Q2 2026 Recap: ServiceNow Beats Everything, Crosses $1B AI ACV, and the Tape Sells the Pop. Maintaining Outperform on a De-Rated Multiple

Results vs. Consensus

YoY View

QoQ View

Revenue Assessment

Subscription growth of +23% CC accelerated from Q1's +19% CC, helped by a full quarter of Armis plus the federal on-prem pull-forward; even discounting both, the organic complex is running above 20% CC at a $15.5B+ run-rate. Demand breadth was the quarter's structural signal: technology workflows closed 50 deals over $1M, security and risk was in 16 of the top 20 with 24 deals over $1M, CRM and industry was in 16 of the top 20, and core business workflows put up 24 deals over $1M on Employee Works demand. By industry, business and consumer services net new ACV grew over 6x YoY, education over 125%, telco and media nearly 40%. The 123 deals over $1M (+~40% YoY) and 32 net additions to the $20M+ customer cohort in twelve months extend the Q1 finding: the largest customers are consolidating spend on the platform faster than the average customer is growing.

Margin Assessment

Non-GAAP operating margin of 29.5% beat guide by 300bp and held flat YoY despite absorbing three acquisitions; that is the operating-leverage story working as promised. The gross-margin line is where the strain shows: non-GAAP subscription gross margin of 80.5% is down 250bp YoY, and the full-year guide was trimmed from 81.5% to 81% on faster hyperscaler migration and accelerating AI consumption. Management's argument, that hyperscaler unit costs fall as volume ramps and AI serving costs are managed by routing work to cheaper models, is credible but is now a multi-quarter proof burden. GAAP operating margin of 4% (from 11%) is the price of the M&A cycle: purchased-intangible amortization ran $219M in the quarter versus $25M a year ago and persists for years, not quarters. The company's FAD commitment to bring SBC under 10% of revenue by 2029 (from 16.5% today) is the other bridge between the two margin presentations that now needs visible progress.

EPS and Below-the-Line Assessment

Non-GAAP EPS of $0.90 (+11% YoY) beat by a nickel on revenue flow-through. Below the line is where the balance-sheet transformation reads through: interest income fell to $70M from $116M as over $3.3B of cash and investments has been consumed year-to-date (alongside new borrowings) to fund the acquisition cycle, including the $7.45B Armis close, and the quarter's GAAP net income of $298M was flattered by a $273M unrealized gain on strategic investments and a $51M valuation-allowance release, both excluded from non-GAAP. The diluted share count of 1,034M was roughly flat sequentially: with no Q2 repurchases, the Q1 ASR's retirement work is done and SBC issuance is again running unopposed.

FCF and Capital Structure Assessment

Q2 FCF of $634M (16% margin) is seasonally consistent, and H1 FCF of $2.3B (29.5% margin) supports the maintained 35% full-year guide. The capital structure is another matter: ServiceNow exited the quarter with $2.08B of short-term debt (including commercial paper), $5.44B of long-term debt, and $6.7B of cash and investments, a swing from roughly $8.6B net cash at year-end 2025 to modestly net-debt today. Goodwill rose from $3.6B to $9.8B and intangibles from $1.1B to $3.8B in six months. Repurchases were zero in Q2 with $4.2B of authorization outstanding, while the stock made new 52-week lows. The company has effectively spent its buyback capacity on Armis and is asking shareholders to accept M&A as the higher-return use of the balance sheet at the exact moment the market is pricing the equity at a historic discount.

Workflow & Portfolio Performance

Security & Risk — The New Second Pillar

The combination of Armis and Veza is pulling the core along with it: security and risk appeared in 16 of the top 20 deals with 24 transactions over $1M, and management now frames the franchise as a ten-figure ACV business growing faster than any of the top ten pure-play cyber vendors. Federal and public-sector demand is a disproportionate driver, with agencies consolidating asset discovery and incident response on the platform.

