What makes Palantir the first stock to commercialize AI?

In the Q2 2025 financial report, Palantir delivered a report card that almost all SaaS practitioners dream of: U.S. commercial business revenue grew by 93% year-on-year, total revenue exceeded $1 billion, adjusted operating profit margin was 46%, free cash flow profit margin was 57%, and the Rule of 40 index reached as high as 94%.

The company is thus referred to as the "first stock of AI commercialization." This title does not stem from being the earliest to talk about AI or having the most funding, but because Palantir is currently the only AI company that has achieved a "high growth rate, high profit margin, and high cash flow" triple loop—it has not only completed the full process from model to product, from product to structure, and from structure to profit, but has also built a complete set of replicable and scalable organizational growth logic.

In Q2, Palantir first surpassed $1 billion in quarterly revenue, with a year-on-year growth of 48%, and revenue growth in the U.S. market reached 68%. Notably, U.S. commercial business revenue surged by 93% year-on-year, increasing its share of total revenue from 23% in the same period last year to 31%.

Government business also grew by 53%, although the growth rate was not as high as that of the commercial sector, it still recorded $426 million in revenue for the quarter. This comparison clearly indicates that Palantir's growth focus has gradually shifted from primarily government clients to the commercial sector, especially AI implementation services for large enterprises.

What drives Palantir's growth is not the traditional software sales model, but a new value chain system built around the implementation of AI products.

Taking customer growth as an example, by the end of Q2, Palantir's total customer count grew by 43% year-on-year, reaching 849; among them, the number of U.S. commercial customers was 485, a year-on-year increase of 64%. This means that Palantir is not only continuously breaking through in "deep cultivation of large clients," but also expanding its market foundation in the "penetration of medium-sized clients."

The Rise of the AI Core Engine

From the breakdown of revenue structure, Palantir recorded a historical high in Total Contract Value (TCV) of $2.3 billion, a year-on-year increase of 140%; among them, commercial TCV was $1.1 billion, a year-on-year increase of 185%, and U.S. commercial TCV reached $843 million, with a staggering year-on-year growth of 222%. Behind this growth curve is the signal that AI is truly entering the underlying processes of enterprises.

For example, with Citibank and Fannie Mae, KYC reviews or fraud detection that used to take days or even months can now be completed in seconds; Lear Corporation, over the past 2.5 years, has supported over 11,000 employees with 175 use cases through Foundry and AIP (Artificial Intelligence Platform), covering multiple scenarios from dynamic balancing on the manufacturing line to administrative process automation The president of Nebraska Medicine stated that the utilization rate of the hospital's discharge rest area has increased by 2100%, "We even measure value using Palantir's unit time."

These cases are not just about functional implementation, but also represent a "second reconstruction" of the clients' own structure and efficiency. This also means that Palantir no longer relies on a large direct sales team for its sales strategy, but instead grows through product effectiveness and customer network word-of-mouth.

CEO Alexander Karp clearly stated: "We are not interested in building a sales system that relies on wining and dining or selling ineffective software. All our value is driven by our products and deployment engineering teams." The conversion of American corporate clients often starts with the phrase, "You have to see how they do it."

The "AI implementation" defined by Palantir is not about deploying an LLM, but rather a systematic understanding centered around Ontology. CRO Ryan Taylor pointed out: "LLM is fragmented intelligence; Ontology is the logic of true understanding. Without Palantir, LLM cannot truly operate in the real world."

Based on this logic, Palantir has taken an extremely firm stance in AI iteration— all AI projects must be premised on Ontology to form a scalable and structured implementation path. At the product level, this understanding is concretely manifested through Foundry and AIP, while at the organizational level, it is embedded in the work of every engineer interfacing with clients.

This advantage is equally significant in the military and government systems. Palantir secured a $218 million contract from Space Force last quarter to build an integrated air and space combat system; the contract cap for the Maven Smart System project was raised by $795 million; and a long-term contract with the U.S. Army, worth up to $10 billion over 10 years, was signed to integrate 75 existing contracts. These long-term deep collaborations are the core embodiment of Palantir transforming AI tools into system platforms and then into operational units.

CEO Alexander Karp does not hide his judgment on Palantir's current position: "This is a Palantir without unnecessary layers, no parasitic organizations, and no ineffective sales; all components are creating value." He repeatedly emphasizes, Palantir is the only AI company that maintains consistency in product, organization, culture, and political stance.

