Getty Images/iStockphoto

News

SAP to add AI agents for catalogs, order management

SAP deepens its footprint in agentic AI.

Don Fluckinger
By
Published: 20 May 2025

The coming year for SAP will be about agentic AI, as the company previewed agents to come.

At this week's SAP Sapphire user conference in Orlando, Fla., Joule agentic AI took center stage for its CX suite. Leading the way was an order management agent with deep hooks into the company's flagship ERP system.

Order management, which hands off order data to ERP when a customer or sales rep puts in an order -- and follows it to shipment and receipt -- is a feature that Salesforce and ServiceNow have added to their CX suites. SAP's order management agent offers human customer service agents a window into order information, shipping tracking details and payment information.

Many SAP customers have deep catalogs of goods and services. A new catalog optimization agent can autonomously perform tasks such as adjusting prices, updating products and optimizing content for search engines.

Order management is a good place to start, Constellation Research analyst Liz Miller said, considering that SAP's customers tend to be large organizations.

"They're dealing with larger enterprises that are dealing with larger complex orders and larger complex sales," she said, "but they're also dealing with larger complex revenue machines and revenue estates from technology."

SAP is stepping up development of its agentic AI, with plans to deploy Joule agents across customer service and the supply chain later this year. By year's end, users will also be able to develop agents with their own custom prompts across SAP with a tool the vendor currently calls "prompt optimizer." The tool is now being tested internally at SAP and is available to a limited number of customers in an early sign-up program.

SAP's agents will be able to identify tasks and gather data, plan to solve them, and use and orchestrate tools -- inside SAP and outside with APIs or other agents -- to execute that plan, said Balaji Balasubramanian, president and chief product officer for SAP Customer Experience and Consumer Industries.

Business Data Cloud key to success

That said, the key to SAP's success is not necessarily SAP's agentic AI, but rather the underlying data, as it is with many AI systems. The company's Business Data Cloud, launched earlier this year, serves a similar purpose to Salesforce's Data Cloud: driving agent activity with normalized data and connecting to data in enterprise apps outside their respective platforms.

"The idea is to take data from SAP systems and non-SAP systems, harmonize it, and have the semantic richness to it," Balasubramanian said. "When agents run on top, then you're guaranteed for this to work in a more reliable manner rather than in a piecemeal manner."

Business Data Cloud makes SAP less of a siloed application stack and more like a composable system that can be deployed across the enterprise, Miller said. An example of this might be a finance department building a service module with an agent that can automate responses to customer billing queries.

Don Fluckinger is a senior news writer for Informa TechTarget. He covers customer experience, digital experience management and end-user computing. Got a tip? Email him.

Dig Deeper on Marketing and sales