At SAP Sapphire in May 2026, CEO Christian Klein stood in front of thirty thousand attendees and reframed the company’s own identity, launching what he called the Business AI Platform, the foundation for a vision of an “Autonomous Enterprise, where agents run the business”, as Klein put it to the keynote audience. That is not incremental product messaging. It is a vendor telling its entire customer base that the commercial model underpinning every SAP contract is changing, and 2026 into 2027 is the window in which that change is actually happening.
What Actually Launched at Sapphire 2026
The Business AI Platform consolidates three previously separate pieces of SAP’s portfolio, Business Technology Platform, Business Data Cloud, and SAP’s AI Foundation, into a single governed environment. Independent analysis of the launch describes the platform as structured around three layers, a context layer built on Business Data Cloud to unify SAP and non-SAP data, a build layer for developing agents, and a governance layer for controlling how those agents operate, offering executive leaders a path beyond isolated AI pilots toward secure, outcome-driven operations at scale.
For any organisation already running BTP, Business Data Cloud, or the AI Foundation independently, this consolidation is not a cosmetic rebrand. It changes what a customer is actually buying when they renew any one of those three components, since the commercial and technical boundaries between them are being deliberately redrawn as part of a single platform story rather than three separate product relationships.
Why SAP Is Moving Away From Per-Seat Pricing Now
The pricing shift is not a minor adjustment to an existing model. SAP CEO Christian Klein is setting up a new unit staffed with hundreds of people specifically to push adoption of the company’s AI features, and plans to overhaul how it charges clients entirely as agentic AI threatens the economics of the traditional per-seat subscription model, moving toward consumption and outcome-based pricing that ties cost to results delivered rather than to how many named users hold a licence.
Klein has been direct about the reasoning in earnings calls throughout 2026, warning that large language models on their own do not understand business data, process, or governance, that raw AI token spend does not reliably mirror business outcomes, and that lock-in to a single frontier AI vendor is a growing concern for enterprise buyers. That framing matters commercially because it is the explicit justification SAP is using to move away from a pricing model that has defined enterprise software for decades toward one where the price is tied to a defined business result.
The Autonomous Suite Sitting on Top of the Platform
Alongside the Business AI Platform, SAP introduced the Autonomous Suite, a set of AI agents built directly on top of the platform to execute end-to-end business processes across finance, supply chain, procurement, and HR rather than simply assisting a human user working through those processes manually. Joule Studio 2.0, part of this rollout, was demonstrated identifying a pricing and purchasing issue with an estimated margin impact of nearly twenty four million dollars and then proposing a validation agent as the fix, generating a product requirements document, technical specification, and orchestrating multiple agents together in response.
That kind of demonstration is a useful signal of direction, but it is also precisely the kind of capability that shifts an organisation’s usage profile most dramatically toward the consumption-heavy end of SAP’s new pricing model. An agent capable of identifying a multi-million dollar margin issue and then autonomously proposing and building a fix consumes a materially different volume of compute and orchestration than a simple assistant answering a policy question, and the pricing conversation needs to reflect that difference explicitly rather than assuming all Joule-branded capability carries the same cost profile.
What Outcome-Based Actually Means in Practice
SAP’s own commentary on this shift has offered a concrete example worth understanding directly. During SAP’s Q2 2026 earnings call, when asked how the company intends to monetise its newly acquired Prior Labs tabular AI technology, Klein confirmed that the module will be embedded into agents and priced based on the outcomes those agents actually deliver, a clear departure from licensing the underlying model or platform access on its own. That is the pattern SAP is signalling across its broader AI portfolio, not an isolated pricing decision limited to one acquired technology.
The practical challenge this creates for buyers is that outcome-based pricing requires a shared, contractually defined understanding of what counts as the outcome being priced, and that definition is where the real negotiation now sits. A vaguely defined outcome metric gives the vendor considerably more room to interpret performance in its own favour than a specific, measurable one negotiated explicitly at signature.
The Three-Stage Maturity Path SAP Itself Is Describing
Independent analysis published shortly after the Sapphire announcement lays out a staged adoption path worth treating as a genuine planning framework rather than marketing language. That analysis describes a Stage One of assisted work happening now, a Stage Two running from 2026 through 2027 in which AI agents execute process segments with human approval gates, for example compressing a finance close cycle from weeks to days while a controller approves exceptions, followed by a Stage Three from 2027 into 2028 and beyond in which AI runs end-to-end processes with humans setting strategy, KPIs, and exception thresholds rather than approving individual transactions.
