Retinol Percentage and the Cosmetic Versus Drug Line

Retinol is one of those ingredients where the formulation question and the marketing question quietly merge into the same legal question. You can have two nearly identical serums, same percentage, same base, and one of them is a cosmetic while the other has drifted into drug territory, purely because of what the label says it does.

It's not really about a single magic percentage

People often ask for "the retinol limit" as if there's one universal number that separates cosmetic from drug. There isn't a single global figure that applies everywhere, and concentration limits for retinoid-family ingredients do vary by market and get revisited periodically, so the specific ceiling in your market is genuinely worth confirming with the current regulator guidance rather than trusting a number from an old blog post or a supplier's marketing sheet. What's stable across markets is the underlying principle: as concentration climbs and as the physiological effect becomes more drug-like, regulators start asking whether the product is still cosmetic in nature at all.

Claims move the line more than percentage does

This is the part formulators underestimate. A retinol product marketed with cosmetic language, softening the look of fine lines, improving the appearance of skin texture, supporting a smoother look, generally stays in cosmetic territory because those are claims about appearance, not claims about treating a disease or altering body structure or function.

The moment the claim shifts toward treating, curing, or preventing a condition, such as claiming to treat acne, cure a skin disease, or reverse a diagnosed dermatological condition, you've made a drug claim regardless of the percentage in the bottle. This is the same logic that governs why a bar of soap with a moisturizing claim is a cosmetic but a bar of soap claiming to treat acne is something else entirely. The chemistry doesn't change. The claim does, and the claim is what regulators evaluate first.

Where the actual risk sits

Three things tend to compound and push a retinol product toward the drug line:

• Concentration climbing toward levels associated with therapeutic effect. Higher percentages correlate with more visible, more drug-like skin changes, which invites more scrutiny.
• Claims framed around treating a named condition rather than describing appearance or texture.
• Directions for use that resemble a treatment protocol (specific dosing language, condition-specific instructions) rather than typical cosmetic use instructions.

Any one of these alone might be fine. Stack two or three together and you've built a strong case for a regulator to reclassify the product, and reclassification isn't a paperwork inconvenience, it usually means a completely different regulatory pathway, different substantiation requirements, and in some markets a different approval process entirely before you can sell it at all.

A practical gut check for your own product

Before you finalize a retinol formula and its marketing copy, ask:

1. Does my label describe appearance and texture, or does it name a condition and promise to treat it?
2. Would a plain reading of my directions for use sound like a skincare routine or a treatment regimen?
3. Have I checked this specific market's current concentration guidance for retinoid ingredients, rather than assuming a figure I've seen used elsewhere?
4. If I'm marketing internationally, am I aware the answer to all of the above can differ by country?

Keeping the formulation and the marketing team aligned

The recurring failure mode isn't usually the chemist getting the percentage wrong, it's marketing writing copy independently and reaching for the most persuasive language available, which often happens to be treatment language. Getting formulation and marketing to review labels together, specifically checking for words that imply treating a diagnosed condition, catches most of the risk before a label ever goes to print.

Cosmetic Comply's screening focuses on ingredient concentration against restricted lists like Canada's Hotlist, which is the formulation half of this question. The claims half still needs a human eye on your actual label copy, since no ingredient screen can tell you whether your marketing language quietly turned a cosmetic serum into an unregistered drug claim.