What a Long List of Plant Extracts Actually Buys You

The rule that makes a long list possible

Cosmetic ingredient declaration in the United States follows an ordering rule that almost nobody outside formulation has read, and it explains the entire phenomenon.

Under 21 CFR 701.3, ingredients other than colour additives present at more than one percent must be listed in descending order of predominance. Ingredients at or below one percent may be listed without respect to order of predominance. Colour additives come last, also unordered.

Read that twice. Above one percent, the sequence is information: each ingredient is present in at least the quantity of the one after it. At or below one percent, the sequence is a choice. A brand can arrange those ingredients however it likes — alphabetically, by marketing appeal, by whichever name it wants a shopper to notice — and it is fully compliant.

So a panel listing eighteen botanicals in a satisfying run is not eighteen ingredients in descending strength. It is, in the overwhelming majority of cases, eighteen ingredients somewhere at or below one percent, arranged for effect. The list is doing exactly what it is permitted to do. It is the reader who assumed it meant something else.

And the assumption is not stupid — it is a reasonable inference from how the top of the list behaves. The rule simply changes partway down, silently, with no marker printed on the label to say where.

Finding the line yourself

There is no dividing line on the panel, but there are reliable landmarks, because a handful of ingredient classes are always used at low percentages. Spot one and you know you have crossed over.

Preservatives are the clearest. Phenoxyethanol is typically used around half a percent to one percent and is capped at one percent in the EU regardless. Ethylhexylglycerin, which usually accompanies it, sits lower still. Sodium benzoate and potassium sorbate are similar. Chelators are lower again — disodium EDTA is generally used around a tenth of a percent. Thickeners and stabilisers such as xanthan gum and carbomer typically run a few tenths of a percent. So do pH adjusters like triethanolamine or sodium hydroxide, and preservative-boosting glycols at the tail end.

When you hit any of those, stop treating the remaining order as meaningful. Everything below is at or under one percent, and the sequence is presentational.

Here is where honesty requires a caveat that most label-literacy content skips: being below the line does not mean being inert. Plenty of genuinely potent materials are used at a fraction of a percent because that is their effective range — retinol, most peptides, tranexamic acid at low doses, and several standardised botanical fractions. Position tells you about quantity, not about potency, and quantity only becomes damning when the ingredient in question needs a substantial dose to work.

Which is exactly the case for most crude plant extracts. Our ingredient decoder will sort a panel into actives, vehicle and functional ingredients so you can see where the line falls without memorising the landmarks.

'Extract' is a description, not a dose

The second gap is bigger than the first and almost nothing on a consumer label closes it.

An INCI name such as Camellia Sinensis Leaf Extract tells you the species and the plant part. It does not tell you the solvent — water, glycerin, ethanol, butylene glycol and supercritical CO2 pull entirely different fractions out of the same leaf. It does not tell you the extraction ratio, which is how much raw plant produced how much extract. It does not tell you the concentration of the constituent that the interesting research was done on. And it does not tell you whether the material was standardised — held to a specified content of a marker compound batch after batch — or simply made and shipped.

Two products can name the same extract on the same line of the panel and differ by an order of magnitude or more in the compound that actually does something. Analytical chemists have documented how difficult it is to even measure phenolic marker compounds reliably in botanical extracts and finished products, because hydrolysis and redox conditions during analysis change what you find. If quantifying it in a laboratory is that fiddly, a shopper reading a plant name has no chance.

There is a third dilution most people never consider. Botanical extracts are usually supplied to formulators already dissolved in a carrier — commonly water, glycerin or butylene glycol — at some fraction of extract solids. When a formulator adds one percent of that supplied material, the actual extract content is a fraction of that one percent, and the plant solids a fraction again. The INCI name looks identical either way.

None of this is deception. It is the honest consequence of a labelling system built to declare identity rather than dose. But it means a botanical name on a panel is roughly as informative about strength as the word 'sauce' is about a recipe. ↗The botanical extracts overview sets out what the category can and cannot deliver.

The extracts that genuinely are the active

Now the concession, and it is a substantial one, because the cynical reading of botanical labelling is as wrong as the credulous one.

Centella asiatica is the clearest case. Its activity sits in a group of triterpenes — asiaticoside, madecassoside, asiatic acid and madecassic acid — and a body of work on topical centella in wound healing has examined mechanisms including collagen synthesis, angiogenesis and antioxidant activity. The pharmaceutical-grade material is a titrated extract standardised to a defined triterpene content, which is why you sometimes see it named as such rather than as a plant. When ↗SKIN1004 Centella Teca Ampoule puts that designation in the product name, it is telling you which grade of material is in the bottle — a claim about standardisation, not a poetic gesture. ↗Centella's evidence page carries the grading.

Green tea is the second. Its polyphenols, principally epigallocatechin gallate, have a photoprotection literature going back to work published in the Journal of the American Academy of Dermatology on cutaneous protection from ultraviolet injury. That is a real mechanism at a real dose, and a formula built around it — as ↗Innisfree Green Tea Seed Hyaluronic Serum is — is doing something different from one that lists green tea nineteenth. ↗The green tea entry is worth reading for where the topical evidence is strong and where it thins.

Licorice is the third. Glabridin, isolated from licorice root extract, was shown in pigment cell research to inhibit melanogenesis and inflammation, and it remains a legitimate component of pigment-directed formulas such as ↗Bioderma Pigmentbio C-Concentrate. ↗Licorice root sets out the constituent-level picture, which is far more useful than the plant-level one.

The pattern across all three is identical and it is the practical lesson of this article. When an extract is genuinely the active, the brand talks about the constituent — the triterpene, the polyphenol, the glabridin — rather than about the plant. When it is decoration, the plant gets a lovely name and eighteen companions.

Reading a botanical label without cynicism or credulity

Four questions settle almost every case, and none of them requires chemistry.

Where is it? Above the preservative and the gums, or below? Above the line, the extract is present in quantity and its position relative to the other ingredients is real information. Below, you are looking at a fraction of a percent arranged for presentation.

How is it named? A standardised material is usually announced, because standardised material costs more and brands that pay for it say so — a named marker compound, a stated percentage, a titrated designation. A plant name alone is a plant name alone.

What does the brand talk about? Marketing copy is unreliable in general, but it is a good tell here specifically. A formula built on centella triterpenes will discuss triterpenes. A formula using centella as garnish will discuss Madagascar.

And how many are there? This is the blunt one. A formula naming twenty botanicals has, at minimum, divided a small budget twenty ways. Something like ↗Purito Centella Green Level Buffet Serum is explicit about being a blend rather than a single-active product, which is a legitimate design — but a blend is a different proposition from a formula built around one well-dosed active, and it should be priced and judged as one.

There is also a cost to a long plant list that rarely gets mentioned. Every additional botanical is an additional set of molecules your immune system has not met, and botanicals are heavily represented among contact allergens — the terpenes that oxidise out of essential oils and plant fractions are among the most common positives in modern patch testing. A twenty-extract formula that produces a reaction is a twenty-variable problem to unpick, where a five-ingredient one is nearly self-diagnosing. If you are layering several botanical-forward products, our conflict checker will flag where the overlaps and the irritant loads stack.

The conclusion is not that plant extracts are marketing. Several of them are among the better-evidenced things you can put on skin, and the industry's interest in them is largely earned. The conclusion is narrower and more useful: the plant name on a panel is an identity claim, the position is a rough dose claim, and only the second one has any arithmetic behind it. Learn where the one percent line falls and most of the mystique evaporates — along with, usually, the reason you were paying extra.