Assessment: The strategic logic (CMDB + identity governance + asset visibility + agentic response in one command center) is coherent and differentiated. The irony that this pitch scaled in the same month a critical vulnerability in ServiceNow's own platform was exploited in the wild is addressed below; it is the pillar's live stress test.

CRM & Industry — $2B ACV, Voice as the Wedge

CRM net new ACV accelerated again both YoY and QoQ, Sales CRM average deal size doubled YoY, and the platform is on track to execute over 2 billion CRM service cases this year. The marquee proof point: a large airline running customer-service voice calls entirely on ServiceNow voice AI agents, live in production and handling 5 million annual calls in year one. A partner closed a full front-office replacement of a major CRM deployment in two months.

Assessment: The Q1 question was whether the 5x NNACV growth could sustain off a small base; the Q2 answer is that the base is now $2B and still accelerating. Voice is the first CRM surface where agentic AI is a categorical advantage rather than a feature, and it plays directly to ServiceNow's deterministic-workflow positioning.

Employee Works — Acquisition Compounding

Employee Works deal volume grew over 150% QoQ, continuing the Moveworks integration's trajectory as the AI front door for the enterprise workforce, with Fortune 500 adoption pulling through broader HR and employee-experience opportunities.

Assessment: Two quarters post-merge, the acquisition-amplifies-the-core pattern is holding. The absent disclosure is ACV; deal-volume growth rates will need to convert to a disclosed revenue line eventually.

ServiceNow AI — From Target to P&L Line

AI ACV crossed $1B with net new AI ACV up over 40% sequentially; deals with 5+ AI products grew 5.5x YoY; first-time agentic-AI buyers grew over 45% YoY, and the share of renewal customers buying agentic AI for the first time doubled both QoQ and YoY. Price uplift on the new AI-native SKUs is landing within the 20-30% framework laid out at FAD, and Pro Plus uplift remains above 30%.

Assessment: The single most important number in the print. A year ago the bear case called Now Assist a feature; the franchise has now compounded through the $250M, $600M, and $1B ACV marks in five quarters and management raised the bar to $1.5B by year-end. Tracking ahead of the FAD target of AI at 30% of ACV by 2030.

Key KPIs

Key Topics & Management Commentary

Overall Management Tone: The most openly market-facing call of this coverage arc. Management structured the entire prepared script as a point-by-point rebuttal of the bear narrative, closing with an explicit plea for a re-rating of the stock, a first after four quarters of letting the numbers speak. The posture was assertive and at moments campaigning rather than reporting; the numbers detail stayed crisp and the only hedged notes came on gross margin and the deliberately conservative back-half guide. Versus Q1's confident-but-conventional posture, this call read as management litigating the share price directly.

1. "We Are Who We Said We Were": Prepared Remarks as Bear Rebuttal

The CEO organized the script around the questions dominating investor meetings: is ServiceNow a cybersecurity company, has AI adoption inflected, is CRM traction real, does the HR position hold, will AI budgets crowd out the platform. Each got a data-backed answer, and the close made the intent unmistakable.

"Our position is real, it is durable, it is compounding. This is the foundation for a re-rating of ServiceNow. Let's get it started today." — Bill McDermott, Chairman & CEO

Assessment: When management starts arguing with the multiple, it usually means the multiple is winning. The substance of the rebuttals was strong; the fact that they were necessary is itself the quarter's story. We note management cannot re-rate a stock; only sustained delivery through the narrative cycle can.

2. ServiceNow AI Crosses $1B ACV, Bar Raised to $1.5B

"ServiceNow AI continued to outperform expectations in Q2 with ACV crossing over $1 billion and net new ACV growth accelerating sequentially, growing over 40% quarter-over-quarter." — Gina Mastantuono, President & CFO

Assessment: The 2026 AI target has moved from $1B (set 2025) to $1.5B (set at FAD) and is now tracking ahead again. Accelerating sequential NNACV at a $1B base is the strongest available counter-evidence to the thesis that agentic AI disintermediates the platform.

