraction has started:

I’ll be checking in periodically. I’ll let you know if any of these come to fruition.

A trustworthy supplement company needs to own the plants it manufactures in so that it has a reputation attached when something goes wrong. It must submit its production lots to a tester outside the company, whose verdict it does not control. And it should publish that testing publicly, even when results may be negative.

Almost no one clears all three, though Thorne comes close.

Start with the format. The people who make gummies say themselves that it is the hardest format to get actives into correctly, and both retail tests above found gummies failing badly. Tablets and capsules are made on pharmaceutical lines by firms that have done it for decades. If both are available to you, buy the tablet version over the gummy.

But that is no silver bullet. You should also look for a third-party test mark. NSF Certified for Sport is the strictest one available, because it screens every production lot rather than certifying a formula once. You can check a specific bottle against its database using the lot number printed on the label. Bear in mind though that the same directory carries Gatorade, Red Bull and 265 separate listings for Celsius energy drinks. All it tells you is that there are no banned substances present, and that the label matches what is actually in the product.

USP Verified is the most thorough formula-level program, and it pulls products off retail shelves once a year to re-test them.

That USP mark is rare though. USP's directory lists 159 verified products across sixteen brands, against 80,000 to 100,000 products on the American market. 99 of those are Nature Made, which is owned by the Japanese pharmaceutical company Otsuka. Looking for the USP seal narrows you to a short list of products from a small number of large companies. They are the ones who pay for the testing.

And be clear about what a seal does. The Pentagon, which has spent years trying to keep service members away from bad supplements, puts it well: "Third-party certification does not mean the product has been evaluated for safety or effectiveness. It means the product content and label match."

You should also know about a small number of organizations that buy supplements at retail and test them without being paid by the companies that made them. They are the closest thing this market has to an independent auditor.

ConsumerLab has tested more than 7,000 products from over 1,000 brands since 1999. It buys everything itself: "CL does not accept product samples from manufacturers or others for Product Reviews." It is funded by member subscriptions rather than advertising. In its March review of 22 multivitamins, seven failed.

The Pentagon runs Operation Supplement Safety. It is free and open to anyone. Its scorecard walks you through seven questions about a product and tells you to walk away if it scores under four. Granted, it’s not the most robust test, but it’s better than nothing.

Before blindly trusting any tester, check how they make their money. Some take a commission on the products they rank, or even charge a fee for accreditation. As we’ve established, a recommendation that pays the recommender is nothing more than an ad.

Given how quickly names change hands in this market, it’s not easy for me to give you an unqualified recommendation to buy from any particular company. What I can share are the small handful of companies that stood out for good reason during the course of this research.

Only three of the forty-eight supplement brands now in the Ledger got the Approved rating. They all have one thing in common: they maintain strict control over their manufacturing process.

Standard Process grows its own ingredients on a certified organic farm in Palmyra, Wisconsin, and makes the supplements 1.4 miles from the fields. Its organic certifier, not its marketing department, states that at least 75% of the raw plant material starts on that farm. It carries no third-party seal of any kind.

Nature Made owns four plants in California and Alabama, and holds nearly two thirds of the products in USP's verified directory. It belongs to Otsuka, a Japanese pharmaceutical company.

NOW Foods owns its plants in Illinois and Nevada and runs its own analytical labs inside them. In January it became roughly 30% employee-owned. It also buys competitors' products at retail and publishes the results, which is where the vitamin C testing above came from.

You will notice that nothing is rated Avoid in the Ledger for now. To rate a brand Avoid I need a documented act by a named company. This industry makes that standard nearly impossible to uphold. When the FDA shut down seven years of production in Edgewood, the recall named ABH Nature's Products; a company no consumer has ever bought from. I could not find a clean list of the brands affected anywhere.

Every brand named here is on The Brand Ledger. When one changes hands, or quietly drops a verification mark, that is where I'll record it.

The gummy was a manufacturing innovation. It made a supplement brand cost almost nothing to start, which explains the flood of new products to the aisles.

That deluge resulted in a market a consumer cannot navigate alone. The verification marks and the independent testers help, and for the products that carry them you can at least confirm that what is on the label is in the jar. For the other ninety-nine thousand, you are taking the brand's word for it.

Working out which compounds are worth ingesting in the first place is a different problem, and a story for another time.

Join the discussion, share product recommendations and suggest topics for me to investigate next in the Worse on Purpose community. Become a member to join.