The retinoid you can actually tolerate

Nearly everyone who abandons a retinoid does so within the first month, and nearly always for the same reason. They used too much of something too strong, too often, and the barrier gave out. The compound was not the problem. What follows is a schedule, a list of things to stop using alongside it, and an honest account of how long the wait is.

The retinoids have a stronger evidence base than almost anything else sold for the face. Tretinoin has been studied in controlled trials since the late 1960s, first for acne and then, after the photoageing work of the 1980s, for the texture and pigment changes that follow sun exposure. Adapalene, a third generation synthetic, behaves predictably at 0.1 per cent and is available without a prescription in several markets. Retinol and retinaldehyde are weaker precursors the skin must convert before they do anything, which is a genuine disadvantage in potency and a genuine advantage in tolerance. All of them act through the same receptors. What differs is speed and the cost of entry. Where people go wrong is dosing. A pea sized amount for the whole face is not a figure of speech, it is roughly the quantity used in the trials, and most people apply three or four times that. Concentration matters far less than the interval between applications. Twice a week at 0.025 per cent tretinoin, or 0.1 per cent adapalene, will outperform a nightly 0.05 per cent that gets abandoned in week three because the skin around the nose is flaking. Build slowly: two nights a week for a month, three for a month, then hold. Many people never need more than four. The purge deserves scepticism. There is a real phenomenon in acne, where existing microcomedones are pushed to the surface faster than they otherwise would appear, and it resolves. It does not explain stinging, tightness or a rash across the cheeks in someone who does not have acne. That is irritant dermatitis, and treating it as a rite of passage is how a barrier gets broken. If skin is sore rather than spotty, reduce the frequency.

None of this works without adjusting what sits either side of it on the shelf. The most common cause of failure is not the retinoid at all, but the three or four other actives someone is still using in the same week, each of them sold as gentle.

Buffering, which means applying the retinoid over a plain moisturiser rather than onto bare skin, reduces irritation without meaningfully reducing effect. The old objection, that it dilutes the dose, has not held up in the comparisons that have been run. Apply to dry skin, wait twenty minutes after washing, and put something bland over the top: a cream with glycerin and a lipid or two, no fragrance, no essential oils, no botanical list running to forty lines. Stop the exfoliating acids while you build up. Glycolic, lactic and salicylic all have their uses, but running them alongside a new retinoid is the most reliable way to arrive at week two with a red, weeping face. The same applies to high concentration vitamin C, physical scrubs, clay masks used more than occasionally, and any device that heats or abrades. Reintroduce one thing at a time after three months, if you still want it. Most people find they do not. The timeline is long and nobody selling anything says so. Acne often looks slightly worse before week six and better by week twelve. Texture and fine lines take longer: twelve weeks is the earliest point at which a change shows in photographs, and the trials demonstrating real dermal collagen change run to a year and beyond. Cost is the part that ought to be reassuring. Prescription tretinoin costs less here than a mid range serum and lasts four months at the correct dose. Over the counter adapalene is comparable. There is no version of this that needs to cost two thousand kroner, and the expensive encapsulated retinols are buying comfort rather than results.