3. Cybersecurity: From Workflow Vendor to Top-10 Cyber Platform

"I'll make it very clear. We now have a 10 figure, cybersecurity business that's growing faster than all the other top cybersecurity companies." — Bill McDermott, Chairman & CEO

Management framed the stack as pre-breach (Armis exposure and vulnerability management), post-breach (the incumbent security-operations franchise), and runtime security, unified by AI Control Tower, with 2.2 billion AI agents entering the enterprise as new identities to govern and Armis already tracking 7 billion devices in real time.

Assessment: The ambition escalated notably: the CEO predicted cybersecurity could become "bigger than ServiceNow is" within a few years and that the company would contend for the top spot from #8. That is either the next decade's growth engine being named in real time or scope creep at a $7B+ acquisition cost. We lean toward the former on the deal-count evidence, but the bar management just set for itself is high.

4. The Armis Balance Sheet: $7.45B Out, $7.5B of Debt On

The quarter's cash flow statement shows the acquisition cycle's full weight: $7,451M paid for business combinations net of cash (the Armis close), funded by $3.94B of senior notes, a term loan drawn and repaid, and roughly $2.1B of net commercial paper. ServiceNow now carries $7.5B of total debt against $6.7B of cash and investments, its first materially levered posture as a public company.

Assessment: The strategic case for Armis was argued convincingly for two quarters; what is new is the financing texture. Interest income has already halved, buybacks stopped, and the company's famous balance-sheet optionality is spent for several quarters. Execution now has to be near-flawless because the cushion is gone.

5. Federal Strength and the Pull-Forward Asterisk

"Part of that outperformance was due to strong U.S. federal demand, which shifted some on-prem revenue from Q3 to Q2. This is simply about timing. I'm very confident in our Q3 and full year guidance." — Gina Mastantuono, President & CFO

Assessment: Credit for volunteering the disclosure unprompted. Federal strength is genuine (the Air Force expansion, agencies consolidating on Armis), but the pull-forward means Q3's optical growth will look softer than underlying demand; the CC growth guide of +20% for Q3 already embeds that. Watch for the Street to misread Q3 seasonality as deceleration.

6. Subscription Gross Margin: the Cost of the AI Ramp

"Notably, even though we had a little pressure on gross margin, we held the operating margin flat." — Gina Mastantuono, President & CFO

The FY26 subscription gross margin guide came down 50bp to 81%, attributed to customers migrating to hyperscaler-hosted instances faster than planned and accelerating AI consumption. Management argued both pressures are front-loaded: hyperscaler unit costs fall with scale, and model-serving costs are actively arbitraged across open-weight, domain-specific, and frontier models.

Assessment: A second consecutive 50bp trim to the gross-margin guide (82% at the FY25 close, 81.5% at Q1, 81% now). Each cut has a plausible story; the trend is the concern. The offset, holding operating margin via AI-driven OpEx efficiency, has worked so far, and the flat 29.5% YoY print while digesting three acquisitions is genuine evidence.

7. AI-Native SKUs and the Product-Led Motion Beyond the Fortune 500

"We have a big announcement coming very shortly, a business model evolution that expands our TAM with AI native products. This new offering will be a conversational service desk experience. No tickets, and AI coded automation. In a new product-led motion, we're targeting the Fortune 500,000." — Bill McDermott, Chairman & CEO

Assessment: A product-led, down-market motion is genuinely new for ServiceNow, whose model has always been enterprise top-down. If the AI-native conversational service desk ships at self-serve price points, it opens a mid-market TAM the company has never addressed and pre-empts the AI-native startups targeting exactly that entry wedge. Execution risk is real (product-led growth is a different muscle), but the strategic read is defensive brilliance: occupy the low end before the disruptors grow up in it.

8. Adoption Inflection: 9x Agentic Production in Nine Months

"The percentage of renewal customers purchasing a genic AI for the first time, doubled quarter-over-quarter and year-over-year. So customers that weren't already on the AI journey are signing up fast." — Bill McDermott, Chairman & CEO

Assessment: The renewal-cohort statistic matters more than the headline 9x: it means AI attach is penetrating the installed base's laggards, not just the early adopters, which is what converts an adoption curve into a durable revenue layer. Management's "customers only want deterministic" framing is the sharpest positioning line the company has landed against the general-purpose agent labs.

