Q3 2026 Recap: ServiceNow Raises Its AI Target Again, Guides cRPO to 18.5%, and Suffers the Worst Session in Its History. Maintaining Outperform on a Trimmed Framework

Results vs. Consensus

Q3 2026 Scorecard

Year-Over-Year Comparisons

Quarter-Over-Quarter Comparisons

Revenue Assessment

Subscription growth of +20.5% CC decelerated 250bp from Q2's +23% CC. Most of that deceleration is mechanical: Q2 carried federal on-prem revenue that management explicitly said belonged in Q3, and Q3 absorbed about 25bp from the new deployment-services reclassification. Both effects are timing, not demand. The less comfortable observation is that every growth rate in this print includes acquired revenue. Management told investors at the Q1 call that Armis would contribute 125bp to full-year subscription growth and 125bp to the Q2 subscription and cRPO guides; it gave no updated contribution for Q3 or Q4, and the 10-Q omits pro forma figures as immaterial. Armis alone at a similar rate would put organic growth near 19% CC; adding Moveworks and Veza, organic subscription growth is more likely in the high teens. Demand breadth remained strong: 19 deals above $5M in net new ACV (+135% YoY), technology workflows with 54 deals above $1M, CRM with 28, core business workflows with 24, and Asset Management crossing $1B of ACV. Business and consumer services led industries with net new ACV up over 3x.

Margin Assessment

The operating line is where this management team keeps its promises. Non-GAAP operating margin of 32.5% beat the guide by 150bp, and the full-year 31.5% guide, up 50bp YoY, now absorbs the dilution from three acquisitions. The gross line tells a different story. Non-GAAP subscription gross margin of 80% is down 300bp YoY and 50bp QoQ, and the professional-services line swung from an 11% non-GAAP gross margin to a 20% gross loss as the company began bundling forward-deployed engineering into large AI deals. The full-year subscription gross margin guide held at 81%, so there was no third consecutive cut; holding it requires Q4 at roughly 80.7% or better, achievable but above the Q3 print. The offset that protects the operating line is OpEx efficiency: management now credits internal AI with more than $1B of annualized value. GAAP operating margin recovered to 9.5% from Q2's 4%, still well below last year's 17%, with $198M of purchased-intangible amortization and $87M of severance in the quarter.

EPS and Below-the-Line Assessment

Non-GAAP operating income grew 17% YoY; non-GAAP net income grew only 5%. The gap is the Armis financing. Interest expense was $81M versus $6M a year ago, and interest income fell to $69M from $115M as cash funded acquisitions, a $121M swing in net interest. After tax at the 21% non-GAAP rate (up from 20%), that swing is worth roughly $0.09 per share, so non-GAAP EPS of $1.00 would have been near $1.09 on last year's capital structure. That is where the two-cent miss against the $1.02 consensus came from: operating income beat, so the shortfall sits entirely below the operating line, where net interest now runs at a cost. GAAP EPS of $0.27 fell 44%: the GAAP-to-non-GAAP operating-margin gap narrowed to 23 points from 25.5 in Q2 but remains far wider than the 16.5-point gap a year ago. Diluted shares of 1,043M were down 0.5% YoY, the residual effect of the Q1 accelerated repurchase; with no buybacks since, SBC issuance is again unopposed.

FCF and Capital Structure Assessment

Q3 FCF of $822M (20% margin) sits in the seasonally soft middle of the year (Q2 was 16%), and the full-year target rose to 36% from 35%, "driven by lower CapEx spend." Year-to-date capex is $346M against $630M in the first nine months of 2025. The balance sheet improved at the margin: total debt of $7.52B ($2.09B short-term, largely commercial paper, plus $5.44B long-term) against $6.96B of cash and marketable securities leaves net debt near $0.56B, narrower than the roughly $0.8B at Q2. Goodwill reached $10.0B. Business-combination outflows were $391M in the quarter, and the call named a tuck-in acquisition of Sweep. Repurchases were zero for the second consecutive quarter, with the $4.2B authorization untouched while the stock ran from $95 to $150 and then fell back to $110.

Workflow & Portfolio Performance

Technology Workflows: The Core Is Not Shrinking

The core IT franchise is the part of the business the displacement narrative says should erode first, and it printed 54 deals above $1M, up from 50 in Q2. ITSM AI deal volume rose nearly 80% YoY, ITOM AI deal volume more than 7x, and Asset Management crossed $1B of ACV, in 13 of the top 20 deals. Management's argument is that every new AI agent is another asset to discover, monitor, and govern, which pulls the CMDB and IT operations deeper into the AI estate rather than routing around it.

"Demand for our core workflows remains robust as every new AI asset and agent introduces greater requirements to manage, monitor, secure and govern the enterprise." — Gina Mastantuono, President & CFO

Assessment: The Asset Management milestone is the quarter's most underrated datapoint: a governance-adjacent module reaching ten figures is direct evidence for the "more agents, more governance spend" thesis. The drop in ITSM and ITOM top-20 presence is mix within a single quarter's largest deals, not a trend we would extrapolate.

