Hair Growth Oils, Graded: Rosemary, Onion, Castor and Pumpkin Seed

Oils are the cheapest and most viral end of this aisle, and the evidence behind them varies more than any other category. One has a randomised comparison, one has a controlled trial for a different condition entirely, and two of the most-recommended have almost nothing we could read. Here they are on the same scale as everything else on this site.

Rosemary oil — 5/10

The best-evidenced oil here, and that says more about the category than about rosemary. Panahi and colleagues randomised 100 people with androgenetic alopecia to rosemary oil or minoxidil 2% for six months and found no significant difference in hair count between the groups at three or six months, with both groups rising at six (Skinmed, 2015). There was no untreated arm, so the trial cannot separate either treatment from time passing.

Two independent 2025 publications read it and were unimpressed: a Canadian expert consensus lists rosemary oil among interventions it does not recommend, and a review of alopecia misinformation names it among remedies touted online without support. The full reading is in what the rosemary oil versus minoxidil trial actually found.

Onion juice — 3/10

Sharquie and Al-Obaidi tested crude onion juice against tap water in 38 people with patchy alopecia areata, applied twice daily for two months. Regrowth of terminal hairs began after two weeks in the onion group; at four weeks 17 of 23 patients showed regrowth and at six weeks 20 of 23, against 2 of 15 in the tap-water group at eight weeks (Journal of Dermatology, 2002).

That is a genuine controlled comparison, which is more than most of this aisle offers, and it is for the wrong condition. Alopecia areata is autoimmune and well known for regrowing spontaneously; pattern hair loss is not the same disease and the trial says nothing about it. Crude juice squeezed from an onion is also not what an onion shampoo contains. Both cost the row points on whether the shelf product matches the trial.

Castor oil — not graded

Castor oil is among the most confidently recommended substances on the internet for hair growth, and we could find no controlled human trial of it for hair loss. A 2026 narrative review in Cureus by a dermatology group at Wayne State University puts it carefully: evidence supports its use in hair care for improving lustre and possibly combating androgenic alopecia by inhibiting prostaglandin D2 synthesis, and further clinical investigation is warranted to establish its efficacy. That is a proposed mechanism, not a measured result, so there is nothing for this scale to score.

The documented downside is more concrete than the upside. The same review lists contact dermatitis and hair felting among rare reported adverse effects, and a 2017 case report in the International Journal of Trichology describes an entire scalp of hair matting into a hard mass immediately after an oil bath with coconut and castor oil — a condition the authors describe as irreversible, with the hair having to be cut off. One case is one case; it is still more than the growth side has.

Pumpkin seed oil — not graded here

Pumpkin seed oil appears in two 2025 reviews of non-prescription actives — Gupta and colleagues in Skin Appendage Disorders, and Bikash in the International Journal of Trichology — as one of the agents with published trials. The studies best known for it are of an oral supplement rather than a topical, which puts them outside what this site covers, and we have not read them at the source. No grade.

What the independent review said about the whole category

The 2025 International Journal of Trichology review went through this exact group — onion juice, coffee extract, saw palmetto extract, pumpkin seed oil, rosemary oil and the branded peptide actives — pulling twenty-four articles and technical files. Its conclusion is the fairest one-paragraph summary of the category we found: most studies have limited sample size, no comparison with standard therapies, non-uniform groups, undisclosed types of alopecia, and conflicts of interest, and these topicals may find a use as an add-on or an alternative for people who cannot tolerate standard therapies.

Growth oils with no named active at all

A large share of this category is sold with no gradeable active whatsoever: a blend of carrier oils, a fragrance, a botanical or two, and a name doing the work. There is nothing to search for and nothing to read, which is different from a weak study and is worth recognising on sight.

Caffeine-containing oils and coffee-extract blends are a common version. The published caffeine evidence concerns a leave-on liquid and a shampoo at stated concentrations, not an unquantified extract in an oil blend, and a 2025 Canadian expert consensus lists caffeine among interventions it does not recommend in any case. The full reading is in the caffeine shampoo evidence, trial by trial.

If you want to use an oil anyway

They are cheap and most are cosmetics, so the calculation is different from a £40 serum. Three honest notes. Essential oils can irritate skin and cause contact allergy, so a patch test on a small area first is sensible. Anything used near the eyes or on a child deserves a doctor's opinion first. And an oil left on a scalp that is already flaking or inflamed can make that worse, which is a real cost even if the oil itself does nothing.

Where oils sit against the medicated shampoos on the same scale is in the best shampoo for hair loss, graded on published evidence.