TL;DR
Salesforce closed 29,000 Agentforce deals and reported $800 million in ARR, but its stock is down 30 percent in 2026 amid the sales SaaSpocalypse. Demos from Williams-Sonoma, UChicago Medicine and SharkNinja turned out to be more works in progress than live deployments.
Salesforce has a problem that no amount of marketing can solve. The company has built its entire story around Agentforce, its AI agent platform, and the numbers look impressive on paper: 29,000 deals closed, $800 million in annual recurring revenue, and a roadmap that promises to replace entire categories of human work. But Wall Street isn’t buying it, and the gap between what Salesforce shows on stage and what customers actually use is widening.
The stock tells the story. Salesforce shares are down about 21 percent in 2025 and another 30 percent in 2026. service companies such as softwarean event the market is calling the SaaSpocalypse. About $285 billion in SaaS market capitalization evaporated in a 48-hour window in February. The logic is simple: if one AI agent does the work of ten employees, why would a company pay for ten?
Salesforce has tried to get ahead of this question by positioning itself as a company that sells agents, not seats. CEO Marc Benioff calls Agentforce “digital labor platform.In its earnings call, the company cites 29,000 deals and an ARR figure as evidence that businesses are buying.
💜 of EU technology
The latest rumblings from the EU tech scene, a story from our wise founder Boris and some questionable AI art. Free in your inbox every week. Register now!
The problem is that showcase patterns keep falling apart under control. At Dreamforce, Salesforce demonstrated a Williams-Sonoma AI agent called Olive, which is supposed to act as an agent chef helping customers plan meals and find products. In practice, Olive struggled with specific questions and recommendations. The more advanced capabilities of the agent are described using the future tense.will soon be able to” as features that are more alive.
A similar pattern emerged at the University of Chicago Medicine. Salesforce introduced the hospital system as the leading Agentforce for Healthcare deployment. The reality was more modest: UChicago Medicine’s first artificial intelligence agent launched in web chat to answer basic questions like parking directions and clinic availability. More ambitious features, including voice-based patient support, were still in development.
SharkNinja, maker of Shark vacuum cleaners and Ninja kitchen appliances, was another major client. Salesforce said the company will use Agentforce to streamline customer service. Bloomberg reported on this 20 percent reduction in support calls as part of the pitch. But the placement described was promising, and the agents “guiding customers through the purchasing process” and “manage revenue,” is no longer a report on results achieved.
This is important because Salesforce is not the only company Overselling AI capabilities. In May, Apple agreed to pay $250 million to settle a lawsuit alleging Apple Intelligence and a smarter Siri overstated what it would offer when it launched the iPhone 16. The agreement included claims that the company’s marketing went further than the technology could initially do.
Salesforce’s financial trajectory adds another layer. Revenue growth has slowed from about 25 percent a few years ago to about 10 percent in fiscal 2026, when the company’s total revenue was $41.5 billion. That’s still a big deal, and the company had a strong fourth quarter with 12 percent growth. But the slowdown is exactly what investors fear when they hear that AI agents will crowd out the number of human users who need software licenses.
The company tried to solve the price issue. Agentforce uses a consumption-based model rather than traditional seat pricing, charging for what Salesforce calls “agent work units.” It consumed about 20 trillion tokens and turned them into more than 2.4 billion such units. The key issue is whether this model can grow fast enough to offset the structural threat to seat-based revenue.
Small clients show both promise and value. The City of Kyle, Texas, deployed Agentforce to manage its 311 service, handling more than 12,000 resident inquiries as of March 2025, with a first-call resolution rate of nearly 90 percent. Bloomberg reported that the city doubled Salesforce spending to $300,000. For a rapidly growing municipality, this can be a reasonable investment. The economics are less clear for enterprise customers, who draw the same scale.
The competitive pressure is real. SAP introduced its autonomous enterprise With over 200 AI agents and Anthropic partnership at Sapphire 2026. ServiceNow, Google, and Microsoft are all building agent platforms. The question is no longer whether AI agents will reshape enterprise software, but whether Salesforce can maintain its position around market pricing.
Benioff responded with characteristic confidence by announcing a new revenue target of $60 billion by fiscal year 2030. It also committed to buy back $50 billion in shares, signaling to investors that the company believes its stock is undervalued. Transforming Slack into an agent platformMore than 30 new AI capabilities and mandatory bundling with every new Salesforce account starting this summer are part of that push.
None of this resolves the underlying tension. Salesforce is asking customers to pay for a future its demos don’t yet offer, while also asking investors to believe that consumption-based AI revenue will replace the seat-based model that built the company. 29,000 deals are real. $800 million in ARR is realistic. But agent AI marketplace rewards results, not announcements, and the space between the two is where Salesforce’s credibility will be tested.