Security & Risk: Big Claims, Fewer Big Deals

Security and risk was again in 16 of the top 20 deals, and security and risk AI net new ACV grew more than 4.5x YoY. The company launched Autonomous Security, a six-solution suite fed by Armis and Veza, and repeated its claim to be "the fastest-growing cybersecurity company at scale." But the count of security deals above $1M fell to 13 from 24 in Q2, and the call gave no update to the $1B+ security ACV disclosed in Q2.

"Our security business is differentiated with the potential to be even bigger in a few years than all of ServiceNow is today." — Bill McDermott, Chairman & CEO

Assessment: One quarter of fewer seven-figure deals does not break the pillar, particularly after a Q2 inflated by Armis-led federal consolidation. But the rhetoric keeps escalating while the disclosure thins, and that combination deserves skepticism until security ACV is updated.

CRM & Industry: The Quarter's Clearest Acceleration

CRM and industry workflows produced 28 deals above $1M, nearly double Q2's 15, and CRM AI net new ACV tripled both QoQ and YoY. The company added Sweep, a tuck-in focused on rapid agentic deployment, and management claims CRM implementations that once took years now take 30 to 60 days. Customer proof points included Jack Henry's CRM modernization and the Bundesliga consolidating fan services from seven channels to one, resolving 89% of inquiries at first contact.

"As enterprises rethink the mandate of their CRM systems, ServiceNow is leading the way. The old CRM was built for sales teams. The new one has to work across your entire business." — Bill McDermott, Chairman & CEO

Assessment: CRM is the pillar where ServiceNow is the disruptor rather than the incumbent being disrupted, and Q3 brought the highest disclosed count of $1M+ CRM deals in our coverage. Implementation speed is the binding constraint in CRM switching; if Sweep compresses deployment times as claimed, the rip-and-replace motion scales.

Core Business and EmployeeWorks: An Acquisition Reaches Scale

EmployeeWorks, the agentic front door built on the Moveworks acquisition, reached $100M of ACV in less than three quarters. Core business workflows were in 15 of the top 20 deals with 24 above $1M, and HRSD AI deal volume nearly doubled. The company added no-code agent building to EmployeeWorks and a vibe-coding developer environment, ToolSmith.

"Our agentic front door EmployeeWorks reached $100 million ACV in less than three quarters." — Bill McDermott, Chairman & CEO

Assessment: This closes the Q2 open question about converting EmployeeWorks deal-volume growth into a disclosed revenue figure. A nine-figure ACV run-rate in three quarters is a fast integration by any enterprise-software standard.

ServiceNow AI: A $1.7B Target and Six Billion-Dollar Businesses

AI net new ACV grew 65% YoY, deals with five or more AI products grew more than 8.5x, and million-dollar AI deals grew 4.5x. Management said six businesses now exceed a $1B run-rate. The new disclosure was consumption: agentic AI consumption in production is up more than 13x year-to-date, and Action Fabric assists grew 500% month-over-month.

Assessment: The AI pillar is the strongest it has been. Its weakness as evidence is that every consumption metric is a growth multiple without a dollar base, which makes it impossible to size how much of the 13x will show up in 2027 revenue.

Key KPIs

Key Topics & Management Commentary

Overall Management Tone: Management was maximally confident and almost entirely forward-looking, leaning on new branding ("Super Intelligence Control Tower"), growth multiples, and partner announcements rather than on the financial bridge investors needed. The prepared remarks gave the cRPO deceleration two sentences; the full bridge emerged only when the second analyst raised the after-hours sell-off directly, as a list of accounting effects delivered firmly but only partly quantified. Versus Q2's campaign-style rebuttal of the bear case, the tone was less defensive and more promotional; it was also less precise.

1. The "Trifecta" and the Super Intelligence Control Tower

The prepared remarks framed the quarter around three claims: a beat across the board, accelerating AI monetization, and a business model shifting toward consumption. The product positioning moved up another level. Q2's "AI Control Tower" is now the "Super Intelligence Control Tower," pitched as the harness that combines orchestration, governance, and security for enterprise AI, with a new AI Passport that attests each agent's permissions and lets the Control Tower revoke them at the infrastructure layer through NVIDIA, Okta, Microsoft, and Google.

"As you'll hear, these Q3 results represent a trifecta moment for ServiceNow, quarterly results beating expectations across the board again, significantly accelerating AI performance at the largest quantum yet, a hybrid business model shifting more to consumption." — Bill McDermott, Chairman & CEO

Assessment: The strategic logic is sound: an enterprise running thousands of agents needs a system of record for what each one may do, and the CMDB heritage is a genuine right to win. The naming inflation is not analysis, however, and investors heard "trifecta" alongside a cRPO guide that decelerated. The market judged the numbers, not the nouns.

2. AI ACV Target Raised to $1.7B+, the Second Raise This Year

The company started 2026 targeting $1B of AI ACV by year-end, raised that to $1.5B at the May Financial Analyst Day, and now targets $1.7B+, with the CEO saying he has challenged the team to exceed it. AI net new ACV grew 65% YoY, and management reiterated that AI is on track, "if not actually ahead of schedule," to reach 30% of ACV by 2030.

