Palantir Q2 2026 Earnings: Growth, Margins and Risk

Palantir Technologies delivered the kind of quarterly report that forces both supporters and skeptics to update their assumptions. Revenue for the three months ended June 30, 2026 rose 93% from a year earlier to $1.935 billion. U.S. commercial revenue increased 149% to $764 million, U.S. government revenue climbed 90% to $809 million, and management raised its full-year revenue outlook to between $8.150 billion and $8.158 billion. The midpoint of that range is almost $500 million above the guidance Palantir issued only three months earlier.

The immediate market response was equally dramatic. Palantir shares, which trade on Nasdaq under the ticker PLTR, rose to $159.28 by 11:53 a.m. Eastern time on August 4, 2026, up $33.63, or approximately 26.8%, from the previous close. The stock traded as high as $160.03 during the session. The rally occurred during a broadly positive day for U.S. equities, but Palantir’s gain was far larger than the advance in the major indexes and reflected the scale of the earnings surprise, the upward guidance revision, and evidence that customer spending was expanding across both commercial and government markets.

The report also produced a headline statistic that traveled quickly across financial media: a “Rule of 40” score of 155. Palantir calculated that figure by adding its 93% year-over-year revenue growth rate to its 62% adjusted operating margin. A score above 40 is commonly used as a rough benchmark for balancing software growth and profitability, so 155 is extraordinary. Yet the figure requires context. It uses an adjusted margin that excludes stock-based compensation and related payroll taxes; the company’s GAAP operating margin was lower, though still exceptional, at 47%. The Rule of 40 is also a snapshot rather than a valuation model, and different investors calculate it with different profitability measures.

The central conclusion is therefore more nuanced than either “Palantir has proved every critic wrong” or “the stock is too expensive to matter.” The quarter provides strong evidence that Palantir has become one of the clearest public-market examples of enterprise AI spending converting into recognized revenue, operating profit, and cash flow. It also shows that the business is no longer accurately described as a government contractor with a commercial experiment attached. At the same time, the share price implies that investors expect years of unusually high growth, strong margins, and successful expansion. Palantir can be executing extremely well while its stock still carries substantial valuation risk.

Last updated: August 4, 2026, 12:00 p.m. ET. Market prices are time-specific and may have changed since publication.

Key Takeaways

• Revenue nearly doubled: Palantir reported second-quarter revenue of $1.935 billion, up 93% year over year and approximately 19% from the first quarter.
• Both U.S. engines accelerated: U.S. commercial revenue rose 149% to $764 million, while U.S. government revenue increased 90% to $809 million.
• The full-year outlook rose sharply: Management raised 2026 revenue guidance to $8.150 billion to $8.158 billion, about $498 million above the midpoint of its prior range.
• Margins were exceptional but definitions matter: GAAP operating margin was 47%; adjusted operating margin was 62% after excluding stock-based compensation and related payroll taxes.
• Contract activity was broad and large: Palantir closed 220 deals valued at $1 million or more, including 98 valued at $5 million or more and 73 valued at $10 million or more.
• The contract categories overlap: The 73 deals of at least $10 million are included within the 98 deals of at least $5 million and the 220 deals of at least $1 million. They are not three separate pools.
• U.S. commercial remaining deal value reached $6.238 billion: That was up 124% year over year, but remaining deal value is not the same as recognized revenue or non-cancelable backlog.
• The Rule of 40 score was 155: Palantir’s calculation adds 93% revenue growth to a 62% adjusted operating margin. Using GAAP operating margin would still produce an unusually high score of roughly 140.
• The market rewarded the report: PLTR rose approximately 26.8% to $159.28 by late morning on August 4, lifting the company’s market capitalization to roughly $409 billion.
• Valuation remains the principal investor risk: At that market value, Palantir traded at roughly 50 times the midpoint of its 2026 revenue guidance, leaving little room for an ordinary slowdown.

What Palantir Reported—and What the Headline Numbers Mean

Palantir’s second-quarter release is striking because nearly every major performance category moved in the same direction. Revenue growth accelerated, U.S. commercial revenue accelerated, U.S. government revenue accelerated, operating margins expanded, cash generation increased, contract value rose, and management raised guidance. Strong software quarters often involve a trade-off: a company may grow quickly while spending aggressively, or it may expand margins as growth slows. Palantir produced rapid growth and unusually high profitability at the same time.

Total revenue increased from $1.004 billion in the second quarter of 2025 to $1.935 billion in the second quarter of 2026. The company also grew approximately 18.6% from the first quarter of 2026, when revenue was about $1.633 billion. Sequential growth of that magnitude matters because it shows the result was not only a comparison against a smaller prior-year base. Palantir added more than $300 million of quarterly revenue in three months.

The U.S. business accounted for approximately 81% of total second-quarter revenue. U.S. revenue reached $1.573 billion, up 115% year over year and 23% sequentially. Within that total, the commercial operation generated $764 million and the government operation generated $809 million. The two segments are now close enough in size that the company’s identity is materially different from its pre-AIP years, when government work dominated the investment narrative.

International revenue can be estimated by subtracting U.S. revenue from the consolidated total. On that basis, revenue outside the United States was approximately $362 million in the second quarter. That was still growth from the prior year, but the international operation was far slower than the U.S. business and increased only modestly from the first quarter. The geographic split is one of the most important restraints on the bullish interpretation: Palantir’s acceleration is extraordinary, but it is concentrated.

