Palantir Revenue: US Commercial Sales Take the Lead

The growth is broad, but the US split is the headline

Palantir's total revenue includes customers beyond the two US categories in the comparison. So you can't say commercial customers now make up more revenue across the entire company. The milestone is specifically that US commercial revenue surpassed US government revenue in Q3.

That distinction matters for how you think about the business. Government deals can be large, long-running, and operationally central - but they often involve procurement rules and timelines that are different from commercial deployments. Commercial growth can suggest Palantir is getting used by a wider range of organizations. It still doesn't prove the company is less dependent on government demand.

Palantir's net income also jumped, more than doubling to $1.13 billion from $476 million a year earlier. The company ended September with $10.4 billion in cash and US Treasury securities. That gives it flexibility to invest and ride out volatility, though cash alone doesn't resolve questions about valuation or how durable growth will be.

Management raised its expectations

Palantir raised its full-year revenue forecast to between $8.530 billion and $8.534 billion, implying 91% growth. The range is extremely tight, so it reads more like a specific outlook than a broad "we're not sure" band. Bloomberg data puts the forecast above analysts' expectations of roughly $8.2 billion.

For US commercial revenue, Palantir now expects more than $3.559 billion for the year - growth of at least 143%. Three months earlier, it forecast $3.42 billion. For Q4, the company expects revenue of about $2.65 billion. In other words, this isn't just a strong quarter on its own; it's also management signaling that the momentum can continue.

A change in customer mix is not the same as a change in business model

Palantir's CEO, Alex Karp, said the company is succeeding without salespeople. That's a big claim, because most enterprise software companies lean on large sales teams to find customers, manage long sales cycles, and expand accounts once they land.

But a CEO's description isn't the same thing as operating evidence. The earnings materials don't tell you how customer acquisition costs compare to other software firms, how much support deployments require, or how much of the growth is from new customers versus existing ones expanding usage. Those details matter if you're trying to understand whether a "sales-light" approach can scale.

Palantir also said it closed 212 deals worth at least $1 million during the quarter, including 64 deals worth at least $10 million. Deal counts show real activity, but they don't reveal the full economics. A big "deal" could mean multiple products, a slow rollout, or a relationship that grows over time. You can't infer long-term value or the exact revenue path from deal counts alone.

Why buyers should care about the trend

For organizations evaluating AI software, faster commercial adoption can be useful context. It suggests the product may be solving problems beyond its original public-sector customer base.

Still, this isn't a recommendation. Buyers should test whether Palantir fits their data environment, security requirements, internal workflows, and budget. They also need to see if it creates value beyond a pilot.

A good way to separate what this quarter tells you: think of it as market validation, not product fit. Palantir's results suggest more US companies are buying. They don't automatically say whether a specific deployment will integrate smoothly, satisfy internal controls, or justify the total cost. The same goes for investors: growth can validate demand without proving the valuation is reasonable.

What the guidance says - and what it does not

The full-year forecast of $8.530 billion to $8.534 billion is the clearest sign that management expects demand to hold through the year. The US commercial forecast - now more than $3.559 billion and raised from $3.42 billion just three months earlier - also matters because it shows management sees more momentum in a segment that was already growing quickly.

But forecasts are conditional. They rely on contracts closing, customers deploying the software, and spending continuing at the expected pace. Palantir's Q3 results are strong evidence of current performance. The outlook is management's view of what follows. Treating the outlook as certain would blur an important line between what happened and what may happen next.

Palantir also talked about sovereign AI. Karp said the shift toward sovereign AI creates a split between countries and organizations that have access to advanced capabilities and those that don't. That's a potential demand driver - governments and businesses may want more control over models, infrastructure, and data used in AI systems. But it's still a broad strategic argument, not a quantified revenue driver in the figures reported for this quarter.

The deployment question sits behind the growth rate

Interest is only the first stage for enterprise software. The real test is whether customers move from experimentation into operational use - and whether they keep expanding after the first deployment.

A company can win deals during a period of high AI interest. Durable growth depends on whether customers keep paying because the software works in real workflows.

What to watch in later quarters: continued commercial growth, renewals and expansion patterns, customer concentration, and evidence that deployments are repeatable. The Q3 report gives strong revenue and deal figures, but it doesn't answer every question about how usage turns into long-term customer economics.

A practical way to read future results is to keep returning to three questions:

• Is growth still broad? Compare commercial and government performance and see whether momentum shows up across more than one quarter.
• Are expectations being met? Watch actual results versus guidance, not just year-over-year percentages.
• Is adoption becoming durable? Look for signs that customers are expanding deployments and staying engaged, not just new deals.

Competition is not limited to traditional software vendors

Palantir has competition from companies building AI models and software, including AI developers that sell directly to businesses. Investors have also previously raised concerns about firms such as Anthropic moving into parts of the software market.

The real competitive question isn't just whose model is best. It's which vendors can deliver a secure, usable system that fits customer workflows and delivers value at a price buyers will keep paying.

That opens the door to multiple competitor types. Customers can choose a model provider, a cloud vendor, an established enterprise software company, an internal tool, or a mix of all of the above. Whether Palantir keeps customers will depend on more than AI buzz: implementation, integration, governance, performance, and total cost all determine whether a platform remains useful after a proof of concept.

Sovereignty and procurement can cut both ways

Karp's comments on sovereign AI point to a real tension. Some governments want more control over technology and data and may prefer domestic providers for sensitive workloads. That preference can create opportunities for local firms and make procurement more complicated for international vendors.

The source report noted that Germany's military was evaluating 30 AI tools, with a prototype expected in 2027. That shows the kind of competition happening in public-sector deployments, but it doesn't tell you who will win or how those evaluations translate into revenue. In markets where sovereignty matters, local rules, security requirements, and politics can weigh as much as technical capability.

Karp also called for nationalizing leading AI labs, describing it as the only realistic way to manage liability. That's his policy position, not a confirmed direction of US policy. For buyers and investors, the key point is simpler: regulation and liability around AI remain unsettled. Changes to rules or procurement requirements could affect vendors and customers differently.

A decision framework for readers

If you're evaluating Palantir as a vendor, start with a bounded use case before you commit to a platform rollout. Test integration work, data access, security controls, user adoption, and measurable cost or time savings. Also ask what changes when the deployment grows from one team to several, since the operational burden often shifts at that point.

If you're looking at Palantir as an investment, separate three layers: reported performance, management guidance, and market valuation. Q3 performance was strong; guidance rose; valuation is still a separate question. You should also consider whether growth rates cool as the company gets larger and competition increases.