"As a reminder, we began the year with a $1 billion AI target by year-end. We raised that target to $1.5 billion at Financial Analyst Day in May. Today, we are raising that target again this time to $1.7 billion plus by year-end." — Bill McDermott, Chairman & CEO

Assessment: A 70% raise to a target within one year is the clearest evidence in the coverage arc that AI is incremental demand, not a feature. The open question, raised in different forms by several analysts, is why $700M of incremental AI ACV is not visible as acceleration in total subscription or cRPO growth. The likely answer is substitution within existing contracts (AI-native SKUs replacing older tiers at a 20-30% uplift), which is accretive but smaller than the headline suggests.

3. The Q4 cRPO Guide and Management's Four-Part Bridge

The Q4 cRPO guide of 18.5% CC was the single number that moved the stock. Management attributed the step-down from Q3's 21% to four factors: early renewals that moved from Q4 into Q3 (contributing about half of Q3's 100bp beat), a higher mix of federal on-prem deals (about 25bp), the new deployment-services arrangements (described only as a "small modest" RPO headwind), and roughly 50bp of growth in termination-for-convenience (T4C) backlog that is excluded from cRPO but will be recognized as revenue.

"We expect cRPO growth of 18.5% on a constant currency basis. This reflects a higher mix of on-prem deals and excludes a rising balance of T4C backlog, both of which contribute to revenue but not cRPO." — Gina Mastantuono, President & CFO

The arithmetic is the problem. If the early renewals are worth about 50bp, adding them to the 25bp on-prem and 50bp T4C effects gives roughly 125bp, which lifts the 18.5% guide to about 19.75%: still 125bp below the Q3 print and below 20%. That figure still includes Armis, which management credited with 125bp of cRPO growth in its Q2 guide. ServiceNow has beaten its cRPO guide by 100-200bp in each of the last four quarters, so the printed Q4 figure will likely land near 19.5-20.5%. The guide is conservative; the direction is still down.

Assessment: Each item in the bridge is individually plausible, and two (on-prem mix and T4C) are genuinely revenue-positive. But a company asking investors to look through a forward indicator should quantify all of it, and the early-renewal effect and the services headwind were left without a Q4 number. We treat underlying organic cRPO growth as running in the high teens. That is still an excellent number for a $13.65B backlog, but it is not the durable +20% CC growth pillar we have underwritten.

4. The Deployment-Services Carve-Out

The quarter introduced a new accounting effect. For select large AI deals, ServiceNow now bundles deployment expertise (forward-deployed engineers) into the platform contract. Under revenue-recognition rules, the value of that delivery is carved out of subscription revenue and booked in professional services. The impact was about 25bp of subscription growth in Q3 and is expected to be about 50bp in Q4. Management said the value returns to subscription revenue when the contracts renew and that it does not expect to include delivery expertise over the long term.

"From an accounting perspective only, the value of those deployments are carved out as subscription revenue, and they're included in professional services rather than subscription revenue. Notably, upon renewal of those contracts, the value of those services return to subscription revenue." — Gina Mastantuono, President & CFO

The cost is visible: non-GAAP professional-services gross margin swung from +11% to a 20% loss, and professional-services revenue grew 10% YoY. Management's early evidence is that supported customers progressed in AI adoption twice as fast over the first five weeks.

Assessment: Strategically defensible: consumption only pays once customers are in production, and buying adoption speed with engineering is rational. Optically costly: the change was introduced in the same quarter as the cRPO deceleration, without a dollar size, a start date, or an end date. It also signals that AI adoption at scale needs more hand-holding than the self-serve narrative implies.

5. The Consumption Pivot: From Seats to Tokens

The CEO described the business model as "shifting more to consumption." Seat-based subscriptions remain the budgeting anchor, but once customers are live, assist consumption drives expansion. Management cited agentic AI consumption in production up more than 13x year-to-date, Action Fabric assists up 500% month-over-month, and an ambition to raise revenue per user by up to 5x.

"And once they're live, the token economics take off, whereby we have a chance to improve our revenue per user by a factor of 5x." — Bill McDermott, Chairman & CEO

Assessment: A move toward consumption is the right answer to seat compression and is the core of our hybrid-pricing pillar. It also makes the reported numbers harder to read: consumption that is not contracted in advance does not sit in cRPO, so the more successful the pivot, the less cRPO describes the business. That is a legitimate reason for the forward indicator to understate demand, and a legitimate reason for investors to demand new disclosure (consumption revenue in dollars) before giving credit for it.

6. Capex Down, FCF Up: The Hyperscaler Trade

The full-year FCF margin target rose 100bp to 36%, explicitly driven by lower capex. Q3 capex was $91M against $235M a year ago. Management positioned this against investor concern about AI infrastructure spending and leverage across the sector.