Profitability was equally notable. Gross profit was $1.639 billion, implying a GAAP gross margin of approximately 84.7%. GAAP income from operations reached $912 million, compared with $269 million in the prior-year quarter. Adjusted income from operations was $1.194 billion. The difference between those figures largely reflects exclusions for stock-based compensation and related employer payroll taxes.

Cash generation tracked operating performance. Cash from operations was $1.216 billion, a margin of approximately 62.8%. Adjusted free cash flow was $1.220 billion, or about 63.1% of revenue. The closeness of those two figures reflects Palantir’s low capital expenditure requirements during the quarter. The company is benefiting from the software economics described by many Palantir supporters: hyperscalers and customers fund much of the physical computing infrastructure, while Palantir sells a high-value software and operational layer.

Yet even the cash-flow numbers need precise wording. Adjusted free cash flow is a non-GAAP measure defined by Palantir. It excludes or adjusts certain items and should not be treated as interchangeable with GAAP net income or as a complete measure of changes in the balance sheet. It is useful, especially alongside cash from operations, but it does not eliminate the need to examine compensation, working capital, taxes, contract terms, and dilution.

Why the Market Reaction Was So Large

Palantir entered the report with a difficult setup. The shares had fallen sharply from their late-2025 record and were down roughly 30% for 2026 before the announcement, according to contemporary market reporting. Investors were concerned about the broader software selloff, the possibility that cheaper and more capable foundation models could weaken application-layer vendors, slowing international growth, political and reputational scrutiny, and a valuation that remained high despite the decline.

Expectations for the business were already elevated. Analysts surveyed by LSEG expected approximately $1.80 billion of second-quarter revenue and adjusted earnings of about $0.35 per share. Palantir delivered $1.935 billion and adjusted diluted earnings of $0.41 per share. The revenue beat was more than $130 million, while management’s third-quarter guidance of $2.160 billion to $2.164 billion was well above the roughly $2.0 billion consensus cited by Reuters and MarketWatch.

The full-year raise was even more consequential. In May, Palantir had projected 2026 revenue of $7.650 billion to $7.662 billion. In August, it increased that range to $8.150 billion to $8.158 billion. The midpoint rose by $498 million, or approximately 6.5%, after only one additional quarter of reported results. Guidance raises of that size are uncommon for a company already operating at multi-billion-dollar scale.

The market was therefore responding to more than an earnings beat. The new outlook changed the expected revenue base for the second half, strengthened the argument that U.S. commercial adoption was moving from pilots into larger deployments, and reduced the immediate credibility of the claim that open models or frontier AI labs were displacing Palantir. It also suggested that the government operation was participating in the acceleration rather than merely providing a stable foundation.

By late morning on August 4, the shares had risen approximately 26.8%. Trading volume exceeded 92 million shares by 11:53 a.m. Eastern time. The move lifted Palantir’s market capitalization to about $409 billion. MarketWatch described it as the stock’s largest one-day gain since February 2024. The rally did not erase all of the year’s earlier decline, but it restored a substantial portion of the lost value in a single session.

The broader market backdrop also mattered. U.S. stocks were higher, Treasury yields were lower, oil prices were declining, and strong corporate earnings supported risk appetite. A favorable tape can amplify a positive earnings reaction. Still, Palantir’s gain was far beyond the indexes, making it reasonable to view the report and guidance as the dominant company-specific drivers.

Correcting the Most Important Errors and Overstatements

The analysis that accompanied the earnings video captured the report’s direction correctly, but several points require correction or tighter framing.

U.S. government revenue was $809 million, not $109 million

One supplied bullet listed U.S. government revenue at $109 million. Palantir’s SEC-filed release reported $809 million, up 90% year over year and 18% sequentially. The transcript itself used $809 million in another passage, indicating that the $109 million figure was almost certainly a transcription or editorial error. It is a material distinction: U.S. government was Palantir’s largest individual revenue category in the quarter.

The 62% operating margin was adjusted, not GAAP

Palantir’s adjusted operating margin was 62%. Its GAAP operating margin was 47%. Both are strong, but they answer different questions. The adjusted measure excludes stock-based compensation and employer payroll taxes related to that compensation. The GAAP figure includes those costs. A balanced analysis should present both rather than describing 62% as the company’s unqualified operating margin.

The full-year guide was above $8.15 billion, not simply “more than $8.2 billion”

The official 2026 revenue range was $8.150 billion to $8.158 billion. Rounding that to “past $8.2 billion” overstates the disclosed range. It is fair to say the guide moved above $8.15 billion or to describe the midpoint as approximately $8.154 billion. The difference may seem small relative to the company’s scale, but earnings reporting should preserve management’s actual range.

The deal counts are nested

Palantir closed 220 deals of at least $1 million, 98 deals of at least $5 million, and 73 deals of at least $10 million. The larger thresholds are subsets of the smaller threshold. It would be wrong to add the categories together. It is also too strong to say that one-third of $1 million contracts “expanded” into eight-figure commitments. The company disclosed the number of deals closed at each threshold during the quarter; it did not state that the 73 eight-figure deals began as smaller contracts and expanded within the same period.

Remaining deal value is not recognized revenue

U.S. commercial remaining deal value reached $6.238 billion, up 124% year over year. That is a useful indicator of contracted opportunity and customer commitment, but Palantir explicitly states that remaining deal value may assume the exercise of customer options and no contract termination. Many contracts can be terminated for convenience. Remaining deal value is therefore broader and less certain than remaining performance obligations under accounting rules.

