A Field Guide to DIY Skincare: What Works, What's Pointless, What Sends You to Hospital
Kitchen skincare is neither the fraud dermatology implies nor the liberation your feed promises. It is three completely different categories wearing the same costume.
Read the preview
Is DIY skincare harmless fun or a hospital trip? Depends on the jar. US rules already recognise oatmeal and glycerin as skin protectants. But plant oils differ — olive oil damaged the skin barrier in one trial while sunflower preserved it. Then there's black salve, which left cancer behind in about a third of treated lesions. Where's the line? The Dermington Gazette sorts the cupboard.
The regulator got there before the internet did
Open the US federal register at 21 CFR 347 — the over-the-counter monograph for skin protectants — and the ingredient list reads like a raid on a pantry. Colloidal oatmeal is there. So are cocoa butter, kaolin, glycerin, topical starch, cod liver oil, lanolin and sodium bicarbonate, each with its own recognised status as an active ingredient rather than a filler. Colloidal oatmeal qualifies at a strikingly low bar: a minimum of 0.007 percent, or 0.003 percent when it is combined with mineral oil.
That detail is worth sitting with, because it dismantles the framing both camps use. Dermatology's public posture toward DIY is a blanket eye-roll. Social media's is a blanket liberation narrative. Neither survives contact with a regulatory document that formally recognises ground oats as a therapeutic active while refusing that status to most of what gets marketed as advanced skincare.
So the real question was never whether a kitchen ingredient can do anything. It is which ones do, at what dose, in what vehicle, and with what capacity to hurt you when the answer is wrong. That is a triage problem, not a morality problem — and triage is what almost nobody writing about DIY actually performs.
Tier one: the short list that genuinely earns its place
Colloidal oatmeal is the strongest case, and it is strong for an unglamorous reason: somebody bothered to test it. Avenanthramides, the phenolic compounds specific to oats, suppress pro-inflammatory signalling, and the clinical work behind them is not a single flattering study but a body of trials in atopic dermatitis, including work in paediatric populations and a trial specifically in Black and African American children with mild-to-moderate disease. A finely milled oat suspension in a warm bath is a legitimate anti-itch measure, and a formulated version like Aveeno Eczema Therapy Moisturizing Cream simply does the same thing with a stable emulsion and a preservative system attached. If you want the mechanism rather than the marketing, our colloidal oatmeal page carries the evidence grading.
Aloe has a narrower but real claim. Meta-analyses of randomised trials in burn injury point to faster healing than some comparators — which is a specific finding about wounds, not a general licence for aloe as a moisturiser. Aloe's evidence thins considerably outside that setting, and a plant leaf snapped open on a windowsill is a very different product from a stabilised, preserved gel such as Cetaphil Soothing Gel-Cream with Aloe.
Honey is similar and gets misquoted the same way. The wound-care literature is built on medical-grade honey — gamma-irradiated, standardised for peroxide activity, applied under a dressing. That is not the jar on your counter, and the gap between them is the entire point. Where honey shows up in a formulated product like the COSRX Full Fit Propolis Honey Overnight Mask, it is functioning as a humectant in a preserved base, which is a modest and honest role. The honey entry spells out where the evidence stops.
And then the finding that should end the phrase 'natural oils are good for skin'. In a forearm-controlled study, volunteers applied olive oil to one arm and sunflower seed oil to the other, twice daily. Olive oil significantly reduced stratum corneum integrity and produced mild erythema, in people with and without a history of atopic dermatitis. Sunflower seed oil preserved barrier integrity and improved hydration in the same volunteers, on the same bodies, over the same weeks. Two plant oils, opposite results. Whatever separates them, it is not naturalness.
Tier two: harmless, pleasant, and doing very nearly nothing
The large middle of DIY is inert. Yogurt masks, cucumber slices, egg white, mashed avocado, rice water, cooled green tea — none of these is going to hurt you, and the standard rebuttal to them is unnecessarily sneering. They feel good. They are cheap. Ten minutes lying still with something cool on your face is a genuinely restorative act, and pretending that has no value is bad faith.
But it is worth being precise about why they do so little, because the reason is transferable. Every one of these fails at least two of the four things a working topical has to solve. Concentration: the lactic acid in yogurt is present at a fraction of a percent, at a pH nowhere near the range where exfoliation happens. Penetration: most of what is interesting in a plant sits inside cell walls that a blender does not open, and what does get out is often a molecule too large or too hydrophilic to cross the stratum corneum. Contact time: a rinse-off mask has minutes, when the actives that matter need hours of residence. Stability: vitamin C in lemon or kiwi has largely oxidised before it reaches your face, which is why formulated ascorbic acid needs a low pH, a chelator, and an opaque bottle.
