The SaaSpocalypse has been brutal.
Over the past year, fear over AI disrupting software businesses has caused a widespread sell-off across enterprise SaaS companies. ServiceNow hasn’t been spared. The stock has been cut over 50% from its 52-week high of $211.48, now trading at around $97.
That’s a 50%+ drawdown for a company that has grown its revenue at 22% annualized over the past five years and more than tripled its operating margin from 4.4% to 13.5% during this period.
The question, then, is whether the SaaSpocalypse fear is warranted for ServiceNow, or whether the market is throwing the baby out with the bathwater. This deep dive is my attempt to answer that question by dissecting the business from the ground up. Part 1 covers what ServiceNow does, who buys it, how it makes money, and why customers almost never leave. Part 2 tackles the big AI debate, the financials, management, valuation, and my conclusion.
Let’s get into it.
What ServiceNow Actually Does
Large organizations are messy. A company with 50,000 employees might have one system for IT tickets, another for HR onboarding, another for customer complaints, another for procurement approvals, and dozens more for everything else. Employees waste hours toggling between applications, emailing back and forth, and manually pushing tasks from one department to the next. McDermott describes this as “the legacy tax,” estimating it at $10 trillion annually for U.S. businesses alone.
ServiceNow replaces that chaos with a single platform. When someone submits a request, the platform automatically determines who needs to do what, in what order, routes the task to the right person or team, tracks it, escalates it if delayed, and reports on the whole process.
Think of it as a unified operating system for how a company runs internally.
The company started by solving IT help desk problems: an employee’s laptop breaks, they submit a ticket, ServiceNow routes it to the right technician. But over the past six years under CEO Bill McDermott, the platform has expanded into multiple internal business functions.
The Four Workflow Families
ServiceNow organizes its products into four categories, each targeting different parts of the enterprise.
At the May 2025 Analyst Day, management disclosed that all four workflow families had surpassed $1 billion in ACV by 2024: CRM and Industry Workflows and Creator Workflows each exceeded $1.4 billion, and Core Business Workflows crossed $1.1 billion. Technology Workflows, the largest and oldest family, was not given a specific figure but is clearly well above that threshold.
Technology Workflows is where ServiceNow started and remains the largest revenue contributor. This includes IT Service Management (the help desk), IT Operations Management (monitoring and automating IT infrastructure), IT Asset Management (tracking hardware and software), Security Operations, and Strategic Portfolio Management. IT Asset Management alone has grown at a 55% compound annual rate since 2020. According to CFO Gina Mastantuono at the May 2025 Analyst Day, fewer than 25% of ServiceNow’s existing customers have adopted it yet. The remaining 75%+ are potential upsell targets without needing to win a single new logo.
CRM and Industry Workflows is ServiceNow’s most ambitious expansion, growing at 30%+ year-over-year. This family surpassed $1.4 billion in ACV in 2024 and includes Customer Service Management, Field Service Management, and Sales and Order Management. Management is explicitly going after legacy CRM displacement. McDermott describes opportunities to “collapse 175 instances of a legacy CRM” onto ServiceNow. Industry-specific solutions for financial services, healthcare, telecom, manufacturing, public sector, and retail add vertical depth.
Source: 2025 Financial Analyst Day
Core Business Workflows crossed $1.1 billion in ACV in 2024 and represents ServiceNow’s push beyond IT into every department: HR, legal, procurement, and facilities. The same workflow automation that originally managed IT help desk tickets now handles employee onboarding, contract reviews, purchase approvals, and building maintenance. In 2024, for the first time, more than half of new customer ACV came from non-technology workflows. As the CFO put it at Analyst Day: ‘We’ve officially busted out of IT. Just to be clear, ServiceNow does not replace a company’s existing HR or payroll systems. It sits on top of them, connecting the tasks that span multiple departments into a single automated workflow.
Creator Workflows and Other surpassed $1.4 billion in ACV in 2024. The pre-built workflow families above cover common problems that every large company faces. But every company also has problems unique to their specific business. A hospital needs to track equipment sterilization cycles. An airline needs to manage crew scheduling exceptions. Creator Workflows gives customers low-code tools to build their own custom applications on the ServiceNow platform, without hiring a development team. Because these custom apps run on the same platform as everything else, they automatically connect to all the other ServiceNow workflows and data already in place.
