Pick up a pre-workout powder and read the Supplement Facts panel. Somewhere on it you may find a line that looks like a brand name, followed by a single number in milligrams, with a list of ingredients indented underneath. That is a proprietary blend. The number is the only amount printed for that whole group.
Now do the arithmetic. Take a blend declared at 200 mg with 18 named herbs inside it, which is an entirely ordinary shape for this kind of product. Split evenly, that is about 11 mg each. The ingredients do not have to be split evenly. But the label still tells you something hard: the last herb on the list cannot possibly be more than about 11 mg, because 17 other ingredients each weigh at least as much as it does and the whole group adds up to 200 mg.
Almost everything written about proprietary blends says the same thing. Federal law lets manufacturers hide individual amounts. That is the practical result, and it is fair to say so. It is not what the regulation says, and the gap between the two is where the useful information lives, because the actual text contains a boundary that most coverage never mentions.
There is no clause that exempts individual amounts
The regulation is 21 CFR 101.36, the rule that governs the Supplement Facts panel. Search it for a sentence saying individual amounts inside a blend need not be declared. There isn’t one.
What happens instead is subtler. The default rule for what the regulation calls “other dietary ingredients,” meaning the herbs, amino acids, botanical extracts and everything else for which the Food and Drug Administration has set no Daily Value, is that each one gets its own weight printed next to its name. Paragraph (c) of the same section, which covers blends, opens by saying that except as specified in that paragraph, all the other listing requirements still apply.
Then paragraph (c) specifies something else. Subparagraph (c)(2) requires only that the blend’s other dietary ingredients be named and listed in descending order of predominance by weight, indented under the blend’s name. Subparagraph (c)(3) requires one quantitative figure, the total weight of all those ingredients, placed on the same line to the right of the blend’s name, followed by a symbol pointing to a footnote reading “Daily Value not established.”
Put those together and the per-ingredient weight requirement is displaced, not repealed. The exemption is manufactured by what the rule demands in its place.
That is not a lawyer’s quibble. It explains why the disclosure you do get takes the specific form it takes, and why that form is more useful than people assume. If the regulation had simply said individual amounts are optional, a label could print a blend name and nothing else. Instead the rule buys the concealment at a price: a total, a complete ingredient list, a fixed order, and a footnote. Those four requirements are the reason the arithmetic later in this article works at all.
The permission itself is older than the regulation. It sits in the statute Congress passed in 1994, at 21 U.S.C. 343(q)(5)(F)(ii), which requires the listing of dietary ingredients to include the quantity of each such ingredient, and then adds an eight-word parenthesis extending that quantity requirement to a proprietary blend of such ingredients. The regulation implements that bracket. It did not invent it.
A blend cannot hide a vitamin
Here is the part almost nobody reports. Subparagraph (c)(1) limits the exception.
Dietary ingredients with an established Reference Daily Intake or Daily Reference Value, which in plain terms means the vitamins and minerals that carry a percent Daily Value on the panel, must still be declared individually, with their own weights, even when they sit inside a blend. FDA confirms the reading in its own labeling guide for industry. Chapter IV, question 34, answers how a blend must be listed and states plainly that any dietary ingredient carrying an RDI or a DRV has to be shown separately with its individual weight.
So the boundary runs like this. A proprietary blend can conceal how much ashwagandha, beta-alanine, bacopa, tribulus or green tea extract is in the bottle. It cannot conceal how much vitamin C, vitamin B6, zinc, magnesium or iron is in there.
That is worth knowing in both directions. If you are reading a multivitamin and the amounts are all printed, that is not a transparency decision by a generous brand. It is the law. And if you are reading a cognition or performance product where the whole sales pitch rests on herbs and amino acids, those are precisely the ingredients the rule allows to disappear into a single number.
It also means a common assumption is backwards. People often suspect a blend is concealing something dangerous. The rule makes that harder than it sounds, because every ingredient has to be named, and the nutrients with upper intake limits attached to them are the ones that still have to show their amounts. What a blend conceals is not usually identity. It is scale.
Why the rule exists, in FDA’s own words
Industry’s account of the exception is that Congress deliberately balanced trade secrets against consumer information. The Council for Responsible Nutrition, a supplement trade association, published a position paper in October 2021 making exactly that case. Its argument is that a blend’s quantities and ratios are a trade secret in the way a recipe is, that full disclosure would let competitors and counterfeiters copy a formula cheaply, and that consumers are not left in the dark because the identity of every ingredient still has to appear, only the individual amounts do not. That is the association speaking for its members, and it should be read as a stated position rather than a neutral finding.
FDA’s own reasoning, published when the rule was finalized in the Federal Register on 23 September 1997, is thinner than that. In the section on proprietary blends, responding to comment 21, the agency wrote: “Yet, by definition, the amounts of the ingredients in the blend are proprietary.” That is an assumption stated as a given, not an analysis. The phrase “trade secret” does not appear anywhere in that section.
