Bakuchiol is usually introduced in one phrase: the plant-based alternative to retinol. It is a good phrase. It is easy to remember, it suggests the benefits without the irritation, and it has done a great deal of work for an ingredient whose human evidence fits comfortably on one page.
I like bakuchiol. I also think it is worth knowing exactly how much weight that one page can carry, because the phrase on the jar is promising rather more than the studies measured.
What it is
Bakuchiol is a compound extracted from the seeds of Psoralea corylifolia, a plant with a long history in traditional Indian and Chinese medicine. Chemically it is not a retinoid. It is not related to vitamin A, it is not converted into retinoic acid, and it does not work through the same receptor in the classic way retinoids do.
The retinol comparison comes from laboratory work. In cell and tissue models, bakuchiol switched on a set of genes that overlapped substantially with the pattern switched on by retinol — collagen-related genes in particular. That is an interesting finding and a reasonable starting point. A similar gene-expression profile in a dish is not the same as a similar result on a face, and it is not evidence that the two can be swapped.
One more detail from the chemistry: the same seeds contain psoralens, compounds that make skin more sensitive to light. Cosmetic-grade bakuchiol is purified to remove them, which is why where the ingredient comes from and how it is refined actually matter here.
The first human study
The study usually cited first was published by Chaudhuri and Bojanowski in the International Journal of Cosmetic Science in 2014. Seventeen people applied a 0.5 percent bakuchiol product twice a day for twelve weeks, and wrinkles, pigmentation, elasticity and firmness were all reported to improve.
It is worth being clear about what kind of study that was. There was no control group — nobody using a plain cream for comparison, nobody using retinol. Everyone knew what they were using. And the authors worked for a company that supplies the ingredient. None of that makes the result false. It makes it a pilot, the sort of study you run to decide whether a proper trial is worth doing.
The trial everyone means
That proper trial came in 2019. Dhaliwal and colleagues at the University of California, Davis, published a randomised, double-blind comparison in the British Journal of Dermatology. Forty-four adults were assigned either a 0.5 percent bakuchiol cream applied twice a day or a 0.5 percent retinol cream applied once a day, for twelve weeks. Facial photographs were analysed for wrinkle surface area and pigmentation, and participants recorded side effects.
Both groups improved significantly on wrinkles and on pigmentation, with no statistically significant difference between them. The retinol group reported more scaling and stinging.
That is a genuinely useful result, and it is the foundation for almost every claim you will read about bakuchiol. It is also a single trial of 44 people, and a few features of it limit how far it reaches:
- "No significant difference" is not "the same". With around twenty people per group, a study can only detect large differences. A real but moderate advantage for either ingredient could easily be invisible at that size. The result is that bakuchiol was not shown to be worse, which is weaker than being shown to be equal.
- There was no placebo group. Twelve weeks of regular moisturising and attention improves photographs on its own. Without a plain-cream arm, we do not know how much of either group's improvement belongs to the active.
- The dosing was not matched. Bakuchiol was used twice a day, retinol once. That is a reasonable design choice, and it is also a reminder that the comparison is between two regimens rather than two molecules.
- Twelve weeks is short for retinoids. Retinoid trials typically report their clearest changes at six months and beyond. A comparison stopped at three months says little about where the two would end up.
The pregnancy question
Bakuchiol is often recommended as the retinol alternative for pregnancy. I understand why: retinoids are generally avoided in pregnancy, and people want something to use instead.
But the reasoning runs on absence, not on data. Bakuchiol is not a retinoid, so the specific concern about retinoids does not transfer to it. That is a fair point. What does not exist is any study of bakuchiol in pregnancy. "Not known to be a problem" and "shown to be safe" are different statements, and during pregnancy the person to ask is the one looking after it.
Where it fits
Here is how I would place it. Bakuchiol has one decent randomised trial suggesting it can produce visible improvements in fine lines and pigment over three months, with less irritation than a retinol of the same percentage. It has a plausible laboratory rationale. It does not yet have the depth of evidence retinoids have built over four decades, and there is no long-term or placebo-controlled human data.
That makes it a reasonable option for people who have tried retinoids properly and could not tolerate them, or who want something gentle and accept that the evidence is thinner. It is not a like-for-like substitute, and the phrase "nature's retinol" claims an equivalence the studies did not test.
If you try it, give it the same twelve weeks the trial did, keep the rest of your routine still, and use sunscreen every day. Unlike retinoids, it is not known to make skin more sensitive to sun — but every result in the photoageing literature depends on it anyway.
Bakuchiol has one good trial: 44 people, twelve weeks, comparable to a 0.5 percent retinol and gentler. That is a promising start, not proof of an equal.
