Journal  /  Which Hair Protein Is Best? The Number the Label Never Tells You

Which Hair Protein Is Best? The Number the Label Never Tells You

Protein treatments are sold as a choice between sources. Keratin because hair is keratin. Wheat because it is gentler. Silk because it sounds luxurious. Rice because it is having a moment. Pick your protein, the aisle suggests, and you have made the decision that matters.

You have not. The source is close to the least informative thing on that label, and the variable that actually decides the outcome is a number that appears nowhere on the packaging. It decides different things for different proteins, and for one of them it decides something more serious than whether the product works.

The short version. Hydrolysed proteins are broken into fragments, and the size of those fragments determines everything. For keratin, fragment size decides whether the treatment strengthens hair or does nothing measurable: mid- and large-weight peptides cut breakage in controlled fibre testing while free amino acids left hair weaker than no treatment at all. For wheat protein, fragment size decides something else entirely, because large acid-hydrolysed forms have caused genuine wheat allergy through the skin. Same shelf, same two words on the label, entirely different consequences. And the molecular weight is never printed.

What "hydrolysed" is actually doing

Whole protein is far too large to interact with a hair fibre in any useful way. So manufacturers break it apart, using acid, alkali or enzymes, into shorter chains. That is hydrolysis, and it is the whole point of the ingredient.

The catch is that hydrolysis is a spectrum, not a setting. You can break protein into fragments of a few thousand daltons, or grind it all the way down to individual amino acids, or stop somewhere in the middle. Every one of those outputs is legally and truthfully declared with the same INCI name.

Controlled work on chemically relaxed hair tested three keratin hydrolysates spanning that spectrum: roughly 221 daltons, about 2,577 daltons, and around 75,440 daltons. On an ingredient list, all three are Hydrolyzed Keratin. In the lab they behaved like different materials.

For keratin, size decides whether it works

The results were not intuitive, which is what makes them worth knowing.

The mid-weight and high-weight fractions raised the fibre's stiffness and reduced breakage at both low and high humidity, with break stress up by roughly 16 to 19 percent against untreated control. That is a real, measured mechanical gain on hair that had been chemically processed.

The smallest fraction penetrated deepest into the cortex and delivered nothing comparable.

And the free amino acid tested as a reference produced hair that was significantly weaker than untreated control. Not neutral. Worse than nothing.

Sit with that last one, because it inverts the story the category tells. "Amino acids are the building blocks of hair" is the most intuitive sentence in hair care, and in the one controlled comparison available, the building blocks made the fibre worse while the intact peptide chains made it better. Whatever is happening, it is not the hair assembling itself from supplied parts. Chain length is doing the work, and the marketing metaphor points the wrong way.

There is a further wrinkle worth noticing: penetration and benefit came apart completely. The fraction that got deepest helped least. So "penetrates the hair shaft," which appears on a great many boxes, is not the selling point it is presented as.

We grade hydrolysed keratin [B] for reducing breakage on chemically processed hair, and [D] for the claim that it repairs or rebuilds hair structure, because reinforcing a fibre and reconstructing it are different acts.

For wheat, size decides something more serious

Wheat protein occupies the same shelf and performs a different, more modest job: it forms a film on the surface, which makes hair feel smoother and look glossier. Genuine, pleasant, temporary, and gone at the next wash. We grade it [C] for that, and [D] for the claim that it strengthens or rebuilds anything, because a coating is not construction.

But wheat carries a history that almost no product mentions, and molecular weight is again the hinge.

Between 2004 and 2010 a Japanese facial soap containing a large acid-hydrolysed wheat gluten became extremely popular. Then users started developing genuine immediate-type wheat allergy: swelling, hives, in some cases anaphylaxis, frequently triggered afterwards by eating wheat. More than two thousand people were formally documented. Repeated washing had allowed the hydrolysed peptides to sensitise them through the skin, and people who had eaten bread safely their entire lives were made allergic to wheat by a face soap.

