Why Brand Strategy Should Start With Demand, Not Categories
Most marketers still begin with a category. Athletic footwear. Skin care. Search engines. Beer. Banking.
Categories are useful for reporting. They let companies calculate market share, identify competitors, organize portfolios, measure penetration, and track performance.
But they can be a poor starting point for strategy.
A category is primarily a supply-side construct. It groups together companies that make or sell similar things. Consumers do not organize their lives that way.
People have things they want to accomplish. Problems they want to solve. Situations they find themselves in. The alternatives they consider can cut across several categories.
That distinction matters when the way demand is satisfied changes. A company can remain very strong in its category and still lose influence over the demand that created the category in the first place.
Nike, Nivea, and Google show three versions of the same problem.
All three have extraordinary awareness, penetration, and availability. By conventional measures, these are very strong brands.
Yet each has faced a harder question:
What if the place where demand is shaped has moved?
Start with demand, not the category
I first made this argument almost twenty years ago.
In Hidden in Plain Sight, published by Harvard Business School Press in 2007, I argued for a demand-first approach to growth. The starting point was not the product, category, or conventional market segment. It was how people actually lived: the activities they carried out, the problems they faced, the episodes they moved through, and the behaviors that were changing. I called the resulting picture a demand landscape.[1]
The American Marketing Association gave the book its Leonard L. Berry Marketing Book Award in 2008.
The distinction was deliberate. In an HBR IdeaCast at the time, I argued that successful companies develop a blind spot when they see the market through existing products, business units, and categories. The first step in the Demand-First Innovation and Growth model was to map what I called the “ecosystem of consumer demand.”
Apple was a useful example. Sony could define the market as portable music players and ask how to build a better Walkman. Apple looked at the demand around music: how people discovered it, acquired it, organized it, carried it, and listened to it. That led to a very different view of the opportunity.
I now use the term demand spaces for these concentrations of demand: a need, a situation, an episode, and the outcome a person is seeking. I have recently argued that brands and capital should be allocated against these spaces instead of assuming that the existing category is the permanent unit of competition.[2]
Mats Georgson has made a related argument. He has repeatedly challenged categories as the basic unit of marketing analysis. People have goals. The alternatives they consider depend on the situation, not on the category a company uses to report its business. He has recently described this as “demand precedence”: demand exists before the products and categories built around it.
The behavioral research points in the same direction. Lawrence Barsalou showed that people form “ad hoc categories” around particular goals. Later consumer research by Ratneshwar, Barsalou, Pechmann, and others showed that personal and situational goals change which alternatives people see as relevant, including alternatives that sit in different product categories.
The implication is straightforward:
A category tells us how companies currently organize supply. A demand space tells us what people are trying to accomplish.
When those two views line up, category analysis works well enough. When they separate, the category can become a trap.
Where is demand being shaped?
A second part of my work helps explain what happens next.
In The Interaction Field, published in September 2020, I argued that companies increasingly create value through interactions among customers, partners, suppliers, communities, and other participants. A simple company-to-customer transaction misses much of what is happening.[3]
That raises a practical question.
Which interactions have disproportionate influence over what people consider, choose, and do next?
I borrow the term control point from Sangeet Paul Choudary’s work on platforms. I interviewed him in 2019 about the shift from pipelines to platforms. His work focuses on the core interaction around which a platform organizes participants and the points through which the platform can shape that interaction. These points can include data, standards, reputation systems, algorithms, and access rules.[4]
The idea applies well beyond platform businesses.
For brand strategy, I would define a control point as an interaction that has disproportionate influence over how demand is interpreted, which alternatives are considered, and what gets chosen.
The control point can be a retailer, an expert, a community, a technology, or an AI agent. In some cases, the brand itself becomes the place where a demand space is organized.
Nike: the wrong control point
Nike is the clearest example of a company that appears to have misread the control point.
Under John Donahoe, Nike accelerated its Consumer Direct strategy. In 2020, the company announced its Consumer Direct Acceleration program, with more emphasis on digital and Nike-owned channels.
The logic was easy to understand.
Why give retailers part of the margin when Nike could sell directly, own more customer data, and control more of the experience?
Nike wanted more control.
The problem was that it appears to have misidentified what mattered most.
Retailers such as Foot Locker and Dick’s Sporting Goods were more than places to complete a sale. They were places where people discovered products, compared brands, and learned what was gaining relevance. They put Nike next to Adidas, New Balance, Hoka, and On. They helped answer a question that comes before “Where should I buy Nike?”
The earlier question is: “What shoe should I buy?”
As Nike pulled back from wholesale, competitors gained more room in those environments. By 2024, Nike was repairing retailer relationships and trying to restore product innovation while On, Hoka, and others had gained ground.