This is the honest version of the DIY critique, and it is not an insult to anyone. Formulation is not a trick played on consumers to justify a markup. It is the difference between having an ingredient and delivering one, and it is genuinely difficult. If you want to see what a real formula is doing rather than what its front label claims, run the INCI list through our ingredient decoder — it is the fastest way to learn the difference between an active and a fragrance.
The part a kitchen structurally cannot do
There is one failure mode in home mixing that no amount of care solves, and it is microbial. Anything containing water grows things. A commercial formula defends itself with a preservative system that has to survive a challenge test — deliberate inoculation with Staphylococcus aureus, Pseudomonas aeruginosa, Candida, Aspergillus — and demonstrate kill over weeks.
Here is the sobering context: in a study sampling ninety-one commercial cosmetics across intact, in-use and near-empty states, roughly 10.6 percent of the surfactant products were contaminated during ordinary consumer use, carrying organisms including Staphylococcus warneri, Staphylococcus epidermidis and Pseudomonas putida. Those products had preservative systems. They had sealed manufacturing. They still picked up organisms from fingers and bathroom air.
A jar mixed at home has none of that defence and starts with a raw agricultural input. This is why the practical rule for home mixing is not 'add a preservative' — dosing and validating a preservative system is harder than the mixing — but 'stay anhydrous or make it fresh'. An oil, a balm, a sugar scrub in a waterless base is a genuinely low-risk home project. A cream, a toner, a hydrosol, anything with water or fruit in it, is a small bioreactor with a two-day life and no way to know when it has turned. And whatever you make, whatever the source, patch test it on a discreet patch of skin for several days before it goes near your face; our patch test tool walks through the timing, because a delayed allergic reaction typically shows up at forty-eight to ninety-six hours, not immediately.
Tier three: the ones that actually cause harm
The dangerous list is short and specific, which is what makes the blanket warnings so counterproductive. Nobody is being hospitalised by oatmeal.
Citrus juice is first, and it is the cleanest case in dermatology because the injury photographs so distinctively. Limes, lemons and several other plants carry furocoumarins; on skin, in sunlight, these produce phytophotodermatitis — a phototoxic burn that blisters in bizarre drip-and-handprint patterns and leaves hyperpigmentation that can persist for months. It is not an allergy, it does not require sensitivity, and it will happen to anyone given enough juice and enough UV. The lemon entry covers why the lightening claim that motivates it is backwards in the first place.
Undiluted vinegar is second. Case reports in the dermatology literature describe chemical burns from topical apple cider vinegar, including a young patient who followed an internet protocol for self-removing moles with vinegar and sustained a chemical burn for it. That case is worth dwelling on, because the mole is the real danger: a caustic applied to a pigmented lesion destroys the tissue a pathologist would have needed to read.
Which brings us to the worst of them. Escharotic 'black salve' — zinc chloride with bloodroot — is sold as a natural cancer cure that supposedly targets only malignant tissue. A five-year Australian retrospective pulled pathology from 409 patients who had treated 475 lesions with it. Persisting cancer was found in 34.2 percent of the treated areas that were not simply benign to begin with, the salve necrosed normal tissue when applied to harmless lesions — refuting the selectivity claim outright — and specimen numbers more than doubled across the study period. People are not just being burned. They are being given a false all-clear over a tumour that is still there, on faces and necks, while the clock runs.
Why people do this, and the answer that respects the reason
It is tempting to file DIY under gullibility. It mostly is not. Three motives account for nearly all of it, and two of them are entirely rational.
Cost is the obvious one. Prestige skincare's markups are real and its price-to-evidence ratio is frequently indefensible. Control is the second: eczema and acne are chronic, appointments are scarce and expensive, and making something yourself is at minimum an action you can take today. The third is distrust, and skincare has earned a great deal of it. Independent testing has repeatedly caught the industry out — labelled SPF values that failed on retest, benzene contamination found by an outside lab rather than by the brands, a regulator that has not cleared most legacy UV filters as generally recognised as safe and effective for want of data. Anyone who concluded from that record that the label is not automatically trustworthy reasoned correctly.
The trouble is that the conclusion gets over-extended. 'Labels sometimes lie' does not license 'therefore an unmeasured kitchen mixture is safer'. The kitchen mixture has no label to lie, no batch testing, no adverse-event reporting and no recall mechanism. It is less accountable, not more.
The genuinely honest answer, and the one that respects the motive rather than scolding it, is that the evidence-backed floor is already cheap. A bland surfactant cleanser like Vanicream Gentle Facial Cleanser, a ceramide-and-petrolatum cream in the mould of CeraVe Moisturizing Cream, and daily sun protection cost less over a year than most single luxury jars, and each of those three steps carries more evidence behind it than the entire DIY canon combined. If what you are trying to escape is being sold to, that route escapes it further than a mixing bowl does. Our answer on natural remedies for eczema and the parallel one on whether natural acne remedies are safer go deeper on the specific swaps.