The Rule of 40 threshold is not an “elite bar” in a formal sense

Forty is a widely used benchmark, not an official certification and not a universal definition of elite performance. The metric also changes depending on whether analysts use EBITDA margin, adjusted operating margin, GAAP operating margin, or free-cash-flow margin. Palantir’s score is exceptional under any reasonable formulation for the quarter, but rhetoric about “quadrupling the goal” can make a heuristic sound more scientific than it is.

A forward earnings multiple near 80 or 90 depends on the estimate period

The stock’s price-to-earnings ratio changes depending on whether the denominator is trailing GAAP earnings, current-year consensus earnings, next-twelve-month earnings, or a later forecast. At $159.28, market data showed a trailing P/E near 179. A forward multiple closer to 90 may be defensible using a later earnings estimate, but it must be labeled with the relevant forecast period. The valuation is expensive under almost any conventional measure; imprecise denominator choices are unnecessary.

The U.S. Commercial Business Is the Main Strategic Breakthrough

Palantir’s U.S. commercial revenue increased from roughly $307 million in the second quarter of 2025 to $764 million in the second quarter of 2026. It also rose 28% from the first quarter. This is the result that most directly supports the claim that Palantir has crossed from an interesting government-focused software company into a broader enterprise platform.

The significance is not merely that commercial revenue grew faster than government revenue. Commercial buyers face different procurement processes, budget cycles, competition, and return-on-investment requirements. They can postpone projects, build internally, choose a cloud vendor’s native tools, select a specialist application, or buy from established enterprise software providers. Winning and expanding commercial work is therefore evidence that customers see operational value beyond Palantir’s defense and intelligence reputation.

The company’s commercial strategy also changes the addressable market. Government opportunities can be large and durable, but they are constrained by appropriations, procurement procedures, security requirements, political priorities, and a limited number of agencies. Enterprise operations span manufacturing, healthcare, energy, insurance, aviation, logistics, retail, telecommunications, financial services, and other industries. Palantir does not disclose a simple product-by-industry revenue table, but its platform is designed to work across data integration, operational decision-making, planning, optimization, and AI-enabled workflows.

Commercial acceleration is particularly important because Palantir’s software is often expensive and implementation-intensive. The strongest evidence of product value is not a demonstration or pilot but an expansion in recognized revenue and contract size. The second-quarter figures suggest that at least some customers are moving beyond experimentation. U.S. commercial total contract value reached a record $2.132 billion, up 153% from a year earlier. Remaining deal value increased 27% in a single quarter.

There is also evidence of expansion within large accounts. At the end of 2025, average trailing-twelve-month revenue from Palantir’s top 20 customers was $93.9 million, up 45% year over year. By the first quarter of 2026, the average had reached $108 million, up 55%. Although the second-quarter release did not foreground a new top-20 figure, the deal and revenue data are consistent with deeper deployment among large customers.

That pattern matters because Palantir’s model is not a conventional low-touch subscription sale. The company often begins with a difficult operational problem, integrates data from multiple systems, builds an ontology that represents the customer’s organization, and embeds applications into decisions and workflows. Once the platform becomes part of production operations, expansion can generate high incremental margins. The same characteristic that makes initial deployment complicated can make a successful deployment difficult to replace.

How AIP Bootcamps Changed Palantir’s Sales Motion

Palantir launched its Artificial Intelligence Platform, or AIP, in 2023 and used short, intensive “bootcamps” to reduce the time between customer interest and a functioning use case. The company’s official bootcamp material describes a process in which customer teams work with Palantir engineers and their own data to build applications in hours or days rather than beginning with a long theoretical consulting project.

The approach addresses a persistent problem in enterprise technology. Senior executives may approve an AI initiative, but enthusiasm can weaken during months of data preparation, procurement, security review, integration, model testing, and organizational negotiation. A bootcamp attempts to create visible operational output before the project loses momentum. The output may be a production-planning application, a maintenance workflow, a supply-chain model, an investigation tool, an agent that works within controlled permissions, or another domain-specific system.

The strategic advantage is speed to proof. A customer does not have to accept a broad claim that generative AI will transform its business. It can test whether a Palantir-powered application improves a measurable process using relevant data. If the application works, the commercial conversation can shift from speculative innovation spending to operational deployment.

The bootcamp is not proof that every pilot becomes a large contract. Palantir does not disclose a simple conversion rate that allows outsiders to trace bootcamp attendance to recognized revenue. Some projects may stall, some may remain narrow, and some organizations may decide that the required cost or complexity is too high. Morgan Stanley analyst Sanjit Singh reported customer hesitation around Palantir’s pricing and total cost among some customers spending more than $5 million, according to Investor’s Business Daily.

Even with those limitations, the revenue trend gives the bootcamp strategy more credibility than it had in 2023. U.S. commercial growth has accelerated across multiple quarters, and the revenue base is now large. The company is not merely reporting a high number of demonstrations; it is reporting recognized revenue, operating income, cash flow, and larger contracts.

Palantir’s “Ontology” Is the Core of the Business Argument

Palantir’s investment case is sometimes reduced to “AI software,” but that label is too broad to explain why the company believes it is differentiated. The more precise concept is the ontology: a software representation of an organization’s entities, relationships, permissions, actions, processes, and operational logic.

