Reading a proprietary blend
A proprietary blend is a lawful way of printing one number where a reader wanted several. It is not a trick and it is not a secret — but it does leave exactly one calculation available to you, and that calculation is the whole of this guide.
A proprietary blend lets a manufacturer declare a single total weight for a group of ingredients and list their names underneath without saying how much of each is present. It is permitted by 21 CFR 101.36(c), it requires the group to be named and the ingredients to be listed in descending order of weight, and it is entirely ordinary.
What it leaves you is the ceiling: the blend total divided by an ingredient’s position in the list is the most that ingredient can weigh. That is an upper bound rather than an amount, and it is enough to rule some claims out even though it can never rule one in.
What the regulation actually permits
In the United States a dietary supplement label has to carry a Supplement Facts panel, and in the ordinary case each dietary ingredient on that panel carries its own quantitative amount per serving. 21 CFR 101.36 is the section that sets this out, and it contains the provision that makes blends possible.
Under that provision a manufacturer may group dietary ingredients that are not otherwise required to be separately declared, give the group a name, declare the total weight of the group, and list the ingredients underneath it without individual amounts. The name is why you see phrases like “energy matrix”, “performance complex” and “proprietary blend” printed on panels at all.
The three things it still requires
- A name for the blend. A group of ingredients cannot be declared anonymously; it has to be identified as a blend or by a proprietary name.
- A total weight. The group carries one quantitative figure per serving, and that figure has to be real. The individual split may be withheld; the total may not.
- Descending order of predominance by weight. The names underneath the total are listed heaviest first. This is the requirement the rest of this guide is built on.
The three things it does not require
- Any individual amount for any ingredient inside the blend.
- Any standardization figure — the percentage of a marker compound, which is what makes an extract comparable to the material used in a trial.
- Any statement of how close two adjacent ingredients are. Position three might weigh four times position four, or a milligram more.
Vitamins and minerals with an established Daily Value are treated differently and generally have to be declared with their own amounts. That is why you sometimes see a panel with a tidy list of vitamins at the top and a single blend line underneath: the part that had to be disclosed was, and the part that did not was not.
Why blends exist
There is a stated reason and there is an effect, and they are not the same thing. Both are worth holding in mind, and only one of them can be checked from outside.
The stated reason: protecting a formula
The argument manufacturers give is trade secret protection. A formulator who spent money arriving at a particular ratio does not want a competitor to read it off the back of the bottle and produce the same product without the development cost. Recipes are not patentable in any useful way, so the blend is the practical alternative to publishing one.
This argument is real. It is also weaker than it sounds, because a competitor with a laboratory and a bottle can get much of the way to a formula by analysis, and because the argument protects the ratio rather than the ingredient list — which is printed anyway, in order, on the panel.
The effect: underdosing becomes invisible
Whatever the intention, the consequence is the same. A blend makes a generous formula and a token one look identical on the shelf. If the marketing names six studied botanicals and the panel prints one total, nothing on the label distinguishes a product with a trial dose of each from a product carrying a dusting of five and a cheap bulk filler as the sixth.
The word for the practice, and why we use it carefully
Fairy dusting is the industry term for including an ingredient in an amount far below anything studied, so that it can be named on the front of the bottle. The word is useful because the practice is real and common.
It is also an accusation, and a blend by itself is not evidence of it. What a blend does is remove the information that would settle the question. A reader is entitled to notice that, and not entitled to conclude the worst from it.
The one calculation a blend leaves you
This is the useful part, and it follows from the descending-order requirement alone. Nothing else about the blend needs to be known.
Suppose a blend declares a total of T milligrams across n ingredients, listed heaviest first. Take the ingredient in position k. Because the list is in descending order, every ingredient above it weighs at least as much as it does, so positions 1 through k together weigh at least k times its weight. Those k positions cannot exceed the total. So k times its weight is at most T, and its weight is at most T divided by k.
| Position | Ceiling | As a share of the total | What it means |
|---|---|---|---|
| 1 | 1,200.0 mg | 100% | Could be the entire blend on its own |
| 2 | 600.0 mg | 50% | At most half |
| 3 | 400.0 mg | 33.3% | At most a third |
| 4 | 300.0 mg | 25% | At most a quarter |
| 5 | 240.0 mg | 20% | At most a fifth |
| 6 | 200.0 mg | 16.7% | At most a sixth |
| 7 | 171.4 mg | 14.3% | At most a seventh |
| 8 | 150.0 mg | 12.5% | At most an eighth |
Scroll this table sideways →
Two features of that table are worth noticing. The ceiling falls away fast at the top and then barely moves: positions six, seven and eight are within fifty milligrams of each other. And the average, if the blend were split evenly, is 150 mg — which is also the ceiling of the last position, because those two figures are the same calculation.
The arithmetic does not care what the ingredients are, how many there are, or what the total is. A 500 mg blend of four things gives ceilings of 500, 250, 166.7 and 125 mg. The method is the same on every label you will ever pick up.
What the ceiling tells you, and what it does not
This is the section most guides leave out, and it is the one that keeps the method honest. A ceiling is a one-directional instrument, and using it in the other direction is the commonest mistake made with it.
What this label does well
- It can rule a dose out. If an ingredient’s ceiling is below the lowest dose ever studied, no split of that blend reaches a studied dose. This is a proof, not a suspicion.
- It works on interactions between positions. Two ingredients with generous ceilings can still be mutually exclusive once you check what each one forces on the positions around it.
- It needs nothing from the manufacturer. Total, position, division. Everything required is already printed on the panel.
- It is exact. The ceiling is not a guess at a likely amount; it is a limit that the regulation guarantees.
What it does not
- It can never rule a dose in. A ceiling above a studied dose means the position permits that dose, not that the dose is there.
