Chinese Skincare Marketing Runs on Ingredient Literacy. That Has a Downside.

The default post is a formulation breakdown

Open Xiaohongshu and look at how a moisturiser gets discussed. The format is not a soft-focus shelf photo with a caption about self-care. It is more often a list: what the actives are, at what percentage, in what vehicle, at what pH, with which conflicts, and whether the user's own barrier tolerated it over four weeks.

The vocabulary has gone mainstream to a degree that has no Western equivalent. 早C晚A — vitamin C in the morning, a retinoid at night — is not an industry term but a consumer hashtag with hundreds of millions of views, and it describes a routine architecture rather than a product. Brands have responded by naming products after it directly, which is a strange thing to see on a box: the marketing has been outsourced to a protocol the audience invented.

The same holds for raw materials. Chinese product names routinely feature the supplier's trade name for an active — the resorcinol derivative sold as a whitening ingredient appears on packaging by its commercial code, not by any invented brand word. That would be unthinkable on a US shelf, where the trend runs the opposite way and the actual INCI is buried under a proprietary "complex".

Whatever you think of the outcome, the informational content of that discourse is higher. A consumer who can ask what the vehicle is has better questions than one who has been sold a feeling.

The regulator is genuinely ahead, and Western readers assume otherwise

Here is the part most Western coverage gets backwards, usually because the assumption runs the other way by default.

China's Cosmetic Supervision and Administration Regulation came into force at the start of 2021, replacing a framework that had stood since 1989. Alongside it, the Cosmetics Efficacy Claim Evaluation Standards took effect in May 2021, and from January 2022 registrants and notifiers have had to evaluate their efficacy claims against that standard and upload an evaluation abstract — product information, the claims evaluated, an overview of methods and results, and conclusions — to a platform the National Medical Products Administration designates. In other words, the evidence summary behind a claim is a public document.

Six categories require more than paperwork. Whitening and freckle-removing, sunscreen, anti-hair-loss, acne removal, nourishing and repairing claims all require human efficacy trials, conducted by testing institutions registered for the purpose in China. For the remaining evaluated claims, a literature review or research data analysis can substantiate the claim instead — a lower bar, but a documented one.

Compare that with the United States, where cosmetic claims require no premarket substantiation filing at all, where "clinically proven" carries no defined evidentiary threshold, and where a brand's data typically becomes visible only if someone sues or a competitor challenges it. This is a straightforward case of a non-Western regulator being ahead of the US on consumer protection, and the reflex to assume otherwise is worth examining. It is also not a blanket endorsement: China's system carries its own burdens, including registration requirements that keep many small foreign brands out entirely. But on the specific question of whether a brand must show its work before making an efficacy claim, the answer differs sharply, and not in the direction most people expect.

Chengfendang: the ingredient party and its built-in failure

The consumer movement behind this has a name — 成分党, the chengfendang, literally the ingredient party, usually rendered in English as "skintellectuals." These are consumers who research formulations before buying, who are unmoved by branding, and who treat learning skincare chemistry as a hobby rather than a chore. They are, by any reasonable standard, a better-informed audience than the one Western beauty marketing addresses.

And the failure mode grew directly out of the strength. When purchasing decisions are made on published percentages, the competitive pressure moves to the percentage. Brands respond by raising the number, because the number is what is being compared. The formulation work that determines whether the number means anything is invisible on a product listing and therefore does not compete.

The clearest illustration is vitamin C. Percutaneous absorption work on L-ascorbic acid found that it must be formulated below pH 3.5 to enter the skin at all; that maximal absorption occurred around a concentration of twenty per cent, above which it did not improve; that tissue levels saturated after three daily applications; and — the detail that undoes an entire product category — that magnesium ascorbyl phosphate, ascorbyl-6-palmitate and dehydroascorbic acid did not increase skin levels of L-ascorbic acid at all. A serum can therefore advertise a high percentage of an ascorbate derivative, be entirely honest about that number, and deliver nothing to the tissue. The percentage is real. The claim built on it is not.