In a manufacturer, an ontology might connect machines, production lines, parts, suppliers, orders, quality events, maintenance schedules, employees, facilities, and financial consequences. In a defense context, it might connect sensors, units, locations, missions, logistics, intelligence reports, permissions, and possible actions. The purpose is not only to store data. It is to create a shared operational model that applications, analysts, decision-makers, and AI systems can use.

This distinction matters because a language model can generate text without understanding which employee is authorized to see a record, whether a proposed action violates a policy, how an operational object is defined, or which real-world system must be updated. Palantir’s argument is that enterprise AI becomes valuable when model output is connected to governed data and permitted action. The ontology is the layer that turns an abstract model response into something an organization can inspect, authorize, and execute.

The company’s four principal platforms support that architecture. Gotham is associated primarily with government, defense, intelligence, and mission operations. Foundry provides data integration, analytics, application development, and operational workflows. Apollo manages software deployment and updates across cloud, on-premises, edge, and other environments. AIP connects large language models and other AI systems to the data, controls, applications, and workflows in Gotham and Foundry.

Palantir’s strongest claim is therefore not that it has built the best general-purpose model. It is that model providers can change while the customer’s governed operational layer remains. Customers can use commercial models, open-weight models, self-hosted models, specialized models, or internally developed systems without rebuilding the entire data and workflow architecture each time.

This is why falling model prices do not automatically weaken Palantir. Cheaper inference and more capable open models may increase the number of tasks that enterprises want to automate. If Palantir remains the controlled interface between those models and real operations, lower model costs could expand usage. The risk is that cloud platforms, enterprise software vendors, data companies, and internal engineering teams build sufficiently capable alternatives that reduce the need for Palantir’s integrated stack.

The Open-Model Debate Cuts Both Ways

Before the earnings report, one bearish narrative held that rapid improvement in models from OpenAI, Anthropic, Google, Meta, and open-weight communities could compress the value of application and orchestration software. If models become capable of writing code, connecting to tools, interpreting data, and building agents, customers may need fewer specialized platforms.

Palantir’s response is that model intelligence is only one component of production AI. Organizations still need data lineage, identity and permission controls, evaluation, monitoring, human approval, auditability, deployment, workflow integration, and a representation of how the business operates. In regulated, classified, safety-sensitive, or operationally complex environments, those requirements can be more important than the choice of model.

The company has reinforced that argument through its work with Nvidia. In 2026, Palantir and Nvidia announced a sovereign AI reference architecture and an engine for deploying Nvidia Nemotron open models in controlled environments. The offering is designed for customers that want to own or control infrastructure, models, data, and operations rather than sending sensitive information to an external model provider.

This is commercially relevant because “sovereign AI” has several meanings. A national government may want domestic control over data and computing. A defense agency may require an air-gapped deployment. A regulated enterprise may need model behavior, permissions, and data location to remain under its authority. A corporation may simply want to prevent proprietary operational knowledge from becoming training material for a third-party system.

Palantir’s positioning is strongest where those concerns are acute. A customer operating a factory, hospital network, energy grid, military system, or regulated financial process may value control more than access to the newest consumer chatbot feature. The company can integrate models without asking the customer to move its operating logic into a model vendor’s environment.

The counterargument is that sovereign AI may become a crowded category. Nvidia sells across the ecosystem. Cloud providers offer private and regulated deployments. Data-platform companies are adding governance and agent capabilities. Open-source orchestration frameworks improve quickly. Large systems integrators can assemble customized stacks. Palantir must continue proving that its integrated approach delivers better outcomes quickly enough to justify premium pricing.

The Government Business Is Accelerating, Not Merely Stabilizing

Palantir’s U.S. government revenue reached $809 million in the second quarter, up 90% from approximately $426 million a year earlier and 18% from the first quarter. That result challenges the idea that government work is a mature, slow-growing base while commercial work supplies all of the upside.

Several structural forces support demand. Modern defense systems generate large volumes of sensor, logistics, intelligence, maintenance, targeting, and operational data. Governments are investing in software that can integrate those inputs, shorten decision cycles, coordinate forces, and deploy AI in controlled environments. Geopolitical conflict and higher defense priorities have increased the urgency of those capabilities.

Palantir’s Gotham platform has long served intelligence and defense customers, while newer offerings extend into battlefield command, logistics, autonomy, industrial planning, and model deployment. Reuters has reported that Palantir and Anduril are working together on technology connected to the U.S. Golden Dome missile-defense initiative. Palantir’s Maven work has also become an important part of the U.S. military’s AI-enabled command and targeting infrastructure.

Government revenue can be attractive because contracts may be large, multi-year, and difficult to displace once software is embedded in critical missions. Security clearances, accreditation, classified deployment capability, procurement experience, and institutional trust create barriers to entry. Palantir has spent more than two decades building those capabilities.

The risks are equally specific. Government contracts may be protested, delayed, reduced, re-competed, or terminated. Budget priorities can change. Revenue recognition may depend on options and task orders. Political controversy can affect procurement. The use of software in military operations, immigration enforcement, law enforcement, surveillance, and public-sector data integration creates legal and reputational scrutiny that ordinary enterprise vendors may not face.

International government work also exposes a tension in Palantir’s sovereign-AI strategy. The company argues that customers should control their data and systems, yet some European governments are wary of relying on an American vendor for sensitive national infrastructure. Reuters reported that France’s domestic intelligence agency planned to replace Palantir tools with technology from French company ChapsVision. Palantir has also faced a legal challenge connected to a London police contract. Sovereignty can support Palantir’s product architecture while simultaneously encouraging governments to prefer domestic suppliers.

