Removing an Ingredient: The S-7 Story

This is the story of an ingredient we took out of S-7, and of how long it took me to be willing to look at it. It is not a flattering story about our process. We are publishing it because the useful part is the part where I was wrong.

The ingredient I liked

Cocamidopropyl betaine — CAPB — is one of the most widely used secondary surfactants in the industry, and it is genuinely good at its job. It stabilises foam, softens the feel of a wash, adds conditioning and body to a formula, and it does all of that at once. It is the reason a lot of sulfate-free shampoos feel like shampoo instead of feeling like soapy water.

I liked it. Through the early and middle stretch of S-7's development it was not a minor ingredient in our formula — it was the main one, at roughly 23% by weight of the surfactant build. Everything else was designed around it.

The problem we hit

There was a low-level itch during testing for a while. Never enough to act on. The kind of thing you notice, file away, and assume is the weather, or the water, or stress.

Then I had a stretch of being properly unwell — a hives-like rash over my body and a general run of bad health. I do not think that episode was caused by anything in this story, and I am not claiming it was. But during it, the itching on my scalp arrived at a completely different intensity than anything before, and it did not stay on my scalp: it showed up wherever the rinse water ran. My scalp felt permanently tight. Then flaking started.

The body illness gradually resolved. Most things improved. The scalp did not. Weeks went by with no sign of it getting better on its own, and that was the part I could not explain away any longer.

Suspecting your own work

So I started actually wanting an answer, and the thought I had been avoiding finally surfaced: surely the thing I am developing is not doing this to me.

I want to be honest about the shape of that thought. It was not a hypothesis. It was something I did not want to be true, phrased as a question. There is a specific kind of blindness that comes from having built something — you will investigate almost anything before you investigate your own work.

What we tested

The method was not clever. I listed every raw material in the formula that could plausibly sensitise someone, then made batches that each left one of them out, and used them on myself. Not on clients, not on colleagues. One person, one variable at a time.

CAPB was not in that queue. I kept it out of the elimination testing until the very end, because I liked it, and because removing it meant rebuilding everything. I want that stated plainly rather than softened: the ingredient I was least willing to test was the one I had the most invested in.

Nothing else changed anything. One by one the candidates came back clean and the itching stayed exactly where it was. Eventually there was nothing left in the queue.

So on March 16 I made a 100 g diagnostic batch with CAPB removed — and with the thickener, the preservative, the fragrance and the protein taken out too, so that there was almost nothing left to be confused by. I wrote the decision rule down before I used it: if the itching stops, CAPB is the cause; if it does not, the remaining surfactants are.

It stopped. Not gradually — it just stopped.

Afterwards I added the hydrolysed protein back into what was left of that batch, on its own, to make sure I had not cleared the wrong ingredient by accident. No itching. That closed it: everything other than CAPB was clean.

What this is and is not

This was one person testing on himself, unblinded, with full knowledge of what was in each batch. There was no patch testing and no clinical diagnosis. I cannot tell you what my immune system was doing, and I am not going to pretend the result is more than what it is.

What supports it: CAPB is a documented contact allergen for a subset of people — that is established, not our finding. The pattern was consistent with a sensitisation-type reaction rather than simple irritation: it started immediately on use rather than building up over days, and it appeared even where only heavily diluted rinse water reached, which is a trace-level exposure.

There is also a wrinkle worth knowing. In many cases of CAPB reaction, the actual culprit appears not to be CAPB itself but an impurity left from its manufacture (dimethylaminopropylamine). Higher-purity grades exist and some people who react to ordinary CAPB do not react to them. We did not go down that road, because sourcing was difficult and because by then I no longer wanted the ingredient in a product I would be washing other people's hair with all day.

What it cost

I should be straightforward about this part. CAPB was not a component I could swap out. It was the axis the whole surfactant system turned on — foam stability, mildness, conditioning, and viscosity were all leaning on it, and no single replacement does all four. Removing it did not damage the formula. It ended it.

Months of work went back to zero, and for a few days I genuinely considered stopping. Not stopping this formula — stopping shampoo development. The thing I had been building was gone and the honest version of the next step was start again.

We started again. The rebuild became the ACD system — cocamidopropyl hydroxysultaine, coco-glucoside and hydrolysed silk protein — and that is what S-7 is today. It took from March to April to get back to a formula worth testing, and considerably longer to get back to one worth selling.

Why a hairdresser hits this first

One more thing, because it explains why this happened to me and not to a customer. Hairdressers are a high-exposure group for CAPB. Not from our own hair — from other people's, ten or twenty heads a day, hands in it every working day. Sensitisation is a function of dose and repetition, and that is a lot of both.

If you are a hairdresser whose hands or scalp have been bothering you and nobody has been able to tell you why, this is worth knowing about. It is not a diagnosis and I am not qualified to give you one — see a dermatologist, and ask about patch testing. But it is a question worth putting in front of them.

Where our products fit

S-7 contains no cocamidopropyl betaine. That is not a marketing position we arrived at by reading the market; it is the residue of the experience above. The cleansing system is sulfate-free and built on cocamidopropyl hydroxysultaine and coco-glucoside instead.

The limits, clearly: a shampoo without CAPB is not hypoallergenic, and we do not claim it is. Every surfactant is capable of bothering someone. If you react to something in shampoo, removing one known allergen improves your odds; it does not make a product safe for you, and it does not tell you what you actually react to. Only testing does that, and the right person to run it is a dermatologist.

If you want to try it, the Travel Size (100 mL) is the full formula, not a thinner sample version.

FAQ

Q. Is cocamidopropyl betaine bad?
A. No, and we would rather not be quoted as saying so. It is a well-established surfactant that the large majority of people use without any problem, and it is in a great many products we would happily recommend. It is a known contact allergen for a subset of people, and one of us turned out to be in that subset. Those are different statements.

Q. Why did it take you so long to test the ingredient you suspected?
A. Because I did not suspect it. I did not want to. That is the actual answer, and it is the reason we wrote this page.

Q. Should I switch shampoos if my scalp itches?
A. Possibly, but do not stop at guessing. Itching has many causes and a shampoo is only one candidate. If it persists, see a dermatologist and ask about patch testing — that gives you a name for what to avoid, which is worth much more than swapping products at random.

Q. Would you use CAPB again in a future product?
A. Not while I am the one washing hair with it every day. That is a personal constraint, not a claim about the ingredient.

Related

Written by Marcy Hasegawa, hairdresser in San Francisco (Hair Caffe). Last updated: September 1, 2026.