Dermatologist's Take
Bakuchiol is a plant compound marketed as a natural alternative to retinol. It is not a retinoid and shares no chemical structure with one; the comparison comes from a lab study showing it shifts some of the same genes. One 44-person trial found bakuchiol and retinol produced similar improvements in wrinkles and dark spots over 12 weeks, with less scaling and stinging on bakuchiol. Read that carefully, though: people applied bakuchiol twice a day and retinol once a day, and a study that small cannot prove two treatments are equal. An independent 2024 review graded the anti-aging evidence for retinol an A and for bakuchiol a C. It is a reasonable option if retinol irritates your skin. It is not a proven equal.
At a glance
- AKA babchi, bakuchi. A single compound from the seeds of Psoralea corylifolia
- 0.5% is what the main trial used; 0.5–1% is the usual range in products
- Works with niacinamide, vitamin C, sunscreen. No known conflicts
- Evidence: Limited — one 44-person head-to-head trial, graded C by an independent review
What is it?
Bakuchiol is a single purified compound taken from the seeds of Psoralea corylifolia, a plant long used in traditional Indian and Chinese medicine. It entered skincare around 2014 and is almost always sold as a retinol alternative.
Chemically it is nothing like a retinoid. Retinoids are built from vitamin A and share a recognisable backbone. Bakuchiol is a phenol with a terpene tail — structurally closer to compounds like resveratrol. The link to retinol is behavioural, not structural: in a lab-grown skin model, bakuchiol shifted many of the same genes in the same direction retinol did. That single finding is where the whole product category came from. It is worth knowing it was published by the company that sells the ingredient.
One thing about the source plant is worth understanding. The same seeds also contain psoralens — the compounds used in medical light therapy, which make skin burn easily in sunlight. Bakuchiol itself is not a psoralen. But purified bakuchiol and raw seed oil are different products, and purity is a genuine safety question here rather than a marketing detail.
How it works
Nobody has established how bakuchiol works in living human skin. That sounds blunt, but it is the accurate position.
The evidence is a gene-expression study done in a lab-grown skin substitute. Bakuchiol and retinol were applied side by side, and the pattern of genes each switched on looked broadly similar. Collagen types I, III and IV rose in that model.
What that does not show is a mechanism. Retinoids work by binding retinoic acid receptors inside the cell — a pathway mapped in detail over decades. No study has shown bakuchiol binds those receptors, and the ingredient's own inventor now describes the effect as happening without direct receptor activation. So the honest summary is that bakuchiol produces a retinol-like pattern in a dish, by a route no one has pinned down.
Subtypes
Three different things are sold under overlapping names, and only one of them is bakuchiol.
- Bakuchiol — the single purified molecule. This is what the research tested. Suppliers aim for 99%+ purity, and the reason matters: lower-grade material can carry psoralens and other sensitizers from the seed.
- Babchi oil / bakuchi seed oil — pressed from the whole seed. It contains bakuchiol, but also psoralens. There is a published case of a person developing a sun-triggered rash after applying babchi seeds to the skin. This is not interchangeable with bakuchiol.
- Psoralea corylifolia seed extract — a whole-plant extract. Same issue. Neither the bakuchiol content nor anything else in it is standardised.
If a label names the plant rather than the molecule, you may not be buying what the studies tested.
Concentrations
- 0.5% — the concentration used in the only trial that compared bakuchiol with retinol, applied twice daily. If you want to match what was actually studied, this is it.
- 0.5–1% — the range the main ingredient supplier recommends for finished products.
- 2% and above — sold, but not studied. No evidence shows that more bakuchiol works better.
A general caution on label percentages: an independent review found that four of fifteen published bakuchiol trials never stated what concentration they used. The literature itself is inconsistent here, so comparing percentages between products tells you less than it appears to.
Conditions it treats
What to expect
In the one trial that ran a full course, changes were most visible at 12 weeks rather than earlier. Give it three months before deciding.
Tolerability is bakuchiol's honest selling point. In the head-to-head trial, people using retinol reported more skin scaling and stinging at every check-in. The bakuchiol group reported more redness at week 4, which settled by weeks 8 and 12. Nobody in either group reported sun sensitivity.
That is not the same as harmless. Allergic contact dermatitis to bakuchiol has been documented in at least three published case reports since 2019, so "natural" should not be read as "cannot react." Mild redness early on is common; a rash that spreads and worsens over days is the pattern that should make you stop.
There is no evidence that bakuchiol causes a purge the way retinoids can — but no study has looked, either.
The evidence3 studies
Studies behind the treatments above. Tap any tag to see what it means.
Bakuchiol 0.5% cream improved wrinkles and pigmentation over 12 weeks with no measurable difference from retinol 0.5% — but bakuchiol was applied twice daily and retinol once daily, and 44 people is too few to prove two treatments are equal.
Moderate evidence
Randomised and double-blind, which is the right design. Graded moderate rather than strong because 44 people is small, the two creams were applied at different frequencies, and a finding of "no difference" in a study this size is absence of evidence rather than evidence of equivalence.
44 people
How many people took part. Bigger studies are less likely to show a result by chance alone.
Uneven dosing
Bakuchiol was used twice a day, retinol once a day. Any comparison between the two arms has to be read with that in mind.
Bakuchiol was graded C for anti-aging evidence quality in an independent review, against A for retinol and vitamin C.
Limited evidence
This is a systematic review of the evidence itself rather than a trial. It applied a standard grading scale to the published studies behind each anti-aging ingredient. Its value here is that the authors have no commercial stake in any of the ingredients they graded.
Review article
A systematic review of 32 studies covering 1,236 patients across a range of cosmetic anti-aging ingredients, graded using the Oxford Centre for Evidence-Based Medicine scale.
The bakuchiol trial literature consists of 15 human studies, of which 12 had no control group and 10 tested bakuchiol mixed with other active ingredients.
Limited evidence
A systematic review of study quality rather than of results. The authors are academic dermatologists with no link to any bakuchiol supplier, which is unusual in this literature.
Review article
Systematic review across Cochrane, PubMed, EMBASE and Web of Science. The studies were too different from one another to be pooled statistically.