Two Demand Engines—but Not Yet a Balanced Global Business

The strongest strategic interpretation of the quarter is that Palantir now has two U.S. demand engines with different customer economics. The commercial business responds to enterprise AI adoption, operational efficiency, competition, labor constraints, and corporate technology budgets. The government business responds to defense, intelligence, logistics, security, geopolitical priorities, and public procurement.

That diversification can reduce dependence on a single cycle. A slowdown in enterprise technology spending does not automatically eliminate defense demand. A pause in one government program does not necessarily stop commercial expansion. Both sides use the same core capabilities—data integration, ontology, deployment, security, AI governance, and operational applications—allowing product investment to support multiple markets.

However, the company is not yet balanced geographically. Approximately 81% of second-quarter revenue came from the United States. The U.S. operation grew 115%, while the estimated international operation grew much more slowly. Palantir’s growth rate would look dramatically different without the U.S. surge.

This concentration may persist because the United States has unusual conditions that favor Palantir: large defense spending, a close relationship between commercial technology and national-security institutions, deeper venture and AI ecosystems, faster enterprise procurement in some sectors, and management’s explicit focus on the domestic market. Europe has stricter data rules, fragmented procurement, stronger preferences for local technology in sensitive systems, and political concerns about U.S. dependence.

The international gap is not fatal to the thesis. A company can create enormous value by dominating a large home market. But it affects the duration of growth. To sustain very high rates after the U.S. base becomes larger, Palantir may need stronger adoption in Europe, Asia-Pacific, the Middle East, and other markets—or it must continue expanding deeply enough within U.S. customers that international weakness remains secondary.

Contract Momentum: Why the Deal Data Matters

Revenue is recognized under accounting rules as Palantir delivers products and services. Contract metrics attempt to show future opportunity before all of that value becomes revenue. In the second quarter, Palantir reported total contract value of $3.373 billion, up 49% year over year. U.S. commercial total contract value was $2.132 billion, up 153%. U.S. commercial remaining deal value reached $6.238 billion, up 124% from a year earlier and 27% from the first quarter.

The company also closed 220 deals valued at $1 million or more. Ninety-eight were valued at $5 million or more, and 73 were valued at $10 million or more. The concentration of large deals supports the argument that customers are moving beyond narrow experiments. An eight-figure commitment usually requires broader executive sponsorship, more workflows, more users, or a longer contract duration than a small departmental pilot.

Still, contract metrics can be misunderstood. Palantir defines total contract value as the potential lifetime value of contracts at execution. It generally assumes that customers exercise available options and do not terminate contracts. Remaining deal value is the total value left on contracts, but it is not identical to remaining performance obligations, which represent contracted revenue that is non-cancelable under accounting standards. Many Palantir contracts contain termination-for-convenience provisions.

This means the deal data should be interpreted as evidence of sales momentum and potential future revenue, not as cash in the bank. A $10 million deal may be recognized over multiple years. An option may never be exercised. A customer may terminate a contract. Implementation timing may shift. Currency, scope, performance conditions, and customer financial health can affect recognition.

The categories also reveal a useful distribution. About one-third of the 220 deals that cleared $1 million also cleared $10 million. That is a large proportion, but it does not prove that smaller deals expanded into larger ones during the quarter. It shows that Palantir was closing many substantial agreements. To assess expansion more directly, investors should combine deal counts with net dollar retention, revenue from top customers, customer count, and recognized segment growth.

The contract disclosures are especially important because Palantir’s revenue recognition can lag the commercial event that creates economic commitment. When customers sign larger and longer agreements, revenue may emerge across future quarters. That can improve visibility even when the exact timing is uncertain. The 27% sequential increase in U.S. commercial remaining deal value suggests that new commitments outpaced revenue recognition during the quarter.

Margins: The 62% Figure Is Real, but It Is Not the Whole Story

Palantir reported $1.194 billion of adjusted income from operations on $1.935 billion of revenue, producing an adjusted operating margin of approximately 61.7%, rounded by the company to 62%. That is unusually high for a business growing 93% year over year. It indicates that a large share of incremental revenue is falling through to profit after the costs included in Palantir’s adjusted measure.

GAAP operating income was $912 million, representing a 47.1% margin. The gap between GAAP and adjusted operating income was approximately $282 million. Stock-based compensation alone was $265 million, or about 13.7% of revenue, up from $160 million in the prior-year quarter. Related employer payroll taxes accounted for much of the remaining difference.

Stock-based compensation is non-cash in the period it is recorded, but it is not economically irrelevant. Equity awards can dilute existing shareholders or require future repurchases to offset dilution. They are also part of employee compensation. Excluding the expense helps analysts understand cash-oriented operating leverage, but including it helps assess the cost of labor to owners.

Palantir’s GAAP result is strong enough that the company does not need to hide behind adjusted accounting. Gross margin was approximately 84.7%, and GAAP operating expenses totaled about $727 million. Sales and marketing expense increased to $340 million from $244 million, research and development rose to $193 million from $135 million, and general and administrative expense increased to $195 million from $163 million. Each major expense category grew much more slowly than revenue.

This is the operating-leverage mechanism behind the quarter. Palantir’s revenue nearly doubled while total operating expenses increased roughly one-third. Gross profit increased by more than $827 million year over year, while operating expenses increased by about $185 million. The difference drove a more than threefold increase in GAAP operating income.