This family also includes two infrastructure products. Workflow Data Fabric connects ServiceNow to data living in other systems (Salesforce, SAP, Workday, homegrown databases) in real time, without copying the data. The data stays where it originally lives, but ServiceNow can read it, use it in workflows, and feed it to AI agents. This appeared in 16 of the top 20 deals in Q4 2025, which tells you how important data connectivity has become to customers. RaptorDB is a faster database engine that sits underneath the ServiceNow platform. As customers run more workflows and more AI agents, the underlying database needs to keep up. RaptorDB more than tripled its net new ACV year-over-year in Q4, including 13 deals over $1 million.
Now Assist is different from the other four families. It is not a separate set of workflows. It is an AI layer that makes all four families smarter. Take any workflow that already exists on the platform: an IT ticket, an HR request, a customer service case. Now Assist adds AI on top: auto-summarizing a case for the next agent, suggesting a resolution based on similar past tickets, answering an employee’s HR question through a chatbot, or handling an entire request from start to finish without human involvement. It surpassed $600 million in ACV by year-end 2025, with a $1 billion+ target for 2026. Net new ACV more than doubled year-over-year in Q4 2025, with 35 deals exceeding $1 million in the quarter alone.
The pricing also works differently. The four workflow families are sold as seat-based subscriptions. Now Assist adds a consumption layer on top: every time an AI agent performs a task, it burns a credit. When credits run out, the customer buys more. This is the assist pack model I describe in the pricing section below.
Who Buys It
ServiceNow sells to large enterprises with thousands of employees and complex internal operations. The customer base is over 8,800 strong, including 85%+ of the Fortune 500.
What’s more telling is how the biggest accounts are expanding. The number of customers paying more than $5M in ACV has grown from 420 to 603 in just two years, a 44% increase. And the $20M+ cohort is growing even faster, at 30%+ year-over-year.
North America still accounts for 63% of revenue with 37% coming from international markets. Global government business grew 80% year-over-year in Q4 2025. The direct sales channel makes up about 78% of revenue, with system integrator and managed service provider partners handling the rest.
And the renewal rate is 98%, sustained for three consecutive years. I’ll come back to this number because it signals how deeply embedded this company is within its customers.
How They Make Money
The revenue model is straightforward: 97% subscription, 3% professional services. Total revenue in FY25 was $13.3 billion, of which $12.9 billion was subscription revenue.
Subscription revenue is recognized ratably over the contract term, typically one to three years. Customers are invoiced annually in advance. Contracts are generally non-cancellable. This creates enormous predictability. The company enters each quarter with most of its revenue already locked in from previously signed contracts.
The pricing model has three layers that stack on top of each other.
Layer 1: Seat-based licenses. Customers choose between Standard, Pro, and Pro Plus tiers, each unlocking more AI and automation capabilities. Upgrading from Standard to Pro generates a roughly 25% price increase. Pro to Pro Plus adds another 30%. Customers who jump straight from Standard to Pro Plus see a roughly 60% price uplift. Pro Plus penetration is still just over 10% of the Pro customer base, leaving substantial room for upsell.
Layer 2: Consumption-based assist packs. On top of seat licenses, customers purchase pre-paid credits that are consumed each time an AI agent performs a task. When credits run out, customers reload. Management describes 50% month-over-month growth in consumption. One customer expanded their assist entitlement 13x upon renewal.
Layer 3: Platform expansion. Customers land with one or two products and expand across workflow families. In 2024, 99% of net new ACV came from multi-product deals, and 86% included 5 or more products. The average customer’s spend accelerates over time. It takes a little over 4 years to go from $1M to $5M in ACV, just over 2 more years to reach $10M, 1.5 more years to hit $15M, and just 1 more year to cross $20M.
I find this expansion dynamic particularly compelling. It’s able to grow by deepening existing relationships. According to the Q4 2025 Investor Presentation, the 2010 customer cohort has expanded roughly 45x, from an initial ACV of about $100,000 to approximately $4,500,000. That is the hallmark of a platform that keeps getting more valuable to its customers over time.
Source: Q4 2025 Investor Presentation
Professional services brought in $395M, just 3% of revenue. This segment intentionally runs at a small loss. The logic: professional services exist to get customers deployed on the platform, which then generates years of high-margin subscription revenue. The company is actively shifting implementation work to its partner ecosystem (Deloitte, Accenture, NTT DATA), with third-party partner costs rising from 10% to 35% of professional services revenue over three years.