9. Headcount-Flat Commitment: the Margin Bridge Made Explicit

"We will complete this cycle of 2026 and start 2027 with the exact same headcount before we did the acquisitions. So you are going to see operating margins and the free cash flow margins of this corporation scale." — Bill McDermott, Chairman & CEO

Assessment: This is the most concrete, gradeable margin commitment management has made in this coverage arc: absorb Moveworks, Veza, and Armis (thousands of employees) and end the year at pre-acquisition headcount through AI-driven internal efficiency. The $62M severance charge shows the mechanism already running. We put this on the commitments-to-watch list with a hard grade date of the Q4 print.

10. FAD Framework: $30B+, Rule of 60+, SBC Under 10%

The May 4 Financial Analyst Day formalized the long-term model: $30B+ subscription revenue by 2030, AI at 30% of ACV, Rule of 60+ (CC subscription growth plus FCF margin), and SBC below 10% of revenue by 2029. The quarter operated at Rule of 56.

Assessment: The 2030 frame requires roughly 14%+ compound subscription growth from here, which current momentum comfortably exceeds; the harder targets are the SBC reduction (16.5% of revenue today, needs to fall by roughly 40% in three years) and holding Rule of 60 through the gross-margin transition. FAD delivered the framework asked of it in our Q1 preview; the market's response, a further 50% de-rating, shows frameworks are not currently the binding constraint.

11. Ecosystem: Governance Layer for Everyone Else's Agents

The quarter's partnership slate all points one direction: NVIDIA's Project Arc secured by OpenShell and governed by AI Control Tower; Microsoft Agent 365 governance extended across the ecosystem; Anthropic as first design partner for Action Fabric, connecting Claude directly to ServiceNow workflows and actions; $1B+ in cumulative AWS Marketplace transactions; Accenture co-selling legacy-cybersecurity migration.

Assessment: The partnership pattern is the platform's strongest structural answer to the displacement narrative: the frontier labs are wiring their agents into ServiceNow's governance and action layer, not around it. Every third-party agent governed by Control Tower converts a potential disruptor into a billable workload.

Guidance & Outlook

Q3 2026 Guide

The guidance texture matters more than the headline. The full-year raise of $15M at the midpoint looks stingy against a quarter that beat by roughly $57M, but the CFO's arithmetic is internally consistent: about half the Q2 beat was Q3 revenue arriving early (no full-year effect), and the full net new ACV outperformance was passed through, with the back half deliberately held prudent "especially in this market environment." The dollar's renewed strength since March adds a ~$35M FX headwind to Q3 cRPO, worth about a point of optical growth. The maintained 31.5% operating margin and 35% FCF margin guides absorb the Armis dilution as promised at Q1.

Implied Q4: The FY26 midpoint implies Q4 subscription revenue of roughly $4.24B, about +22.5% reported YoY, meaning the guide assumes no material deceleration into year-end despite the conservatism framing.

Street at: Consensus sat near $15.75B before the print; the new $15.77B midpoint nudges above it. Expect estimates to drift up on cRPO flow-through rather than the guide itself.

Guidance style: The pattern is now five consecutive quarters of beat-the-high-end against guides management itself calls prudent. The conservatism is a feature: in a market punishing any wobble, ServiceNow has chosen to keep the bar where it cannot miss.

Analyst Q&A Highlights

Autonomous L1 Agents: Adoption Proof and the Monetization Path

The opening question asked for a progress report on the Level 1 ITSM autonomous-agent product cycle launched at Knowledge, and for early monetization data. Management's answer was the most quantified agentic-AI disclosure to date: 40+ customers live, autonomous closure of the large majority of service requests, and resolution times collapsing from days to minutes, with the CFO tying the consumption model directly to the 30%-of-ACV-by-2030 AI target.