The financial decline was substantial. Nike’s fiscal 2025 revenue fell 10 percent. Nike Direct revenue fell 13 percent, including a 20 percent decline in Nike Brand Digital. Wholesale revenue fell 7 percent.
DTC did not cause all of Nike’s problems. Product innovation, fashion cycles, China, and stronger competitors also mattered.
But the control-point mistake still matters.
Nike treated the transaction as the control point. The more valuable control point sat earlier, where people discovered, compared, and chose.
Nike gained more control at the end of the purchase process while weakening some of the interactions that created demand for Nike in the first place.
Nivea: when the control point moves
Nivea presents a different problem.
For decades, Nivea occupied an unusually strong demand space around trusted, accessible, everyday skin care.
The blue tin, mass distribution, and decades of familiarity reinforced that position. If someone wanted a trustworthy product for basic skin care, Nivea was an obvious answer.
But more skin care decisions now begin with a specific problem.
Consumers ask: What should I use for pigmentation? How do I repair my skin barrier? Which ingredient works against visible aging? What is right for sensitive skin? What would a dermatologist recommend?
Once the consumer starts there, the competitive field changes.
The consumer is not initially choosing among brands in a category called “skin care.” She is trying to solve a problem. Ingredients, dermatologists, creators, pharmacies, search engines, and specialist brands can all shape which solutions are considered.
Beiersdorf’s own numbers make the question worth asking.
Nivea grew only 0.9 percent organically in 2025. Beiersdorf’s Derma business, led by Eucerin and Aquaphor, grew 11.7 percent. In the first half of 2026, Nivea declined 6.8 percent while Derma grew another 7.8 percent. Beiersdorf has put Nivea into an 18-month turnaround program.
Short-term factors matter, including trade disruption, destocking, weather, and regional conditions. Beiersdorf also says Nivea sell-out remained positive in the first half of 2026. The numbers do not prove a single strategic cause.
Still, compare how the two brands enter demand.
Eucerin increasingly enters through a skin problem and a scientific answer. Ingredients such as Thiamidol and Epicelline give the brand authority around specific needs. Beiersdorf has also expanded Eucerin into new geographic and category white spaces.
Nivea historically entered through trusted everyday care.
The question is whether enough valuable skin care demand has moved toward specific problems, ingredients, and expert authority that Nivea’s traditional control point has become less important.
Nike weakened a control point it already had. Nivea may show something more difficult to see: the brand can remain famous, trusted, and widely available while the control point moves somewhere else.
Google, AI, and the control point upstream
AI makes the point even clearer.
Google Search has one of the strongest market positions ever built. Its penetration, awareness, and access are enormous.
If the unit of analysis is the search-engine category, Google looks almost unassailable.
But “search engine” is the category. It is not the demand.
People wanted to find something, understand something, decide something, or get something done.
Search was one way to satisfy those demands.
ChatGPT, Gemini, Claude, and other AI systems offer another way.
The old path was often: intent → search → links → websites → comparison → decision.
The emerging path can be: intent → AI → interpretation → alternatives → recommendation → decision → action.
Google itself now describes Gemini on Android as part of an “intelligence system” that can take a user’s intention into action.
OpenAI is moving in the same direction. In commerce, ChatGPT can help a user describe a need, compare alternatives, and decide what to buy instead of simply returning a page of links.
The competition is larger than a contest between AI models.
It is a contest to become the place where a person starts with an intent: “Help me accomplish something.”
Whoever becomes the habitual starting point can gain an extraordinary position between human intent and action.
Why availability can give false comfort
This is where conventional brand measures can give management false comfort.
Byron Sharp, Jenni Romaniuk, and the Ehrenberg-Bass Institute have made an important contribution by showing the role of mental and physical availability. Mental availability concerns whether a brand comes to mind in buying situations. Physical availability concerns whether it is easy to find and buy.
Both matter.
But they answer a narrower question than many managers assume.
They tell us how available a brand is within a set of buying situations.
They do not tell us whether those buying situations are changing, whether people are entering the market through a different demand space, or whether someone else now shapes the choice earlier.
Nike did not become unknown. Nivea did not disappear from shelves or memory. Google does not have an awareness or access problem.
Their problem is elsewhere.
A brand can have enormous mental and physical availability and still lose influence over the interaction where demand is being formed.
This is why I think the debate over salience versus meaning, or availability versus differentiation, starts too late. Brand strategy should begin further upstream, with demand and the situations in which a brand becomes relevant.
Availability matters once the buying situation is defined. Strategy also has to ask whether the buying situation itself is changing.
The danger of measuring yesterday’s market very precisely
This is one reason successful companies can be surprised.
Categories are easy to measure. Market share, penetration, awareness, and distribution are all visible and familiar.
The numbers can look reassuring.
They can describe the existing market very precisely while missing the formation of a different one.
A category view asks: How strong are we against the companies that sell what we sell?