- It says nothing about the actual amount. Position two of a 1,200 mg blend has a 600 mg ceiling and could be 5 mg. Nothing on the label would change.
- It is defeated by standardization. A milligram of a 10% extract and a milligram of a 1% extract are the same weight and not the same dose. Blends rarely state which they are.
- It assumes the total is the whole story. Excipients and carriers inside an extract count toward its weight, so a 300 mg extract is not 300 mg of the plant.
The honest summary is that the ceiling is a filter rather than a measurement. It removes claims that cannot be true and leaves everything else exactly as uncertain as it was. That is a modest result, and it is the only rigorous one available from a blend.
The three questions to ask of any blend
Asked in this order, these three take a minute and settle most of what a blend can be made to give up.
Is the total large enough for the number of names under it?
Divide the total by the number of ingredients. If a blend of twelve botanicals declares 400 mg, the average is 33 mg, and almost nothing in the botanical literature has been studied at 33 mg a day. The list is longer than the total can support.
Which named ingredient has the best-known dose, and where does it sit?
Find the ingredient you can look up a trial dose for, and check its ceiling against that dose. If the ceiling is below the dose, the blend cannot deliver it. If it is above, the blend is allowed to deliver it and you still do not know whether it does.
What does the last position have to be?
The last ingredient has the smallest ceiling and is the cheapest place to put something included for the name alone. Work out its ceiling, then ask whether anything is known to do anything at that weight.
A fourth question is worth asking of the seller rather than the label: is there a certificate of analysis for the lot, with per-ingredient amounts on it? A manufacturer is not obliged to provide one for a blend, and a seller who does has answered everything above at once.
One thing not to conclude
A blend is not evidence of bad faith. It is a lawful labeling choice used by cautious formulators and careless ones alike, and treating it as proof of underdosing is the mirror image of treating the marketing as proof of adequacy.
What a blend is, reliably, is a reduction in what you can check. The reasonable response is to check the things it leaves intact, which is what the three questions above do.
A worked example: the SodaPeak panel
The label this site is about declares one blend, Test Boost Proprietary Blend, at 1,200 mg per 2-capsule serving, shared by eight ingredients with no individual amount for any of them. Every step above applies to it directly.
Question one: is the total large enough? 1,200 mg divided by eight is 150 mg on an even split. For a botanical blend that is a reasonable size rather than a token one, and it is larger than most blends of this kind. The total passes the first question, which is what makes the rest interesting.
Question two: the best-known dose, and its position. Tongkat ali sits second, so its ceiling is 600 mg, and the lowest dose in its published trials is 100 mg a day for twelve weeks. The position allows it six times over. That is a pass, and it is still not a statement that 100 mg is present.
Run the same check on nettle root and the answer reverses. It sits fourth, so its ceiling is 1,200 mg divided by four, which is 300 mg. The lowest dose in its trials is 360 mg a day. No split of this blend, however favorable, reaches it. That is the ceiling ruling a dose out, which is the one thing it can do conclusively.
Question three: what the last position has to be. Boron amino acid chelate sits eighth, ceiling 150 mg. The human studies used about 60 mg of a 5% chelate to give 3 mg of elemental boron, so the position allows that two and a half times over. The label states no elemental boron figure, so the conversion is ours and it is an assumption about a trade grade rather than a fact about this product.
And then the interaction the ceilings reveal
Saw palmetto sits third and its studied dose is 320 mg. Descending order puts positions one and two at 320 mg or more each, so the top three account for at least 960 mg and leave at most 240 mg for the last five.
Boron sits eighth and needs about 60 mg. Descending order puts positions four to seven at 60 mg or more each as well, so those five positions need at least 300 mg. Three hundred is more than two hundred and forty, so saw palmetto and boron cannot both be at a studied dose. The full derivation is on the proof page.
None of that required anything the manufacturer did not print. It is one total, eight positions and a division, and it is available to anyone holding the bottle.
References
The sources behind the numbers on this page.
- U.S. Government Publishing Office21 CFR 101.36 — Nutrition labeling of dietary supplements, including the proprietary blend and descending-order requirements — Code of Federal Regulations · source
- Chinnappan, Rahmat, Mohamed et al. — 2021Efficacy of Eurycoma longifolia (Physta®) on testosterone and quality of life in men aged 50 to 70 — Food & Nutrition Research · DOI 10.29219/fnr.v65.5647 · source
- Leisegang, Finelli, Sikka, Panner Selvam — 2022Eurycoma longifolia and male infertility and hypogonadism: a systematic review and meta-analysis — Medicina (Kaunas), volume 58 · DOI 10.3390/medicina58081047 · PMID 36013514 · source
- Safarinejad — 2005Urtica dioica for treatment of benign prostatic hyperplasia: a prospective, randomized, double-blind, placebo-controlled, crossover study — Journal of Herbal Pharmacotherapy · DOI 10.1080/J157v05n04_01 · PMID 16635963 · source
- Bent, Kane, Shinohara et al. — 2006Saw palmetto for benign prostatic hyperplasia — New England Journal of Medicine · DOI 10.1056/NEJMoa053085 · PMID 16467543 · source
- Nielsen, Hunt, Mullen, Hunt — 1987Effect of dietary boron on mineral, estrogen, and testosterone metabolism in postmenopausal women — The FASEB Journal, volume 1 · DOI 10.1096/fasebj.1.5.3678698 · PMID 3678698 · source
- NIH Office of Dietary SupplementsBoron — Health Professional Fact Sheet — ods.od.nih.gov · source
See the SodaPeak packs
One, three or six bottles. The price is on the seller’s checkout page, because it is the one figure this site was never given — and a figure we were not given is not a figure we will print.
Shipping and the guarantee term are shown at checkout. This site publishes no number it was not able to verify.