"Against the backdrop of growing investor scrutiny around AI-related infrastructure spend and leverage, ServiceNow is positioned to capture the AI opportunity while reducing its CapEx and expanding free cash flow margins." — Gina Mastantuono, President & CFO

Assessment: The FCF raise and the gross-margin pressure are two sides of one decision. In Q2 the company attributed its gross-margin cut to customers moving to hyperscaler-hosted instances; the same migration reduces ServiceNow's own data-center capex. Cost moves from below the FCF line to above the gross-margin line. Net, it is a good trade for shareholders (lower capital intensity, no compute constraint, the "Rule of 57"), but it is not free margin.

7. Subscription Gross Margin: 80% in the Quarter, 81% Held for the Year

Our Q2 commitments list flagged that a third consecutive cut to the 81% subscription gross margin guide would break the pattern claim. There was no cut. But the Q3 print of 80% (non-GAAP) is down 300bp YoY and 50bp QoQ, and the year-to-date figure is 81% only after rounding.

"We continue to expect subscription gross margin of 81% and operating margin of 31.5%, up 50 basis points year-over-year." — Gina Mastantuono, President & CFO

Assessment: Holding the guide is a pass, not a reversal. The drivers (hyperscaler hosting, AI inference costs, purchased-intangible amortization in GAAP cost of revenue) all persist into 2027. We model subscription gross margin flat to down slightly next year and continue to rely on OpEx leverage for operating-margin expansion.

8. Headcount: The Flat-by-2027 Commitment Is Tracking

In Q2 the CEO committed to begin 2027 at the pre-acquisition headcount despite absorbing Moveworks, Veza, and Armis. On this call he said total headcount "has actually slightly declined." The CFO described the mix: quota-bearing sellers up double digits, sales operations down significantly, and AI engineering hires offset by fewer support and overhead roles. Severance of $87M in Q3 (after $62M in Q2) is the mechanism showing up in the reconciliation.

"And so while you might see total sales and marketing headcount come down, actually quota-bearing heads are up double digits, while sales operations, for example, is down significantly." — Gina Mastantuono, President & CFO

Assessment: This is the most concrete margin commitment management has made, and it is ahead of schedule one quarter before the grade date. Growing quota capacity while shrinking total headcount is the operating-model proof that internal AI is real, and it supports the FAD framework's Rule of 60 path even if top-line growth settles in the high teens.

9. Flow Ships: The Product-Led Motion Goes Live

The Q2 call promised an AI-native conversational service desk for the "Fortune 500,000" within weeks. It shipped in October as Flow: self-serve, consumption-billed, deployable the same day, and running outside the core platform. Management positioned it as removing the need for stand-alone AI-native service desks.

"For the Fortune 500,000, we introduced Flow. There are many enterprises that want leading domain expertise but aren't ready for formalized ITSM. Flow fills the gap, a new conversational service desk that is self-serve and consumption build." — Bill McDermott, Chairman & CEO

Assessment: Commitment delivered on time. Flow is strategically important as a pre-emptive defense against AI-native challengers at the low end, but it will be immaterial to revenue for several quarters, and no adoption metrics were given. We will look for a customer count at Q4.

10. The Ecosystem: Every Lab Wired Into the Control Tower

The partner slate expanded again. OpenAI and ServiceNow announced a ChatGPT plugin built on Action Fabric, with planned security solutions using OpenAI cyber models and a Codex integration. The new Control Tower Alliance's founding members include Anthropic, AWS, Cisco, NVIDIA, Okta, and OpenAI. Google Gemini Enterprise now calls ServiceNow workflows headlessly, and NVIDIA's OpenShell enforces AI Passport permissions at the infrastructure layer.

"And if you think about it, as Amit said, ServiceNow basically is the execution layer. And I believe we're the execution layer for the entire AI economy." — Bill McDermott, Chairman & CEO

Assessment: The partnership pattern remains the best structural answer to the displacement narrative: the frontier labs are routing enterprise actions through ServiceNow, and each call burns assists. The risk is the inverse: when the user interface moves to ChatGPT or Gemini, ServiceNow's brand recedes behind someone else's front door, and pricing power depends entirely on the work being impossible to perform elsewhere.

11. Federal: The Biggest Quarter of the Year, and a cRPO Headwind

The public sector delivered in its largest quarter, with federal deals above $1M up more than 30% YoY, and management expects federal net new ACV growth to accelerate in Q4. The same strength explains two of the four cRPO headwinds: on-prem deals recognize revenue at a point in time and add little to cRPO, and federal contracts carry termination-for-convenience clauses that keep backlog out of cRPO.

"We expect federal net new ACV growth to accelerate further in Q4 as agencies increasingly see ServiceNow as the AI operating system for the enterprise." — Gina Mastantuono, President & CFO

Assessment: Federal strength is a genuine positive and a genuine source of reported-metric noise. Investors who model cRPO as a pure demand signal will keep misreading a company whose fastest-growing customer segment is structurally under-represented in it.

