C-suite should make AI data management the 2026 ERP priority

Vendors push data products and partnerships

ERP vendors are making it clear through their product strategies that they see data as the key to realizing the potential of AI to usher in a new age of ERP that communicates in natural language, enables end-to-end process automation across corporate boundaries and executes most tasks on behalf of users -- all while fulfilling its traditional role of hosting business transactions and records.

SAP, Oracle, Workday, Infor and other vendors bolstered their data management and integration offerings last year.

SAP, which had been criticized for having a scattershot strategy, drew praise for introducing Business Data Cloud (BDC), a SaaS data management and analytics platform for unifying SAP and external data, embedding it with technology from Databricks, a cloud data lakehouse platform. It later released a new version, BDC Connect, to provide secure, bidirectional links between BDC and partner platforms, including Databricks, Google Cloud and Snowflake. A key feature is zero-copy sharing, which keeps data in the owner's SAP system while allowing other companies to access it from their data platforms. Mueller said the capability is important because it helps companies maintain ownership of data and avoid having to move it around.

Meanwhile, Oracle consolidated its data management and AI development offerings in a new AI Data Platform and introduced a lakehouse that uses AI to automate data management. Analysts said the moves help prepare data for AI and minimize the need for integration between ERP modules.

Workday announced a similar, partnership-driven, zero-copy offering called Workday Data Cloud. Infor has had lakehouses for more than a decade, according to Mueller, originally to support integration, reporting and analytics strategies that predate the generative AI era that began with ChatGPT's introduction in late 2022.

AI demand driving ERP cloud migrations

The AI data consolidation trend is also hastening the ERP cloud migration decision for some organizations.

Jon Reed

Reed said the modernization planning that often leads to the cloud also sets up the data discussion as organizations look to rationalize their data architecture and align it with those of their ERP vendor and its partners.

"You get to have that data conversation in a much more aggressive way than you ever did in the past," he said. "It's got to be a different kind of data conversation, too, because no one at the board level in your organization has any appetite for a multi-year data cleansing initiative for its own sake. It's going to have to be something much more iterative, where you're able to prove value and ROI all along the way."

The desire to implement AI can also spur better conversations across departments during ERP modernization, he said, but today's AI doesn't inherently solve siloing and other nontechnical problems.

"If you're a well-run organization with good data, process flows and human collaboration, then your AI projects are probably going to take you that much further. If you're not a well-run organization, have bad data and processes, trying to run agents on top of that is just going to make your situation worse."

Holger Mueller

Mueller agreed that an AI data strategy probably means moving the organization's data to the cloud. One reason is AI usually has to run in the cloud, and it needs the elasticity that is cloud's defining feature. "We figured out years ago that it's better to let an AWS worry about their servers and capacity to get you that elasticity," he said.

Data architecture decisions are also closely tied to cloud migration strategies, which often involve hybrid configurations that keep some ERP modules on-premises while adding new SaaS applications.

"The question is how often do you have to update the data up and down," Mueller said. That setup is challenging to maintain, especially when software needs upgrading. "It's much easier to put everything in a data lakehouse in the cloud."

Taking the first steps toward an AI data strategy

Reed stressed the importance of alignment between the organization's data strategy, that of its ERP vendor -- current or prospective -- and the vendor's data partners. IT buyers should ask detailed questions about the vendor's data strategy to gauge its maturity, compare it to the organization's strategy and see if the two already have data partners in common. "Understanding the pricing and policy implications of accessing that data is really important as well," he said.

While it's good that demand for ERP modernization and AI is forcing organizations to have more vigorous data conversations than before, senior strategists should use quick wins to show progress that supports their case, Reed said. "A lot of the vendors have answers to that. I know some finance planning vendors, for example, that will try to get you up and running with an initial use case in a matter of weeks."

Greenbaum emphasized lining up internal support. "Look for champions in the organization who really think about end-to-end process and have the authority and permission to start building allyships around breaking down the silos in the enterprise."

David Essex is an industry editor who creates in-depth content on enterprise applications, emerging technology and market trends for several Informa TechTarget websites.