EdgeBytes: How ServiceNow and SAP Are Rewiring Enterprise Value from AI Hype to Throughput | 3.09.26

Hey everyone — welcome back to EdgeBytes.

​

What jumped out at me this past week were a few announcements from ServiceNow and SAP that, taken together, fan the flames of a marked shift happening in enterprise software: the market is moving away from AI as a feature and toward AI as an operating model for value realization. ServiceNow is pushing autonomous, workflow-native AI into telecom and government, while SAP is reorganizing around end-to-end customer value with a new services and support portfolio and a new Chief Customer Officer role for Thomas Saueressig, effective April 1, 2026. That combination matters because it tells us where the enterprise stack is heading next: toward governed execution, measurable outcomes, and tighter accountability from sale to adoption to expansion.

​

Let’s start with ServiceNow. At Mobile World Congress, the company framed Autonomous CRM as a system of action, not just a system of record. In telecom, that means unifying sales, service, and fulfillment workflows with autonomous agents that do the work in real time. ServiceNow cited internal CX research showing that 75% of telecom customers rate service as less than great, 51% would switch because of poor or slow service, and reps often have to bounce across three to five disconnected systems just to resolve a single issue. Bell Canada’s early results give that story more credibility: ServiceNow says Bell improved customer response time by 25%, while Bell also reported 90% positive feedback on AI accuracy from case managers.

​

That is a Value Physics story in plain sight. You’re going to hear a lot more about the Value Physics framework in the coming months. It’s essentially a simplified model we’ve come up with that borrows fundamental concepts from Physics to characterize how companies accelerate, capture, and sustain value from enterprise AI. The simple equation states that Enterprise Value equals [(Acceleration − Friction) / Mass] × Velocity.

​

A core premise of Value Physics is that enterprise value does not compound when AI is merely informative. It compounds when AI removes friction from the workflow itself. Bell’s result is not impressive because “AI was added.” It’s impressive because intake, triage, and case handling moved closer to clean execution. Less swivel-chair work. Fewer routing errors. Faster response. Better utilization of human judgment where it actually matters. In Value Physics terms, that is not experimentation; that is throughput improvement due to decreased friction, with measurable operational leverage.

​

The same pattern shows up in government, where the stakes are different but the logic is identical. ServiceNow announced EmployeeWorks, which combines Moveworks’ conversational AI and enterprise search with ServiceNow Employee Center, and it introduced Autonomous Workforce for government cloud environments, including GCC and NSC. The message is that agencies do not need more point tools; they need AI connected directly to workflows, with authority, governance, and human oversight built in. That is why ServiceNow keeps using the phrase “AI control tower.” It is trying to own the orchestration layer, not just the assistant layer.

​

And the proof points matter. ServiceNow says its first out-of-the-box AI specialist is a Level 1 IT Service Desk specialist for tasks like password resets and software access, designed for FedRAMP High, IL4, and IL5 environments. It also cites the City of Raleigh, whose virtual agent reportedly achieved a 98% deflection rate and saved the equivalent of a full month of time. Whether you are in public sector or not, the takeaway is broader: AI adoption gets real when organizations constrain scope, define authority, and instrument outcomes. That is another crucial Value Physics principle. Unbounded AI creates noise. Governed AI creates margin.

Now shift over to SAP, because SAP’s announcements are less flashy on the surface but maybe even more consequential strategically.

On March 3, SAP introduced a redesigned Services and Support portfolio with three tiers: Foundational Success Plan, Advanced Success Plan, and Max Success Plan. The company explicitly tied the move to transparency, speed, flexibility, business continuity, AI-driven innovation, and continuous value from SAP Business Suite investments. Foundational is included with every cloud solution, Advanced adds specialized expertise and AI-assisted guidance, and Max adds premium strategic engagement, including dedicated success plan managers and customer-specific prototype development to accelerate offerings such as SAP Business AI.

Then, one day earlier, SAP announced the creation of a new Customer Value Group and expanded Thomas Saueressig’s role to Chief Customer Officer. That new board area combines Customer Success with Customer Services and Delivery.

Christian Klein’s quote is the real headline: “In a business where adoption and renewal define success, the lines between selling and delivering disappear.” That is one of the clearest statements I’ve seen from a major enterprise software CEO acknowledging that recurring revenue economics have changed the operating model. In other words, revenue quality now depends on post-sale execution quality.

