Scaling Global Organizations in the Age of AI with ServiceNow Chairman and CEO Bill McDermott

Summary

• McDermott’s rebuttal to the “SaaS apocalypse” is economic and operational: generated code and language models do not by themselves replace a deterministic platform that executes and closes cases. He says replicating a simple ServiceNow application with a language model would cost 10 times more after engineering labor, GPU infrastructure, tokens, and rebuilding are counted. The decisive distinction is, “AI thinks. Workflow acts.”
• The software most exposed to AI is narrow, departmental, and low on the CEO’s priority list—not deeply embedded systems spanning functions or holding critical records. ServiceNow’s claimed moat includes more than 85 billion workflows in flight, 7 trillion platform transactions, and integrations with roughly 800 substantial systems of record. McDermott’s warning is equally blunt: businesses tolerate human error, but “they never will forgive software for making a mistake.”
• ServiceNow is positioning itself as an AI control tower connecting language models, hyperscalers, systems of record, and enterprise data rather than competing with every component. Moveworks supplies an “agentic front door,” Veza covers human and non-human identity, and Armis extends visibility across IT and operational technology; McDermott says the businesses were integrated in 20 days. His security rationale is that cybercrime would rank as the world’s third-largest economy and represents “a $1 trillion a month problem.”
• AI’s clearest financial effect may be operating leverage: ServiceNow already has agents managing 90% of customer-service cases, leaving only 10% involving people. McDermott expects dramatically fewer net-new hires over five years even as the company grows, with human investment concentrated in engineering, judgment, trust, and customer relationships. “If an agent can do it as good or better,” he says, the economic decision is easy; he expects 2.2 billion agents to enter the workforce within a couple of years.
• Faster implementation could expand software consumption and create more, shorter services projects. Large ServiceNow customers can now go live in under 30 days, and McDermott argues systems integrators gain more projects that can be completed much faster. Against Sarah Guo’s figures of roughly 20% growth and more than $13 billion in trailing revenue, his key forward indicator is rising agent assists and consumption—the layer that could add a new shareholder-value engine.
• Enterprise adoption remains uneven, but some customers are moving from open-ended experimentation toward prescriptive execution and measurable ROI. McDermott says only 11% of Brazilian companies had progressed beyond experiments, while U.S. adoption leads. Financial institutions are moving quickly, while public-sector and healthcare organizations are addressing different modernization needs. Buyers now want vendors to know their business, specify the transformation, and “get me there fast,” with AI ultimately expected to lift revenue per employee, margins, and potentially revenue growth.
• McDermott’s leadership thesis is that accelerating AI should make human agency more valuable, not less. “It is fast, but it’ll never move this slow again,” he tells leaders, while insisting AI should “make the human ambition greater.” His operating rule remains the lesson of his teenage deli: “the customer and the customer alone determines whether you win or lose.”

Deep dive

Not yet available upstream; scheduled sync will retry.