The same response contains the sentence that defines the whole arrangement: “If a manufacturer wishes to voluntarily list the weights of ingredients, it is free to do so,” and FDA went on to say it was not requiring such a disclosure for other dietary ingredients in a blend.
The descending-order rule reads, in context, like a consolation prize. FDA had turned down a request to let a lesser ingredient be listed first, reasoning that doing so would create a false impression about the others, and that fixing that impression would require the weight of each ingredient. Having already decided not to require those weights, the agency concluded that ranking by weight was the only way left to avoid misleading anyone.
There is one more piece, and it complicates the industry defence. Trade associations point out that FDA can see the exact formula during a manufacturing inspection. A 2026 student Note in the University of Georgia’s Journal of Intellectual Property Law, by Anna Marie Saunders, makes the point that the formulations producers submit are generally treated as trade secrets and typically fall under Exemption 4 of the Freedom of Information Act, so the public cannot get at them. The Note is a law-student work rather than peer-reviewed research, and it concludes that the current framework has settled into a workable balance that tilts slightly toward innovation. Even on that reading, the oversight is real and permanently invisible.
What FDA enforces, when it enforces at all
Proprietary blends are not an unpoliced corner of the label. On 6 July 2017, FDA’s Denver district office sent a warning letter to a Utah supplement manufacturer citing the blend rules repeatedly. The letter is the clearest available picture of where the line sits.
Four separate failures were charged. On several products the required total weight for the blend was simply missing. One product declared two trademarked sub-blends inside a larger blend without naming what was in either of them, which FDA cited as a failure to list the name of each ingredient of the blends. That same product also had its other dietary ingredients out of descending weight order. And five products omitted the symbol in the percent Daily Value column that points to the “Daily Value not established” footnote.
Now look at what that list leaves out. FDA checked whether the total was printed, whether every ingredient inside the blend was named, whether the order was right, and whether the asterisk was there. A company can satisfy all four and still tell you nothing at all about dose. The agency polices the format. It does not police whether anything inside the blend is present in an amount that does anything.
One detail is genuinely useful: a blend inside a blend whose contents are not named is a violation. If a panel shows a trademarked complex with no ingredient list under it, that is not a stronger trade secret. It is a labeling problem.
The arithmetic the label hands you
Because the rule requires a total weight and a descending rank order, it hands you an upper bound on every ingredient except the first. This is the most useful thing in the subject, it needs no laboratory, and almost nobody uses it.
The rule in one line: for an ingredient in position k on the list, the maximum possible amount is the blend’s total weight divided by k.
Why that works: everything above position k weighs at least as much as position k, so those k ingredients together account for at least k times the weight of ingredient k, and they all have to fit inside the printed total.
Worked through on that 200 mg, 18-ingredient blend:
- The first ingredient could be anything up to 200 mg.
- The second cannot exceed 100 mg.
- The fifth cannot exceed 40 mg.
- The tenth cannot exceed 20 mg.
- The eighteenth cannot exceed about 11 mg.
A four-step check you can run in a shop:
- Find the blend’s total weight. It is required to be there, on the same line as the blend name.
- Count the ingredients indented underneath it.
- Locate the ingredient you bought the product for, and note its position in the list.
- Divide the total by that position. That is the ceiling.
Then compare the ceiling to the amount used in the research you have read about that ingredient. The researchers who documented the problem in pre-workout products put the same logic more bluntly: a product with an 8 g serving cannot contain both a 6 g dose of citrulline and a 4 g dose of beta-alanine, whatever the blend is called.
Measured, or calculated? The evidence is two different things
“Fairy dusting,” the practice of adding a token amount of an ingredient so it can be named on the label, is real and documented. But the evidence for it comes in two grades, and conflating them would be dishonest.
The first grade is measurement, and it only exists for products that declare an amount. Article 347 reports what happened when Pieter Cohen and colleagues ran that test on 57 sports supplements in 2023, and the short version is that 23 of them held none of the botanical their label named. A declared amount can turn out to be wrong. An undeclared amount cannot be checked at all.
The turkesterone rows are the clearest. One product labelled as containing an extract standardised to 10 percent turkesterone at 500 mg per serving measured 0.01 mg, which is 0.02 percent of the labelled quantity. Of the eight turkesterone products tested, four contained none at all. Thirteen products listing a Rauwolfia vomitoria extract contained no detectable Rauwolfia alkaloids. Halostachine products labelled at 40 mg and 100 mg measured between 1.3 percent and 3.0 percent of label.
The second grade is arithmetic, and that is what we have for blends. A 2019 study in Nutrients by Andrew Jagim, Patrick Harty and Clayton Camic transcribed the Supplement Facts panels of the 100 best-selling pre-workout products as of October 2018. The average product listed 18.4 ingredients, with 8.1 of them inside a proprietary blend at undisclosed quantities. Across all 100 products, 44.3 percent of all ingredients sat inside a blend. In the 58 products that used at least one blend, the blends held 14 ingredients on average, about 64 percent of everything listed.
