Salesforce has reported robust financial results for its fourth quarter and full fiscal year, while aggressively addressing mounting investor fears that the rise of artificial intelligence (AI) agents could undermine the traditional software-as-a-service (SaaS) business model. The company announced a significant share buyback and dividend increase alongside the introduction of new AI performance metrics.
The cloud software giant posted quarterly revenue of $10.7 billion, a 13% year-over-year increase. For the full fiscal year, revenue reached $41.5 billion, up 10% from the previous year, a figure bolstered by its $8 billion acquisition of data management firm Informatica in May 2024. Net income for the year was $7.46 billion.
Confronting the 'SaaSpocalypse' Narrative
During Wednesday's earnings call, CEO Marc Benioff directly confronted the term "SaaSpocalypse," which has been used to describe investor anxiety that AI agents could make per-user subscription models obsolete. Benioff mentioned the term at least six times, attempting to downplay the threat.
"You’ve heard about the SaaSpocalypse? And it isn’t our first. We’ve had a few of them," Benioff stated. He later added, "If there is a SaaSpocalypse, it may be eaten by the Sasquatch because there are a lot of companies using a lot of SaaS because it just got better with agents."
Financial Measures and a Theatrical Earnings Call
To bolster investor confidence, Salesforce announced a new $50 billion share repurchase programme and increased its quarterly dividend by nearly 6% to $0.44 per share. The company also projected strong future revenue of between $45.8 billion and $46.2 billion for the coming year.
Departing from a standard financial presentation, the earnings call featured pre-recorded video testimonials from three customer CEOs, including the heads of SharkNinja, Wyndham Hotels and Resorts, and SaaStr, all praising Salesforce's new AI agent products.
Introducing 'Agentic Work Units'
Central to its AI defence was the introduction of a new performance metric: Agentic Work Units (AWU). Salesforce executives argued that measuring raw AI "tokens" processed—which reached 19 trillion last quarter—was insufficient for enterprise value.
"You can ask it a question and it can write you a poem, but that’s not really all that valuable in the enterprise world," said Salesforce President and CMO Patrick Stokes. The AWU metric is designed to track when an AI agent completes a verifiable business task, such as writing to a customer record, rather than merely generating text.
Architectural Vision and Competitive Stance
Salesforce presented its own architectural blueprint for the future of enterprise AI, positioning SaaS platforms like itself at the top of the technology stack. This vision places AI model providers, like OpenAI, at the bottom as commoditised "work engines."
This presentation is seen as a direct rebuttal to OpenAI's recent "Frontier" platform launch, which depicted its own systems owning the core stack, potentially relegating data platforms like Salesforce to a secondary role.
The company's remaining performance obligation (RPO), representing contracted revenue not yet recognised, stands at over $72 billion. Looking ahead, Salesforce's strategy appears focused on convincing the market that its integration of AI agents will reinforce, not replace, its core SaaS offerings.