ServiceNow COO Zavery tells William Blair that cross-department workflow data is the real moat against AI disruption. The trade hinges on AI attach rates in Q2 earnings.
ServiceNow (NOW) President, Chief Product Officer and COO Amit Zavery appeared at the 46th Annual William Blair Growth Stock Conference on June 3, 2026. The session, hosted by analyst Arjun Bhatia (who disclosed personal ownership of NOW shares), opened with a question that frames the central market debate in enterprise software. “Explain why workflow orchestration, the position that ServiceNow has, is a tailwind to – or benefits from AI rather than is disrupted by AI,” Bhatia said.
The question captures a binary choice for investors. One camp argues generative AI will disrupt incumbent application layers by automating tasks directly. The other sees AI as a platform-level tailwind for vendors that own workflow data and integration logic. For a generalist audience, the answer determines whether NOW trades as a compounder or a value trap.
A naive interpretation holds that AI agents can replace many manual IT and HR workflows. A chatbot that resolves password resets or approves leave requests could bypass ServiceNow’s interface entirely. Under this view, the company’s premium valuation rests on a threatened moat.
Zavery’s argument inverts that logic. Workflow orchestration generates structured metadata – approval chains, SLA breach histories, ticket resolution patterns, escalation rules. That corpus is proprietary, real-time, and tied to enterprise permissions. A standalone AI model without access to that context produces generic responses. ServiceNow’s platform provides the authorization layer, the process logic, and the historical data that make AI actions safe and accurate.
Practical rule: the value shifts from the interface to the data pipeline that feeds the AI.
Generative AI models need context to produce useful actions. A chatbot resolving an IT ticket needs to know the user’s role, asset assignment, prior incident history, and the escalation policy. ServiceNow stores all of that in a unified data model spanning IT, HR, customer service, and facilities. The more workflows run through the platform, the richer the training signal becomes. That creates a self-reinforcing cycle. More data improves AI responses. Better responses drive higher adoption. Higher adoption generates more data.
Critics point to Microsoft Copilot and Salesforce Einstein as competing AI layers that could bypass ServiceNow’s interface. The rebuttal from Zavery’s presentation centers on silo fragmentation. Copilot works well within Microsoft’s ecosystem. Einstein lives inside Salesforce. Neither connects an IT ticket to an HR leave request or a facilities work order on a single record. ServiceNow argues that end-to-end workflow ownership prevents that fragmentation. An AI agent that can resolve an IT incident but cannot escalate to HR for a related employee issue creates friction. The company’s bet is that enterprises will pay for the unified orchestration layer rather than stitching together multiple AI silos.
Key insight: ServiceNow is positioning its platform as the operating system for enterprise AI actions – the middleware that decides which model gets called, with what permissions, and what audit trail.
Conference presentations alone rarely move stocks. This one matters because it stakes the company’s strategic narrative before quarterly earnings force the issue. ServiceNow’s stock currently reflects considerable uncertainty. The AlphaScala Alpha Score on NOW stands at 60 out of 100, rated Moderate, placing it in the Technology sector. That score captures a business with strong fundamentals held back by an unresolved AI disruption narrative.
Sell-side consensus expects ServiceNow to sustain 20%+ subscription revenue growth through fiscal 2027. Achieving that requires AI products – Now Assist, AI Search, Proactive Actions – to convert into incremental seat expansion, not just replacement of manual workflows at lower prices. The bull case depends on land-and-expand accelerating as AI makes the platform stickier. The bear case depends on competitors collapsing the price of workflow automation toward zero.
Two data points will determine whether Zavery’s argument holds or breaks.
Risk to watch: a sharp slowdown in enterprise IT spending could drag all growth stocks lower, regardless of AI differentiation. ServiceNow also faces execution risk as it integrates Moveworks (acquired in January 2026) – a chatbot vendor whose technology must fold into the core platform without alienating users.
The William Blair session does not resolve the AI debate. It reframes the risk-reward equation. The question for ServiceNow is whether workflow orchestration becomes a tollbooth for enterprise AI – collecting fees every time a model needs to execute an action inside a company’s operations. That outcome would justify a premium multiple. The opposite outcome – AI models that bypass the platform entirely – would compress the multiple toward single-digit growth software.
ServiceNow currently sits in a no-man’s land: too expensive for a value approach, too high-beta for conservative growth accounts. The trade is binary until execution data emerges. Traders watching the AI software theme should treat NOW as a beta to the AI-moat thesis rather than a pure-play AI winner. A miss on the next quarterly attach rate could trigger a 10-15% drawdown. A beat that shows accelerating enterprise adoption could close the multiple gap with high-growth peers.
For a broader view of the SaaS valuation selloff and how ServiceNow fits into the risk event, see SaaS Apocalypse Panic: ServiceNow Risk Event Watch. For alternative readings of AI disruption across tech, read SA Analysts Flag Meta, ServiceNow as Undervalued Tech.
The practical takeaway: monitor cRPO growth and AI product revenue disclosures. Those metrics will tell you whether Zavery’s conference argument becomes a quarterly earnings story or a fading narrative.
Prepared with AlphaScala research tooling and grounded in primary market data: live prices, fundamentals, SEC filings, hedge-fund holdings, and insider activity. Each story is checked against AlphaScala publishing rules before release. Educational coverage, not personalized advice.