A demand view asks: What is the customer trying to accomplish? What alternatives can satisfy that demand? Who shapes the decision?
That produces a different competitive map. It also shows where the control points sit.
The strategic question
I would put the argument this way.
Categories tell us how companies organize supply. Demand spaces tell us what people are trying to accomplish. Control points tell us where that demand is shaped into choice.
Brand strategy has to understand all three.
The mistake is to assume that enormous penetration, mental availability, and physical availability mean that a brand controls its future.
Nike suggests otherwise. Nivea may be showing the same problem in a different form. The contest between Google and AI could make the point on a much larger scale.
Strong brands rarely disappear overnight. Awareness remains. Distribution remains. Distinctive assets remain familiar.
What can change much faster is the importance of the interaction they dominate.
Consumers start somewhere else. They ask a different question. They trust a different expert. A new intermediary interprets their intent before the brand enters the picture.
By the time conventional brand metrics show the problem, the strategic change may already have happened.
So the question I would ask management is not simply: How mentally and physically available is our brand?
I would ask: Where is demand being formed? Who shapes it there? Do we still have influence over the interactions that matter?
That is where the future value of the brand is being decided.
References and notes
- Erich Joachimsthaler, Hidden in Plain Sight: How to Find and Execute Your Company’s Next Big Growth Strategy, Harvard Business School Press, 2007. The book received the American Marketing Association’s 2008 Leonard L. Berry Marketing Book Award. https://www.ama.org/the-leonard-l-berry-marketing-book-award/
- For recent applications of demand spaces, see Erich Joachimsthaler, “What AI Teaches Us About Strategy and Brands,” March 6, 2026; “Rethinking Brand Allocation for Economic Profit,” March 27, 2026; “The Henkel Olaplex Deal: What You Need to Know,” March 31, 2026; and “Capital Allocation Doesn’t Create Value. Future Demand Does,” June 22, 2026. https://vivaldigroup.com/what-ai-teaches-us-about-strategy-and-brands/
- Erich Joachimsthaler, The Interaction Field: The Revolutionary New Way to Create Shared Value for Businesses, Customers, and Society, PublicAffairs, September 2020. https://vivaldigroup.com/wp-content/uploads/2020/07/Interaction-Field_Intro.pdf
- Sangeet Paul Choudary, discussion with Erich Joachimsthaler, Vivaldi, “Sangeet Paul Choudary: The New Rules of Business,” February 6, 2019. https://vivaldigroup.com/future-platform-thinking-interview/
- Mats Georgson, writings on categories and “demand precedence,” including his LinkedIn discussion of why conventional categories do not explain real consumer choice. https://www.linkedin.com/posts/matsgeorgson_categories-are-useless-for-explaining-real-activity-7400178530036105216-U2fV
- Lawrence W. Barsalou, “Ad Hoc Categories,” Memory & Cognition, 11(3), 1983, pp. 211–227; S. Ratneshwar, C. Pechmann, and A. D. Shocker, “Goal-Derived Categories and the Antecedents of Across-Category Consideration,” Journal of Consumer Research, 23(3), 1996, pp. 240–250; S. Ratneshwar, L. W. Barsalou, C. Pechmann, and M. Moore, “Goal-Derived Categories: The Role of Personal and Situational Goals in Category Representations,” Journal of Consumer Psychology, 10(3), 2001, pp. 147–157.
- Byron Sharp, How Brands Grow, Oxford University Press, 2010; Jenni Romaniuk’s work on mental availability and Category Entry Points; and Ehrenberg-Bass Institute materials on mental and physical availability. https://marketingscience.info/news-and-insights/mental-availability-is-not-awareness-brand-salience-is-not-awareness
- Nike, “Consumer Direct Acceleration,” 2020; Reuters reporting on Elliott Hill and retailer relationships, September 20, 2024; and Nike fiscal 2025 results. https://www.reuters.com/business/retail-consumer/nikes-new-ceo-may-look-fix-retailer-ties-sales-revival-push-2024-09-20/ ; https://investors.nike.com/investors/news-events-and-reports/investor-news/investor-news-details/2025/NIKE-Inc–Reports-Fiscal-2025-Fourth-Quarter-and-Full-Year-Results/default.aspx
- Beiersdorf, full-year 2025 and first-half 2026 results. Nivea grew 0.9 percent organically in 2025 versus 11.7 percent for Derma; in the first half of 2026, Nivea declined 6.8 percent while Derma grew 7.8 percent. https://www.beiersdorf.com/newsroom/press-information/all-press-releases/2026/08/03-half-year-results-2026
- Google, Android and Gemini strategy, 2026; OpenAI, product discovery in ChatGPT, 2026. https://blog.google/products-and-platforms/platforms/android/android-show-io-edition-2026/ ; https://openai.com/index/powering-product-discovery-in-chatgpt/