That staging is directly relevant to renewal timing. An organisation currently negotiating a multi-year BTP or S/4HANA agreement is very likely to move from Stage One into Stage Two of this framework during the life of that same contract, which means pricing and governance terms negotiated today need to anticipate a meaningfully different usage pattern eighteen to twenty-four months from now, not just reflect current consumption.
The Investor Pressure Behind the Pricing Change
The pricing shift is also happening against a backdrop of genuine investor scrutiny worth understanding, since it affects how much room SAP has to move on commercial terms in any individual negotiation. SAP’s share price fell more than thirty five percent from the start of 2026 through late June, and the company’s own operating profit guidance for the year was revised downward to account for the dilutive impact of recent AI-focused acquisitions including Dremio and Prior Labs. That combination of share price pressure and near-term margin dilution gives SAP a genuine commercial incentive to demonstrate that its AI investments are translating into cloud backlog growth and pricing power, which is worth keeping in mind when evaluating how firmly SAP’s account teams hold their position on AI-related pricing during any given negotiation.
Why SAP Itself Has Already Admitted a Course Correction
It is worth noting directly that SAP’s own executives have publicly acknowledged this was not a smooth, linear strategy from the outset. At Sapphire in Madrid, CEO Christian Klein told press and analysts that SAP changed direction on its AI messaging approximately eight or nine months earlier, after recognising that a focus on AI technology itself was missing the mark with customers who were actually looking for demonstrated business value rather than technical capability.
That admission is useful context for any organisation evaluating how much weight to give SAP’s current roadmap commitments. A vendor that has already pivoted its AI messaging once within the space of a year is one whose pricing model and platform boundaries should be treated as still evolving rather than settled, and contract terms signed now should build in enough flexibility to accommodate a further pivot rather than assuming the current three-layer platform structure is permanent.
What This Means for Existing BTP and RISE Customers Specifically
Organisations already committed to a multi-year RISE with SAP or BTP Enterprise Agreement should not assume the outcome-based pricing shift only affects new purchases. As SAP folds Business Data Cloud and AI Foundation capability into the unified platform, existing customers are likely to encounter the new commercial structure at their next renewal or expansion point rather than being grandfathered indefinitely into the older per-seat or credit-based model. Confirming directly with the account team how an existing agreement’s committed spend or CPEA allocation maps onto the new platform structure, well before the renewal date arrives, avoids discovering the answer for the first time inside a renewal quote with limited room to negotiate.
What This Means for a Contract Being Negotiated Today
Organisations negotiating BTP, Business Data Cloud, or S/4HANA terms during this transition period should treat the outcome-based pricing shift as a live variable in the negotiation rather than a future consideration to address at the next renewal. Specific points worth raising directly include how any newly introduced outcome-based or consumption-based pricing components interact with existing committed spend under a current CPEA or BTP credit allocation, whether outcome metrics are defined precisely enough in the contract to be independently verified rather than left to SAP’s own reporting, and what contractual protection exists if SAP’s platform boundaries or pricing structure change again during the contract term, which recent history suggests is a real possibility rather than a remote one.
Building in an explicit review checkpoint at twelve or eighteen months, rather than waiting for the full contract term to elapse before revisiting pricing structure, is a reasonable request given how quickly this specific area of SAP’s commercial model has moved during 2026 alone.
Conclusion
SAP’s shift toward outcome-based and consumption-based AI pricing is not a distant future consideration. It is happening now, was signalled explicitly by SAP’s own CEO throughout 2026, and is staged by SAP’s own commentary to reach a materially more autonomous, and more heavily consumption-priced, state by 2027 and beyond.
Organisations negotiating any SAP contract touching BTP, Business Data Cloud, or AI Foundation capability during this window should treat the pricing model itself as unsettled, build explicit outcome definitions and review checkpoints into whatever gets signed now, and revisit those terms on a shorter cycle than a typical multi-year enterprise agreement would normally warrant, given how much SAP’s own AI commercial strategy has already moved within a single year.
