Article

Building the AI-Ready Enterprise

Five insights from SAP Sapphire 2025.

By Florian Braun, Juan Francisco (Paco) Jimenez, Numan Waheed, and Tony Pagliarulo

Published on 六月 02, 2025
min read


The Bain delegation just returned from a whirlwind two weeks at SAP Sapphire in Orlando and Madrid. The central theme—“the business AI flywheel”—underscored how enterprise applications are generating mission-critical data that fuels AI embedded into business processes. SAP is positioning its suite-as-a-service strategy to bring clean, contextualized data to life, enabling smarter decision making through Joule, its generative AI copilot. As we reflect on the keynotes, breakouts, and client conversations, five major themes stand out. 1. AI is delivering real value—if the foundation is in place. Agentic AI is gaining traction, and companies should start planning integration into high-impact workflows such as sales and operations planning, category management, financial reconciliation, customer service, and pricing. AI tools from across the ecosystem are also accelerating ERP programs by automating legacy-heavy tasks like process discovery, code conversion and remediation, data mapping, and system testing. Enterprise success depends on a robust data architecture, with orchestration and control layers that span SAP and non-SAP systems. As clients plan their AI roadmap, determining how SAP-native and hyperscaler AI capabilities fit into the broader architecture will be critical to enabling flexibility and scale. It’s also becoming increasingly clear that leadership teams want to “see it, not hear it.” Validated, referenceable use cases with measurable business outcomes are crucial. 2. Clean, connected data is the launchpad for AI at scale. The enterprise data layer is the bottleneck and the enabler for AI, automation, and decision intelligence. Emerging leaders treat data as a strategic asset, not an IT afterthought. SAP’s Business Data Cloud, Digital Discovery Assessment, and enabling tools like LeanIX are helping companies rationalize and govern data estates. And unified, governed data across the enterprise is essential: No usable data = no usable AI. 3. Process redesign is where the value gets captured. Leading organizations are pairing technology programs with process redesign to embrace clean core, modernize their ways of working, and boost agility, freeing up resources to invest where it truly matters. In parallel, these organizations leverage this journey to increase their data integrity and standardization. Process mining tools help identify pain points, but leadership judgment is key to prioritization. One helpful development: SAP Signavio is evolving from a process map to a transformation governance backbone. 4. Business and organization adaptation remain the hidden barriers. While technology is rapidly advancing, organizational readiness lags—and becomes the critical enabler of success. Emerging leaders are adopting an agile delivery model, enabling rapid build-operate iterations on digital capabilities. Capturing the full value of AI requires reimagining end-to-end processes and the operating models that support them. That also means talent strategies must include retraining and reprogramming the workforce at all levels to work with and design AI-driven systems. Separate but related: As “out-of-the-box AI” use cases become common, organizations must develop the judgment to buy vs. build and integrate solutions without fragmenting the stack or duplicating efforts. 5. Choose the right channel to deliver AI. As organizations integrate AI into their enterprise stack, the focus is shifting from who provides the model to how AI is embedded effectively into business processes. Proximity to data matters: Embedding AI closer to core workflows can improve efficiency, reduce integration friction, and accelerate value realization. Platforms like SAP will aim to support a wider range of AI orchestration and deployment options and hold the potential to help architect and manage a network of AI agents. Many companies are exploring hybrid models—leveraging SAP’s native tools where advantageous while maintaining flexibility to integrate other platforms and third-party or hyperscaler AI solutions. As the landscape matures, organizations will need clear frameworks to evaluate total cost of ownership, interoperability, agility, and long-term business value across the enterprise. Building a strong data foundation, understanding the implications for your operating model, and developing a robust change plan to ensure adoption are some of the key AI prerequisites that Bain helps our clients get right. Ultimately, the goal is not just to deploy AI but to ensure it is embedded in the right places, aligned to business outcomes, and supported by an architecture that enables long-term agility and control. Learn more about how Bain can help with ERP transformation.

The Bain delegation just returned from a whirlwind two weeks at SAP Sapphire in Orlando and Madrid. The central theme—“the business AI flywheel”—underscored how enterprise applications are generating mission-critical data that fuels AI embedded into business processes. SAP is positioning its suite-as-a-service strategy to bring clean, contextualized data to life, enabling smarter decision making through Joule, its generative AI copilot.

As we reflect on the keynotes, breakouts, and client conversations, five major themes stand out.

1. AI is delivering real value—if the foundation is in place.

Agentic AI is gaining traction, and companies should start planning integration into high-impact workflows such as sales and operations planning, category management, financial reconciliation, customer service, and pricing. AI tools from across the ecosystem are also accelerating ERP programs by automating legacy-heavy tasks like process discovery, code conversion and remediation, data mapping, and system testing.

Enterprise success depends on a robust data architecture, with orchestration and control layers that span SAP and non-SAP systems. As clients plan their AI roadmap, determining how SAP-native and hyperscaler AI capabilities fit into the broader architecture will be critical to enabling flexibility and scale.

It’s also becoming increasingly clear that leadership teams want to “see it, not hear it.” Validated, referenceable use cases with measurable business outcomes are crucial.

2. Clean, connected data is the launchpad for AI at scale.

The enterprise data layer is the bottleneck and the enabler for AI, automation, and decision intelligence. Emerging leaders treat data as a strategic asset, not an IT afterthought. SAP’s Business Data Cloud, Digital Discovery Assessment, and enabling tools like LeanIX are helping companies rationalize and govern data estates. And unified, governed data across the enterprise is essential: No usable data = no usable AI.

3. Process redesign is where the value gets captured.

Leading organizations are pairing technology programs with process redesign to embrace clean core, modernize their ways of working, and boost agility, freeing up resources to invest where it truly matters. In parallel, these organizations leverage this journey to increase their data integrity and standardization. Process mining tools help identify pain points, but leadership judgment is key to prioritization. One helpful development: SAP Signavio is evolving from a process map to a transformation governance backbone.

4. Business and organization adaptation remain the hidden barriers.

While technology is rapidly advancing, organizational readiness lags—and becomes the critical enabler of success. Emerging leaders are adopting an agile delivery model, enabling rapid build-operate iterations on digital capabilities. Capturing the full value of AI requires reimagining end-to-end processes and the operating models that support them. That also means talent strategies must include retraining and reprogramming the workforce at all levels to work with and design AI-driven systems. Separate but related: As “out-of-the-box AI” use cases become common, organizations must develop the judgment to buy vs. build and integrate solutions without fragmenting the stack or duplicating efforts.

5. Choose the right channel to deliver AI.

As organizations integrate AI into their enterprise stack, the focus is shifting from who provides the model to how AI is embedded effectively into business processes. Proximity to data matters: Embedding AI closer to core workflows can improve efficiency, reduce integration friction, and accelerate value realization. Platforms like SAP will aim to support a wider range of AI orchestration and deployment options and hold the potential to help architect and manage a network of AI agents. Many companies are exploring hybrid models—leveraging SAP’s native tools where advantageous while maintaining flexibility to integrate other platforms and third-party or hyperscaler AI solutions. As the landscape matures, organizations will need clear frameworks to evaluate total cost of ownership, interoperability, agility, and long-term business value across the enterprise.

Building a strong data foundation, understanding the implications for your operating model, and developing a robust change plan to ensure adoption are some of the key AI prerequisites that Bain helps our clients get right. Ultimately, the goal is not just to deploy AI but to ensure it is embedded in the right places, aligned to business outcomes, and supported by an architecture that enables long-term agility and control.

Learn more about how Bain can help with ERP transformation.