SAP runs SAP: Proving the Autonomous Enterprise starts with a unified AI engagement layer
Scaling to 110,000 employees as “customer zero”
SAP’s vision for the Autonomous Enterprise is clear: People set direction, and AI executes across business domains. But before asking customers to make that shift, SAP wanted to prove it internally. With the Joule Work capability, SAP has moved from fragmented AI tools to a single, intent-driven workspace where employees, assistants, and agents work together across business processes.
| Industry | Region | Company Size |
| High tech | Walldorf, Germany | 110,000+ employees |
unique active users in October 2026.
employee-built skills automating business tasks.
business systems and data products connected.
VP and Global Head of Enterprise AI, Corporate Processes, and IT; SAP SE
Shifting from fragmented tools to a single engagement layer
SAP has run on its own software for more than 50 years, including investing in and applying AI across its business. Yet, as AI tools proliferated, SAP’s 110,000 employees faced the same challenges its customers faced. They grappled with a fragmented landscape of scattered tools, shadow AI risk, and no consistent way to connect AI with the calendars, inboxes, and business systems where work actually happens.
Employees were also uncertain about which tools to use. Some teams relied on third-party solutions, and others adapted developer-focused tools not designed for business use. Consumer AI tools increased the risk of unsanctioned accounts, creating challenges for data privacy, security, and compliance—particularly for a company operating across more than 158 countries.
SAP’s strategic answer was not just another tool but a new way of working: the Autonomous Enterprise. Here, the Joule solution serves as the engagement layer that brings together data, workflows, and AI agents across SAP and third-party software systems. Instead of navigating between applications, employees express what they want to accomplish. Joule surfaces the right context, coordinates agents, and executes—while keeping humans in control of the decisions that matter.
SAP went all in on that vision and decided to prove the model as customer zero.
Customer Office and SAP Experience, Corporate Processes, and IT; SAP SE
Collaborating to move from prototype to 10,000 users in nine weeks
The breakthrough started on an ordinary Friday. Convinced that the company needed a single home for AI at work, SAP’s chief AI officer, Jonathan von Rüden, spent the evening and the following weekend coding the very first version of Joule Work.
That prototype did not stay a side project for long. By the following Monday, Sebastian Steinhäuser, COO and a member of the Executive Board of SAP SE, had recognized the prototype’s potential and brought together teams from across the company to support its development.
In May 2026, the corporate processes and IT (CPIT) team, SAP’s internal IT organization, became the first customer, testing and fine-tuning the capability in real business scenarios alongside product engineers for SAP Business AI Platform. After five weeks, the team rolled it out to the first 10 users, then 100, then 1,000. By the end of the initial rollout, 10,000 employees were using Joule Work just nine weeks after the first line of code was written. Tens of thousands of feedback messages followed through a closed-loop service—each one reviewed, classified, and addressed individually to evolve the product directly.
COO and Member of the Executive Board, SAP SE
Creating business value from adoption to autonomous processes
By early August, SAP had rolled out Joule Work to all employees without a single critical incident, making it one of SAP’s fastest-adopted internal tools. But adoption alone does not make the case for the Autonomous Enterprise. What matters is how work itself changes.
At SAP, sales teams have moved from fragmented customer views to an AI-enabled deal engine. Through a Model Context Protocol–enabled connection to the SAP 360 Customer solution, Joule Work brings together insights from nearly 100 underlying data products. Account teams now get a single, comprehensive view of opportunities, deal pipelines, and the installed base in one intent-driven interaction—replacing what previously required navigating multiple applications and manually assembling information. The result is accelerated deal preparation and faster progression through the sales cycle.
Meanwhile, finance and HR teams have shifted from manual retrieval to real-time steering. Employees use Joule Work to access, retrieve, and analyze operational data through natural-language requests. Tasks that previously required switching between reporting tools and spreadsheets are now handled in a single workspace, supporting faster insights and real-time decision-making.
Finally, across the enterprise, employees have already built around 2,000 skills. CPIT introduced self-paced learning paths and a gamified challenge sponsored by executive leadership, inviting any employee to build skills or connectors that solve real business problems. This fostered a growing ecosystem of skills—from automated email triage to document analysis to meeting preparation—each one a small step toward autonomous process execution.
Enterprise AI, Corporate Processes, and IT; SAP SE
Progressing from engagement layer to autonomous domains
Joule Work is the first tangible proof point of the Autonomous Enterprise, but this is just the beginning. SAP’s road map extends the model into five autonomous domains: Autonomous Finance, Autonomous Spend, Autonomous Supply Chain Management, Autonomous HCM, and Autonomous CX. In each domain, specialized AI assistants and agents execute comprehensive business processes within clearly defined governance boundaries, with Joule Work as the consistent engagement layer across them.
SAP Business AI Platform provides the enterprise-wide foundation that connects agents and assistants with the business context, semantically rich data, and enterprise-grade governance they require to act reliably at scale. Adoption and rollout as part of the SAP Early Adopter Care program is yielding promising first results on scalability, adoption, and value.
As SAP’s customer zero, CPIT has learned three things it now shares with customers beginning their own journey.
First, it advises customers to start with the engagement layer. Give employees one place to work with AI—and consolidate before you automate.
Second, customers should prove value in one domain, and then expand. They can focus on a single business area, measure outcomes, and use evidence to earn the mandate for the next domain.
Third, customers should scale through community, not mandates. Employee-built skills and peer recognition create organic adoption in addition to clearly expressed C-suite support and sponsorship.
Want to know more about SAP?
SAP runs SAP: Fueling human insights and AI innovation with data products (SAP Online Customer Story)