Q: "I wanted to ask you about your Level 1 ITSM product cycle. I know you talked about it at knowledge and went generally available you pilot customers, all that good stuff. My question to you is, what are your options over the last 3 months? What's working with that product cycle, what's the feedback in terms of how you can make it better."
— Gabriela Borges, Goldman Sachs

A: "Today, 40-plus customers are using it already getting a lot of good benefits. We're talking about closing down close to 80% to 85% of the service request without having any human interaction. ... So for example, some of these requests might take 2 days for humans to get to our result. We're doing that in 20 minutes."
— Amit Zavery, President, CPO & COO

Assessment: The 80-85% autonomous-closure claim is the product's core economic argument: it monetizes labor substitution, not software seats. If that number holds at scale it rewrites the company's pricing power; it also deserves scrutiny, as it is a vendor-reported figure from an early-adopter cohort.

Quality of the Beat: Pull-Forward Mechanics and Guide Prudence

The most analytically pointed exchange pressed on why a 150bp revenue beat produced only a $15M full-year raise. Management decomposed the beat cleanly and defended the back-half conservatism as deliberate positioning for the environment.

Q: "Just help us understand the relatively modest full year subscription revenue raise versus the Q2 outperformance? Is there more conservatism other dynamics we should be thinking about?"
— Tyler Radke, Citi

A: "We talked about the fact that about half of the beat is a pull forward of on-prem for a quarter, right? So that won't impact Q3 versus Q2. From a full year revenue guide, we passed along all of the net new ACV overachievement. But yes, I'm always looking, especially in this market environment, at a conservative prudent guide, right? There's certainly upside, but we feel really good about the guide right now."
— Gina Mastantuono, President & CFO

Assessment: A direct, arithmetic answer with no dodge. The disclosure discipline here (volunteering the pull-forward rather than banking the optical beat) is why this management team's guides retain credibility five quarters into the beat streak.

AI Pricing Under Competitive Cross-Fire

A recurring line of questioning asked whether competitors' outcome-based pricing experiments would pressure ServiceNow's hybrid seat-plus-consumption model, and what happens to AI pricing as the module count multiplies. Management's answer rejected outcome-based pricing as unmeasurable in practice and defended hybrid predictability as the customer-preferred structure, with the CFO adding that realized uplifts are holding the FAD framework.

Q: "Your pricing is token-based, assist based and competitors like Salesforce offer outcome-based that could introduce pricing pressure. ... do you think on top of the value -- can you also offer price advantage versus competition? What happens to pricing of AI given all these differences between you and competitors?"
— Tal Liani, Bank of America

A: "Hybrid pricing structure, which we introduced where it's a combination of license and usage directly applies to giving the customer predictability and flexibility. And it does -- when you use more, that means you are getting good outcomes, otherwise, you would not use more. So outcome is already tied in the underlying assumption of usage."
— Amit Zavery, President, CPO & COO

Assessment: The philosophical defense is sound (usage embeds outcomes), and the 20-30% realized uplift data supports it. The unstated risk is deflation per unit of AI work as model costs fall; management's answer is that it prices the resolution, not the tokens, which holds only as long as resolutions stay differentiated.

Are AI Budgets Crowding Out Software Spend?

The dominant macro topic in Q&A, raised repeatedly in different forms: peers and neighboring vendors have reported customers delaying purchases as capital shifts to AI hardware and model spend, and the questioning asked directly whether ServiceNow sees elongating sales cycles. Management's denial was categorical, arguing the platform sits inside the AI budget rather than beside it, and pointed to elongating contract durations as counter-evidence.