Net income attributable to common stockholders was $1.062 billion, or $0.41 per diluted share. The net-income margin of approximately 54.9% exceeded the operating margin because Palantir earned $77.5 million of interest income and $91.8 million of other income, while recording only $15.4 million of income-tax expense. Those items are legitimate components of GAAP profit, but they mean the net margin is not a pure representation of software operations.

The tax line deserves attention in future periods. Palantir’s effective tax rate in the quarter was low relative to the U.S. statutory rate, reflecting its tax attributes and accounting position. As profitability persists and deferred tax assets are used or remeasured, cash taxes and reported tax expense can change. Investors should not assume the second-quarter relationship between pre-tax and after-tax income will remain constant indefinitely.

Cash Flow and Balance-Sheet Strength

Cash from operations was $1.216 billion in the second quarter, up sharply from the prior year. For the first six months of 2026, Palantir generated $2.115 billion of operating cash flow, compared with $850 million in the first half of 2025. The increase reflected higher net income, stock-based compensation, and growth in contract liabilities, partly offset by a $434 million increase in accounts receivable.

Accounts receivable reached $1.485 billion at June 30, up from $1.042 billion at the end of 2025. A growing receivables balance is not surprising when revenue accelerates, but it is worth monitoring. Cash collection can trail recognition, especially in large government and enterprise contracts. The first-half cash-flow statement shows that the increase in receivables absorbed more than $400 million of cash.

Deferred revenue and customer deposits increased as well. Current deferred revenue rose to $579 million from $409 million at year-end, while current customer deposits increased to $452 million from $357 million. Those balances can support cash flow because customers pay before all related revenue is recognized.

Palantir ended the quarter with $2.030 billion of cash and cash equivalents and $7.379 billion of marketable securities. The company described cash, cash equivalents, and short-term U.S. Treasury securities as approximately $9.2 billion. The balance sheet contained $11.679 billion of assets and $1.794 billion of total liabilities.

The large liquid-asset position gives Palantir several strategic advantages. It can invest through economic cycles, fund product development, recruit employees, support classified and sovereign deployments, make acquisitions or investments, and avoid financing pressure. Interest income also contributes meaningfully to reported profit.

At the same time, the balance sheet is small relative to the equity valuation. Roughly $9 billion of cash and securities is valuable, but it does not transform a $409 billion market capitalization into a conventionally valued software company. The investment thesis still depends overwhelmingly on future operating performance rather than existing financial assets.

What a Rule of 40 Score of 155 Actually Tells Investors

The Rule of 40 is a shorthand used to compare growth and profitability in software. The basic formula adds a company’s revenue growth rate to a profitability margin. A company growing 30% with a 15% margin scores 45. A company growing 10% with a 35% margin also scores 45. Both clear the threshold, although the businesses may have very different risk profiles and valuation prospects.

Palantir’s calculation was straightforward: 93% year-over-year revenue growth plus a 62% adjusted operating margin equals 155. Using the exact reported figures produces approximately 154.7. Using GAAP operating margin instead would produce about 140. Using adjusted free-cash-flow margin would produce roughly 156. The result is exceptional under each version.

Why does the number matter? High growth often requires heavy sales spending, implementation costs, research investment, cloud expense, and hiring. High margins often occur after a company matures and growth slows. A very high Rule of 40 score signals that a company is avoiding that usual trade-off, at least during the measured period.

Palantir’s score also supports the idea that its software sits above the most capital-intensive layer of AI. Microsoft, Amazon, Alphabet, Meta, Nvidia customers, data-center operators, and governments are spending heavily on chips, networks, power, memory, and facilities. Palantir uses that infrastructure but does not carry the same capital-expenditure burden. Its second-quarter purchases of property and equipment were modest relative to revenue.

However, the metric has four major limitations.

First, there is no single accepted profitability measure. Some investors use free cash flow, some use EBITDA, some use adjusted operating income, and some use GAAP operating income. Exclusions can materially change the result.

Second, one quarter can be unusually favorable. Revenue growth can benefit from contract timing, easy comparisons, large deals, or concentrated demand. Margins can benefit from expense timing and non-operating conditions. A durable Rule of 40 profile across years is more meaningful than a single peak.

Third, the formula values one point of growth the same as one point of margin, even though growth and margin have different implications. Bessemer Venture Partners has proposed a “Rule of X” framework that weights growth more heavily for some companies. Other analysts argue that the Rule of 40 hides differences in quality, retention, gross margin, capital intensity, and dilution.

Fourth, the Rule of 40 does not determine what a stock is worth. A company can post an extraordinary score and still produce disappointing shareholder returns if the market price already assumes even better results. The metric describes operating performance; valuation converts that performance into an investment outcome.

Historical context is also useful. Palantir’s 155 is among the strongest current large-cap software readings, but it is not necessarily the highest score in software history. Pandemic-era companies such as Zoom produced extreme combinations of growth and cash flow, and other software businesses have occasionally posted scores above 100. The appropriate conclusion is not that Palantir has broken financial mathematics. It is that the company delivered an unusually strong combination at a much larger revenue base than many high-growth software companies achieve.

Guidance: The Raise Changes the Second-Half Math

Palantir’s 2026 revenue guidance moved from $7.650 billion to $7.662 billion after the first quarter to $8.150 billion to $8.158 billion after the second. The midpoint increased from $7.656 billion to $8.154 billion. Because first-half revenue was $3.568 billion, the new outlook implies approximately $4.586 billion of revenue in the second half.

