When will Agentforce make its mark on Salesforce earnings? Wait ‘til next year

It was the moment during the Salesforce earnings call on Wednesday when the needle scratched across the record.

Amy Weaver, Salesforce president and CFO, was talking about Agentforce and said: 

“We are incredibly excited about the customer momentum we are seeing. However, the adoption cycle is still early as we focus on deployment with our customers. As a result, we are assuming a modest contribution to revenue in fiscal ’26. We expect the momentum to build throughout the year, driving a more meaningful contribution in fiscal ’27.”

Overall, Salesforce earnings were a disappointment, as was the company’s guidance for the 2026 fiscal year. And the markets are responding as markets do to such news — unkindly.

However, the idea that Salesforce customers are uninterested in Agentforce doesn’t hold water, at least not yet. There’s a lot of work to do before businesses trust AI agents to talk to customers and supply employees with accurate information. 

For Salesforce customers, that work is being done by Data Cloud. 

The Salesforce earnings release did not show that its customers are shying away from Data Cloud or AI in general. 

• Data Cloud and AI annual recurring revenue were up 120% year-over-year to $900 million.
• Nearly half of the Fortune 100 are both AI and Data Cloud customers.
• All of Salesforce’s top 10 wins in Q4 included Data and AI.
• 5,000 Agentforce deals closed since October, and 3,000 were paid. 

AI agents using bad data is a bad idea

As you likely know, there are more than 14,000 martech tools on the market. As anyone I’ve talked to in the last couple of months will tell you, the tools get a lot of attention, but I believe the data does all the work.

Without quality data, the tools just deliver junk. Feed them bad data, and they deliver more junk. 

Now imagine AI agents with access to bad data and little human oversight. Does this sound fun? No, it does not. 

Last week, I spoke to Gabrielle Tao, Salesforce’s senior vice president of product management. Tao has a long history in the CDP space and currently works on Data Cloud at Salesforce.

I asked how Salesforce customers and prospects are preparing their data for a world where AI agents can be trusted to do their thing.

Tao said many businesses she talks to have unified data. That’s a good start.

“The problem is actually their disconnect between their unified data and their front-end business user and applications,” Tao said. “A sales rep will never log into any back-end data warehouses where the unified data architecture is constructed. A marketing user creating a campaign will also never log into a data warehouse. So, how do you serve all of the greatness of the unified data to these business applications and users, and by extension, the agents? That unified meta-data-level connection is what Data Cloud provides.”

This is a process for many organizations. And Salesforce was one of them. 

Salesforce’s app for its Dreamforce conference back in September was a special-purpose, quick-start AI agent use case. It could tell users about sessions or where to grab some food, but beyond that, it didn’t offer much. For example, you couldn’t pepper it with questions about Data Cloud after viewing a session description.

Moving beyond the quick-start use cases is a big step for many Salesforce customers. 

“In that next step, they realize, beyond that quick-start use case, that’s when they need to bring their enterprise data to full bear, and that’s when we have that middle layer, unified metadata, mapping, governance, augmented AI and analytics discussion,” Tao said.

I’m not predicting Agentforce will be a goldmine for Salesforce or a bust, either. I’m saying it’s the early innings, and there’s a long way to go. But evidence that customers are investing in better data is a good start.

The post When will Agentforce make its mark on Salesforce earnings? Wait ‘til next year appeared first on MarTech.