Almost everything sold for ageing skin is sold as repair — creams that promise to undo what has already happened. I have never found that framing useful, because undoing is the hardest and least reliable thing skincare does.
Protecting is easier, cheaper, and it compounds. That is what I mean by skin longevity: not a different shelf of products from anti-ageing, but the same products used with a different goal. You are not fixing damage. You are keeping it from adding up.
Which means two things have to happen together, and most routines only do one. You need the actives that genuinely rebuild — a retinoid, vitamin C (see My 5 Curated Vitamin C Products), sunscreen. And you need a barrier strong enough to tolerate them. Actives without barrier care is how people end up irritated and quitting in week three.
Why the barrier does the heavy lifting
Your skin holds itself together with fats — ceramides, cholesterol and fatty acids, in roughly fixed proportions. It makes fewer of them as you get older, and every harsh cleanser, over-exfoliation and cold dry winter takes more out.
When that layer thins, water leaves and irritants get in. Skin feels tight, stings when you apply things, flushes for no reason. And here is the part that matters for longevity: that is exactly the state in which you cannot use a retinoid. The barrier is not a nice-to-have alongside your anti-ageing routine. It is the thing that decides whether you get to have one.
There is one number worth keeping in mind. Dermal collagen falls by roughly 30% in the first five years after menopause, then around 2.1% per year for the fifteen years after that (Brincat et al., 1987). That is a steep curve, and it is the clearest argument I know for starting the protective habits early rather than reaching for repair later.
Dr.Different Vitalift-A, 0.05% retinal
Retinal sits one conversion step away from retinoic acid, where retinol sits two — so you get more activity per unit of irritation. 0.05% is a genuine starting strength, roughly equivalent to 0.1% retinol, from a brand founded by a Korean dermatologist with a stabilisation process built for the molecule. Start twice a week.
Isntree Hyper Retinol EX 1.0 Serum
1% retinol with 0.3% retinyl palmitate and 0.6% bakuchiol, buffered with niacinamide, peptides and ceramides. It is well built, and it is still 1% retinol — two to three nights a week to begin with, and never in the same routine as acids or pure vitamin C.
Purito Seoul Timeless Bloom Bakuchiol Serum
1% bakuchiol with acetyl hexapeptide-8. The reference study is Dhaliwal et al., 2019: 44 patients over twelve weeks, 0.5% bakuchiol against 0.5% retinol. Both significantly reduced wrinkle surface area and hyperpigmentation with no statistical difference between them — and the bakuchiol group reported no scaling or stinging, where the retinol group did. One small trial, so I would not claim it settles the question. But if reactivity is your limiting factor, this is where I would go.
Aestura Atobarrier 365 Cream
Ceramide NP with cholesterol and a fatty acid complex, from Amorepacific's dermatological line, which began life as a hospital barrier cream in 1982. This is the workhorse of the whole routine and the least interesting product to talk about. It is also the one I would protect if the budget got cut.
Rejuran Turnover Ampoule
PDRN is salmon DNA fragments, and the injectable evidence for it is genuinely good. Topically it is a harder claim — PDRN molecules are larger than the collagen I dismiss further down, so getting them anywhere useful is the whole problem. What makes this the one I would pick is that Pharma Research make the injectable, publish the concentration at 0.5%, and state what they do to shrink the molecule, where most PDRN serums disclose none of that. I use it as soothing recovery after I have overdone the retinal — not as an injection in a bottle. More on the ingredient here.
SKIN1004 Madagascar Centella Hyalu-Cica Water-Fit Sun Serum SPF50+ PA++++
The filter list is the reason I use this one. Uvinul A Plus and Tinosorb M handle UVA — the deeper wavelength that breaks collagen down, which is the whole ageing question — and both stay stable in sunlight instead of degrading on your face by lunchtime. Under that sits centella at 9,800 ppm and niacinamide, with no fragrance or alcohol, which matters a lot when you are reapplying every day. And it is watery enough that you will actually use a full amount. That last part is not a small thing: it is the difference between the SPF on the label and the SPF on your face.
The routine
Sunscreen is not one product on this list — it is the reason the list exists.
| Morning | Evening |
| Gentle cleanse Hydrating toner or essence Vitamin C serum Ceramide cream SPF50+ PA++++ |
Cleanse Retinal or bakuchiol — not both Ceramide cream on top |
On sunscreen: roughly a third of a teaspoon for the face alone, and reapply if you are outside for any length of time. Under-application is the most common reason sunscreen underperforms — the SPF on the label is measured at 2 mg/cm², and almost nobody applies that much.
On retinoids: if you are new to either the retinal or the bakuchiol, apply your ceramide cream first and the active over the top. It cuts irritation noticeably and costs you very little efficacy.
And never in the same session: a retinoid plus AHA or BHA plus pure vitamin C. Pick one lane per night.
Consistency beats strength, reliably. Twelve weeks is the shortest window in which any of the studies above measured anything at all. Three nights a week for a year does more than seven nights a week for a month followed by a peeling face and a drawer.
About collagen — what is worth it and what isn't
- Topical collagen: no. Native collagen is around 300,000 daltons. The rough working threshold for crossing the outer layer of skin is 500 daltons. A collagen cream is a decent moisturiser and nothing more — it sits on the surface until you wash it off. If one makes your skin feel better, that is the humectants and emollients doing the work.
- Oral collagen peptides: a qualified yes. A 2023 systematic review and meta-analysis of 26 randomised controlled trials, covering 1,721 participants, found hydrolysed collagen significantly improved skin hydration and elasticity against placebo. A separate 2023 double-blind trial found 1,650 mg daily over twelve weeks improved hydration, wrinkling, elasticity and desquamation. That is a real signal. The caveats are also real: a substantial share of these trials are industry-funded, the effect sizes are modest, and the measurements are instrumental rather than "you will see it in the mirror".
- What I would spend on first: the things that make your own collagen — retinoids, vitamin C as the required cofactor, and above all sunscreen, which stops the enzymes that break collagen down from being switched on in the first place. If supplements are still within budget after that, the evidence is decent enough that I would not argue. Before that, they are the wrong order of operations.
Preserving is cheaper than repairing. Start with the sunscreen.
Build your skin profileSources: Brincat M et al., "Skin collagen changes in post-menopausal women receiving oestradiol gel," Maturitas, 1987. Dhaliwal S et al., "Prospective, randomized, double-blind assessment of topical bakuchiol and retinol for facial photoageing," British Journal of Dermatology 180(2):289–296, 2019. de Miranda RB et al., "Effects of oral collagen for skin anti-aging: a systematic review and meta-analysis," Journal of Cosmetic Dermatology, 2023. Bos JD and Meinardi MMHM, "The 500 Dalton rule for the skin penetration of chemical compounds and drugs," Experimental Dermatology 9(3):165–169, 2000. Concentrations quoted are taken from each brand's published ingredient list.
