Salesforce Inc. (CRM) is heading into its third-quarter earnings report with a peculiar problem: management keeps talking about record-breaking AI momentum, but the people actually implementing Salesforce technology aren't feeling it.

It's the kind of disconnect that makes investors nervous. On one side, you have CEO Marc Benioff championing the company's "Agentic Enterprise" transformation with impressive metrics. On the other, you have partners—the consulting firms and system integrators who deploy Salesforce in the real world—quietly reporting contracting pipelines and customers stuck in what they're calling "pilot purgatory."

Management's Story: The AI Machine Is Humming

The numbers Salesforce put up in Q2 look genuinely impressive. The company's Data Cloud and AI portfolio grew 120% year-over-year to reach $1.2 billion in annual recurring revenue. That's not nothing.

Benioff has been vocal about the shift from chatbot experiments to actual production deployments, pointing to a 60% quarter-over-quarter jump in customers moving past the pilot stage. The company has signed over 12,500 Agentforce deals, with 6,000 of them generating revenue.

Bulls see this as the beginning of a flywheel effect that will drive long-term margin expansion. The narrative is compelling: AI isn't just hype anymore, it's turning into real revenue.

What Partners Are Actually Seeing on the Ground

Here's where things get messy. Partners—meaning the third-party consultants, system integrators, and software vendors who implement Salesforce for customers—are telling a different story.

According to a Guggenheim partner survey, the net score of partners exceeding their targets dropped from 26 in the prior quarter to 16. That's a significant deterioration in sentiment, and it's happening right as management is declaring victory.

The phrase "pilot purgatory" keeps coming up. Companies are experimenting with AI, sure, but they're not committing to large-scale deployments. Wells Fargo analysts note that even when Agentforce gets adopted, it's often limited to one or two isolated use cases rather than becoming a company-wide solution.

The culprit? Data quality. Partners consistently cite "dirty data" as the major roadblock preventing effective AI deployment. You can have the best AI tools in the world, but if your underlying data is a mess, those tools aren't going to deliver much value.

What Wall Street Expects from Q3

Expectations are deliberately low, which could actually work in Salesforce's favor if they can beat them. Analysts are forecasting Q3 revenue of approximately $10.27 billion, representing about 8.7% year-over-year growth—hardly the explosive acceleration you'd expect from a company supposedly riding an AI wave.

Earnings are expected at $2.58 per share. Non-GAAP operating margins should hold steady around 34.1%, reflecting the cost discipline Salesforce has maintained.

One wildcard: the Informatica acquisition that closed in mid-November is expected to add roughly $350 million to Q4 revenue, which could make it harder to parse out organic growth when the company issues guidance.

The real question investors want answered is whether Benioff's agentic vision can bridge the gap between enthusiastic press releases and the contracting pipelines that partners are reporting from the trenches.

The Stock Has Been Punished

Salesforce (CRM) closed 0.81% higher at $234.71 on Tuesday, adding another 0.55% in after-hours trading. But zoom out and the picture looks rough: the stock has declined 29% year-to-date and maintains weak price trends across short, medium, and long-term timeframes.

That kind of decline reflects genuine uncertainty about whether the AI narrative will translate into sustained growth or whether customers will remain stuck in experimentation mode while competition intensifies.

This earnings report isn't just about beating or missing numbers. It's about whether management can convince investors that what they're seeing from the executive suite matches what's happening on the ground. If partners keep reporting gradual adoption and data quality headaches while management touts record momentum, that credibility gap becomes a real problem.