Niacinamide is one of skincare’s better generalists. It has also become a percentage arms race nobody asked for. Barrier support, visible oiliness, uneven tone, signs of photoaging: there is some human evidence behind all of it, which is unusually good company for a cosmetic ingredient to keep. Then the shelf starts at 5%, moves to 10%, and eventually arrives at 20% as though we were upgrading storage.
The number is easy to be seduced by, and that is not a character flaw. It is the only thing on the label that behaves like a spec — everything else is adjectives — so it becomes the thing you shop on.
(General cosmetic reading, not a treatment plan. Persistent acne, rosacea and diagnosed pigment disorders deserve a real assessment.)
The evidence is real, and it is scattered across completely different jobs
| Concentration | Studied for | The study behind it |
|---|---|---|
| 2% | Facial sebum | A randomized study of 2% niacinamide. |
| 4% | Inflammatory acne | Topical nicotinamide studied at 4% in inflammatory-acne trials. |
| 5% | Visible signs of facial aging | A 12-week randomized split-face study in 50 women: a moisturizer containing 5% niacinamide versus its control moisturizer. |
| 20% | — | The number on the shelf. A manufacturer may hold finished-product data at that concentration. The figure printed on the front is not it. |
Different outcomes, different populations, different formulations. They do not quietly merge into one sentence that says “5% is the proven dose,” and they certainly do not add up to “20% must be four times better.” Three good studies about three different questions are still three different questions.
Which means the honest first move is not choosing a percentage. It is choosing a row.
Twenty percent is a label number, not an efficacy multiple
Many of the familiar cosmetic studies live in the 2% to 5% range. Retail products can go a great deal higher. A manufacturer may well have useful finished-product data for a higher concentration, but the percentage on its own does not tell you the effect is larger.
Four times the concentration is not four times the clinical result. Skincare is inconvenient that way, and the product grid will never say so, because the grid can only sort what it can measure. It is also worth checking whether you needed a new bottle at all — niacinamide is already in most of the routine, usually at the concentrations that were actually studied.
Barrier support is useful, and it is not a force field
Niacinamide turns up in a lot of moisturizers because there is mechanistic and clinical rationale tied to epidermal barrier function. That makes it a sensible supporting ingredient, and a genuinely good one.
It does not make the formula irritation-proof. A routine running retinoids, exfoliating acids and an enthusiastic cleanser does not become civilized because niacinamide showed up in the moisturizer. You cannot buy your way out of an over-active routine by adding one more product to it, though most of us try that first, because it is the option that feels like progress.
Oil, pigment and acne are not one job called “multitasking”
Look back at the table. A 2% sebum study is relevant to visible oiliness. A 5% photoaging study is relevant to the outcomes it actually measured. A 4% acne study is condition-specific clinical evidence. These are not interchangeable senses of the word balances, however hard the copy works to make them sound like one.
If a product promises meaningful treatment of persistent acne or a diagnosed pigment disorder, the evidence standard has to match the condition — not the ingredient’s general good reputation.
Compatibility charts matter less than whether the formula suits you
Internet rules tend to treat ingredient names as though they react inside a spreadsheet. In practice pH, vehicle, the other actives, the solvents and your own barrier decide whether a combination is comfortable to live with.
If a niacinamide product stings, burns or leaves you persistently red, the answer is not to defend the molecule’s reputation on its behalf. Stop, simplify, and reassess the finished formula. The ingredient is not on trial. The bottle is.
Pick the row, then pick the bottle
Name the job — oiliness, barrier support, visible aging — find the concentration that was studied for it, and buy something at that strength you will still be using in three months. If you are reaching for 20% because 10 did not work, the number was never the variable that failed you. Niacinamide is allowed to be a good ingredient without also being a personality.
Checked against peer-reviewed dermatology literature. Reviewed August 2026.
Continue: The Verdict Index
Source file
What this article is standing on
- Niacinamide: a review on dermal delivery strategies and clinical evidence (Drug Delivery and Translational Research, 2024)The synthesis of niacinamide's clinical evidence and how delivery affects it. It supports several modest effects at conventional concentrations and does not establish a dose-response case for twenty percent — the specific claim this article is testing.
- Cochrane: Topical azelaic acid, salicylic acid, nicotinamide, sulfur, zinc and fruit acid for acne (2020)Covers topical nicotinamide for acne at low to moderate certainty. Trial concentrations sit well below the headline percentages sold at retail.
- FDA: Summary of Cosmetics Labeling RequirementsA cosmetic label carries no concentration declaration, so a stated percentage on the front of a bottle is the brand's own claim rather than a regulated disclosure.
- FTC: Health Products Compliance GuidanceSubstantiation an advertiser must hold before a stronger claim implies a stronger result. The standard behind "if twenty percent is better, show the trial".