Reference Guides From This Desk

A proprietary blend is a heading, a single weight and a list of names underneath. The labelling rule allows it, so the figures are lawful, and it is a common shape in this category: three of the four groups on the tub this desk sells are built that way.

The total is a real number and it earns its place. Added across the groups, it tells you how much of a serving is dietary ingredient rather than flavouring, and it sets a hard upper bound on every name inside. What it withholds is the split — and the split is what decides whether any one of those names is present in the sort of amount a trial used.

The first guide works that through: the ceiling, the floor, what the order of the names buys you, and the five things a label prints when it has the split to print. Read it here.

A capsule is filled by a machine, sealed and counted. A scoop is filled by the person holding it, from a powder that settles in transit and packs differently on the day the tub is new — a difference that reaches further than most buyers expect.

A powder is tasted, so it carries sweetener, flavour and an acid inside the same serving the actives are in. It also holds far more than a capsule, which is why grams appear on a scoop label where milligrams appear on a capsule one. And a live-culture count on a powder is almost always quoted at the time of manufacture, which is a different question from what is alive when you open it.

The second guide takes those in turn and ends with a checklist and a score. Read it here.

1. The form and the serving. A scoop, a capsule, a gummy and a dropper are not comparable by weight, and a serving of two capsules is not the same purchase as a serving of one.
2. Whether an amount is printed per name. Where a blend heading carries the only weight, every comparison you can make is a ceiling.
3. What the claims actually say. A supplement may make structure and function claims and no disease claims, and the agency’s own example of a permitted one is that fibre maintains bowel regularity.
4. How the live cultures are described. NCCIH reports a review of 53 studies that could not identify which species or strains are most likely to help, so a count on its own settles very little.
5. The window, and who pays return postage. A long window costs nothing to print — the postage clause is where its real shape shows.

Somebody holding a tub in one hand and a phone in the other, trying to work out whether the panel on the back supports the sentence on the front. That reader needs two things: the vocabulary, and the arithmetic that is honestly available.

They are also written for somebody comparing two products in the same aisle. Most of what separates a good label from a poor one there is not the ingredient list, which tends to converge, but how much of the formula the label is willing to show — and that is checkable in a minute once you know where to look.

Neither guide needs you to have bought anything. Both end on the product this desk sells, because a checklist nobody applies to their own shelf is not worth writing.

Three places and nowhere else: the printed label and its Supplement Facts panel, the seller’s own pages with the date they were read, and public sources that were opened and checked for the sentence attributed to them.

Sources are preferred in one order. Systematic reviews and meta-analyses first, then randomised controlled trials, then government and institutional fact sheets such as MedlinePlus on dietary fibre or the NIH selenium fact sheet, then mechanism reviews. Where an ingredient has no modern trial record, that is what the page says.

Every citation carries the amount the study actually used — the dose is the part of a trial that decides whether it applies to the tub in front of you.