Smith Collection/Gado/Getty Images
Salesforce $CRM walked into earnings priced like an AI bystander and walked out with the kind of record quarter that leaves very little to argue with. Revenue roughly matched forecasts, profit was miles above them, the sales backlog grew double digits, and the AI and data business closed in on a $1.4 billion run-rate. In after-hours trading, the stock jumped more than 5%.
Heading into the print, the setup looked very different. The stock was down about 30% for the year and trading near the bottom of its 52-week range, with a multiple that had been squeezed down to “slightly better than the S&P” for a company that used to live in the premium SaaS stratosphere. The bar on revenue was modest — roughly 9% growth to about $10.27 billion — but the subtext was loud: Prove the AI story is real, or get treated like an old utility that’s desperate to glom onto agentic AI hype.
Salesforce’s headline numbers cleared that bar. Easily. Third-quarter fiscal year 2026 revenue landed at $10.3 billion, up 9% year over year. But the real punchline was the EPS: Non-GAAP EPS came in at $3.25 versus expectations in the high-$2.80s, a gap wide enough to stop even the crankiest model-tuner mid-keystroke.
Current remaining performance obligation (cRPO), Salesforce’s favorite proxy for near-term demand, rose 11% to about $29.4 billion; total RPO climbed 12% to roughly $59.5 billion. Inside the building, that cRPO number is being held up as evidence of a “powerful pipeline of future revenue,” as Benioff said in the press release, something the company clearly wants investors to latch onto.
Non-GAAP operating margin came in around the mid-30s, GAAP margin a bit above 21%. Operating cash flow grew in the high teens, free cash flow in the low 20s, and Salesforce sent roughly $4 billion-plus back to shareholders between buybacks and dividends. For a stock that’s been trading like a problem child, the P&L looked conspicuously like a well-adjusted adult.
Guidance nudged the company in the same direction — the right one for markets. Salesforce pushed full-year fiscal 2026 revenue guidance up to about $41.5 billion at the midpoint, implying 9–10% growth, and raised its outlook for operating cash-flow growth into the low-to-mid-teens. The longer-range promise — more than $60 billion in organic revenue by fiscal 2030 and a combined growth-plus-margin “50” — stayed in place, now backed by one more quarter that doesn’t obviously undercut it. This isn’t a hypergrowth story, but it’s not a stall-out story, either.
Enter AI. Agentforce and Data 360 together are now running at nearly $1.4 billion in ARR, up more than 100% from a year ago. Agentforce on its own has crossed the half-billion mark and is growing several hundred percent year over year. Salesforce says it has closed more than 18,000 Agentforce deals since launch, with something north of 9,000 actually paid, and its LLM gateway has processed trillions of tokens. Data 360 is ingesting tens of trillions of records per quarter, with “zero copy” data — the bit that lets Salesforce sit on top of a customer’s data rather than yank it in — growing even faster.
To make the whole thing feel less like AI theater, Salesforce is also flashing a few adoption badges, claiming that nearly 90% of Forbes’ Top 50 AI companies now run on Salesforce, typically with an average of four clouds in play, and that about half of all Agentforce and Data 360 bookings come from expansion inside existing customers — the kind of cross-sell math Wall Street understands instinctively.
And because this is Salesforce, the AI story has a name: the “Agentic Enterprise,” a phrase repeated with enough conviction that you can practically hear the trademark application fluttering through legal. Agentforce and Data 360, the company insists, aren’t add-ons — they’re “momentum drivers” tied to the company’s long-term revenue and profit targets.
One wrinkle to this AI story is Informatica. Salesforce’s acquisition formally closed this quarter, and the company now expects Informatica to contribute about 80 basis points to full-year revenue and roughly three points to fourth-quarter growth. It’s a neat way to thicken the data-and-AI story — and a neat way to remind investors that some portion of that growth is freshly purchased, not conjured from organic demand.
Still, the scale problem hasn’t disappeared. A $1.4 billion AI-and-data run-rate inside a business guiding to more than $41 billion in annual revenue is meaningful, but it’s not transformational — yet. The core business still looks like a grower with healthy margins and a buyback habit, now with a fast-compounding AI layer bolted on. The question the quarter doesn’t answer is how quickly that layer can move from “nice second engine” to something that actually changes the slope and scope of the whole company.
For one night, at least, the narratives had to stand down for the spreadsheet. Salesforce delivered a quarter that makes it harder to dismiss as dead money in the AI age — without magically making it the next great growth stock. The numbers finally looked as good as the pitch, and the investors are nodding along.
