There's a classic scene in The Matrix where Morpheus asks Neo: "Do you want the blue pill or the red pill?" The market looked at Salesforce's results Wednesday night, saw better-than-expected numbers, and chose the panic pill anyway.
Down 4.5% in after-hours. Stock hitting $183, damn near its 52-week low.
What the hell else does this company need to do?
The numbers that should have been enough
Let's get to it. Revenue of $11.2 billion in the fourth fiscal quarter of 2026 — 12% growth year-over-year, beating the consensus of $11.18 billion. Adjusted earnings per share of $3.81, 77 cents above what the Street expected. Year-over-year, adjusted EPS grew 37%.
Agentforce, the AI platform that's Marc Benioff's big bet on the future, has already closed more than 29,000 deals since launch and turned into an $800 million annual recurring revenue business. Amazon, Ford, AT&T, Moderna, GM, and Pfizer are on the client list. CEOs from SharkNinja and Wyndham Hotels got on the earnings call to say the thing actually works — cutting labor costs and generating millions in incremental revenue.
This isn't a sales pitch. These are real customers talking with real money.
And still, the market said: "I don't care."
The wart nobody can ignore
Here's where the devil lives in the details, and it's where the real investor needs to pay attention instead of just reading the headline.
cRPO — contracted revenue expected over the next 12 months — grew 13% on a constant currency basis. Looks nice, right? Except 4 percentage points of that growth came from the recently closed Informatica acquisition. Organically? 9% growth. Single digits.
And the market wanted double digits.
That's the narrative eating Salesforce from the inside: the thesis that generative AI is going to devour traditional enterprise software — and that the company can't grow Agentforce and the legacy business at the same time. The organic 9% didn't kill that thesis. It fed it.
To make matters worse, GAAP margins — the ones that follow actual accounting principles, no makeup — compressed year-over-year and came in below estimates. Non-GAAP margins looked pretty, of course. But as Taleb would say, when everybody's using non-GAAP to hide their scars, the smart investor grabs a magnifying glass and stares at GAAP.
And the guidance? Revenue and GAAP operating margin for the new fiscal year came in "a hair below." In an environment where the stock is already down 27% on the year — after a miserable 2025 — "a hair below" is the same as pouring gasoline on the fire.
The desperation play that actually makes sense
Salesforce bought back $4 billion in shares during the quarter and announced a new repurchase program of up to $50 billion. With a market cap of $180 billion, that's equivalent to 27% of the company.
Read that again: twenty-seven percent.
This isn't signaling. This is the board screaming: "The stock is too damn cheap!" And at 14.5x the midpoint of projected earnings, with double-digit revenue growth, it's hard to argue they're wrong — if you believe the business isn't going to be swallowed by cheaper AI agents.
The elephant in the room called "AI eating software"
This is the existential fear. And no single quarter is going to resolve it.
The market needs proof — quarter after quarter — that Agentforce isn't just a cosmetic patch, but the real evolution of the business model. That Salesforce can be the platform that delivers the disruption instead of being the victim of it.
For now, what we have is a company that beats revenue and earnings estimates, has heavyweight clients validating the product, is buying back shares like the world ends tomorrow — and still can't convince the market.
Sometimes Mr. Market is rational. Sometimes he's a paranoid who's watched The Terminator one too many times.
The $180 billion question: are you going to buy the fear, or wait for the fear to prove you right?