​

This is exactly where Value Physics becomes useful. The old software model let companies celebrate bookings while implementation friction, low adoption, and weak renewal economics remained someone else’s problem. The new model does not. In a cloud-and-AI economy, value leakage after the contract signature destroys enterprise value just as surely as poor sales execution before it. SAP’s organizational redesign is, in effect, an admission that customer lifetime value, expansion, and renewal are operational variables, not just commercial ones. The handoff model is dying. The full-journey accountability model is rising.

So what is the bigger pattern we’re seeing here? It appears 2026 is shaping up to be the year enterprise vendors stop competing primarily on AI features and start competing on who can govern outcomes across complex processes. ServiceNow is betting that the winning architecture is workflow-centric, agentic, and cross-system. SAP is betting that the winning architecture is lifecycle-centric, tiered, and tightly aligned to adoption and customer success. Different starting points, same destination: fewer silos, more orchestration, and much more pressure to prove value in production.

​

I would say that ServiceNow has near-term momentum where work is fragmented, rules-based, and cross-functional: telecom service operations, public sector workflows, employee service, and other environments where orchestration is worth more than front-office glamour. Its pitch is strong because it connects AI to work, not just to content. And with more than 80 billion workflows running annually on its platform, the company can credibly argue that it already has the process fabric needed to operationalize agentic AI at scale.

​

SAP’s advantage is different. It sits closer to the transactional and operational core of the enterprise, so if it can truly align selling, delivery, services, support, adoption, renewal, and expansion under one customer-value operating model, it strengthens its ability to turn installed-base trust into durable AI monetization. The risk, of course, is execution complexity. Reorgs sound elegant on paper. They only create value if incentives, data, accountabilities, and customer metrics actually converge in the field. Still, the direction is the right one, because the market has moved beyond “Who sold the AI?” to “Who made the AI stick?”

​

For CEOs, my  recommendation is straightforward: stop funding AI as a collection of isolated experiments and start governing it as a portfolio of workflow interventions. Ask one hard question every time an AI project is proposed: where, exactly, does this remove friction from a revenue-producing, cost-bearing, or risk-sensitive process? If the answer is vague, it is not a transformation investment; it is a science project. CEOs should also collapse the artificial divide between pre-sale promise and post-sale delivery.

SAP’s move is a reminder that value realization is now a board-level growth issue, not a customer-success afterthought.

For CIOs, this is the moment to architect for governed interoperability, not tool sprawl. ServiceNow’s telecom and government announcements both underline the same principle: AI works best when it sits on top of orchestrated workflows, connected data, clear authority boundaries, and measurable escalation paths. CIOs should prioritize process observability, identity and access control, model governance, auditability, and clean integration over shiny standalone copilots. The winners will not be the firms with the most AI agents. They will be the firms with the fewest unmanaged ones.

​

For CFOs, the implication is even sharper. Do not let AI ROI be reported as anecdote. Demand instrumented metrics tied to cycle time, first-contact resolution, case deflection, labor reallocation, renewal quality, gross margin protection, and working-capital improvement where relevant. Bell’s 25% response-time improvement and Raleigh’s 98% deflection rate are the kind of operational metrics that can be translated into financial narratives. CFOs should insist that every major AI initiative have a value-realization baseline, a ramp profile, and a governance owner. In Value Physics terms, capital should flow toward initiatives where adoption, execution discipline, and economic capture reinforce each other.

​

The bottom line is this: the market is beginning to reveal what mature enterprise AI looks like. Mature enterprise AI is not chatbot theater. It is governed action. It is fewer handoffs. It is cleaner process design. It is accountability across the customer lifecycle. It is measurable operating leverage. ServiceNow is pushing that logic from the workflow layer inward. SAP is pushing it from the customer-value layer outward. Both are converging on the same truth: in the enterprise AI era, value does not come from intelligence alone. It comes from intelligence that is embedded, governed, adopted, and economically captured.

That’s all for now. Thanks for listening to EdgeBytes, signal over noise from The Enterprise Edge. And keep an eye out for updates on the Value Physics book coming out in early 2027. Take care everybody!