Use the pantry for what the monograph already vindicated: an oat bath when you itch, an anhydrous balm when you are dry. Leave the water-based experiments, and treat the citrus, the vinegar and anything sold as a natural cancer cure as what they are — the small set of DIY practices with a documented body count of injuries behind them.
References
- ARTICLE: Colloidal Oatmeal Part I: History, Basic Science, Mechanism of Action, and Clinical Efficacy in the Treatment of Atopic Dermatitis — Journal of Drugs in Dermatology, 2020
- A 1% colloidal oatmeal OTC cream is clinically effective for the management of mild to moderate atopic dermatitis in Black or African American children — The Journal of Dermatological Treatment, 2023
- Effect of olive and sunflower seed oil on the adult skin barrier: implications for neonatal skin care — Pediatric Dermatology, 2013
- Effects of Aloe vera on Burn Injuries: A Systematic Review and Meta-Analysis of Randomized Controlled Trials — Journal of Burn Care & Research, 2024
- Microbiological study of cosmetic products during their use by consumers: health risk and efficacy of preservative systems — Letters in Applied Microbiology, 2006
- Chemical Burn from Vinegar Following an Internet-based Protocol for Self-removal of Nevi — The Journal of Clinical and Aesthetic Dermatology, 2015
- Chemical burn from topical apple cider vinegar — Journal of the American Academy of Dermatology, 2012
- Persisting Cancer in Black Salve Treated Skin Lesions: Results of a Large 5 Year Retrospective Analysis of Australian Histopathology Specimens — Integrative Cancer Therapies, 2023
- New Insights Concerning Phytophotodermatitis Induced by Phototoxic Plants — Life, 2024
Every cited link is verified to resolve to the paper named. Where a source could not be independently confirmed, it is listed without a link rather than pointing you somewhere wrong.
Frequently asked
Is an oatmeal bath actually doing anything, or does it just feel nice?
It is doing something, and it is one of the few DIY practices with formal regulatory backing — colloidal oatmeal is a recognised skin protectant active in the US OTC monograph, at a minimum of 0.007 percent. The clinical evidence sits in atopic dermatitis specifically, including paediatric trials. Use finely milled colloidal oat rather than whole oats, which mostly sink.
If plant oils are natural, why does this article say one of them damages skin?
Because 'natural' is not a mechanism. A forearm-controlled trial applying olive oil to one arm and sunflower seed oil to the other found olive oil significantly reduced stratum corneum integrity and produced mild erythema, while sunflower seed oil preserved barrier integrity and improved hydration. The fatty acid profile of a given oil, not its provenance, determines what it does.
Can I just add a preservative to a homemade cream and make it safe?
Not reliably. Preservative systems are dosed against a challenge test — deliberate inoculation with specified bacteria and fungi, then measured kill over weeks — and the correct level depends on pH, water activity and what else is in the formula. Even fully preserved commercial cosmetics picked up organisms during ordinary consumer use in about one in ten sampled surfactant products. Staying anhydrous is the workable home rule.
Why is lemon juice singled out when so many products contain citrus extract?
Concentration and photoactivation. Furocoumarins in raw citrus juice, applied undiluted and then exposed to sunlight, cause phytophotodermatitis — a phototoxic burn with blistering and months of residual pigmentation. Cosmetic citrus extracts are used at low levels and are frequently furocoumarin-reduced, which is a different exposure entirely.
Is black salve really that dangerous if it removes the lesion?
The removal is the danger. An Australian five-year pathology series across 409 patients found persisting cancer in 34.2 percent of treated areas, and the salve necrosed healthy tissue when applied to benign lesions, disproving the claim that it targets malignancy selectively. A destroyed lesion also destroys the specimen a pathologist needed, so a tumour can keep growing under an apparent cure.
Each shelf lists every product in the catalogue built on the ingredient, graded live against your profile — the shelf stays current when a single pick would not.
- 304 products on the shelf Colloidal Oatmeal The monograph-recognised active and the trial evidence behind it.
- 117 products on the shelf Honey (Mel) Where the medical-grade wound literature stops and the counter jar begins.
- 545 products on the shelf Aloe Vera Burn-healing evidence versus the general moisturiser claim.
- 5 products on the shelf Lemon Extract (Citrus Limon) The phototoxicity mechanism behind the most common DIY injury.
- 2,542 products on the shelf Glycerin
- 218 products on the shelf Petrolatum (Petroleum Jelly)
- 233 products on the shelf Sunflower Oil (Helianthus Annuus Seed Oil)
Products named in this piece
Independently reviewed, with a verdict per concern — the real products behind this piece.
Listen next
All episodesPrint it out
The Print RoomFree posters and handouts on the same ground, ready to print or share.