ServiceNow Plunges as Slower Contract Bookings Puncture Enterprise AI Optimism

Imagine paying for a gym membership where the gym works out for you, then cuts your monthly bill because you spent less time on the treadmill. That paradox is the existential dread hanging over enterprise software right now. On Oct. 29, ServiceNow fell roughly 26%, on track for its steepest one-day drop since its 2012 market debut. The software giant beat its own guidance, raised its full-year cash targets, and hyped massive corporate interest in artificial intelligence. Wall Street dumped it anyway.

To understand why, you have to understand what ServiceNow actually sells. It is the digital nervous system for corporate plumbing. When your company laptop breaks, when HR onboards a new hire, or when security flags a suspicious login, that request routes through ServiceNow. Chief information officers buy the software on multi-year contracts, paying a set annual price for every employee seat. ServiceNow is the landlord of the corporate help desk, sitting quietly in the background while collecting rent on every worker with a corporate email address.

Over the last two years, ServiceNow pitched itself as the ultimate AI shovel seller. The logic sounded bulletproof: if companies deploy armies of autonomous software agents to handle mundane workflows, someone has to supervise them, assign their permissions, and track their output. ServiceNow built the control tower. The company raised its year-end target for AI annual contract value to $1.7 billion, a 70% jump from its forecast at the start of the year. Production use of its agentic AI tools surged more than 13-fold year to date, while subscription revenue rose 21% to $3.99 billion. By traditional software metrics, the engine purred.

The crash came down to a single forward-looking metric: current remaining performance obligations, or cRPO. Think of cRPO as the contracted revenue a subscription business is legally guaranteed to recognize over the next twelve months. It is Wall Street’s speedometer. ServiceNow closed the third quarter with $13.65 billion in cRPO, up 20.5%. But for the fourth quarter, management guided that growth down to 18.5%, falling short of the roughly 19.4% analysts expected. In high-multiple software, even a faint tap on the brakes feels like hitting a brick wall.

Management pointed out that early customer renewals pulled business out of the fourth quarter and into the third. They also shifted certain complex AI deployment projects out of subscription revenue and into lower-margin professional services. That accounting change trimmed roughly 25 basis points off third-quarter subscription growth, with another 50-basis-point drag coming in the fourth.

ServiceNow's CFO conceded on the call that cRPO is a "nuanced accounting metric" that doesn't capture all the momentum in the business. Citi's Tyler Radke called the quarter "a little bit uncharacteristic for ServiceNow standards," pointing at the lack of upside. He also said the payoff from engineers embedded inside customer offices will take longer than investors want.

The panic spread quickly across the cloud. In Thursday trading, the broader software ETF slid roughly 4%, with peers like Salesforce, Workday, and Palo Alto Networks dropping between 4% and 5%. Investors suddenly confronted the uncomfortable reality that AI might not be an incremental revenue driver after all. If digital bots do the work of three IT coordinators, the company buying the software needs fewer human seats.

Oppenheimer's Brian Schwartz kept a Buy rating and a $165 target. The gap he conceded is real, though. He argues ServiceNow's AI momentum "counters a disruption thesis" and still leaves no clear path to organic reacceleration.

The bear case argues that ServiceNow's AI numbers are simply attach revenue dressed up as a second act. A 65% jump in net new AI contract value sounds impressive until you realize it is merely an upsell into captive accounts that already pay millions every year. Meanwhile, the core contract pulse is decelerating, and shunting AI deployments into professional services proves these systems still require expensive manual consulting rather than self-serve code. If autonomous software actually eliminates administrative jobs, the per-seat pricing model that built this empire will contract. A 98% renewal rate proves clients stay, but it says nothing about how many seats they will renew when bots do the typing. The bears will be proven right if cRPO growth stays trapped in the high teens next quarter or professional services revenue outpaces subscription billings.

For the stock to re-rate upward, contract growth must climb back over the 20% mark, and that $1.7 billion in AI bookings must translate into high-margin recurring software rather than billable engineering hours. Trading around 22 times forward earnings, the equity is no longer priced like an untouchable growth compounder. What we are watching next quarter is whether fourth-quarter cRPO growth holds the 18.5% line or slips deeper into the teens.

The market has stopped paying for the promise of enterprise AI and started penalizing the friction of delivering it.

Disclosure: The Signal holds no position in NOW. Positions may change. This is not financial advice.