Salesforce reported 11.3 billion dollars in quarterly revenue, up 11 percent from a year ago, and adjusted earnings per share of 5.90 dollars, apparently blowing past the roughly 3.27 dollars analysts expected. On the surface, that looks like one of the cleanest beats of the entire earnings season. It is not quite that simple.
Buried in the company's own reconciliation tables is a single line item, gains on strategic investments, that totaled 2.613 billion dollars this quarter, versus just 6 million dollars a year ago. Salesforce's own footnote states plainly that this gain added 2.43 dollars to GAAP earnings per share and 2.53 dollars to adjusted earnings per share, and unlike many companies with large equity-investment swings, Salesforce does not exclude this gain from its adjusted, non-GAAP results. Strip it out, and adjusted earnings per share comes to roughly 3.37 dollars, a real beat of the roughly 3.27 dollar consensus, but closer to 3 to 4 percent than 80 percent. GAAP earnings per share, once the gain is removed, comes to roughly 1.86 dollars, which is actually below the 1.96 dollars Salesforce reported in the year ago quarter, when no comparable gain existed.
None of this means the quarter was weak. Revenue, contracted backlog, and several AI product metrics all show real growth. It means the single number most likely to anchor headlines, the 5.90 dollar adjusted EPS figure and its apparent 80 percent beat, is doing far more work from a one-time, non-operating, fair-value gain than from the underlying business.
Growth That Leans More on Acquisitions Than the Headline Suggests
Revenue growth itself needs a similar adjustment. The reported 11 percent increase includes a 456 million dollar contribution from Salesforce's Informatica acquisition. Strip that out, and organic revenue growth comes to roughly 6.4 percent, a meaningfully softer number than the double-digit headline implies. That distinction matters even more once Salesforce's own guidance notes are read closely: management's own disclosure attributes this quarter's full-year revenue guidance increase to roughly 100 million dollars of organic growth, 200 million dollars from two pending acquisitions expected to close next quarter, and a 100 million dollar currency headwind netted against both. In other words, the raise is disclosed by the company itself as roughly half acquisition-driven, not primarily a signal that organic AI demand has reaccelerated, even though that is the story dominating the evening's reaction.
Profitability tells a similarly mixed story. GAAP operating margin compressed 230 basis points from a year ago, to 20.5 percent, even as adjusted operating margin held essentially flat at 34.1 percent. No source available for this article identifies a specific driver of that GAAP margin compression, whether integration costs from recent acquisitions, higher amortization of purchased intangibles, or something else, and this publication is not in a position to assign a cause it cannot verify.
The Product Metrics That Are Genuinely Strong
Set the accounting questions aside and Salesforce's AI product numbers are real and growing fast. Combined annual recurring revenue from its AI and data products, primarily Agentforce and Data Cloud, approached 4 billion dollars, more than tripling from a year ago. Agentforce alone crossed 1.5 billion dollars in annual recurring revenue, also more than tripling. The company said customers generated 3.2 billion units of agentic work during the quarter, a 97 percent increase from the prior quarter, and that its Data Cloud platform ingested 104 trillion records, up 355 percent from a year ago. Current remaining performance obligation, a forward-looking bookings measure, grew 14 percent on a constant currency basis, ahead of revenue growth itself, which is generally read as a sign that near-term demand is outpacing what has already been recognized as revenue.
On the call, an analyst from Goldman Sachs pressed management on how Salesforce plans to monetize AI agents that customers deploy outside its packaged software, through direct programming interfaces rather than its standard user interface. The chief financial officer's answer, according to the available call material, pointed to premium-edition upgrades and flexible pricing as levers, but explicitly acknowledged that pricing for this kind of deployment is still being worked out. That is a meaningful admission. It suggests Salesforce has strong usage numbers for its AI products but has not yet settled on how much of that usage translates into durable, predictable revenue, particularly for the deployment model many of the most technically sophisticated customers are likely to prefer.
Separately, the company announced an expanded partnership with Anthropic the same evening, extending its AI assistant integration further into its platform, and its chief executive used a same-day media appearance to directly reject the idea that AI assistants pose an existential threat to software companies like Salesforce, arguing that customers still consume AI through packaged applications rather than choosing underlying models directly.
What the Market Still Has to Sort Out
Shares rallied roughly 13 percent in extended trading, a larger move than the roughly 8 percent options market had priced in ahead of the print, even though the stock remained well below its 52-week high and down meaningfully for the year to date. That gap, a rally this large on a stock still off its highs, suggests the market was pricing in downside risk that the quarter's headline numbers appeared to erase, without most of the coverage that shaped the initial reaction accounting for how much of the erasure came from an investment gain rather than operating performance.
The burden of proof going forward is narrower and more specific than the headline suggests. Investors should watch whether organic revenue growth, stripped of Informatica and the two pending acquisitions, actually reaccelerates in the coming quarters, whether GAAP operating margin stabilizes or continues compressing, and whether Salesforce settles on a monetization model for AI agents deployed outside its own interface. The bull case strengthens if Agentforce and Data Cloud ARR growth continues at anything close to its current pace while organic revenue growth genuinely picks up. The bear case strengthens if the next several quarters show that this quarter's acceleration was substantially a function of acquisitions and a one-time gain rather than a durable shift in how fast the core business can grow on its own.