Niacinamide runs the same trap in reverse. The most-cited clinical work used 5% twice daily over twelve weeks in fifty subjects and found improvements in fine lines, hyperpigmented spots, blotchiness, sallowness and elasticity. That is the concentration the evidence describes. Products at double or quadruple it are making a tolerability decision, not delivering a documented upgrade, and the flushing and irritation that come with high-percentage niacinamide have no corresponding benefit in the literature.

Retinol is the third. Tolerability, not potency, is the limiting factor in almost every real-world retinoid outcome, because the routine people abandon in week three delivers nothing. A high-percentage retinol serum like ↗Skintific 2.5% Retinol Skin Renewing Serum is competing on a number in a category where the number is close to the least important variable. Meanwhile a product named for the consumer protocol itself, ↗Proya Early C Late A Dual Anti-Aging Serum, and one named after a raw-material trade code, ↗Skintific SymWhite 377 Dark Spot Moisture Gel, show how completely the naming convention has absorbed the ingredient-first logic.

What a percentage cannot tell you

If the number is a poor proxy, it is worth being specific about what is missing from it, because "formulation matters" is a thought-terminating phrase rather than an explanation.

Four things determine whether a stated concentration does anything. The vehicle, which decides whether the molecule reaches viable tissue or sits on the stratum corneum. The pH, which for acids and for ascorbic acid determines what fraction is in the form that penetrates — this is the same arithmetic that makes a high-percentage acid at the wrong pH inert and a low-percentage one at the right pH effective. Stability, since several of the most-marketed actives oxidise, and a serum that has gone amber is delivering degradation products. And the rest of the formula, which can either support penetration or block it entirely.

None of those four appears on a product page. All four are legible from a full ingredient list if you know what you are looking at, which is precisely what the Ingredient Decoder is for — paste the list and get every INCI name translated with allergen and comedogenic flags, rather than hunting for the one number the brand chose to print.

The other thing a percentage hides is interaction. The 早C晚A architecture exists because putting an acidic vitamin C serum and a retinoid in the same step is a bad idea, and the consumer protocol solved that by separating them across the day — which is a genuinely good piece of folk engineering. But routines have grown well past two actives, and once you are running four, the pairwise interactions are not something anyone reasons about reliably. The Conflict Checker exists to do that arithmetic, and it is the natural companion to this style of buying. ↗The vitamin C entry covers the pH and derivative question in full, ↗niacinamide covers where its evidence actually sits, and ↗retinol covers the tolerability curve that percentage-shopping ignores.

Import the literacy, leave the metric

The right conclusion is not that Chinese skincare consumers are being fooled. They are considerably harder to fool than the audience Western beauty marketing has cultivated, and the regulatory backdrop is stricter in a way that would materially improve US shelves if it were copied.

What is worth importing is the habit: reading the ingredient list first, expecting a brand to be able to name its actives and their concentrations, treating "clinically proven" as a question rather than a conclusion, and being willing to learn a little chemistry rather than outsourcing the decision to a bottle's aesthetic. That is a strictly better consumer posture than the one most Western readers have.

What is not worth importing is the scoreboard. A percentage is one input among four, it is the only one brands can escalate cheaply, and escalating it degrades tolerability — which is the variable that actually determines outcomes, because the routine you can sustain beats the routine you abandon. The moment a category starts competing on a number, that number stops carrying information.

The products worth looking at from this market tend to be the ones not playing that game. ↗Winona Face Moisturizing Cream is built around barrier lipids and a botanical anti-irritant rather than a headline percentage, which is a formulation-led proposition. ↗Skintific 5X Ceramide Low pH Cleanser advertises its pH, which is exactly the variable the percentage arms race trains people to ignore. And ↗Proya Source Force Repairing Essence leads on a lipid and ferment blend rather than a number.

The literacy is the achievement here, and it is real. The number is just the first metric a well-informed market happened to reach for, and like every metric that becomes a target, it stopped measuring the thing it was chosen to measure.