Empowering frontline workers through AI is another clear line in Palantir's strategy. Palantir is advancing product deployment in factories, hospitals, shipyards, and other scenarios through the AI FDE (Forward Deployed Engineer) system; on the product side, systems such as the Warp Speed industrial operating system and MRP production line scheduling module are built based on AIP and have been deployed in multiple defense industrial bases A customer said: "What originally took a day for production line scheduling now only takes an hour."

Winning with a Structural Rather than Scale-Oriented Organizational Model

For Palantir's growth, what outsiders may first notice are the numbers—over 90% year-on-year growth in U.S. commercial revenue, quarterly revenue surpassing $1 billion, and record high TCV. However, what truly supports these explosive performances is a set of organizational models and cultural frameworks that are completely different from most software companies.

Palantir did not grow through a large sales system, but rather through continuously accumulating user trust based on product effectiveness and deployment methods; it did not establish scale by following traditional software industry experiences, but chose an "extremely non-mainstream" path from the very beginning, organizing resources entirely around value.

A typical example is the sales approach. CEO Alexander Karp clearly stated that Palantir's sales do not rely on a large team of 10,000 people, nor do they involve flying to meet CIOs every quarter, wining and dining clients, or creating internal promotion chains in a "status banquet-style" sales manner. He described traditional software sales as "selling a tool that is basically useless but once used, cannot be discarded," and then reorganizing processes around that tool, which Palantir firmly opposes.

Instead, Palantir adopts a "value-driven sales" approach, with product engineers on the front lines and actual results as the basis for trust. Palantir's greatest sales leverage comes from recommendations by existing customers, especially within U.S. enterprises where talent flows frequently; a user who previously introduced Foundry at one company will, upon moving to another company, make it a priority to "bring them in." This is the ultimate form of a "product-oriented organization": sales emerge from a network structure naturally extended from product effectiveness, rather than a sales funnel artificially constructed.

The direct effect of this model is a rapid increase in entry barriers. Customers will deploy a complete Ontology model and AIP architecture from day one, no longer engaging in a slow trial-and-error purchasing process, but quickly integrating around value objectives. This change is also reflected in customer feedback.

CRO Ryan Taylor stated that more and more customers are proactively asking during the initial stages of projects: "How do we deploy this across the entire enterprise?" rather than the previous approach of "let's first do a POC and see if it works."

Alexander Karp summarized this value delivery mechanism as: "We are downstream of our own unit economic model, sharing this model with customers." In other words, Palantir itself is an organization with a very strong unit economic return (such as a Rule of 40 of 94), and it "spills over" this capability to generate returns to customers through its products and organizational methods.

To this day, Palantir still:

• Has a very small sales team
• Maintain extremely low management levels
• Have the most heterogeneous and technically discerning talent structure
• Adopt a collaborative approach centered around "deployment engineers"
• Have a fully autonomous product stack, from Foundry to Ontology to AIP

Karp pointed out that many of Palantir's most critical business breakthroughs come from the United States, stating that we are currently in a "window period of American-style industrial and AI technology explosion," and "we possess the kind of organizational flexibility and commercial penetration paths that are not available in Europe."

At the same time, Palantir is actively conveying the empowerment mechanism of AI to blue-collar workers. He made a particularly noteworthy observation: in the wave of AI, the value created by many uneducated users using Palantir tools even exceeds that of college graduates. This is not emotional language, but a systematic understanding derived from product practice.

He also mentioned that Palantir is formulating a "Labor Cooperation" plan, intending to collaborate with labor unions in the United States to empower blue-collar users with more control and benefits from AI tools. This is also a tangible practice of Palantir's belief that "AI should enhance human agency."

On this basis, he reiterated the importance of the "unit economic model": "What we can provide to customers is not just functionality, but also a unit economic model structure similar to ours. Once this structure is established, customers can achieve growth and profit efficiency similar to the 93%, 94%, and 68% we demonstrated today within their industries."

This idea is also reflected in the customer lifecycle. Ryan Taylor added that what Palantir is seeing now is not customers slowly starting and then expanding linearly, but rather "rapid deployment from the beginning, followed by a quick promotion across the organization."

