Snowflake and SAP are bringing their technologies closer together in a partnership designed to make enterprise data more accessible, connected, and useful for AI. The collaboration combines the Snowflake AI Data Cloud with SAP’s Business Data Cloud (SAP BDC), giving organizations a unified way to use their structured business data alongside external datasets, and to build AI-driven solutions with enterprise-grade governance and speed.
Strengthening the Enterprise Data Fabric
The joint offering enhances SAP’s open data ecosystem by embedding the Snowflake AI Data Cloud directly within SAP BDC. Customers can now bring together semantically rich SAP application data and Snowflake’s scalable data processing, AI, and collaboration capabilities. The goal is to let enterprises activate AI with full business context, avoiding the need for complex data duplication or manual integration work.
To support this, Snowflake and SAP have announced two new products: SAP Snowflake, a certified SAP Solution Extension, and SAP Business Data Cloud Connect for Snowflake, a bidirectional cloud service.
With SAP Snowflake, customers can work with semantically modeled SAP data that carries built-in business meaning directly in Snowflake. The integration happens in near real time through a zero-copy connection, allowing organizations to build AI applications that combine SAP and non-SAP data while maintaining unified governance and performance.
SAP BDC Connect for Snowflake extends the same flexibility in the other direction. Companies already using Snowflake can link their existing instances with SAP BDC to gain seamless, real-time access to SAP data products. The zero-copy architecture means the data stays in place while remaining instantly available for analytics, model training, and application development.
How Enterprises Are Using the Partnership
Early adopters across industries including retail, manufacturing, healthcare, life sciences, and financial services are already using the new integrations to improve operations, planning, and customer experience.
1. Predictive supply chains
In supply chain management, enterprises can publish real-time inventory and shipment data from SAP as a data product in Snowflake. By combining it with external datasets—such as third-party logistics or weather information from the Snowflake Marketplace—organizations can create a “live logistics” data product that maps operational data with external conditions. AI agents built within Snowflake can then predict potential disruptions, such as weather-related port closures, and automatically model the downstream impact on inventory and production. Those insights can be surfaced back in SAP through SAP Joule, the company’s AI copilot, helping planners prevent stockouts and improve on-time delivery.
2. Continuous financial planning
Financial and sales data from SAP can be shared to Snowflake and blended with marketing and CRM data to form a “unified performance” dataset. With this consolidated data, analysts can build predictive revenue models and use natural language queries to test different business scenarios such as adjusting marketing spend or changing sales assumptions. The resulting forecasts can then flow back into SAP BDC, enabling analysts to perform “what-if” analyses directly inside their planning tools using Joule.
3. Personalized customer engagement
Customer transaction and service data from SAP can be combined in Snowflake with call-center transcripts and identity data from third-party sources to create a comprehensive “customer 360” dataset. Snowflake’s AI models can analyze sentiment and predict churn, producing “next best action” recommendations for customer-facing teams. Those predictions are returned to SAP applications, where Joule presents churn-risk alerts and suggested retention actions in real time.
A Model Built on Measurable Value
Snowflake’s business model is consumption-based as it earns revenue only when customers use its products. Company executives see that as a natural fit for enterprise AI, where adoption is closely tied to demonstrated returns.
“Our model ties directly to the value customers realize,” an executive said. “We focus on replacing existing systems, reducing costs, and increasing flexibility through Snowflake Intelligence. If customers aren’t using what we build, we don’t get paid.”
This usage-driven model has shaped how Snowflake approaches AI rollouts. Rather than large, upfront implementations, the company works with customers to create pilot projects that prove value before expanding to production scale. “That step-by-step approach keeps us focused on outcomes,” the executive explained.On the question of whether enterprise AI has entered a speculative bubble, Snowflake leaders remain pragmatic. “There’s a lot of enthusiasm in the market, but we stay focused on what AI can actually do for customers,” one said. “Snowflake Intelligence was designed to make agentic AI practical, so that data analysts spend less time writing manual queries and more time building data agents that solve business problems. The focus is on production use, not hype.” Customers such as AstraZeneca are already using the combined platforms to modernize operations and data management.
