A proprietary blend is a line on a Supplement Facts panel that gives one combined weight for a group of ingredients and no individual amount for any of them.
It is entirely legal, it is disclosed as what it is, and it removes the only calculation a buyer could have performed. All three of those are true at once, which is why the argument about it never resolves.
What is a proprietary blend on a supplement label?
It is a permitted labelling format. The rules allow a manufacturer to declare a blend by its total weight and to list the ingredients inside it in descending order by weight, without stating how much of each is present. So a panel can read Recovery Blend 1,200 mg followed by eleven names and nothing else.
Two things are disclosed by that line and they are worth naming precisely. The total, and the ranking. Everything else — every individual dose, every ratio, every comparison to another product — is withheld, and withheld lawfully.
Are proprietary blends legal?
Yes, and not by oversight. The Supplement Facts panel is governed by 21 CFR 101.36, Nutrition labeling of dietary supplements, and paragraph (c) of that section provides for a proprietary blend by name. The FDA’s own Dietary Supplement Labeling Guide answers the question how must I list proprietary blends directly, and cites 21 CFR 101.36(b)(2) and (c) beneath it. This is a written permission, not a gap somebody found.
Which is worth holding on to, because the word legal does a great deal of unearned work in most arguments about this. The rule does not say a manufacturer may withhold doses. It says a group of ingredients may be declared together as one blend, and then it sets out, line by line, what that declaration has to contain. The withholding is a consequence of the format rather than the object of it — and a statement about a label is not a statement about the contents. Two different questions, and only one of them is settled by the word legal.
What does a proprietary blend actually hide?
One thing, and only one: the weight of each individual ingredient inside it. Everything else about that line stays on the label — the blend’s name, the total weight of the group, every ingredient in the group by name, and the order they sit in by weight. A blend is not an undisclosed ingredient list. It is a disclosed ingredient list with the quantities taken out.
Four things the panel still has to print, and three of them are useful to you:
- The name. The term Proprietary Blend, or an appropriately descriptive term, or a fanciful name — which is to say, anything a marketing department can think of. This is the one that is not useful.
- The total weight of the ingredients in the blend, on the same line as that name.
- Every ingredient in it, listed underneath, in descending order of predominance by weight.
- A footnote against those ingredients reading Daily Value Not Established.
And the exception, which changes how you read the rest of the panel. An ingredient carrying a Reference Daily Intake or a Daily Reference Value — a vitamin, a mineral — has to be listed separately from the blend, with its own weight declared. A blend therefore cannot hide the magnesium, or the zinc, or the vitamin C. It can only hide the ingredients that have no Daily Value at all: the botanicals, the amino acids, the enzymes, the extracts. Which is, on most panels of this kind, the same list as the reasons printed on the front of the bottle.
Why do brands use proprietary blends?
The stated reason is recipe protection: a competitor who can read every dose can copy the formula. That reason is real and it is weaker than it sounds, because an analytical laboratory can reverse a formula from the finished product anyway. The practical effect is different from the stated one.
The practical effect is that under-dosing becomes undetectable. If eleven ingredients share 1,200 mg and the first two are cheap fillers, the last nine can be present in amounts too small to matter and the panel will look identical to one where they are not. The format does not prove that has happened. It makes it unprovable either way, which is the point of the criticism.
Can I work out the doses from the order they are listed in?
No. The order gives you bounds, not doses, and for most of the list the bounds are too wide to be worth anything. Descending order by weight, plus one total, is enough to put a ceiling over every ingredient in the blend and a floor under exactly one of them — the first. Everything named after the first has no floor at all, and that is the whole of the problem in a sentence.
The arithmetic is worth doing once, because it is the only arithmetic a blend leaves you. For a blend of n ingredients totalling T milligrams:
- The ingredient listed k-th can be no more than T ÷ k. The k ingredients at or above it in the order each weigh at least what it weighs, so if it were any heavier those k would exceed the total between them.
- The first-named ingredient is the only one with a lower bound: at least T ÷ n. If the largest were smaller than the average, nothing in the list could add up to the total.
- Nothing puts a floor under any of the others. The last name on the list can be present in whatever quantity was needed in order to print it.
Run that on the panel at the top of this page. Recovery Blend 1,200 mg, eleven ingredients. The first name is somewhere between 1,200 ÷ 11 — about 109 mg — and very nearly the whole 1,200 mg, an elevenfold range. The third name is capped at 400 mg and has no floor. The eleventh is capped at about 109 mg and has no floor either. Eleven names, one number, and not one of the eleven is a dose.
The trade has a word for what that permits — pixie-dusting, a token amount of an expensive ingredient carried mainly so that its name can appear on the panel. The honest thing to say is that a blend does not prove it has happened. It removes the means of showing that it has not, which is a different complaint and a fairer one.
What does a blend make impossible?
Every comparison worth making. Cost per gram of the active. Dose against the amount used in the research a brand is gesturing at. Elemental arithmetic on a mineral. Stacking totals across a cabinet. All of them need one individual figure, and a blend is the decision not to print one.
| The question | With a blend |
|---|---|
| How much of the active? | not computable |
| Cost per gram of the active? | not computable |
| Elemental figure for a mineral? | not computable |
| Daily total across four products? | not computable |
| Ranking of the ingredients | disclosed |
That table is the whole argument. A blend is not a lie and it is not evidence of one. It is the removal of arithmetic, and for anyone whose method is arithmetic that is the same thing as an answer.