Q: "There have been a couple of recent indications from tech vendors at organizations have become more consumed with hardware and AI spending, and this is having a knock-on effect on other aspects of IT spend. So just to ask this directly, have you seen any impact to sales cycles with ServiceNow?"
— Multiple analysts incl. Gregg Moskowitz, Mizuho; Matthew Hedberg, RBC

A: "I have not. In fact, the key here ... is this has really become a C-suite story. ... So many of them have such a fragmented enterprise that they need to take advantage of our workflow automation above the mess and click in that AI engine into this unlivable platform and really begin running. So the higher you go in the C-suite, the more relevant we are."
— Bill McDermott, Chairman & CEO

Assessment: This is the exchange the whole tape was listening for, one day after a smaller workflow peer blamed a miss on exactly this dynamic. The categorical "I have not," backed by 123 seven-figure deals and lengthening durations, is as strong a denial as the data permits. It is also the claim with the most riding on it: if cycles do elongate, this quarter's transcript becomes the reference point.

Gross Margin: Hyperscaler Mix and AI Consumption Costs

A focused exchange on the tension between accelerating AI consumption and the trimmed gross-margin guide. Management framed both drivers as good problems with front-loaded costs, and emphasized that operating margin held despite them.

Q: "How are you managing the tension between an increase in consumption mix and better token pricing going forward? And how should we think about the upside versus down risk to gross margin from here?"
— Brad Zelnick, Deutsche Bank

A: "Both good reasons for a little bit of short-term pressure on margins. Notably, even though we had a little pressure on gross margin, we held the operating margin flat. ... we absolutely think that mid and longer term, we have some opportunity to continue to increase."
— Gina Mastantuono, President & CFO

Assessment: The answer leaned on scale economics and model arbitrage (routing work across open-weight, domain, and frontier models). Credible, but this is now the second consecutive quarter the gross-margin guide moved down while the operating-margin guide held; the offset is coming from OpEx, which has a floor.

How Far Does the Cybersecurity Ambition Go?

A strategy question probed the boundaries of the security buildout: what is organic versus acquired from here, and how the company thinks about converging cyber and observability. Management described the current portfolio as an organic story on an acquired foundation and explicitly disclaimed ambitions to cover every security category.

Q: "You have Vesa and Armis, which are intriguing assets, but there's a lot of areas where you don't play or don't participate. ... how do you think about revisiting your strategy also on the observability side?"
— Keith Bachman, BMO

A: "We're building the world's most integrated end-to-end security platform. I think that's the way you should think about it. ... What I laid out today was an organic story."
— Bill McDermott, Chairman & CEO

Assessment: The "organic story" framing is the balance-sheet reassurance the leverage question needed: management is signaling the M&A cycle is complete and the security franchise now compounds on what was bought. The friendly nods to the incumbent cyber vendors read as deliberate channel diplomacy while the company takes their budget.

AI Control Tower: Right to Win in a Crowded Control Plane

The final strategic exchange asked why ServiceNow wins the AI-governance control plane against every platform vendor claiming the same ground, and requested customer proof. Management grounded the answer in the CMDB heritage (governance requires a system of record of everything being governed) and cited adoption evidence.

Q: "Obviously, that's a huge opportunity and a critical function. But obviously, also a very competitive space -- could you talk a little bit about your right to win there?"
— Adam Wood, Morgan Stanley

A: "We build a control tower with that kind of foundation where we would be able to discover anything an enterprise is running which includes now AI agents, any devices which are connected to it. And it's a very heterogeneous. This work across multiple domains. It's not really for ServiceNow assets, but everything else."
— Amit Zavery, President, CPO & COO

Assessment: The 500+ Control Tower customers in the product's first six months (cited in the fuller answer) is the fastest zero-to-adoption ramp of any product in the company's history, and heterogeneity (governing non-ServiceNow agents) is the moat claim that matters. The Q1 read that Control Tower is the strategic high ground has strengthened.

