REWE's AI Strategy: From Core Ops to Org Culture
REWE Digital leaders detail their multi-year AI roadmap, distinguishing between core business automation and generative AI enablement. The analysis covers their stage-gate prioritization model, the AI Hub's role in cultural transformation, and the strategic shift toward agentic commerce and hyper-personalization in retail.
Strategic Divergence in AI Implementation
REWE Digital’s approach to artificial intelligence demonstrates a critical distinction between core business automation and organizational enablement. While traditional machine learning models drive invisible efficiencies in demand forecasting and assortment planning, the introduction of generative AI necessitated a parallel strategy focused on cultural transformation. This dual-track approach allows the company to capture immediate financial value from operational AI while simultaneously preparing its workforce for broader technological adoption.
Governance and Prioritization Framework
The company employs a rigorous stage-gate process for AI use case selection, led by a board comprising business unit heads, IT leadership, and the group executive board. This structure ensures that only initiatives with validated business cases and strategic alignment proceed to development. Crucially, REWE manages these initiatives as products rather than projects, forming interdisciplinary teams that are financially self-sustaining. This model prevents the common pitfall of AI projects ending without long-term operational integration, ensuring that AI capabilities are embedded into the core business fabric.
Organizational Enablement and the AI Hub
Recognizing that technical deployment alone is insufficient, REWE established the AI Hub as a center of excellence for enablement, communication, and change management. The Hub’s mandate is to provide direction and support, ensuring employees understand not just how to use AI tools, but what they achieve. This is operationalized through a multiplier program that trains internal ambassadors within specific business units. These ambassadors, allocated approximately two hours per week, serve as localized experts, bridging the gap between central AI strategy and daily operational needs. This decentralized model accelerates adoption by addressing specific use cases within their native contexts.
Future Outlook: Agentic Commerce
Looking ahead, REWE identifies agentic AI as the next major frontier. The company anticipates a shift from reactive digital commerce to proactive, hyper-personalized experiences driven by AI agents. These agents will integrate customer preferences, real-time data, and inventory status to create a unified commerce experience. This transition requires not only technological advancement but also a redefinition of customer interaction standards, emphasizing the need for high-quality, reliable AI interactions to maintain trust and satisfaction.
Key insights
-
REWE distinguishes between core business AI, which is strictly ROI-driven and managed via stage-gate processes, and generative AI, which requires a broader enablement strategy to drive cultural adoption.
Impact: This dual-track approach allows companies to balance immediate financial gains with long-term organizational readiness, preventing adoption bottlenecks caused by cultural resistance.
-
Managing AI initiatives as self-financing products with interdisciplinary teams, rather than temporary projects, ensures long-term sustainability and integration into core operations.
Impact: This model reduces the risk of AI initiatives failing post-pilot by embedding ownership and financial accountability directly into the product lifecycle.
-
The AI Hub functions as a center of excellence for enablement, focusing on providing direction and support to employees, rather than just deploying tools.
Impact: By prioritizing human understanding over tool availability, organizations can significantly increase the effective utilization rate of AI technologies across the workforce.
-
A multiplier program training internal ambassadors within business units decentralizes AI expertise, allowing for faster, context-specific adoption and support.
Impact: This approach reduces the burden on central IT teams and ensures that AI solutions are tailored to the specific needs and workflows of each department.
-
REWE anticipates a shift toward agentic commerce, where AI agents proactively manage customer preferences and interactions, moving beyond reactive digital channels.
Impact: Early adoption of agentic AI in retail could create significant competitive advantages in customer retention and personalization, redefining industry standards.
Action items
-
Establish a cross-functional AI governance board that includes C-suite, IT, and business unit leaders to prioritize use cases based on validated business cases.
Impact: This ensures that AI investments are aligned with strategic goals and have clear paths to measurable financial returns.
-
Transition AI initiatives from project-based management to product-based management, creating interdisciplinary teams with financial ownership.
Impact: This fosters long-term sustainability and integration of AI capabilities into core business processes, avoiding the 'pilot purgatory' trap.
-
Create an AI Hub or center of excellence focused on enablement, communication, and change management to support employee adoption of generative AI.
Impact: This addresses the human side of AI transformation, ensuring that employees have the skills and confidence to use AI tools effectively.
-
Implement a multiplier program that trains internal ambassadors within specific business units to provide localized AI support and guidance.
Impact: This decentralizes AI expertise, accelerating adoption by addressing specific use cases within their native operational contexts.
-
Develop a strategic roadmap for agentic AI, focusing on hyper-personalized customer experiences and proactive AI interactions.
Impact: Positioning the company ahead of the agentic commerce trend can drive significant customer engagement and retention in the digital retail space.
Quotes
“Innovation alleine nicht reicht. Technologien, Plattformen und so weiter, das ist natürlich alles wichtig. Daten, ganz, ganz wichtig. Aber ich glaube, sie entfalten wirklich nur Wirkung, wenn wir Mensch und Organisation auch befähigen, damit auch sinnvoll umzugehen und zu verstehen.”
“AI first eben nicht bedeutet, dass wir möglichst viele Tools, verschiedene Tools anbieten und einführen oder auch möglichst schnell Dinge automatisieren, die ersten sind, die irgendwie den Top-Use-Case für den Kunden am Markt haben, sondern es bedeutet vor allem auch bewusst zu entscheiden, wie wir arbeiten wollen, wie wir mit unserem Kunden interagieren wollen”
“Wir glauben, dass natürlich jetzt erstmal ein bisschen mehr Zeitaufwand notwendig ist, um dieses ganze Wissen sich anzueignen. Wir gehen aber davon aus, dass eben durch verschiedenste Formate die Zeit sich dann in Grenzen halten würde, wenn diese Person dann als Ambassador fungiert.”