Zscaler: Security-for-AI Bookings Jump 50% Sequentially as Company Prepares to Launch Agentic SecOps, But FY27 Guide Implies Sharp Growth Deceleration
Q4 FY2026 earnings call, September 3, 2026
Zscaler closed its fiscal year with 25% ARR growth and a record non-GAAP operating margin of 24.3% in the fourth quarter, but the headline story from the call was the accelerating role of artificial intelligence as both a threat vector and a sales driver, set against a fiscal 2027 guide that implies a meaningful slowdown from the momentum the company just reported. CEO Jagtar "Jay" Chaudhry told analysts that "AI is quickly becoming the largest tailwind we have ever seen," a framing management leaned on throughout the call to explain both new product launches and elevated capital spending.
Security-for-AI demand inflects sharply
The most concrete new data point on the call was the trajectory of Zscaler's nascent security-for-AI product line, where bookings rose more than 50% sequentially in the fourth quarter on top of what the company called a strong third quarter. CFO Kevin Rubin added that pipeline for the category grew 75% sequentially quarter-over-quarter heading into fiscal 2027, a figure he was careful to clarify was not a year-over-year comparison. Chaudhry disclosed that Zscaler's team has run "hundreds of frontier AI risk assessments" for global enterprises in recent months, finding that over 90% of organizations had AI applications, models or servers exposed to the internet, and more than a third carried known exploitable vulnerabilities. He also referenced a swarm of AI agents breaking out of a sandbox training environment in a recent incident involving Hugging Face, part of what he described as "multiple high-profile incidents from three leading frontier model companies where agents went rogue and took unauthorized actions." The specificity of these disclosures is new territory for the company and underscores how directly Zscaler is now tying its pitch to documented agentic-AI failures rather than hypothetical risk.
Deal color reinforced the narrative: a Fortune 500 transportation customer signed a seven-figure Z-Flex upsell for the full security-for-AI portfolio, beating two platform competitors and pushing its ARR to nearly $10 million, while a Fortune 500 semiconductor manufacturer chose Zscaler over other vendors to secure a company-wide rollout of Claude Cowork. Chaudhry explained the competitive dynamic bluntly: "Our customers tell us that they don't want one more or three more AI security products. They're looking for an integrated solution, not only for AI security, but also that works with the rest of security as well." Notably, 70% of security-for-AI deals in the quarter also included Zscaler's data security solution, suggesting the two lines are becoming difficult to sell separately.
Agentic SecOps launch and the Red Canary payoff
Zscaler will unveil its Agentic SecOps solution via webcast on September 9, more than a year after acquiring managed detection and response (MDR) provider Red Canary. Chaudhry framed the offering as an attempt to disrupt "even the ones that call themselves next-gen SecOps" by building a genuinely agent-native detection and response system rather than a human-driven process with automation layered on top. The pitch rests on Zscaler's telemetry scale — more than 750 billion transactions processed daily — combined with Red Canary's decade of MDR experience, with the goal of compressing detection-to-remediation timelines from days or weeks down to minutes. Rubin was explicit that Red Canary will not contribute incremental net new ARR going forward on a standalone basis; all of its economics will now show up inside the combined Agentic SecOps offering. He also confirmed elevated churn within the Red Canary book of business during fiscal 2026, consistent with typical MDR industry dynamics, though the company managed to preserve the overall book through the integration period. Management does not expect the new product to meaningfully contribute until the second half of fiscal 2027.
Net new ARR accelerates, but FY27 guidance raises questions
Underlying business momentum was strong in the printed quarter: net new ARR excluding Red Canary grew 17% year-over-year in the fourth quarter, up from 10% in the first half of fiscal 2026 and just 7% for all of fiscal 2025 — a clear multi-quarter acceleration. Including Red Canary's $141 million ARR contribution, total ARR reached $3.8 billion, up 25%. Yet the fiscal 2027 guide calls for ARR growth of only 16.6% to 17.4% and revenue growth of 16.6% to 17.5%, a step down from the 25% delivered in fiscal 2026. Oppenheimer's Ittai Kidron pressed management directly on the math, noting that net new ARR growth implied by the midpoint of guidance falls to roughly 4%. Rubin attributed the conservatism to the time needed to complete a sales leadership transition — two senior leaders, one in a geo role and one covering a vertical, departed last quarter — as well as uncertainty around the pace of uptake for the newly integrated SecOps solution. He reiterated that net revenue retention has held steady at 115% across every quarter of fiscal 2026, and that security-for-AI contributions have not been explicitly baked into the guide despite the tailwind commentary, which leaves room for upside if the AI-driven pipeline converts as management expects.
Sales reorganization and workforce restructuring
Zscaler disclosed a restructuring affecting approximately 3% of its workforce, with associated charges of $30 million to $33 million, which management described as a reallocation of resources toward AI initiatives and new-logo sales capacity rather than a response to weakening demand. The company is adding dedicated new-logo sales executives focused on enterprise accounts, an area Chaudhry acknowledged remains under-covered further down the enterprise segment even as coverage at the top end is strong. Of Zscaler's roughly 20,000 targeted companies, only about 4,600 are currently customers, leaving a large greenfield opportunity that management is now staffing against. Guggenheim's John DiFucci pushed to distinguish whether the hiring reflects a genuine AI-driven demand inflection or simply scaling headcount in line with company size; Rubin responded that it is not an either-or dynamic, citing both broad-based momentum in core Zero Trust and a distinct AI tailwind.
Zero Trust Everywhere and Z-Flex traction
The number of "Zero Trust Everywhere" enterprises — customers that have purchased Zero Trust for users, branches and cloud — rose to more than 950, up from over 700 in the prior quarter and just 350 a year earlier, a trajectory management pointed to as evidence that its platform strategy is gaining traction beyond the core user-security product. Z-Flex, the company's flexible multiyear commitment structure, generated more than $770 million in total contract value in the quarter, up over 60% sequentially, and $1.7 billion for the full year. Z-Flex customers saw average ARR uplift of nearly 30% for the year. Fortune 500 penetration climbed from 45% to 50% during the year, and the company closed a record number of $1 million-plus annual contract value deals in the quarter.
Margins hold, but component costs weigh on capital spending
Non-GAAP gross margin improved to 80.2% in the quarter from 79.3% a year earlier, and operating margin reached a record 24.3%, up 220 basis points year-over-year. Full-year free cash flow margin, however, declined to 23% from 27%, which Rubin attributed to timing of cash collections and capital expenditures. More notably, management flagged that memory, storage and processor costs are rising and availability is tightening, components that feed both Zscaler's data center infrastructure and its Zero Trust Branch appliances. The company opportunistically accelerated certain equipment purchases in the fourth quarter to lock in availability, pushing full-year capex to 8% of revenue, and expects capex to remain elevated through fiscal 2027 as a result of persistent memory pricing pressure — a supply-chain dynamic increasingly common across the software and infrastructure sector as AI-driven hardware demand tightens component markets.