What happens before an active ingredient goes into a cream

At a raw materials trade show, someone hands you a folder. Inside there’s a new active ingredient, a name ending with a registered trademark symbol, and a percentage.

The percentage almost always means the same thing: an improvement of something by some percentage over a few weeks. This is the moment when I have to decide whether that number actually means something, and it’s work no one sees because it happens two years before the product reaches a pharmacy.

[INDICE:h2]

The folder

Presentations for active ingredients all look alike. There’s the mechanism of action explained with a colourful diagram, a drawn molecule, a bar chart with two columns—before and after—and an upward-pointing arrow.

The first thing I look at isn’t the tallest bar. It’s the small line beneath the chart, written in eight-point type, where the important information is: how many people, for how long, and above all, on what.

Very often, that line says “in vitro”. It means the result was obtained using cells in culture, in a plate, in direct contact with the active ingredient. It’s useful data: it shows that the molecule does something at a biological level. But between a cell in a dish and a fibroblast beneath the stratum corneum of a real person, there are several layers of skin in between—and it is precisely their job to let nothing through.

The question that brings the conversation to a halt

When the data comes from people, I ask the question I’m really interested in, and I always phrase it the same way.

Was this study conducted on the pure active ingredient or on the finished product?

It sounds like a subtle distinction, but it changes everything. An active ingredient tested at five per cent, in a vehicle specifically designed to help it penetrate, at an optimal pH, is not the same active ingredient used at 0.2 per cent in a cream that also needs to be stable, fragrant, pleasant to use, and shelf-stable for thirty months.

Three times out of four, there is no answer. Not because they’re lying to me: because that data often doesn’t exist. Testing the finished product costs far more than testing the active ingredient, and raw-material suppliers have no reason to do it.

“An active ingredient with excellent supporting literature may do absolutely nothing in the wrong formula. It isn’t the ingredient that works: it’s the formula.”

Then you test it—and that’s where almost all of them die

Let’s say the data convinces me. At that point, the long part begins: the active ingredient goes into a prototype, and we see what really happens.

Very unromantic things happen. An active ingredient that works beautifully at pH six but loses half its activity in a formula that has to stay at pH five. A plant extract that changes colour after three weeks in a heat chamber—and no one sells a product that turns yellow. A texture that feels perfect on the first day but separates after a month. An active ingredient that conflicts with another active ingredient we need more.

And then there’s the worst-case scenario, the one no one talks about: the prototype works, it’s stable, it looks good—but getting the active ingredient to the required concentration would put the final price beyond what the market will bear. Then you have two choices: use it at a token dose so you can list it on the label, or don’t use it at all. I choose the second option, which is why some products I would have liked to make do not exist.

Most of the active ingredients I evaluate die here. Not because they’re bad, but because there is an enormous distance between interesting and usable, and that distance can only be measured by testing.

Good suppliers exist—and you recognise them immediately

I don’t want to give the impression that this is an adversarial relationship, because it isn’t. There are suppliers who do excellent work, and you can recognise them by two things.

First: they also tell you what their active ingredient does not do. They tell you at what concentration it loses effectiveness, what it is incompatible with, and under what conditions it degrades. When someone lists the limitations of their product, I trust everything else in the presentation more.

Second: when they don’t know something, they say they don’t know it. It’s the rarest and most valuable answer. “We don’t have data on the finished product” is a perfectly acceptable sentence. What isn’t acceptable is pretending the question was never asked.

Why the catalogue is small

Every now and then, someone points out that other brands launch ten new products a year, while we launch far fewer.

The reason is all here. For me, a new product means months of evaluating data, developing prototypes, conducting stability tests, making corrections, and running tests. It also means accepting that some of those months will lead nowhere, because the formula doesn’t hold up or because the result doesn’t justify the price.

I don’t believe this is the only way to make cosmetics. But I do believe it’s the only way to look a pharmacist in the eye when they ask me whether a product works, and give them an answer more precise than “yes”.

What does this mean for you, the person buying?

This article isn’t meant to turn you into a formulator. It’s meant to give you one thing: when you read a percentage on a package or in a post, ask yourself what it was measured on.

If it isn’t stated, that doesn’t necessarily mean the product doesn’t work. It means that percentage, on its own, isn’t telling you anything. It’s the same distinction I use when I look at a fashionable ingredient: I wrote about it in relation to PDRN, where the gap between the available data and the story being told is particularly wide.

And if you want to find your way among the molecules yourself, the cosmetic ingredients guide is where I would start.