Management guided third-quarter revenue to $2.160 billion to $2.164 billion. At the midpoint, that would represent approximately 11.7% sequential growth from the second quarter. Subtracting first-half revenue and the third-quarter midpoint from full-year guidance implies about $2.424 billion of fourth-quarter revenue. That would represent another sequential increase of roughly 12.1%.

These are calculations based on the midpoint, not formal fourth-quarter guidance. Actual quarterly distribution could differ. Still, the math shows why the market viewed the full-year raise as evidence that momentum was not expected to fade immediately. Management’s outlook assumes continued growth through the rest of the year.

Palantir also raised U.S. commercial revenue guidance to more than $3.424 billion, representing growth of at least 134%. The prior outlook was more than $3.224 billion. This $200 million increase came after the segment generated $764 million in the second quarter and supports the view that commercial expansion is continuing into production.

Adjusted income from operations is now expected to reach $4.889 billion to $4.897 billion for the year. At the midpoint, that would equal about 60% of guided revenue. Adjusted free cash flow is expected to be between $4.5 billion and $4.7 billion, implying a midpoint margin of roughly 56%.

Management continued to expect GAAP operating income and GAAP net income in every quarter of 2026. That statement matters because Palantir historically relied more heavily on adjusted measures and reported losses during earlier stages of its public-company life. Sustained GAAP profitability improves financial quality, index eligibility, and the credibility of operating leverage.

The Bull Case After Q2 2026

The strongest bullish interpretation begins with the quality of the operating result. Palantir is not asking investors to accept a distant promise that AI revenue will eventually appear. It reported almost $2 billion of quarterly revenue, more than $900 million of GAAP operating income, more than $1 billion of GAAP net income attributable to common stockholders, and more than $1.2 billion of operating cash flow.

The second element is acceleration at scale. Year-over-year revenue growth rose to 93% even though the comparison base exceeded $1 billion. U.S. commercial revenue grew 149% on a base of more than $300 million. U.S. government revenue grew 90% on a base above $400 million. The company also added more than $300 million of revenue sequentially.

The third element is dual-market strength. Many AI companies are exposed primarily to model training, consumer subscriptions, advertising, cloud infrastructure, or one category of enterprise spending. Palantir sells into commercial operations and national-security systems. The two markets have different budgets and demand drivers but use related technology.

The fourth element is model neutrality. Palantir does not need one frontier lab to remain permanently dominant. Its platform can connect commercial, open-weight, self-hosted, and internally developed models to governed data and workflows. If model intelligence becomes abundant, the scarce assets may shift toward proprietary data, operational context, permission systems, evaluation, and deployment.

The fifth element is financial leverage. Palantir’s gross margin and low capital intensity allow revenue growth to produce operating income and cash. The company does not need to finance the full semiconductor and data-center buildout that supports AI. Its balance sheet provides additional flexibility.

The sixth element is evidence of larger deployments. The contract thresholds, U.S. commercial TCV, remaining deal value, and top-customer expansion suggest that at least some customers are standardizing around the platform. If Palantir becomes part of an organization’s operating system, replacement becomes expensive and risky.

The seventh element is management’s raised outlook. Palantir did not merely beat the quarter and maintain guidance. It increased full-year revenue expectations by nearly half a billion dollars and projected third-quarter revenue well above the prior consensus. That suggests visibility beyond a single period.

The Skeptical Case Has Shifted, Not Disappeared

The strongest skeptical argument is no longer that Palantir lacks a real business. The second-quarter numbers make that claim difficult to sustain. The better skeptical case is that the stock price discounts a very long period of exceptional performance and that several components of the current growth rate may normalize.

First, U.S. concentration is extreme. About 81% of revenue came from the United States, and the estimated international business grew far more slowly. A company valued as a global AI platform eventually needs either broader geographic adoption or deeper domestic penetration than ordinary enterprise software vendors achieve.

Second, commercial pricing may become a constraint. Palantir often addresses high-value problems, which can justify large contracts. But expensive deployments invite scrutiny, competitors, internal development, and pressure to prove measurable savings. Customer references reported by analysts include enthusiasm about product capability as well as concern about total cost.

Third, contract metrics contain assumptions. TCV and RDV are useful but can include options and cancellable arrangements. Investors should not capitalize every dollar of remaining deal value as though it were guaranteed revenue.

Fourth, stock-based compensation remains meaningful. The $265 million quarterly expense represented nearly 14% of revenue. Palantir is producing enough GAAP profit to absorb it, but dilution and compensation economics remain relevant to per-share value.

Fifth, government growth is exposed to politics and procurement. Defense demand may remain strong, yet specific programs can be delayed or contested. Public criticism over surveillance, immigration, military targeting, and data integration can lead to litigation, regulatory review, employee concerns, or procurement resistance.

Sixth, AI technology changes quickly. Palantir currently benefits from complexity: customers need help connecting models to data and operations. If model providers, cloud platforms, and enterprise software companies make that integration dramatically easier, Palantir’s differentiation could narrow. The company must maintain a lead in deployment, governance, ontology, security, and time to value.

Seventh, extraordinary growth creates difficult comparisons. A 149% segment growth rate cannot continue indefinitely as the denominator expands. Even if the business remains excellent, a decline from triple-digit growth to a lower rate can pressure a valuation built on exceptional acceleration.

Finally, the equity valuation is severe. At roughly $409 billion and a 2026 revenue midpoint of $8.154 billion, Palantir traded at about 50 times current-year guided sales. The ratio would be slightly lower after subtracting net cash, but not enough to alter the basic conclusion. The market capitalization was also about 84 times the midpoint of adjusted operating-income guidance. Those multiples require sustained growth and margins well beyond a normal software cycle.

