Category Language: Why NIVEA's Strength Isn't Producing Growth

NIVEA declined by 7.9% in 2026.

That is the reported number for one of the strongest brands in its category. Enormous familiarity. Reach almost nobody in skincare can match. Shelf presence in markets where challenger brands have to buy their way in.

Sounds counterintuitive. It isn't.

Familiarity is not differentiation. A brand can be the first name you think of and still give you no reason to pick it up today. Those are two different assets, and most brand tracking only measures the first one.

What the Double Jeopardy Test showed

We ran a first Double Jeopardy Test across mass-market skincare. NIVEA landed very close to the generic category position.

The test asks one question: is your stated position distinct from what a large language model would propose for your category?

That sounds like a gimmick until you think about what a language model actually does. It is trained to predict the most likely next answer. Feed it a category and it returns the average of everything ever written about that category. So if an AI would have written your positioning unprompted, you are not sitting on a strategy. You are sitting on the category mean, and you paid an agency for it.

This is the practical edge of what I have called the Double Jeopardy Law of AI. The model regresses to the mean. The organisation regresses to the mean, because it approves what most people in the room agree with. Two forces, same direction. Generic advice is not neutral. It is a warning sign, because growth comes from exceptional customer understanding, not commonly known customer understanding.

One caveat, up front and not in a footnote: this is an outside-in diagnostic built on public brand communication, not on internal NIVEA data. I would not treat it as an answer. I would treat it as a question worth testing.

The category language

Here is what mass-market skincare sounds like:

• clinical proof
• dermatologist credentials
• ingredients
• hydration
• before-and-after
• “for every skin type”

All sensible. Every one of those claims is defensible, testable, and signed off by legal. That is exactly the problem. Almost everyone says the same thing.

I call this category language: the shared vocabulary a category converges on because each individual claim is reasonable. Fluency in it buys credibility. It cannot buy differentiation, because the words belong to the category and not to the brand.

And there is a second assumption buried under all of it, the one that actually does the damage. Category language assumes the shopper is running a rational proof comparison in front of the shelf.

She isn't.

In everyday skincare the decision takes seconds. It is made from memory. From feel. From self-image. From what seems right in that moment. Nobody is weighing dermatologist credentials against hydration claims at 8:40 on a Tuesday. The decision has already happened by the time the rational mind shows up to justify it, which is the whole reason we measure this with implicit methods instead of asking people.

The competitor nobody is measuring

That changes the growth question entirely.

The real competition may be less “NIVEA vs. Dove vs. Garnier”. It may be:

Look at that list. Two of those three are not brand choices at all. They are a non-decision and a default.

Now look at your brand tracker. It reports share of a market defined as people who bought something in the category. The skip does not appear. The confusion does not appear. The shopper who reached for whatever felt safe shows up as a brand preference, when what you actually measured was the absence of a reason to think.

Most brand trackers are built to answer a question your growth does not depend on. That is not a tracking bug. It is a framing error, and it is expensive, because it makes the category look saturated when a large part of it was never activated.

What that much trust could be spent on

Here is where the story turns from a problem into an opportunity, and it is a bigger one than another product claim.

A brand with NIVEA's reach could own something the category has left lying on the floor: guidance without complexity.

Helping people feel they know what their skin needs. Without turning skincare into a ritual, a science project, or another source of judgement.

Read the three alternatives above again. “I don't know what I need” is not an objection to a product. It is an unmet job, sitting in plain sight, and the category answers it with more ingredient names. Decades of trust, currently being spent to say what everyone else is saying.

That is the kind of hypothesis I would pressure-test next. Not announce. Test.

Garnier could be next

None of this is a NIVEA problem. The same diagnostic flags the same pattern next door at Garnier: the same category language, the same short distance to the generic position.

Which tells you it is a category problem. When every brand optimises the claims it can prove, the whole category converges, and convergence is how a market full of strong brands ends up growing slower than its own population. One brand breaks out of that. The rest split the middle.

This is also where it differs from the premium-stretch question I wrote about earlier. There, the issue was whether a mass brand's associations can carry it upmarket. Here, the brand has not left its own category. It is standing exactly where it belongs, saying exactly what its neighbours say.

So: strip the logo off your last campaign. Hand it to your two closest competitors. Could either of them run it unchanged?

If the honest answer is yes, familiarity is not your asset. It is your anaesthetic.

I'm reading the replies.

Category language: frequently asked questions

What is category language in marketing?

Category language is the shared set of claims every brand in a category uses because each one is individually sensible. In mass-market skincare that means clinical proof, dermatologist credentials, ingredients, hydration, before-and-after results and suitability for every skin type. Each claim is defensible. None of them separates one brand from another, because almost everyone makes them. Fluency in category language buys credibility and costs differentiation.

Why can a strong, familiar brand still lose market share?

Because familiarity and differentiation are different assets. Familiarity gets a brand into the consideration set; it does not give a shopper a reason to reach for it today. A brand that spends its familiarity restating claims the whole category makes is recognised and interchangeable at the same time. Reported figures put NIVEA down 7.9% in 2026 despite enormous reach and shelf presence, which is what that pattern looks like in a sales report.

What is the Double Jeopardy Test?

The Double Jeopardy Test checks whether a brand's stated position is distinct from what a large language model would propose for that category. A language model is trained to predict the most likely answer, so it reproduces the category average. If an AI would have written your positioning unprompted, you are standing on the average. SUPRA offers the test free and ungated.

How can a brand find out whether it is stuck in category language?

Start outside-in: run the Double Jeopardy Test on public brand communication to see how far the position sits from the generic category answer. That result is a hypothesis, not a verdict. Confirm or kill it with implicit research into what actually drives choice, SKU-level price-demand curves, occasion-based portfolio analysis, and creative tested as a growth variable rather than as an execution detail. Related terms are collected in the SUPRA glossary.

Dr. Frank Buckler advises leadership teams on growth decisions that depend on why customers choose. He founded SUPRA, holds a PhD in marketing and was among the first to apply causal AI to marketing questions. His latest book is THE TOP 5%.