12. Capital Allocation: A Second Quarter With No Buybacks

The cash flow statement shows no repurchases in Q3, after none in Q2. The remaining $4.2B authorization sat idle while the stock closed between $91.94 and $158.96 between the two prints. Capital went instead to debt service, acquisitions ($391M of business-combination outflows in the quarter; the call named the Sweep tuck-in), and rebuilding liquidity. Neither the CEO nor the CFO mentioned buybacks; the CEO instead stressed that the company has no capital or compute constraints.

"We're not compute constrained. We don't have capital constraints. We're activating a hybrid business model." — Bill McDermott, Chairman & CEO

Assessment: "No capital constraints" and zero buybacks at $110 are hard to hold together. Net debt is now only about $0.56B, so the balance sheet can support repurchases. At the post-print price, a resumption would be the most credible signal management could send that it believes its own Q4 bridge.

Guidance & Outlook

Q4 2026 Guide

Full-Year 2026 Guide

The subscription guide is better than its $15M raise suggests. The prior range implied Q4 subscription revenue of about $4,244.5M at the midpoint; the new Q4 range centers on $4,250M even after absorbing a roughly 50bp (about $17M) deployment-services reclassification. On a CC basis, Q4 subscription growth is guided to accelerate to 21.5% from Q3's 20.5%, helped by an easier Q4 2025 comparison (+19.5% CC) and federal on-prem timing. The CFO said the full-year raise passed through all of the FX benefit and half of the Q3 beat, holding back the other half only because of the services carve-out. The cRPO guide is the exception, and it is the one investors read as the 2027 indicator.

Implied Q4: The 36% FCF margin on roughly $16.25B of total revenue implies about $5.85B of full-year FCF, or roughly $2.7B in Q4 after $3.12B year-to-date: a Q4 FCF margin in the low 60s, against 57% in Q4 2025. Demanding but consistent with Q4 collections seasonality and the capex reduction.

Street at: The Q4 subscription guide landed roughly in line with expectations. The cRPO guide did not: Street expectations for Q4 cRPO growth ranged from about 19% to 20%, putting the 18.5% guide 50-150bp short. For 2027, published estimates of subscription growth are moving toward 18%, from closer to 20% before the print, and consensus FY27 non-GAAP EPS sat near $5.00 going in. Expect FY27 revenue estimates to come down modestly and EPS estimates to absorb the full interest cost of the acquisition debt.

Guidance style: ServiceNow has beaten the high end of its subscription guide in each of the six quarters we have covered, and its cRPO guide by 100-200bp in each of the last four. The guide is conservative by construction. The issue this quarter is not credibility of the number but the size of the step-down from the prior print, which no amount of historical beat cadence fully closes.

Analyst Q&A Highlights

Reconciling the Q4 Guide With the Growth Narrative

The second question of the call raised the after-hours sell-off directly and asked whether the nuances behind the Q4 guide would normalize in 2027. The CFO's answer produced the entire cRPO bridge discussed above, plus the disclosure of the services carve-out, and closed with a general caution that cRPO is an imperfect leading indicator.

Q: "Obviously, the stock is down pretty meaningfully after market on some of the, I think, the 4Q guide. But any constant kind of comments on your confidence in growth going kind of either staying high or going higher, particularly even for next year as some of these nuances normalize and they don't necessarily impact subscription revenue as much as maybe the cRPO deceleration would imply?"
— Aleksandr Zukin, Wolfe Research

A: "With respect to cRPO, we saw strong early renewals in Q3 that contributed about half of the beat in Q3. Now that has a negative impact in Q4 because we pulled those deals out of Q4 and into Q3. So that growth is already in Q3."
— Gina Mastantuono, President & CFO

Assessment: A thorough answer on mechanics and an incomplete one on magnitude. Two of the four effects carry a Q4 number; the early-renewal and deployment-services effects do not, and no 2027 growth framework was offered. Investors were left to do the arithmetic themselves, and the result was below 20%.

When Does AI Momentum Reach the Top Line?

The defining strategic question of the call asked when accelerating AI metrics would show up as acceleration in total subscription revenue and cRPO, noting that cRPO looks generally stable and modestly decelerating on an organic basis. The CEO's answer pointed to 2027 and to implementation speed as the gating factor.

Q: "So my high-level question, Bill, is just when do you think the AI momentum will start to translate to overall top line acceleration, cRPO acceleration?"
— Sanjit Singh, Morgan Stanley

A: "You will absolutely see in some now, but also in 2027, as Gina said, the true acceleration because it's all about the customer getting these implementations and getting these tokens activated in their workflows or their business processes."
— Bill McDermott, Chairman & CEO

Assessment: The most consequential forward statement of the call: management has now promised visible top-line acceleration in 2027. That becomes the commitment the stock will be graded on. It is a strong claim to make on the same call that guided the leading indicator down 250bp.

Is AI Cannibalistic or Accretive to Spend?

A related question asked directly what happens to total ServiceNow spend among the most advanced AI adopters. The CEO answered at the level of macro productivity; the President and COO gave the only customer-level answer, describing usage expanding three to four times after the first use cases go live.