Karp also mentioned that Palantir does not reject increasing the sales team, but the premise is always "a product-centric organizational logic." He added, "We are not interested in building a sales team that relies on eloquence and misleading to close big deals without delivering actual value. Instead, we will continue to establish a structure that allows customers to entrust us with organizational restructuring based on trust." As he said, "This is trust brought by structure, not coercion brought by scale."

Regarding talent, Karp put forward a penetrating judgment of "de-credentialization." He repeatedly emphasized that at Palantir, background, school, and family are no longer standards for evaluating talent; the only credential is the value proven at Palantir.

The Trio of Cash Flow, Profit Margin, and Growth

Palantir's financial report for this quarter not only provided impressive results on the revenue side but also achieved breakthroughs in profit and cash flow, validating that its AI commercialization model can not only "scale" but also "make money." CFO David Glazer's statement after the financial report broke down the financial performance in great detail, confirming the true meaning of the "unit economic model" emphasized by CEO Karp: Only when the growth rate, profit margin, and cash flow all enter the fast lane does it mean that the company has truly stepped into the core area of the "AI structured dividend."

First, looking at the growth indicators, Q2 total revenue was USD 1.004 billion, a year-on-year increase of 48%, which not only exceeded market expectations but also grew by 14% compared to Q1. Compared to the same period last year, Palantir's quarterly growth rate increased by 21 percentage points, indicating that the growth curve is steepening quarter by quarter. The U.S. market continued to drive global growth, achieving revenue of USD 733 million in Q2, a year-on-year increase of 68% and a quarter-on-quarter increase of 17%. This growth rate is highly consistent with the global large model implementation pace and indicates that Palantir has shifted from an AI infrastructure role to an AI operating system role, becoming an indispensable part of customer businesses.

In terms of profit margin, Palantir also performed impressively. Adjusted operating profit was USD 464 million, with an operating profit margin of 46%, exceeding the upper limit of the guidance provided last quarter by nearly 300 basis points. This result also drove the Rule of 40 score from 83% in Q1 to 94% in Q2, setting a historical record.

Looking further at the net profit level, GAAP net profit was USD 327 million, with a net profit margin of 33%; adjusted earnings per share were USD 0.16, higher than market expectations. It is worth mentioning that Palantir is still conducting equity incentives at a high ratio, but even so, its GAAP net profit performance has shown sustainability, which is an important sign that Palantir is one of the few that has truly achieved AI commercialization profitability.

In terms of cash flow, Palantir's operating cash flow for this quarter was USD 539 million, with adjusted free cash flow reaching USD 569 million, and a free cash flow rate as high as 57%. Such cash recovery capability is extremely rare among global SaaS companies, providing Palantir with ample strategic flexibility.

Regarding customer structure, Palantir's total number of customers reached 849, a year-on-year increase of 43% and a quarter-on-quarter increase of 10%. Among them, the number of U.S. commercial customers grew by 64%, indicating that the implementation of AI applications is accelerating out of the technical personnel circle and gradually becoming a core tool at the operational, management, and even strategic levels.

At the same time, the value of Palantir's major customers continues to grow. The average revenue contributed by the top 20 customers over the past 12 months reached USD 75 million, a year-on-year increase of 30%. This not only reflects customer stickiness but also indicates that the product is deeply integrated into core systems in a platform-like form, becoming part of the digital infrastructure of enterprises In addition, some operational metrics released by Palantir are worth noting:

• Net revenue retention rate is 128%, an increase of 400 basis points from the previous quarter;
• Total remaining contract value reached $7.1 billion, a year-on-year increase of 65% and a quarter-on-quarter increase of 20%;
• Remaining performance obligations (RPO) are $2.4 billion, a year-on-year increase of 77% and a quarter-on-quarter increase of 27%;

These data indicate: Not only has the willingness of existing customers to pay increased, but new customers have also entered the renewal and expansion cycle, extending the overall customer lifecycle.

Against the backdrop of AI Palantir being generally "overvalued but profitability is far off," Palantir is charting a very rare path: it not only achieves over 45% revenue growth but also realizes over 45% profit margin and over 50% free cash flow rate.

The synergistic breakthroughs in these three metrics constitute a genuinely operable "AI operating system" template, providing solid underlying support for its high valuation and recognition in the capital market.

Risk Warning and Disclaimer

The market carries risks, and investment should be cautious. This article does not constitute personal investment advice and does not take into account the specific investment objectives, financial situation, or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article are suitable for their specific circumstances. Investment based on this is at one's own risk