What They're NOT Saying

1. The actively exploited platform vulnerability. A critical pre-authentication remote-code-execution flaw in the ServiceNow AI Platform (CVE-2026-6875, publicly disclosed July 13, in-the-wild exploitation confirmed days before the call) received no direct acknowledgment in the prepared remarks and drew no analyst question. For a company that spent half the call selling itself as a top-10 cybersecurity vendor, the silence on its own security event is the report's most conspicuous omission. Patches shipped ahead of disclosure and hosted instances were remediated first, but self-hosted customers remain the exposed tail, and the reputational collision with the security pitch is unaddressed.
2. Zero Q2 buybacks at the lows. $4.2B of authorization sat unused while the stock made 52-week lows; the cash went to Armis. The Q1 call celebrated a $2B ASR at ~$190; no one explained why $94 merited nothing. The honest answer is the balance sheet is spent, which is exactly why the question deserved airtime.
3. The GAAP margin collapse. GAAP operating margin fell from 11% to 4% and GAAP EPS declined 22% YoY; neither number was spoken on the call. The 25.5-point GAAP-to-non-GAAP operating-margin gap is the widest in company history and will persist while $219M/quarter of purchased-intangible amortization runs through.
4. The cost of the debt. $7.5B of borrowings arrived this quarter; no interest-expense guidance, blended cost of debt, or deleveraging timeline was offered. Interest income already halved YoY.
5. FX. The ~$35M Q3 cRPO headwind from renewed dollar strength appears in the release's outlook text but was not mentioned once on the call.
6. Seat and ARPU disclosure. Management asserted active seats are rising and durations lengthening, both directionally supported by RPO, but the company still discloses no seat counts, ARPU, or consumption-revenue split, leaving the seat-compression debate unfalsifiable with public data.
7. Veza specifics. Armis and Moveworks got operating metrics; Veza's contribution, integration status, and deal traction went unquantified for a second consecutive quarter.

Market Reaction

• Pre-print setup: NOW closed July 22 at $95.46, down 6.5% on the session as a smaller workflow-software peer's miss (attributed to customers delaying purchases amid AI-market changes) dragged the whole group, compounding a week of headlines about the platform vulnerability. The stock entered the print down 37.7% YTD and 50.4% over the trailing twelve months, versus the S&P 500 up 9.5% YTD; the trailing 30 days were +2.6%. The 52-week closing range: $83.00 to $199.24.
• After-hours move: Shares rose solidly on the release, with the cRPO beat (matching Q1's pace against a deceleration guide) cited as the key metric in early coverage.
• Reaction session (in progress at publication): The stock opened higher and traded to an intraday high of $100.25, up 5.0%, before fully surrendering the gain; as of midday it sat near $93.90, down 1.6%, below the pre-print close.

The fade is the analytical event. A quarter that beat the high end of every guided metric, crossed $1B of AI ACV, and re-committed the long-term framework bought roughly two hours of appreciation. This is what a narrative-dominated regime looks like: the tape is not pricing ServiceNow's results, it is pricing the proposition that agentic AI makes per-seat enterprise software structurally obsolete, and no single print can falsify that proposition. The entry setup did the print no favors either: the prior session's 6.5% peer-contagion drop meant part of the after-hours pop was mere retracement. What ends this regime is repetition (several more quarters of 20%+ CC growth with AI ACV compounding) or a sector event that breaks the displacement narrative externally. Until then, fundamentals are accumulating in the price like a coiled spring rather than being paid for on delivery.

Street Perspective

Debate: Is the "AI Displaces SaaS" Thesis Right About ServiceNow?

Bull view: The bull camp argues ServiceNow is the anti-fragile name in the selloff: agents need governance, context, and a system of action, which is exactly what the platform sells, and the quarter's evidence (AI ACV over $1B, frontier labs integrating into Control Tower, first-time agentic buyers up 45%) shows AI is a demand engine, not a substitute. Some desks call the software selloff disconnected from checks with IT buyers.

Bear view: The bear camp contends the disruption is early, not absent: model labs are moving up the stack, AI-native startups price at a fraction of enterprise SaaS, and budget reallocation toward AI infrastructure is already lengthening cycles at weaker vendors, with stronger ones next. On this view the multiple compression is rational anticipation, not panic.

Our take: The bull argument is winning on evidence and losing on flows. Nothing in this print, or the last four, supports displacement at ServiceNow specifically; everything in the tape says the market will not pay for that distinction yet. We side with the evidence and accept that the timing of recognition is unknowable.

