NOW · Research
ServiceNow sells agents by the seat, and customers want it that way
ServiceNow's third quarter tests a question we parked: will superintelligence agents replace the seats software companies charge for? On this print, no. Agents are adding dollars on top of seats, renewals held at 98%, and the CFO says customers are not asking for outcome pricing.
ServiceNow's third quarter tests a question we parked: will superintelligence agents replace the seats software companies charge for? On this print, no. Agents are adding dollars on top of seats, renewals held at 98%, and the CFO says customers are not asking for outcome pricing.
The weak spot is the forward guide, not the seat.
| Scorecard | Read |
|---|---|
| Thesis | ServiceNow owns the workflow record, and agents that run through it are priced per seat plus metered use, so the gate charges more per user instead of losing users [1][2] |
| What proves it wrong | The 10-Q shows falling seat counts or price per seat, or renewals slip below 98% |
| Gate test | Owns the record: yes. Outcome pricing: no, by customer choice [2] |
1. What customers are buying
Subscription revenue was $3,988M, up 20.5% in constant currency [1]. Superintelligence net new ACV rose 65%, and the year-end agent ACV target rose to $1.7B from $1.0B at the start of the year [1][2].
CFO Gina Mastantuono said customers prefer the hybrid of seats plus consumption and are "not asking at all for outcome pricing" [2]. Renewal rate held at 98% [2].
In Q2, the CEO said "50% ... of our net new business is already non-seat based" [4].
Q3 did not update that figure.
Part of the growth is bought. If Armis added the roughly 175 basis points guided last quarter, organic subscription growth was about 18.75%, our calculation, not a company figure [1][3].
2. Why the guide, not the seat, is the problem
cRPO, the contracted revenue due in the next twelve months, was $13.65B, up 21% in constant currency and ahead of the roughly $13.56B the Street expected [1][5]. The Q4 guide is 18.5% [1].
Subscription revenue is guided to 21.5% [1].
The CFO gave three reasons. First, early renewals: "we saw strong early renewals in Q3 that contributed about half of the beat in Q3. That has a negative impact in Q4 because we pulled those deals out of Q4 and into Q3" [2].
Second, on-premises mix: "Strong public sector demand is driving higher on-prem mix in Q4 ... that is driving a 25 basis points headwind in Q4 cRPO, but will absolutely hit revenue" [2].
Third, deployment services bundled into select large deals are carved out of subscription, about 50 basis points of Q4 subscription growth [2].
Federal demand is real: US federal deals over $1M grew more than 30%, and the CFO expects "federal net new ACV growth to accelerate further in Q4" [2]. No source we read published a Street number for the Q4 cRPO guide.
3. The cost of the lift
The stronger counterpoint is margin. Non-GAAP subscription gross margin fell to 80% from 83% a year ago, and GAAP subscription gross margin fell to 73.5% from 80% as amortization of acquired intangibles rose to $152M from $32M [1].
Management tied the 81% full-year guide to customers running on hyperscaler partners and faster agent adoption [4]. Our read: agents that lift revenue per seat also lift compute cost per seat.
Stock pay was $630M, up 28%, or 15.3% of revenue, our calculation [1]. Interest expense rose to $81M from $6M after the deal debt [1].
Free cash flow after capex rose to $748M from $578M, our calculation, and the full-year margin target rose to 36% [1].
What to watch
- Seat counts, price per seat or renewal rate in the 10-Q [2]
- The Q3 share of net new business that is not seat based [4]
- Q4 cRPO against the 18.5% guide, and whether Q1 cRPO rebounds once the pulled renewals lap [1][2]
- Subscription gross margin against the 81% full-year guide [1]
- Stock pay as a share of revenue [1]
Stock Profiles & Comparable Companies
NOW Stock Profile →More Published Research
View All →Oracle's $20B cap leaves out the $288B it signed in data center leases
At its Investor Day on Oct 28, Oracle capped new money at $20B through fiscal 2030 and said new contracts will not need incremental capital from Oracle. Both promises leave out leases.
Thoma Bravo paid 8.4 times revenue for the map of who can open what
Varonis is being taken private. On Oct 15, 2026 it signed an all-cash merger with Proofpoint, a Thoma Bravo portfolio company, valued at about $6.3B and expected to close in the first half of 2027.