Q: "But when you look at your early adopters and your most advanced AI customers, what are you seeing in terms of how their overall spend with ServiceNow is changing as they've adopted Now Assist and your AI capabilities?"
— Arjun Bhatia, William Blair

A: "Many of the customers typically start with one or two use cases where we help them or they decide like incident management or triaging or resolution planning. And once they see that go live, we've seen that go up to 3x, 4x usage, but also new use cases emerge because now they're seeing it in the production environment."
— Amit Zavery, President, CPO & COO

Assessment: The question asked about spend; the answer described usage. Neither executive gave a net-expansion figure for AI adopters versus non-adopters, which is the single disclosure that would settle the cannibalization debate. Its absence is itself informative.

How to Model the Deployment-Services Investment

A recurring line of questioning sought detail on the new forward-deployed services: when the decision was made, how broadly it applies, and what it does to revenue recognition and margins. The CFO said it applies to a select number of large deals, quantified the subscription impact at 25bp in Q3 and 50bp in Q4, and said it is not intended to be permanent.

Q: "I'm curious, I guess, when was that decision made? And I'm just curious how we should think about modeling that? Is this something that will be broadly available?"
— Samad Samana, Jefferies

A: "And we don't expect to include this delivery expertise over the long term. As with other new technology adoption cycles, it's really about optimizing the first wave of adopters and leveraging the know-how in a repeatable and scalable way."
— Gina Mastantuono, President & CFO

Assessment: The "when" was not answered, and there is no end date or dollar size. If the program runs through 2027, its subscription headwind compounds before the renewal tailwind arrives. We would want a disclosed annual services commitment before modeling a clean reversal.

Outcome-Based Pricing Pressure

The opening question asked whether customers are pushing for the outcome- or value-based pricing that AI-native competitors are adopting, and what it would mean for the financial model. Both executives said customers prefer the predictability of the hybrid seat-plus-consumption model and are not asking for outcome pricing.

Q: "Bill, we increasingly hear about AI-native applications companies implementing outcome or value-based pricing. And I was just curious to what extent you're seeing customers asking you to go down this path?"
— Brad Zelnick, Deutsche Bank

A: "And then, Brad, from the impact on the financial model, as Bill said, our customers are very focused and quite happy with the hybrid model that we have today and are not asking at all for outcome pricing."
— Gina Mastantuono, President & CFO

Assessment: A categorical answer, consistent with Q2. The second half of the question, the financial impact if the market did move, went unanswered beyond "we'd be ready," so we cannot size the scenario.

Where the Internal Productivity Gains Are Landing

A question on internal AI productivity asked how headcount reductions concentrated in sales and marketing square with the need for revenue growth. The CFO separated quota-bearing capacity (growing) from support roles (shrinking) and quantified the internal value at more than $1B annualized.

Q: "Can you help us understand how that's playing out because it looks like a lot of it is actually focused in maybe front office product and sales and marketing."
— Peter Weed, Bernstein

A: "And so while you might see total sales and marketing headcount come down, actually quota-bearing heads are up double digits, while sales operations, for example, is down significantly."
— Gina Mastantuono, President & CFO

Assessment: The best-quantified answer on the call, and it directly supports the headcount commitment. Double-digit growth in quota capacity is also a demand signal: management is adding sellers going into 2027, which is inconsistent with the bear reading of the cRPO guide as a pipeline warning.

Monetizing Agents That Arrive Through Someone Else's Front Door

The final question asked how Action Fabric monetizes when work arrives through ChatGPT or other external agents rather than ServiceNow's own interface. Management said each external call consumes assists, so the company is paid for the work regardless of where the request originates.

Q: "We wanted to ask on Action Fabric monetization, particularly with the OpenAI like partnership expansion announcement that came out with earnings tonight."
— Matthew Calitri, Needham

A: "It's just a call coming from ChatGPT. And then the work which we were doing typically through our user experience now happens through this interface, and we get full value for it from the monetization perspective because you're burning down assist for every work you'd ask ServiceNow to do."
— Amit Zavery, President, CPO & COO

Assessment: The clearest statement yet of the headless-platform economic model. If it holds, external agents are a distribution channel, not a threat. The claim the CEO attached to it, that this model will be "substantially larger in a very short period of time than the entirety of ServiceNow is today," is not a forecast we would underwrite without numbers.

Control Tower: Bought Before Production, Not After

A question on AI governance timing asked whether customers deploy a control layer now or wait until agent populations scale. Management said Control Tower is purchased ahead of production deployment, which makes it a leading indicator of later consumption.

Q: "Are they doing that now? Or are they waiting until they get more deployment in production?"
— S. Kirk Materne, Evercore ISI

A: "So it's coming in first before AI goes in production. And the volume of usage just starts going up because as you start building out more AI as you get confidence, we are really helping them manage it."
— Amit Zavery, President, CPO & COO

Assessment: Control Tower net new ACV up more than 6x YoY is consistent with that sequencing. If governance is purchased first, its growth is a reasonable proxy for the AI consumption pipeline, which is a more useful leading indicator than cRPO for the AI layer.