What makes that study damning is the subset it could check. Where beta-alanine’s amount was printed, in 57 products, exactly one carried a dose above the threshold the authors identified for an effect. Where creatine’s amount was printed, only 29 percent reached the minimum studied quantity. The disclosed doses were mostly too small to matter. There is no reason to assume the concealed ones are better.
What does not exist is a laboratory study that takes a named proprietary blend apart and reports what each ingredient measured. As far as this research could establish, no such study has been published, and the reason is structural: there is no per-ingredient claim to measure the result against. Fairy dusting is proven analytically in products that declare amounts. In products that do not declare amounts, it is unverifiable by design. Any article that blurs those two is overstating the evidence.
One more thing no mark will solve for you. Third-party certification programmes verify that a product matches its own label. A blend declared as a 200 mg total, containing those ingredients in that order, is a conforming product. Certification checks the claim a manufacturer chose to make. It does not require the manufacturer to make a more informative one, and no US state disclosure law requiring individual amounts inside a blend was located. The companion article on third-party testing covers what those programmes do check.
What the opacity costs the research
The consequences are not only at the till. A 2018 review in the Journal of the International Society of Sports Nutrition by Patrick Harty and colleagues, including Jagim, notes that comparing one multi-ingredient product with another, or with a single ingredient tested alone, is difficult and often impossible because blends do not disclose specific amounts. That journal belongs to a society with substantial industry membership, which is worth naming, and it makes the criticism more notable rather than less.
Follow that through. If a trial finds a blended product works, nobody can say which ingredient did it or at what dose. If a trial finds it does nothing, the maker can argue the dose was wrong without ever saying what it was. A proprietary blend does not only protect a formula from competitors. It protects a formula from being tested.
When a blend becomes a clinical problem
Undisclosed amounts stop being an abstraction the moment someone needs to assess a risk.
If you take a prescription medicine, a clinician weighing a possible interaction with a supplement needs to know what is in it and how much. With a blend, the second half of that question has no answer on the label. The practical step is to bring the actual container to the appointment rather than a remembered product name, so the ingredient list and the blend total can be read directly. Pharmacists do this assessment routinely and it is a reasonable thing to ask for.
Get medical attention promptly, rather than waiting, if you develop chest pain, a racing or irregular heartbeat, fainting, severe agitation, yellowing of the skin or eyes, dark urine, or persistent vomiting after starting any multi-ingredient product. Keep the container. The ingredient list and lot number are the only record of what was taken, and in products built on blends they are frequently the only record anyone has.
The one thing worth doing
The division is the thing to carry away.
Find the blend total and the number of ingredients. Divide the first by the position of the ingredient you care about. That number is the most anything beneath the first line can be, and it comes from the two disclosures the law does require. In a great many products on the shelf, the answer lands far below any amount that has been studied, and you can establish that in the aisle without a laboratory, a certificate, or a call to the manufacturer.
Sources
- Cornell Legal Information Institute. “21 CFR 101.36 – Nutrition labeling of dietary supplements.” law.cornell.edu
- Cornell Legal Information Institute. “21 U.S.C. 343 – Misbranded food.” law.cornell.edu
- US Food and Drug Administration. “Dietary Supplement Labeling Guide: Chapter IV. Nutrition Labeling.” fda.gov
- US Government Publishing Office. “Food Labeling; Statement of Identity, Nutrition Labeling and Ingredient Labeling of Dietary Supplements, Final Rule, 62 FR 49826 (23 September 1997).” govinfo.gov
- US Food and Drug Administration. “Warning Letter: Professional Botanicals, Inc., 6 July 2017.” fda.gov
- Council for Responsible Nutrition. “Proprietary blends in dietary supplements protect trade secrets and foster innovation (October 2021).” crnusa.org
- University of Georgia School of Law. “Saunders AM. “Pills, Powders, and Proprietary Blends: The IP and Regulatory Battle Over Dietary Supplement Transparency.” Journal of Intellectual Property Law 33(1):291, 2026 (student Note).” digitalcommons.law.uga.edu
- Nutrients. “Jagim AR, Harty PS, Camic CL. “Common Ingredient Profiles of Multi-Ingredient Pre-Workout Supplements.” Nutrients, 2019. PubMed. DOI: 10.3390/nu11020254.” doi.org
- JAMA Network Open. “Cohen PA, Avula B, Katragunta K, Travis JC, Khan I. “Presence and Quantity of Botanical Ingredients With Purported Performance-Enhancing Properties in Sports Supplements.” JAMA Network Open, 2023. PubMed. DOI: 10.1001/jamanetworkopen.2023.23879.” doi.org
- Journal of the International Society of Sports Nutrition. “Harty PS, Zabriskie HA, Erickson JL, Molling PE, Kerksick CM, Jagim AR. “Multi-ingredient pre-workout supplements, safety implications, and performance outcomes: a brief review.” JISSN, 2018. PubMed. DOI: 10.1186/s12970-018-0247-6.” doi.org