Is AI Killing SaaS — Or Just Supercharging Salesforce?

Salesforce entered this earnings print squarely at the center of the AI versus SaaS debate. Investors have spent the past year asking whether generative AI agents will erode traditional seat-based software models, compress growth, and shift value away from application vendors toward model providers and infrastructure. Against that backdrop, Salesforce’s results were less about the quarter itself and more about whether AI is disrupting — or reinforcing — the core SaaS model.

On the surface, the numbers were solid. Adjusted EPS of $3.81 handily beat expectations of $3.04. Revenue of $11.2 billion grew 12% year over year and was essentially in line with consensus. Full-year revenue reached $41.5 billion, up 10%. The company generated $15 billion in operating cash flow and returned more than $14 billion to shareholders.

But the more important metrics were under the hood.

Current remaining performance obligation (cRPO) came in at $35.1 billion, up 16% year over year and ahead of expectations. Total RPO reached $72 billion, up 14%. These backlog metrics suggest demand is intact, though not accelerating dramatically. Organic cRPO growth around 9% constant currency was slightly above guidance but not a material upside surprise — and that modest beat is what initially weighed on the stock.

However, there’s nuance here. Management and analysts both flagged that cRPO may increasingly understate demand as the business shifts toward consumption-based Flex Credits, particularly for Agentforce. Roughly 50% of Q4 Agentforce bookings were tied to Flex Credits, which don’t flow through traditional cRPO in the same way. In other words, the old SaaS yardsticks may not perfectly capture the new monetization model.

The headline AI story is Agentforce.

Agentforce annual recurring revenue reached $800 million in Q4, up 169% year over year and sharply above prior levels. Salesforce closed 29,000 Agentforce deals, up 50% quarter over quarter. Usage metrics were also notable: nearly 20 trillion tokens consumed and more than 2.4 billion “agentic work units” delivered. That framing is important — Salesforce is not positioning AI as an add-on chatbot, but as embedded productivity inside enterprise workflows.

Marc Benioff leaned into this theme aggressively. “This is not our first SaaSpocalypse,” he said, pushing back on the narrative that AI threatens the core SaaS model. Instead, he argued Salesforce has “rebuilt” itself into the operating system for the “Agentic Enterprise,” where humans and AI agents work together on one platform.

The message was clear: AI is not cannibalizing Salesforce — it is a tailwind.

Still, investors are wrestling with the core business. Growth in Tableau, Marketing Cloud, and Commerce Cloud remains soft. Organic growth is sub-10%. While large deal momentum improved — wins over $1 million were up 26% year over year and wins over $10 million up 33% — the beat magnitude was not overwhelming. Management reiterated confidence in a reacceleration in the second half of FY27, driven by improving net new AOV trends, but the street wants to see that inflection materialize.

Guidance reflects that tension.

Q1 revenue guidance of $11.03–$11.08 billion was slightly ahead of consensus, and adjusted EPS guidance topped expectations. But FY27 revenue guidance of $45.8–$46.2 billion (10–11% growth) came in roughly in line to slightly below consensus at the midpoint. That tempered initial enthusiasm.

Where Salesforce leaned heavily positive was capital allocation. The company authorized a massive $50 billion share repurchase program, replacing prior authorizations, and raised its dividend. Benioff openly described current share prices as attractive. For a stock down roughly 28% year to date, that move signals management confidence in long-term fundamentals.

The broader debate, however, is whether AI changes the structure of SaaS economics.

If AI agents reduce seat counts or compress pricing, SaaS multiples should contract. If, instead, AI expands workflows, increases automation, and deepens integration, it could expand wallet share and stickiness. Salesforce’s data gravity argument is central here: enterprises already run critical workflows on its platform. Embedding agentic AI inside those workflows may strengthen switching costs rather than weaken them.

Agentforce’s monetization model is also evolving. Management described three approaches: premium SKUs, expanded seat coverage, and consumption-based Flex Credits for customer-facing use cases. That hybrid model blurs the traditional recurring revenue lens but potentially increases long-term ARPU.

Shares initially traded down roughly 3–4% in premarket, reflecting lingering skepticism. But as the pre-market session progressed, the stock surged toward $191, echoing the pattern seen in Workday the prior day — a weak initial reaction followed by a sharp reversal. That reversal in Workday sparked a rally in the broader software ETF (IGV). If Salesforce sustains this rebound, it could catalyze another rotation back into software names that have been punished in the AI transition.

The key question is whether this marks the beginning of a broader shift in sentiment.

If Salesforce — one of the largest and most entrenched enterprise software companies — can demonstrate that AI agents are additive rather than destructive, it could reframe the AI vs. SaaS debate. Early metrics suggest real adoption momentum, particularly in large enterprise deals and government contracts, including a notable 10-year U.S. Army agreement.

Still, execution matters. Investors will want proof that Agentforce growth translates into sustained double-digit organic growth at the consolidated level. Until that reacceleration becomes visible, valuation multiples may remain compressed.

For now, the quarter suggests AI is not dismantling Salesforce’s moat. Instead, the company is attempting to absorb the AI wave into its platform and monetize it through consumption and premium offerings.

Whether this is the beginning of a software rebound — as we saw with Workday — or just another volatile chapter in the AI transition will depend on whether growth visibly inflects in the back half of FY27.

But one thing is clear: Salesforce is no longer defending SaaS. It is trying to redefine it.