The molecules your body already makes

Read the back of enough supplement labels and they start to look alike. A list of names, a number beside each one, and an implied promise that the bigger numbers are the better ones.

It is a reasonable instinct, and for a large share of what this category sells it is the wrong model.

Two kinds of ingredient
Almost everything in a supplement falls into one of two groups, and they behave differently.

The first is the things your body cannot make. Vitamins and most minerals. That is what essential means in nutrition: not important, but impossible to synthesise, so it has to arrive from outside. For these the amount matters directly, because whatever does not come in is simply not there.

The second is the things your body makes itself. Creatine. Glutathione. The fatty acid amides involved in sleep. The coenzymes that run your metabolism. You are producing all of them right now, and were doing so long before you had heard of a supplement.

Those two groups need different questions asked of them.

A production line, not an empty tank
For anything in the second group there is a pathway. Raw materials go in, enzymes act on them, the molecule comes out the other end. That pathway has a rate, and the rate depends on what is available to it.

Glutathione is the clearest example. It is the body's principal antioxidant, assembled inside the cell from three amino acids. How much a cell can make comes down largely to two things: how much cysteine is available, and the activity of the rate-limiting enzyme that joins the first two pieces together [1].

Creatine works the same way, as an earlier piece in this series set out. Glycine and arginine combine to form guanidinoacetic acid, which is then methylated into creatine.

So the useful question for these molecules is not how much can be forced in. It is whether the line is running short of anything.

Why flooding is the wrong model
There is a second reason the bigger-number instinct fails here, and it is the more interesting one. Your body regulates what it makes, because a system that produces something needs a way to stop producing it.

Creatine shows this directly. In a twelve-week double-blind, placebo-controlled trial of 458 adults, taking 3g of creatine a day lowered circulating guanidinoacetate by around 11%, while it drifted slightly upward in the placebo group [2]. Guanidinoacetate is the body's own intermediate on the way to making creatine. Supply the finished molecule from outside and the production line slows down in response.

That is not an argument against creatine. It is a demonstration that these molecules sit inside a regulated system rather than a tank you fill. Melatonin shows the same principle through a different mechanism, where the problem is not reduced production but receptors that stop responding to a signal arriving far larger than any the body sends [3].

Two mechanisms, one conclusion. With a molecule the body already produces and already regulates, more is not reliably more.

What that means for how we formulate
It is why precursors keep appearing across the range rather than the finished molecules.

The NAD+ patch delivers NMN rather than NAD+ itself. There is a size reason too, since NAD+ is too large to cross skin readily under the 500 dalton rule [4], but the pathway reason stands on its own. The NAC in the same patch supplies cysteine, the input glutathione production depends on most.

The creatine patch delivers guanidinoacetic acid alongside glycine, arginine and the B vitamins that keep the methylation step supplied. That is the same intermediate the trial above found suppressed by creatine itself, which is the point: feeding a pathway and feeding it the end product are not the same intervention. In one randomised crossover trial, four weeks of GAA raised creatine levels in muscle and brain tissue more than an equivalent amount of creatine did [5]. Collagen Pro delivers amino acids and vitamin C rather than collagen, for the reason set out in that piece.

The signalling molecules follow the same thinking. Sleep Deep contains palmitoylethanolamide, a fatty acid amide the body produces itself [6], and oleamide, first identified accumulating in the cerebrospinal fluid of sleep-deprived mammals [7]. Neither is a foreign compound being introduced to the system. Both are molecules the system already uses.

Where the principle ends
Every useful idea has an edge, and this one has two worth naming.

The first is that precursor is not a magic word. A pathway can be well supplied with raw material and still be paced by enzyme activity rather than by what you feed it, and the language of supporting production has become a convenient way for products to avoid thinking about that. Used properly it is a design rule, and a strict one: it tells you what belongs in a formulation and what the formulation has to carry alongside it, which is why the creatine patch carries methylation cofactors rather than guanidinoacetic acid on its own.

The second is that it does not transfer to the first group. For vitamins and minerals the quantity is the point, because the body makes none of it, and no amount of pathway reasoning substitutes for the right amount arriving.

What the distinction gives you is the right question to ask of any ingredient. For something your body cannot make, ask how much. For something it makes already, ask what the production line needs.

Most of the category never asks either one.