SAP reshapes AI strategy: new teams and pay-per-use instead of subscriptions

SAP is beginning one of the most notable shifts in its product strategy in recent years. CEO Christian Klein is assembling a new AI division with hundreds of employees and preparing the transition from traditional subscription to a model where customers pay for actual AI feature usage.

Why Change the Model

For SAP, this is not a cosmetic price update, but an attempt to align the entire business with new market logic. Historically, enterprise software was sold based on the number of users, seats, or modules. But when AI assistants and agents start performing routine operations themselves, such a scheme no longer accurately reflects the product's value. One well-integrated AI scenario can automate the work of several employees, and in such a situation, charging SAP simply for system access becomes increasingly inconvenient and less compelling.

The company faces pressure from two sides simultaneously. Externally, competition from generative AI platforms intensifies, promising quick results without heavy implementations. Internally, SAP is forced to accelerate its own restructuring: in early March, Klein already redistributed management responsibilities to focus more intensely on product and artificial intelligence. Now the next step is launching dedicated engineering teams in July that will work closer to customers and help bring AI scenarios to real-world usage.

What Clients Will Get

The new model implies not only a different price list, but also a different implementation approach. SAP is planning to create so-called forward-deployed engineering teams—mixed teams of consultants and developers who will build customized AI solutions together with the customer for specific business processes. This is not about a "magic button" on top of ERP, but about embedding AI in finance, procurement, supply chains, HR, and other areas where context, data, and control matter.

• Dedicated teams for collaborative AI scenario development with the customer
• More customized solutions for specific processes, not just standard features
• Transition to payment for actual AI usage, not just user count
• Focus on AI assistants and agents embedded in existing business systems

For customers, this may be more convenient than the traditional subscription, especially if AI is used unevenly or initially launched in only a few processes. SAP already demonstrates in its own materials that it is moving toward a usage-driven model with AI units and access to Business AI through Joule. The idea is simple: if AI accelerates quarter closing, automates sales proposal preparation, or removes some manual work, the price should be tied to the volume and effect of usage, not to an abstract number of licenses.

Where the Main Risk Lies

For SAP, this shift is painful precisely because it touches the foundation of the old enterprise software economy. Subscriptions provide predictable revenue, while a consumption model makes income more variable and ties it more closely to proven results. Against this backdrop, the company still has to carry the legacy of its previous transition—from on-premises software to the cloud, which took years and remains a sensitive topic for customers, partners, and investors.

There is a second risk as well: embedding AI into a real business process is far more complex than demonstrating a polished demo. SAP itself noted in late March that AI already enters more than two-thirds of its recent cloud deals, but this does not yet mean massive industrial impact. For an agent to work in finance or supply chains, it needs quality data, reconciliation rules, audit trails, access controls, and predictable processes. If SAP cannot quickly prove the value of the new scheme, some customers may keep their SAP core but build the AI layer with external models and contractors.

What This Means

SAP effectively recognizes that AI changes not only the interface of enterprise software, but also its monetization. If Christian Klein's experiment succeeds, the market will receive a strong signal: in the age of AI agents, selling enterprise products purely by user count is no longer sufficient, and those vendors will win who can link AI to concrete business results.