Japan convened a dedicated safety committee in 2011. The risk tracked with large, acid-hydrolysed peptides, and expert guidance now holds wheat protein acceptable in cosmetics when the fragments are kept small, under roughly 3.3 kilodaltons.

This is why we grade wheat protein [D] on safety, and the grade needs reading precisely. It does not mean wheat protein is usually dangerous, and it is not a reason to panic about a conditioner you have used for years. It means the ingredient has a proven, documented capacity to do something to you rather than merely for you, which is a category of risk the phrase "natural protein" actively conceals. Anyone with an existing wheat or gluten allergy should take individual medical advice rather than ours.

The symmetry nobody points out

Put the two together and the pattern is almost too neat.

For keratin, low molecular weight is the useless end of the range. For wheat, low molecular weight is the safe end. The same physical property governs both ingredients and governs them in opposite directions, and in neither case is the number disclosed.

That is the honest state of this category. Not that protein treatments are a scam, and not that they are miracle repair. They are real, modest, size-dependent conditioning agents sold on a label that omits the only specification that would let you predict anything.

Side by side, on the grades

Two flat noes on repair, and a shared no on the one thing a shopper would most like to know.

The verdict, laid bare

Protein treatments earn a place in a routine for hair that has been bleached, relaxed or otherwise chemically pushed, and they earn nothing at all for hair that has not.

The reason is mechanical rather than philosophical. These ingredients reinforce a fibre that has lost structure. Where no structure has been lost, there is nothing to reinforce, and a good conditioner delivers most of the same feel for less money. If your hair has never been processed and a protein treatment left it feeling stiff or straw-like, that is not a mysterious "protein overload" so much as a coating applied to a fibre that did not need one.

Where the label defeats you, behaviour is the only available instrument. If a protein product makes your hair feel stronger and better in a way that outlasts a single wash cycle, its fragment profile probably suits your hair. If it makes hair rigid or brittle, it does not, and no amount of reading the ingredient list will tell you why. That is unsatisfying, and it is more useful than pretending the source name predicts the outcome.

The label names the protein. It is the size of the pieces that decides what happens, and nobody prints that.

FAQ

Which protein is best for hair, keratin or wheat? For reducing breakage on chemically processed hair, keratin has the better evidence, with measured tensile gains in controlled fibre testing. Wheat protein is a surface film-former: it improves feel and shine without a comparable mechanical result. But both are outperformed as a variable by fragment size, which neither label discloses.

Do protein treatments repair damaged hair? No. Hair above the follicle is dead keratin with no capacity to heal, so nothing applied to it can restore it. Protein treatments reinforce an existing fibre and wash out; they do not rebuild the structure that bleaching or relaxing destroyed.

Are amino acids better than peptides for hair? The one controlled comparison available says the opposite. Hair treated with a free amino acid was significantly weaker than untreated control, while intact peptide fractions of larger molecular weight were stronger. The "building blocks" framing is intuitive and the data does not support it.

Is hydrolysed wheat protein safe? In the small-fragment forms used in cosmetics today, it is generally well tolerated. The documented sensitisation cases involved large acid-hydrolysed peptides, and guidance now holds the ingredient acceptable when fragments are kept under roughly 3.3 kilodaltons. Anyone with a known wheat or gluten allergy should seek individual medical advice.

Does my hair need protein if it has never been bleached? Usually not. The benefit depends on there being lost structure to reinforce. On virgin hair, most of what you would notice comes from conditioning, at a much lower price.

What is protein overload? The term describes hair that feels stiff or brittle after protein products, and it is community vocabulary rather than a defined finding in the literature. A plausible reading is a coating applied to hair that had no structural deficit to address. Stopping the product resolves it.

---

We do not sell hair products, and no brand pays us to move an assessment. The sensitisation history described here is reported from the published record for awareness and is not medical advice; anyone with a wheat or gluten allergy should seek individual medical guidance.

Full evidence breakdown: Hydrolyzed Keratin · how we grade.