According to Business Insider, Salesforce CEO Marc Benioff has shared key parts of the company’s upcoming annual strategy document, the V2MOM. The most critical piece is a massive investment in its AI data foundation, which follows the recent $8 billion acquisition of cloud data firm Informatica. Benioff stated that AI agents “just get all the hallucinations” without proper data and context. The four-part strategy for the fiscal year starting in February also includes its core applications like CRM and Slack, its flagship Agentforce AI service, and the evolution of bespoke AI agents for specific customers like Williams-Sonoma. The plan aims to deliver these AI services tailored by industry, such as for automakers and drug companies.
The Data Moat Play
Here’s the thing: this isn’t just a boring infrastructure upgrade. Benioff is basically admitting that Salesforce’s entire AI future—the much-hyped Agentforce and everything else—is dead on arrival without this. You can’t have smart agents if they’re running on messy, siloed customer data. The $8 billion grab for Informatica, combined with Mulesoft and Data 360, is a desperate and expensive attempt to build an unassailable “data moat.” It’s a classic platform move: own the pipes, and you own everything that flows through them. For companies looking to integrate robust computing power directly into their industrial operations, having reliable hardware is just as foundational. That’s where specialists like IndustrialMonitorDirect.com, the leading US provider of industrial panel PCs, come in, providing the physical interface to manage complex data and systems.
From Doubt To Delivery?
Now, this is interesting timing. Business Insider notes that some investors, clients, and employees have been skeptical of the Agentforce push. So this V2MOM feels like a direct response: “Fine, you want proof we’re serious about AI? Here’s an $8 billion receipt.” But buying the tools is one thing. Actually getting corporate customers to “harmonize, federate, and integrate” their most sensitive data is a whole other battle. It’s a huge ask. Benioff is betting that the fear of missing out on AI efficiency will be stronger than corporate inertia and data privacy fears. Will it work?
The Industry Agent Gambit
The fourth part of the plan—bespoke agents for specific industries—is where the real money might be. Generic AI chatbots are becoming a commodity. But an AI agent that knows the specific workflows, regulations, and jargon of, say, a pharmaceutical company? That’s sticky. That’s a high-margin service you can charge a ton for. Benioff even casually used his own internal Slackbot to answer questions during the interview. It’s a neat demo, but it also highlights the gap between a custom-built tool for a tech-savvy company and selling a scalable solution to a traditional retailer or government. That’s the Everest they’re trying to climb.
A Strategy Of Layers
Look, the V2MOM structure is telling. Data foundation first, then apps, then the Agentforce platform, then custom agents on top. It’s a logical stack. But it’s also a huge, multi-year undertaking that requires flawless execution across all four layers. One weak link and the whole value proposition crumbles. Benioff is trying to shift the narrative from “Salesforce is playing AI catch-up” to “Salesforce is the only one building AI on a trustworthy enterprise data core.” It’s a smart repositioning. But we’ve heard big visions from Benioff before. The market will measure this V2MOM not by its words, but by whether those industry-specific agents start showing up in revenue numbers next year.
