A percentage only helps you if it refers to one named active at a known strength. When a number adds together different ingredients — or counts things that are not retinoids at all — it stops being a dose and becomes a marketing figure. Two products can both show a large number and deliver very different amounts of real retinoid.
HPR (hydroxypinacolone retinoate) is a genuine retinoid that appears to bind skin's retinoid receptors directly, without needing to convert first. That is chemically interesting. It does not automatically make a product stronger or better than plain retinol, and head-to-head evidence at matched doses is limited.
A useful habit is to look for one active with one clear number. If a label will not give you that, treat the big figure as advertising, not information.
| Key active | HPR + retinol (true dose undisclosed) |
| Texture | Silicone serum |
| Best for | See below |
| Approximate price | $65–$85 |
It is a well-textured serum and people like using it. The silicone base is elegant, and it does not sting the way a genuinely high-dose retinol would.
HPR (hydroxypinacolone retinoate) is a real ingredient, and it is interesting: it appears to bind retinoid receptors directly without needing conversion. In principle that is attractive.
There is nothing dangerous in the bottle.
Irritation is expected, not a sign it is "working." Redness, flaking and stinging in the first weeks are common with any real retinoid. Start twice a week and build up. If it is still angry after a few weeks, the dose is too high for you.
The "6.5% retinoid blend" is the least meaningful number on this page, and here is how it is built:
Roughly 5% is a "retinoid ester complex," 1% is a "liposomal retinol blend," and 0.5% is blue-green algae described as having "retinoid-like activity."
Blue-green algae is not a retinoid. Adding it to a retinoid percentage is not a rounding error — it is a category error. And the two genuine components are themselves complexes, not pure actives: the actual HPR is reported at around 0.5%, and the real retinol is a fraction of that 1% "blend."
So "6.5%" is not a potency figure. It is an addition of things that should not be added. Compare: Drunk Elephant says 1.0% and means 1.0%.
The HPR evidence is thinner than the marketing implies. There is no independent, randomized, matched-dose trial of HPR versus retinol. The claim that it matches retinol's benefits without the irritation is plausible and unproven.
Worth knowing about the brand: Sunday Riley entered into an FTC consent order over employees posting fake reviews of its products on Sephora between 2015 and 2017. That is a fact about how this company markets.
$85, and it contains denatured alcohol.
Water, Propanediol, Dimethyl Isosorbide, Triheptanoin, Dimethicone, Glycerin, Pentylene Glycol, Hexyldecanol, Disiloxane, Cetyl PEG/PPG-10/1 Dimethicone, Diheptyl Succinate, Propylene Carbonate, Hydroxypinacolone Retinoate, PPG-24-Glycereth-24, Phenyl Trimethicone, Retinol, Ubiquinone, Honey Extract, Sodium Hyaluronate Crosspolymer, Algae Extract, Zingiber Officinale Extract, Opuntia Ficus-Indica Fruit Extract, Bisabolol, Butyrospermum Parkii Butter, Phospholipids, Saccharomyces Cerevisiae Extract, Lecithin, Sodium Acrylates Copolymer, Hydrogenated Polyisobutene, 1,2-Hexanediol, 4-T-Butylcyclohexanol, Polyglyceryl-10 Stearate, Caprylyl Glycol, Helianthus Annuus Seed Oil, Sorbitol, Cetylhydroxyproline Palmitamide, Polysilicone-11, Benzoic Acid, Dehydroacetic Acid, Sodium Phytate, Glyceryl Polyacrylate, Polysorbate 80, Hydroxyphenyl Propamidobenzoic Acid, Stearic Acid, Brassica Campestris Sterols, Pvp, Capryloyl Glycerin/Sebacic Acid Copolymer, Sodium Benzoate, Alcohol, Potassium Phosphate, Decyl Glucoside, Phenoxyethanol, Hexylene Glycol, Potassium Sorbate, Tocopherol
1. Human Skin Aging and the Anti-Aging Properties of Retinol (https://pubmed.ncbi.nlm.nih.gov/38002296/) — Backs the general mechanism/efficacy framing that topical vitamin-A retinoids act on skin via retinoid signaling and improve texture, fine lines and dermal/epidermal thickness, i.e. a 'real retinoid' works through retinoic-acid receptor pathways. 2. Retinoid stability and degradation kinetics in commercial cosmetic products (https://pubmed.ncbi.nlm.nih.gov/33206444/) — Confirms hydroxypinacolone retinoate is a real, notably stable cosmetic retinoid, and that measured retinoid content was lower than declared in some products, supporting the caution that a serum's true active dose can be undisclosed/uncertain. 3. The synergistic effect of retinyl propionate and hydroxypinacolone retinoate on skin aging (https://pubmed.ncbi.nlm.nih.gov/36762391/) — Shows HPR is a genuine retinoid that engages retinoid receptor (RARB) gene expression and gives anti-aging benefit while causing markedly less irritation than an equivalent concentration of retinol, supporting that HPR is real, interesting, and gentler than high-dose retinol. 4. Prospective, randomized, double-blind assessment of topical bakuchiol and retinol for facial photoageing (https://pubmed.ncbi.nlm.nih.gov/29947134/) — Randomized trial in which retinol users reported more facial skin scaling and stinging, supporting that redness, flaking and stinging are common, expected effects of a real retinoid in the early weeks.