Salesforce's AI agent business has grown more than 200%, a sign that the company is finding paying customers for its AI-driven enterprise software. The jump puts the business at the center of Salesforce's strategy, but the road ahead is not without friction. The company is changing how it charges for the technology, and there are open questions about how far the business can expand.
Growth that highlights enterprise appetite
The triple-digit growth reflects a broader push by companies to automate tasks that once required human workers. Salesforce's AI agents handle things like customer inquiries, lead scoring, and data entry, all inside the same cloud platform many businesses already use. That built-in base is a big reason the adoption curve has been steep.
Customers aren't buying a brand-new tool. They're adding a feature to a system that's already running. That makes the decision easier for buyers and shortens the sales cycle. The growth also signals that enterprise buyers are moving past pilot projects and putting AI agents into real workflows, which is exactly the kind of traction Salesforce wants to show.
Pricing model shifts and the revenue challenge
But the 200% number comes with a catch. Salesforce is shifting its pricing model for AI agents. The old way, a flat subscription per user, is giving way to usage-based pricing, where customers pay according to how many tasks the agent completes or how much data it processes.
That model can make revenue more variable. When usage is high, the money flows. When customers pull back or find cheaper alternatives, the numbers could fall just as quickly. The shift also makes it harder for Salesforce to predict quarterly revenue with the same confidence it had with subscription seats. Whether the 200% growth translates into sustained revenue depends on how consistently customers keep using the agents at scale.
Expansion concerns and the limits of scale
There are also concerns about how far the business can expand. As the customer base grows, Salesforce needs to keep up with infrastructure, data handling, and model training. Every new AI agent deployment adds load to the company's systems, and any slowdown in performance could push buyers to look elsewhere.
Competition is another factor. Large cloud platforms and smaller startups are all selling AI agents. Salesforce has the advantage of deep enterprise relationships, but it's not the only option. The question is whether the company can defend its position as the market gets more crowded. A misstep in pricing or infrastructure could give customers a reason to switch.
The next earnings report will be the first real test of whether this 200% is a one-time spike or a new baseline. The company has to show that the growth isn't just a short-term boost from early adopters, but that the AI agent business can keep expanding without the pricing model dragging it down.
