The evidence is mixed. A 2006 study found lower Sebumeter-measured sebum excretion with a 2% niacinamide gel in its Japanese arm. Its US split-face arm found a lower casual sebum level after six weeks, but not a significant sebum-excretion difference versus untreated skin (Draelos, Matsubara & Smiles, 2006).
That supports a narrow possibility of less measurable surface oil in a similar formula (Draelos, Matsubara & Smiles, 2006). It does not guarantee a matte finish, smaller-looking pores or a treatment result.
What the 2% trial actually measured
The Japanese study enrolled 100 healthy women; 98 completed. Fifty used a 2% niacinamide hydrophilic gel and 50 used a placebo twice daily for four weeks. Sebumeter and Sebutape readings were taken at baseline, week 2 and week 4 (Draelos, Matsubara & Smiles, 2006).
The US study enrolled 30 women in a randomized split-face design for six weeks. Casual sebum level fell on the treated side at week 6, but the sebum-excretion-rate result was not significantly different from the untreated side. The authors note that increased attention to cleansing may have lowered both sides (Draelos, Matsubara & Smiles, 2006).
The important limitation is the disagreement inside one paper: the same nominal concentration did not produce the same endpoint across populations and designs (Draelos, Matsubara & Smiles, 2006).
The formula matters
The tested gel contained 2% niacinamide plus 1% panthenol. A 10% serum is not the studied formula, and a moisturizer, sunscreen or exfoliating serum can change feel, shine and tolerance even when niacinamide is present (Draelos, Matsubara & Smiles, 2006).
A separate randomized proof-of-concept study evaluated a multi-ingredient biomimetic moisturizer in oily, blemish-prone women and used a 4% niacinamide gel cream as a positive-control regimen. That is formulation evidence, not a dose-response experiment of pure niacinamide. Crossref metadata lists GlaxoSmithKline as the funder, another reason not to treat this as independent ingredient evidence (Santos-Caetano et al., 2020; Crossref metadata).
Sebum is not the same as visible shine
Sebumeter output, casual surface sebum, sebum excretion after cleansing and a person’s visual impression of shine are related but different endpoints. A Sebumeter reading does not itself measure a product’s visual finish, which can depend on the vehicle and lighting (Draelos, Matsubara & Smiles, 2006).
Use the evidence for what it says: a possible, formula-specific change in measured oil over several weeks (Draelos, Matsubara & Smiles, 2006). Do not turn it into “niacinamide stops oil,” an acne claim or a universal high-percentage recommendation.
What remains uncertain
No robust dose-response trial comparing common niacinamide strengths on visible shine was located. The primary study also had Procter & Gamble research-and-development affiliations, so the commercial relationship belongs in the evidence context (Draelos, Matsubara & Smiles, 2006).
If a formula causes persistent burning, flushing, swelling or a spreading rash, stop and seek clinical advice. This page does not diagnose or treat an oily-skin condition.
Sources
- Draelos, Matsubara & Smiles — 2% niacinamide sebum trial
- Accessible article PDF
- Santos-Caetano et al. — niacinamide-containing moisturizer study
- Crossref — Santos-Caetano funding metadata
- Oily skin review
This page summarizes published research. It is not a product review or medical advice.