Niacinamide and NAD⁺ are not the same molecule, and only one of them has a dermatology literature. Niacinamide — the ingredient already in half the serums on your shelf — is a small vitamin B3 amide with vehicle-controlled trials behind it. NAD⁺ is a dinucleotide more than five times its weight, and the published human evidence for putting it on skin is one uncontrolled case series in which it was delivered by needles.
If you are looking at a bottle labeled NAD, the useful first move is to read the ingredient list rather than the front. What each of these molecules is in the first place is covered in what nicotinamide actually is and what NAD⁺ is; this page is about which of them has been shown to do anything to skin.
Size is the first problem, not the last one
The stratum corneum is a barrier optimized to keep things out, and the working rule in dermatology is that a compound needs a molecular weight under about 500 daltons to get through it. That threshold was argued from three separate lines of evidence — essentially all common contact allergens sit below it, as do the standard topical pharmacological agents and every drug used in transdermal delivery systems [1].
Niacinamide’s formula weight is about 122 daltons. NMN is about 334. NAD⁺ is about 663 — above the line, and carrying two phosphate groups that make it strongly hydrophilic as well as large. That does not make topical delivery impossible; the rule is a heuristic, and formulation science exists precisely to bend it. It does mean that a serum listing NAD⁺ is asking you to assume something about penetration that nobody has demonstrated on a label.
You can watch the industry conceding the point. A 2026 paper developing elastic cationic liposomal nanogels for topical NMN opens by stating that NMN’s topical application is hindered by high hydrophilicity, poor skin permeability and limited stability, and reports a penetration enhancement of 22.7% and 16.0% for its two formulations relative to free NMN [2]. That is a laboratory permeation result for the smaller of the two molecules, not a clinical outcome — and it exists because the plain ingredient does not get in well.
What niacinamide has actually been shown to do
Real, repeatedly, and at a modest scale. In a randomized, double-blind, placebo-controlled study, 24 women aged 19 to 27 with axillary hyperpigmentation were treated with 4% niacinamide, 0.05% desonide or placebo. Both actives produced significant colorimetric improvement against placebo, with a good to excellent response in 24% of cases for niacinamide against 6% for placebo — and desonide, a topical steroid, outperformed it [3].
A double-blind, randomized, placebo-controlled study in 67 women aged 30 to 60 with signs of aged skin found that a niacinamide formulation improved stratum corneum water content, reduced transepidermal water loss and reduced sebum [4]. And in a double-blind, placebo-controlled challenge, 40 female panelists pretreated on back sites for two weeks with 5% niacinamide or vehicle and then exposed to 1.5 minimal erythemal doses of solar-simulated radiation showed reduced erythema and reduced skin-surface inflammatory biomarkers on the treated sites [5].
Barrier function, water loss, sebum, pigmentation, a blunted sunburn response. Those are the endpoints, and they are worth having. What none of them is, is a structural anti-aging result — and none of those studies measured skin NAD⁺ at all, so nothing in them supports the story that niacinamide works because it raises NAD⁺ in your face.
The strongest vitamin B3 result in dermatology is a pill
This is the finding the category should be built on, and it almost never appears in NAD skincare copy, presumably because it does not sell a serum. In a phase 3, double-blind, randomized trial, 386 participants who had had at least two non-melanoma skin cancers in the previous five years took 500 mg of oral nicotinamide twice daily or placebo for twelve months. The rate of new non-melanoma skin cancers was 23% lower in the nicotinamide group (95% CI 4 to 38, p = 0.02). Actinic keratoses were 11% lower at three months, 14% at six, 20% at nine and 13% at twelve (p = 0.001). Adverse events did not differ, and there was no evidence of benefit once nicotinamide was stopped [6].
That is a hard clinical endpoint, in a high-risk population, from a cheap generic vitamin taken by mouth. It is also specific: it was chemoprevention in people with a history of skin cancer, not a complexion claim for a healthy 40-year-old, and the benefit evaporated on discontinuation. Quote it for what it is.
What we searched for topical NAD⁺, and when
In September 2026 we queried PubMed for randomized controlled trials of nicotinamide adenine dinucleotide applied topically, cutaneously or intradermally. The search returned two records, and neither is a vehicle-controlled trial of topical NAD⁺: one is the oral nicotinamide riboside nerve-fiber study, the other a 2007 trial of a nicotinic acid derivative. Run alongside it as a known-good control, the same query shape for topical niacinamide returned 28 randomized controlled trials, so the database was reachable and the query form works.
The one human study of topical NAD⁺ we could locate is a prospective, single-center case series in 36 Korean women with mixed-type melasma. Over 21 weeks they received five sessions of a microneedling therapy system immediately followed by application of a sterile NAD⁺ booster. Mean Melasma Area and Severity Index fell from 16.8 ± 5.2 to 6.9 ± 3.1 (p < 0.001), a 59.2% improvement; blinded assessors rated 83.3% of patients improved; adverse effects were transient erythema and edema resolving within 48 hours [7].
Read the design before the number. There was no control arm, no vehicle and no split-face comparison, and the NAD⁺ was applied immediately after microneedling — a procedure that improves melasma on its own and that also breaches the barrier this whole page is about. The study is classified level IV evidence, and its authors say in their own conclusion that the approach warrants further investigation in randomized designs such as split-face or vehicle-controlled trials. They are right, and until someone runs one, the honest description of topical NAD⁺ is untested rather than proven.
The one thing that did raise NAD⁺ in skin
It was not NAD⁺, and that is the most instructive detail here. A placebo-controlled study of myristyl nicotinate — a lipophilic nicotinic acid derivative engineered to deliver nicotinic acid into skin without the flushing that plain niacin causes — reported that it increased skin cell NAD by 25% (p = 0.001), increased stratum corneum thickness by roughly 70% (p = 0.0001) and epidermal thickness by roughly 20% (p = 0.001), and reduced transepidermal water loss by about 20% relative to placebo on cheeks (p = 0.012) and arms (p = 0.017) [8].
To get NAD⁺ up inside skin, the researchers had to design a small, fat-soluble precursor that could cross the barrier. Nobody achieved it by dissolving NAD⁺ in a serum, and that asymmetry is the whole argument of this page in one experiment.
How to buy, given all that
Read the INCI list, not the front of the bottle. If it says niacinamide, you are buying an ingredient with vehicle-controlled human data for barrier function, water loss, sebum and pigmentation — usually at 2% to 5%, which is where the trials sat. If it says NAD⁺ or nicotinamide adenine dinucleotide, you are buying a molecule above the conventional penetration threshold with no controlled trial behind it, generally at a much higher price.
And keep the routes separate in your head. Skin is a different question from taking a precursor orally, which is a different question again from injecting it — the evidence does not transfer between them in either direction, and what has and has not been established overall is the right place to start if you are weighing the category as a whole.