Debate: Prudent Guide or Quiet Warning?

Bull view: The modest $15M raise is mechanical (half the beat was timing) plus deliberate conservatism in a market that executes any miss; the implied Q4 of ~$4.24B still assumes no deceleration, and five straight high-end beats say the guide is a floor.

Bear view: Skeptics read the small raise plus the second consecutive gross-margin trim as management quietly building cushion against softening pipeline, with the federal pull-forward flattering Q2 at Q3's expense.

Our take: The CFO's decomposition was transparent and arithmetically complete, and cRPO (the forward indicator) beat by 200bp; a company hiding softness does not volunteer its pull-forwards. Prudence, not warning, but the gross-margin trend deserves its own watch line.

Debate: The Cyber Pivot — Second Act or Expensive Distraction?

Bull view: A $1B+ security business growing fastest among the top ten, with Armis pulling core deals along (16 of the top 20), validates the ~$8.8B acquisition outlay; security spend is the most recession- and AI-proof budget in enterprise IT.

Bear view: The company levered a previously pristine balance sheet, paused buybacks, and accepted years of amortization drag to enter a market defended by focused incumbents, and its own platform just suffered an actively exploited critical vulnerability, undercutting the security-vendor credibility it is spending billions to build.

Our take: Lean bull on the strategy, watchful on the execution. The deal-count evidence of pull-through is real and immediate. But the vulnerability episode shows the reputational stakes have risen: a security vendor's own breach economics are categorically worse than a workflow vendor's, and ServiceNow has chosen to hold itself to the higher standard.

Debate: Where Is the Buyback?

Bull view: Armis at strategic value beats buybacks at any price; the headcount-flat commitment restores FCF trajectory, and repurchases resume in 2027 against a de-levered balance sheet and a (presumably) higher stock.

Bear view: Management repurchased $2B at ~$190 and nothing at $94, the reverse of value discipline; with FCF committed to debt service and integration, the shareholder-return story is dead money for four to six quarters.

Our take: The bear critique of sequencing is fair, but backwards-looking. From here, every incremental dollar of the remaining $4.2B authorization deployed below $100 is dramatically more accretive than the Q1 ASR; management should be pressed on pace at the Q3 print, and resumption would be a strong signal.

Model Update Needed

Valuation: At ~$94, market cap is roughly $97B and enterprise value roughly $98B against ~$5.7B of FY26E FCF (guide) and ~$7.1B of FY27E FCF: approximately 17x and 14x EV/FCF respectively, and ~24x FY26E non-GAAP EPS, for a business guiding +21% CC subscription growth at Rule of 56. These are multiples the market last afforded ServiceNow in the 2022 rate shock, and they now embed permanent growth impairment. PT framework reset: Base $150 / Bull $190 / Bear $80 (from $280/$320/$190), built on ~22x / ~28x / ~13x FY27E FCF. The reset reflects the multiple regime, not the fundamentals: we are marking our framework to a market that has repriced the entire software complex, while the base case still assumes only partial multiple recovery. From ~$94: base +60%, bull +102%, bear -15%. Up/down roughly 4:1.

Thesis Scorecard

Overall: The standing thesis strengthened on every operating pillar and was simultaneously overwhelmed by the one variable outside management's control: the multiple regime. Two new risk factors enter the framework this quarter: the AI-displaces-SaaS narrative as a persistent valuation ceiling, and platform-security incidents as a tail risk that now carries strategic (not just operational) cost given the cybersecurity positioning.

Action: Maintaining Outperform with HIGH conviction. For long-horizon capital, this is the most attractive entry point in the coverage arc: a Rule-of-56 compounder at ~17x FCF with its AI product line compounding through targets. The honest caveats: the catalyst path is repetition rather than any single event, the balance sheet has no slack for error, and the narrative regime can persist longer than conviction usually survives. Grade the headcount-flat commitment at Q4; watch buyback resumption and the gross-margin line at Q3.