Collagen Scores a Zero as a Protein — and Doesn't Become Your Skin. Here's Why It Still Works

Rate collagen as a protein and it scores a flat zero. Protein quality (PDCAAS) is capped by the amino acid a protein is most short of, and collagen contains no tryptophan at all — so on that scale it lands on 0 (Paul, Leser & Oesser 2019). And the headline promise — that swallowing collagen tops up the collagen in your skin — isn't how digestion works. Eaten collagen is broken down into amino acids and small peptides before anything is absorbed; it does not travel to your face intact. Your body decides what to build.

So it's fair to ask whether the whole category is a con. It isn't — and this is where the honesty cuts both ways. In pooled randomized trials, hydrolyzed collagen peptides produce a real, if modest, improvement in skin elasticity (effect size 0.72) and hydration (0.63) across 26 studies and 1,721 people (Pu 2023), plus a smaller signal for knee-osteoarthritis pain (Luo 2023). The plausible mechanism isn't deposition — it's signaling: a collagen-derived peptide called Pro-Hyp appears in the blood and nudges your own skin cells to make more collagen (Shigemura 2009). One honest caveat we won't skip: much of this evidence base is manufacturer-funded, and in the highest-quality independent trials the skin effect shrinks.

The one mistake that wastes your money

Two completely different products are sold under the single word "collagen," and they are not interchangeable. For skin and hair, you want hydrolyzed peptides at studied doses — roughly 2.5–10 g a day (Proksch 2014). For osteoarthritis, the evidence points to a different molecule entirely: undenatured type-II collagen (UC-II) at just 40 mg a day, which in one trial beat both placebo and glucosamine-plus-chondroitin on knee-pain scores (Lugo 2016). Buy the peptide tub hoping to fix arthritic knees, or the 40 mg UC-II pill hoping to firm your skin, and you've bought the wrong thing.

The cheap upgrade the label leaves out

Your body can't build collagen without vitamin C — it's the rate-limiting cofactor for the enzymes that assemble it — and most powders don't include any. So the highest-leverage move is simple: take collagen peptides with a vitamin-C source, at the studied dose, and match the form to your actual goal. That's how we scored the field — form-and-dose match to the job first (30%), then dose accuracy, third-party testing and cost per serving. One clear pick per goal.