What They're NOT Saying

1. A dollar base for consumption. The call offered 13x consumption growth, a 500% month-over-month assist increase, and a "5x" revenue-per-user aspiration, but not the dollar value of consumption revenue or its share of subscription revenue. Growth multiples on an undisclosed base cannot be modeled.
2. The full cRPO bridge in basis points. On-prem mix (25bp) and T4C backlog (about 50bp) were quantified; the Q4 effect of early renewals and of the deployment-services contracts was not. The missing pieces are exactly the ones needed to judge whether underlying growth is above or below 20%.
3. Acquired versus organic growth. Management quantified Armis at 125bp of growth in the Q1 full-year guide and the Q2 quarterly guide. No contribution was given for Q3 actuals or the Q4 guide, and the word "organic" appeared in the call only in an analyst's question and in a reference to "organic innovation."
4. A 2027 framework. The CEO promised "true acceleration" in 2027 without any quantitative frame: no preliminary subscription growth range, no margin direction, and no reiteration of a cRPO trajectory.
5. Buybacks. The words "buyback" and "repurchase" did not appear on the call. The cash flow statement shows zero repurchases for a second consecutive quarter, with $4.2B of authorization idle and net debt near $0.56B.
6. Security ACV. Q2 disclosed a $1B+ security business. Q3 offered growth multiples and a top-20 deal count but no updated ACV, while security deals above $1M fell to 13 from 24.
7. Platform vulnerabilities. The critical pre-authentication flaw in the AI Platform disclosed in July (CVE-2026-6875), and an August advisory covering three further maximum-severity (CVSS 10.0) flaws, went unmentioned in a call that pitched the company as the governance and security layer for enterprise AI. The firm that reported in-the-wild exploitation in July later revised its report (the activity matched a published proof of concept), and the company has said it saw no evidence of such activity on instances it hosts; the silence is a positioning gap, not evidence of new damage.
8. The cost of debt. Interest expense of $81M a quarter (from $6M) and a $121M YoY swing in net interest cost roughly $0.09 of non-GAAP EPS. No interest-expense outlook or deleveraging plan was offered.

Market Reaction

• Pre-print setup: NOW closed October 28 at $150.13, down 2.0% YTD against an S&P 500 up 13.1% YTD, and down 20.0% over twelve months. The trailing 30 days were +14.2%, and the stock had closed as high as $158.96 on October 23. From the Q2 pre-print close of $95.46, it had rallied 57% in three months, to within a dollar of our $150 base-case target. The 52-week closing range entering the print was $83.00 to $189.26.
• After-hours move: Shares fell about 14% within minutes of the release, and the decline deepened to roughly 17-18% through the evening as the call failed to reframe the cRPO guide. Pre-market indications on October 29 were down 18-19%.
• Reaction session (October 29): The stock opened at $120.91 (-19.5%), traded between $108.68 and $121.50, and closed at $109.67, down 26.9% (-$40.46), on 76.4M shares, 6.7x the 30-day average. The S&P 500 fell 0.3%. It was the largest one-day percentage decline since the stock listed in 2012; the second-largest was the 17.7% drop after the Q1 2026 print in April.
• Follow-through (October 30): Shares traded lower again in early-afternoon trading, around $105, roughly 4% below the reaction close and about 31% below the 2025 year-end close.
• Peers: Software sold off in sympathy, but far less. The software sector ETF fell roughly 3.5-4% on October 29, Salesforce and Workday about 4-5%, and Atlassian about 7%. Alphabet fell more than 12% on its own results the same day, and the Nasdaq Composite fell about 1.4%.

Priced for acceleration. The setup did the print no favors. From the Q2 pre-print close to the October 23 peak, NOW rose two-thirds, and at $150.13 it traded at about 30x consensus FY27 non-GAAP EPS of roughly $5.00, against about 22x at the reaction close. The rally rested on multiple expansion (relief after the August software rebound, rotation toward cash-generative software in October, and a wave of target-price increases into the print) rather than on estimate revisions. A stock re-rated on the expectation that AI would soon lift aggregate growth was then handed a guide that pointed the other way.

One metric decided the session. The stock opened down 19.5% and closed within a dollar of the session low, with no intraday recovery, on 6.7x normal volume. The accelerating Q4 subscription guide (21.5% CC) and the raised FCF target went unrewarded because investors now treat cRPO as the 2027 revenue indicator, and the Q4 guide sat 50-150bp below where the Street had it. The two-cent EPS miss, mechanical as it was, removed the clean-sweep framing that had carried the stock through earlier prints. Management's bridge arrived as a list of accounting effects, and the market discounted it as such.

Idiosyncratic, not sector-wide. Software peers fell 3.5-7%; NOW fell 27%. That spread says the market repriced ServiceNow's own growth path, not software broadly. The stock was down another 4% in early-afternoon trading on October 30. The last record decline, 17.7% after the Q1 print in April, was followed by an 87.5% rally to the October peak ($84.78 to $158.96), which shows how far sentiment overshoots in both directions in this regime. The question beneath both moves is the one we framed at Q2: whether AI shows up in aggregate growth. The market has now named the test, and it is the Q4 cRPO print.

