AB InBev is bigger than Coca-Cola. Most people do not know that.
I met the Head of Innovation Europe at AB InBev some time ago. Three centuries of continuous operations, 500+ global brands — Stella Artois, Corona, Budweiser, Leffe, Jupiler. A machine compounding since 1717, built on a brewery originally founded in 1366.
And they are trying to innovate. Genuinely.
The Structural Problem Has a Name
Clayton Christensen called it the Innovator's Dilemma. Large companies are structurally optimized to serve existing customers, protect margins, and reduce variance. That is what management science is engineered to do. Innovation requires the exact opposite: tolerating variance, operating under deep uncertainty, and sometimes cannibalizing your own core revenue.
Michael Tushman and Charles O'Reilly named the organizational response: Ambidexterity — the capacity to exploit the existing business and explore new opportunities simultaneously.
The strategic question is never whether to do both. The question is how you structure the separation.
AB InBev chose structural ambidexterity. The exploit engine and the explore engine run in separate organizational units, with different operating models, different risk tolerances, and different definitions of success. That is the right architectural call at this scale. But it creates a profound integration problem.
Three Horizons, Actually Operationalized
McKinsey's Three Horizons framework is frequently used as a passive planning slide. AB InBev actually runs it as an operating model:
Open Innovation: The Outside-In Flow
Henry Chesbrough's Open Innovation model distinguishes two directions of knowledge flow:
- Outside-in: Absorbing external ideas, startups, and technologies before competitors do.
- Inside-out: Letting internal IP flow outward through licensing or spinoffs.
AB InBev's model is almost entirely outside-in, and deliberately so:
- ZX Ventures pulls external entrepreneurs inward, funds them, and builds new businesses backed by AB InBev's massive distribution infrastructure.
- BeerGarage brings startups and external research talent inside the firm to co-develop proofs of concept.
- Acquisitions represent outside-in innovation in its most capital-intensive form: buying proven models rather than attempting to replicate them under uncertainty.
Nokia's historical corporate venturing data between 1998 and 2002 is instructive: 70% of new ventures were discontinued, 21% were re-absorbed into the core business, and 16% created new IP. A third of the portfolio generated measurable enterprise value. That is the expected return distribution of a functioning innovation portfolio, not a failure rate.
The Integration Problem Structural Ambidexterity Creates
Here is what does not get discussed enough in boardrooms:
Structural ambidexterity solves the exploration problem. It does not solve the integration problem.
The explore unit and the exploit unit speak fundamentally different languages. They move at different speeds. They measure success with opposing metrics. A ZX Ventures squad measuring weekly active users and a traditional brand manager measuring national volume share are not naturally aligned.
The interface between the two units is where most corporate venturing value gets destroyed. Ideas that survive the explore phase quietly die at the handover. This is not a personality clash — it is a structural fault line: the ambidextrous organization is only as good as the translation layer between its two operating systems.
What the Conversation Revealed
One observation from the discussion stayed with me. When I asked about innovation at AB InBev, the conversation quickly focused on new products for end-consumers. Understandable. But innovation covers a wider surface: business model innovation, process redesign, logistics, data strategy. AB InBev invests in all of this. It just did not surface naturally in conversation.
That gap between structural investment and cultural vocabulary is not a criticism. It is a pattern. In most large organizations, the word innovation collapses to product in everyday speech. The broader definition lives in the strategy decks and the labs, not in how people describe their work to a visitor.
The same week, I met the CEO of Brussels Beer Project. A scale-up, not a giant. Same industry, completely different ambidexterity posture. That contrast deserves its own piece.
For now: structural ambidexterity is the right architecture for a company AB InBev's size. The question worth asking in your own organization is not whether you have innovation labs. It is whether you have a functioning translation layer between your exploit engine and your explore engine. Most do not.
- Christensen, C.M. (1997), The Innovator's Dilemma, Harvard Business School Press.
- O'Reilly, C.A. & Tushman, M.L. (2004), "The Ambidextrous Organization," Harvard Business Review 82(4), 74-83.
- Chesbrough, H. (2003), Open Innovation: The New Imperative for Creating and Profiting from Technology, Harvard Business School Press.
- Baghai, M., Coley, S. & White, D. (1999), The Alchemy of Growth: Practical Insights for Building the Enduring Enterprise, McKinsey & Company.
Bridge the gap between exploration and exploitation.
brainTerms.ai acts as the automated translation layer: converting exploratory innovation bets into structured, board-ready unit economic roadmaps that traditional executive operating models can execute.