Palantir just posted its first $2 billion quarter, and the stock still slipped. According to TechCrunch AI, the company reported third-quarter revenue of $2.31 billion on Monday, but shares traded slightly down on Tuesday because its enterprise sales showed signs of slowing. Demand isn’t the open question here. Staying power is.
The quick read
- Revenue: $2.31 billion, up 96% from a year earlier and 20% from the previous quarter. Analysts expected $2.18 billion.
- Adjusted earnings per share: 44 cents, against expectations of 41 cents.
- The milestone: U.S. commercial revenue hit $984 million, up 148% from a year ago. It passed U.S. government revenue ($953 million) for the first time.
- The warning sign: Total contract value from U.S. enterprises fell to $1.46 billion from a record $2.13 billion the quarter before. That’s a drop of about 31%.
- The outlook: Palantir raised its full-year revenue forecast and now expects U.S. commercial revenue “in excess of $3.559 billion, representing a growth rate of at least 143%.”
Why enterprises are signing up
Palantir’s pitch comes down to control. Its AI software lets companies run frontier models and custom-train open-weight models (models with publicly released weights) without sharing their data and business processes with outside AI labs.
The other draw is what’s now called forward-deployed engineering, where Palantir’s own engineers work directly with customers to customize the software. Palantir has offered this for years, long before the term caught on.
That combination has changed the shape of the business. A company long tied to government work now makes more from U.S. corporations than from the U.S. government, if only by $31 million.
The catch: fast deals, short terms
Enterprises are moving quickly, but many are signing short-term trials. CEO Alex Karp described “much quicker interaction, much quicker to deployment, shorter duration of the contract but then an ability to expand that contract dramatically” on the earnings call.
This goes well beyond Palantir. TechCrunch AI notes that corporations will now skip the usual bureaucracy to try AI software, but they’re also more willing to cancel it. Easy in, easy out.
Karp isn’t worried, or at least he doesn’t sound it. “We’ve gone from being the most attractive person in the physics department to being the captain of the football team, kind of thing,” he said of the wave of prospective customers. He added: “once you go Palantir, you don’t go back, and that’s just how it is, and you see it in our numbers.”
The numbers don’t prove that yet, as TechCrunch AI points out.
My read on the contract drop
What stands out here is that the two stories can both be true. Shorter contracts shrink total contract value on their own, because each deal locks in less future money. So the 31% drop could mean slower sales, shorter terms, or a mix of both, and the figures reported here don’t separate them.
It’s also one quarter measured against a record, so I wouldn’t call it a trend yet. Still, Karp’s expansion story has to show up in later results. A short contract is a low-commitment date, and nobody knows how many turn into marriages until renewal time.
What this means for you
- If you buy enterprise AI: Short trials are now normal, so use them. Push for short initial terms, and decide up front what results would justify expanding.
- If you sell AI software: A fast signature tells you little about retention. Track how many pilots expand and how many quietly end.
- If you follow the stock: Watch U.S. enterprise contract value next quarter. It’s the number that fell, and many enterprises historically sign their biggest tech purchases in the final quarter of the year.
That makes Palantir’s next report the first real test of Karp’s confidence. If contract value rebounds, this dip will look like a blip. If it doesn’t, the shift to shorter deals may start to look like hesitation. You can find the full report at TechCrunch AI.