Street Perspective

Debate: Is the cRPO Deceleration Real or Accounting?

Bull view: The bull camp argues every item in the bridge is mechanical: federal on-prem deals and termination-for-convenience contracts produce revenue that cRPO cannot see, early renewals simply moved backlog between quarters, and the services carve-out returns to subscription at renewal. ServiceNow has beaten its cRPO guide by 100-200bp in each of the last four quarters, and the Q4 subscription guide accelerates. On this view the gap between cRPO and revenue reverses over 2027.

Bear view: The bear camp contends that a new caveat every quarter is itself the signal. Organic growth is already in the high teens once Moveworks, Veza, and Armis are stripped out, early renewals leave less room for the usual Q4 beat, and even a conservative guide points to deceleration. Some desks have cut their 2027 subscription growth estimates to around 18%.

Our take: Both camps are partly right. The bridge is real but incomplete, and our adjusted figure (about 19.75% including Armis) still decelerates. We treat underlying organic growth as high-teens and accept that the durable +20% claim is now at risk. A Q4 cRPO print at or above 19.5% CC would settle the debate in the bull camp's favor; a print near 18.5% would settle it the other way.

Debate: Is AI Big Enough to Move the Total?

Bull view: The year-end AI ACV target has risen 70% in ten months, AI net new ACV grew 65%, and agentic consumption is up 13x. Management has promised visible acceleration in 2027 as deployments go live and consumption scales, and the forward-deployed engineering program shortens the time to that inflection.

Bear view: Even at the $1.7B target, AI ACV would be about 10% of a subscription base running near $17B annualized, so even rapid AI growth adds only a few points to the total, and much of it replaces older SKUs at renewal. The skeptics add that seat-expansion freezes do not show up in a 98% renewal rate, and that if the work surface moves to ChatGPT or Gemini, the AI lab owns the user.

Our take: The arithmetic favors the bears for the next two to three quarters: a 10% layer growing 65% cannot by itself offset a high-teens organic core. The bull case is a 2027 claim that depends on consumption, which is precisely the revenue stream the company does not yet size in dollars. We hold the AI pillar ON TRACK on its own terms and grade the acceleration claim separately.

Debate: Was the Rally Wrong, or the Sell-Off?

Bull view: At about 22x consensus FY27 EPS and roughly 16x FY27 FCF, the stock again prices a business in structural decline, which the retention, AI metrics, and capital efficiency refute. History favors patience: the April record drop was followed by an 87.5% rally, and post-print target cuts left most ratings unchanged.

Bear view: The July-to-October rally was multiple expansion with almost no estimate revisions, and the sell-off simply returned the multiple to where the fundamentals put it. On this view the stock faces a period of multiple compression until cRPO re-accelerates, and high-teens organic growth does not support the premium the bulls want.

Our take: The rally priced acceleration before the evidence; the sell-off prices high-teens growth for years. We think the second is closer to wrong than the first. At $110, the base case needs only modest multiple recovery on unchanged FY27 FCF, but we have lowered conviction because the catalyst is now a single metric on a single date.

Model Update & Valuation Framework

Valuation: At the October 29 close of $109.67, market capitalization is about $114B on 1.043B diluted shares, and enterprise value is about $115B including roughly $0.56B of net debt. That is about 19.7x FY26E FCF of ~$5.85B, about 16.4x FY27E FCF of ~$7.0B, and roughly 28x FY26E non-GAAP EPS of ~$3.90. PT framework trimmed: Base $140 / Bull $180 / Bear $80 (from $150/$190/$80), built on ~21x / ~27x / ~12x FY27E FCF. The trim reflects slower FY27 growth implied by the cRPO exit rate, worth about one turn of multiple; FY27 FCF is essentially unchanged because the higher FCF margin offsets slower revenue growth. From $109.67: base +28%, bull +64%, bear -27%. The bear case is unchanged: the stock closed at $83.00 in April on fundamentals no stronger than today's.

Thesis Scorecard Post-Earnings

Overall: The standing thesis weakened on one pillar and strengthened on four. AI monetization, CRM, customer expansion, and hybrid pricing all confirmed, and the headcount commitment that anchors the margin pillar is tracking ahead of schedule. But the first bull pillar, durable +20% CC subscription growth, moves to at risk: the Q4 cRPO guide, adjusted generously, points to organic growth in the high teens entering 2027, and management's promise of "true acceleration" in 2027 is now the explicit test. The bear case on narrative regime has proven two-sided: it lifted the stock 57% without new fundamentals and then removed 27% in a day.

Action: Maintaining Outperform, with conviction lowered from high to moderate. At roughly 16x FY27E EV/FCF, the stock again prices a business whose growth is impaired, while the AI metrics, retention, and capital efficiency say otherwise. The 2027 acceleration claim is the gradeable commitment: if Q4 cRPO prints at or above 20% CC and the company frames 2027 subscription growth at 20%+ in January, the base case is conservative. If Q4 cRPO prints near the guide, the growth pillar breaks and the rating moves to Hold. Watch for buyback resumption below $110 as the cheapest signal of management's own conviction.