What is a proprietary extract, and is it the same thing as a blend?
Not the same thing, and the difference is which fact has been removed. A blend is several ingredients sharing one weight, so what is missing is the quantities. A proprietary extract is a single ingredient with its own line and its own weight, so the quantity is printed — what is missing is how it was made, and how much of the compound anyone is buying it for is inside it. A blend withholds quantity. An extract withholds composition.
They fail different tests, which is why the two words should not be read as the same warning. An extract can be compared on weight against another extract and cannot be compared on strength, because two lines carrying the same weight may be nothing alike inside. A blend cannot be compared on either. And they stack: an extract named inside a blend is the case where neither number exists — its weight is swallowed by the blend total, and its composition was never printed to begin with.
Are proprietary blends always a red flag?
Not automatically. A flavour system, a gum blend or a preservative package is a genuine recipe and nobody needs its arithmetic. The flag goes up when the active sits inside the blend — when the reason a person is buying the product is the ingredient whose dose has been withheld. That is a specific test rather than a general suspicion.
Which produces one line worth keeping: if the ingredient on the front of the bottle is inside the blend on the back, the front of the bottle is not a claim anyone can check. Walk, or accept that the purchase is made on trust rather than on a number.
What does “clinically studied” mean on a label with a blend on it?
Usually that an ingredient was studied somewhere, at some dose, in some population — not that the product in your hand was studied, and not that it holds the dose that was. Those are three separate claims set in the same typeface, and the only one a buyer can check is the third, and only when the dose is on the panel. Put “clinically studied” and a proprietary blend on the same label and the claim becomes unfalsifiable by construction: the study named a number and the panel declines to.
Our own range has the same problem with the numbers left in, which is the only version of it a reader can do anything about. The trial the Under Load protocol is built on used 15 g of gelatin with vitamin C an hour before loading, in eight healthy men over three days, and measured serum PINP — a blood marker of collagen synthesis, not a tendon. The 15 g arm roughly doubled that marker against placebo and a 5 g arm landed in between. (Shaw G, Lee-Barthel A, Ross ML, Wang B, Baar K. Am J Clin Nutr 2017;105(1):136–143.) Eight men and a blood marker is a thin floor to build a protocol on, and it is the floor there is.
The Supplement Facts panel on Under Load declares a 10 g serving. The 15 g figure is a directions statement on this site — three scoops, 30 to 60 minutes before training — and it is never a serving declaration. Three numbers, all printed, and anyone can see for themselves that two of them are not the same. That is the entire benefit of printing a dose, and it runs one way only: it lets a reader find the gap. It does not close it.
Does TARE use a proprietary blend?
No. Under Load has one ingredient, and one ingredient has nothing to be blended with. Base Work is creatine monohydrate. Night Work is a magnesium glycinate capsule. There is no second active anywhere in the range to put inside a blend, so the answer costs us nothing to give — which is the correct way to weigh it. It is arithmetic about a short range, not evidence about our character.
What does TARE print?
None of the three TARE formulas contains a proprietary blend, and the reason is duller than a policy: not one of them contains a second active to blend. Under Load is collagen peptides. Base Work is creatine monohydrate. Night Work is magnesium glycinate. One ingredient each, one figure each, nothing on any of the three panels whose quantity is withheld.
That is not a boast about generosity, and it is a weaker demonstration than it used to be. The product this section pointed at until August 2026 was the one place in the range a blend could plausibly have hidden — a multi-ingredient capsule in the category where a blend is the default — and it was retired for reasons that had nothing to do with its panel. Declining to blend four ingredients is a decision. Having only one ingredient to declare is arithmetic. Both leave the reader able to run the procedure above; only the first was ever a test of us.
One limit on that, in the same breath, because the paragraph above is broader than our paperwork. What is settled is the active: there is one in each product and its quantity is printed. The full Other Ingredients line for the three panels — the flow agents, and the shell the Night Work capsule is made of — is not something this site publishes yet, because we do not hold it in writing from the manufacturer, and a capsule shell is an ingredient whether or not anyone calls it an active. So the claim that survives is the narrow one: no active in this range sits inside a blend. Not: we have published every word of three panels. We have not, and until we do, that gap is ours to name rather than yours to discover.
What the absence of a blend does not prove
Printing every dose says nothing about whether the doses are any good. A fully disclosed panel can be fully disclosed and under-dosed, and the reader can then see that it is — which is the whole benefit and also the whole limit. Disclosure enables the judgement; it does not make it.
And it says nothing about what is in the capsule. A printed figure is a claim about composition until a document measures it. All three of our certificates report a measured result against a specification. The magnesium document does so on a raw-material powder lot, in the laboratory's own serving, with no identity test, so it still cannot vouch for what is in the capsule. There is a second split on top of that one: the collagen was assayed by the company that makes it, on the lot our own tub was filled from, while the other two documents come from an independent laboratory but cover raw material rather than a finished container of ours. We name which is which at the top of every panel rather than letting one word cover three products. The full transcriptions are on Testing & Certification.
More in this cluster: the sixty-second label procedure, what a certificate of analysis reports, and the elemental arithmetic a blend makes impossible.