Revenue Backlog and Visibility
One of ServiceNow’s strongest financial attributes is its revenue visibility. At year-end 2025, the company had $28.2 billion in Remaining Performance Obligations (RPO), which is the total dollar value of signed contracts where the service hasn’t been delivered yet. Of that, $12.85 billion is current RPO (revenue expected within 12 months). RPO grew 26.5% year-over-year. This backlog represents roughly 2+ years of revenue already under contract.
Source: Q4 2025 Investor Presentation
Competitive Position
The investment case for ServiceNow depends heavily on whether its competitive advantages are durable.
Switching Costs: Deeply Embedded
The 98% renewal rate, sustained for three consecutive years, is a strong data point.
Source: Q4 2025 Investor Presentation
ServiceNow does not sit in one department. In 2024, 86% of net new ACV came from deals including 5 or more products. A typical large customer might run IT ticketing, HR onboarding, customer service, security operations, and custom-built apps all on the same platform, sharing the same data, the same automation engine, and the same user interface.
Ripping out ServiceNow would mean simultaneously replacing five or more mission-critical systems and re-integrating everything from scratch. This is what makes that 98% renewal rate so sticky. It’s not just that customers are satisfied. It’s that leaving would be enormously painful.
The expansion data reinforces this. As noted earlier, customer spending accelerates over time. The average customer who reaches $5M in ACV takes only 2 more years to reach $10M, then 1.5 years to reach $15M, and 1 more year to cross $20M. The pace of expansion accelerates, which means customers are becoming more dependent on the platform over time, not less.
Platform Breadth: The Cross-Functional Moat
Most enterprise software companies are systems of record: Salesforce stores customer data, Workday stores employee data, SAP stores operational data. ServiceNow is different. It is a system of action that sits on top of those systems and orchestrates the work that moves between them. That is why it can span IT, HR, customer service, security, legal, and procurement on a single platform without needing to replace the underlying systems those departments already use. In some cases, it goes further, consolidating fragmented point solutions and internal tools that were doing the job poorly.
This creates an internal network effect within each customer. Each additional workflow a customer deploys increases the value of every existing workflow because they share the same data, the same automation engine, and the same AI capabilities. An IT incident that affects customer service can automatically trigger a customer notification workflow, which can trigger an HR escalation if staffing is needed. No combination of point solutions can replicate this cross-functional orchestration without expensive custom integration.
The Q4 2025 earnings call provided a concrete example of the consolidation mode: a leading European telecom provider consolidated 7 internal systems onto ServiceNow CRM, reducing costs by 30%, cutting order-to-fulfillment cycle time by 25%, and resolving 20% more work orders on the first request. This kind of cross-functional consolidation is ServiceNow’s most powerful competitive argument, and the hardest for competitors to match.
Data Asset: 20 Years of Workflow Intelligence
ServiceNow has been running enterprise workflows for two decades. The platform now processes 80 billion workflows and 6.4 trillion transactions annually across 8,800+ enterprises. The 10-K describes this institutional knowledge of how enterprises operate as something that “cannot be readily replicated.”
This matters enormously for AI. The effectiveness of ServiceNow’s AI agents depends on understanding how enterprise workflows operate in practice. A new entrant can build a clever AI model, but without two decades of workflow execution data across thousands of enterprises, their AI agents won’t know how to route an approval through a complex org structure or predict which IT incidents will cascade into outages.
The data moat feeds the AI moat.
Distribution: Relationships That Took Decades to Build
ServiceNow’s direct sales force has relationships with 85%+ of the Fortune 500. The partner ecosystem includes all three hyperscalers (AWS, Azure, GCP), major systems integrators (Deloitte, Accenture, EY, KPMG), and technology providers including NVIDIA. These relationships cannot be replicated quickly.
The sales efficiency data is worth noting: management reported the sales efficiency multiple at 4.6x in FY24, up from 3.8x the prior year. This is unusually high for a company growing 20%+ and suggests the go-to-market engine is becoming more efficient, not less.
So the business is deeply embedded in its customers, expanding within accounts at an accelerating pace, and protected by switching costs that make leaving enormously painful. But the biggest question hanging over ServiceNow isn’t about its current moat. It’s about whether AI strengthens that moat or erodes it. That’s where Part 2 picks up.
Disclosure: I hold a position in ServiceNow at the time of publishing this article (this is a disclosure and NOT A RECOMMENDATION).