Valuation: Why Great Results Do Not End the Debate

Valuation is the bridge between business performance and shareholder return. A company can report excellent results and still be a poor investment from a particular starting price. Conversely, a company can report modest results and generate strong returns if expectations were sufficiently low. Palantir’s second-quarter report exceeded expectations, but the post-earnings price reset those expectations upward.

At $159.28 per share, Palantir’s market capitalization was approximately $409 billion. Dividing that value by the midpoint of 2026 revenue guidance produces a price-to-sales ratio near 50. A simple enterprise-value-to-sales ratio would be somewhat lower after accounting for cash and securities, but it would remain far above most mature software companies.

Price-to-earnings measures require more care. Market data at the time showed a trailing P/E near 179. Forward P/E estimates vary with the analyst model, forecast period, and post-earnings revisions. A multiple near 90 can refer to a later earnings year, while a trailing measure is much higher. The exact number matters less than the broad message: Palantir is priced for substantial earnings growth.

A high multiple can be rational when growth is rapid, margins are rising, retention is strong, the market is large, and competitive advantage is durable. Palantir currently checks many of those boxes. The problem is that each must remain true for a long time. If revenue compounds at exceptional rates and margins stay high, today’s sales multiple can decline quickly as the denominator grows. If growth slows earlier than expected, the multiple can compress even while revenue and profit continue rising.

Consider the implied scale. A company valued above $400 billion must eventually generate tens of billions of dollars of annual revenue or extraordinarily large recurring cash flows to support conventional long-term return expectations. Palantir’s guided 2026 revenue is just over $8 billion. The gap is not impossible to close, but it requires years of execution.

The second-quarter report improves the probability that Palantir can grow into a portion of its valuation. It does not eliminate the valuation risk. In fact, a 27% one-day rally increases the amount of future success embedded in the price.

How Palantir Compares With the Rest of the AI Software Market

Palantir is often grouped with several different categories of companies, but none is a perfect comparison. It competes with cloud platforms, data-infrastructure vendors, analytics specialists, defense contractors, systems integrators, enterprise application companies, and customers’ own engineering teams. The breadth of those overlaps is one reason the company can address a large market. It is also why simplistic peer comparisons can mislead.

Cloud platforms provide infrastructure, not the complete operating model

Amazon Web Services, Microsoft Azure, and Google Cloud supply the computing infrastructure on which a large share of modern enterprise software runs. They also offer data warehouses, model-training services, foundation models, security products, orchestration tools, and application-development frameworks. Those capabilities overlap with portions of Palantir’s offering, especially as the cloud companies expand higher in the software stack.

Yet the cloud vendors generally benefit when customers use more computing, storage, networking, and managed AI services. Palantir’s sales proposition is different. It attempts to map an organization’s operational reality, connect disparate data sources, encode permissions, and enable software-assisted decisions. The cloud provider may host the workloads; Palantir seeks to become the layer through which managers, analysts, operators, and AI agents interact with the business.

That distinction is strategically attractive because Palantir can benefit from the hyperscalers’ capital spending without financing the same physical infrastructure. Microsoft, Amazon, Google, Meta Platforms, and other large technology companies are investing heavily in data centers, accelerators, networking, energy, and model development. Palantir’s capital requirements are much lighter. Its challenge is to prove that its software captures a durable share of the economic value created on top of that infrastructure.

Data platforms solve adjacent problems

Snowflake and Databricks are more direct software comparisons in some enterprise accounts. Both help organizations centralize, process, govern, and analyze large volumes of data. Both have expanded into AI development and deployment. Customers may use them alongside Palantir, instead of Palantir, or as part of a broader architecture that assigns different jobs to each platform.

The central competitive question is where the most valuable control point will reside. One possibility is that the data platform becomes the natural home for AI applications because it already holds or governs the data. Another is that application-specific vendors build the best workflows for particular industries. Palantir’s argument is that the ontology and operational layer become the decisive control point because they connect raw data to the people, assets, permissions, and decisions that define the organization.

That claim is plausible, but it is not automatically true for every customer. Some businesses can satisfy their needs with a cloud data warehouse, conventional business-intelligence tools, and internal software. Others may prefer industry-specific applications. Palantir is most compelling where the data is fragmented, the decisions are consequential, access controls are complicated, and the value of solving the problem is large enough to justify a substantial deployment.

Systems integrators are competitors and complements

Consulting firms and systems integrators have historically earned large fees by connecting enterprise systems, cleaning data, redesigning workflows, and managing technology transformations. Palantir can compete with that model by delivering a more standardized software platform and compressing deployment time. It can also work with integrators that help customers implement or scale the software.

AIP bootcamps sharpen this competitive contrast. The bootcamp is designed to demonstrate a functioning use case with the customer’s own data in a short period. That approach reduces the risk that a project spends months in planning before anyone sees an operational result. It also creates a direct path from demonstration to a broader contract.

The economic advantage depends on repeatability. If every deployment still requires extensive bespoke engineering, the business will resemble a high-margin services company more than a scalable software platform. If Palantir can reuse software, ontology patterns, security models, and implementation knowledge across customers, incremental revenue can grow faster than labor costs. The 62% adjusted operating margin suggests substantial operating leverage, although stock-based compensation and other excluded costs must remain part of the analysis.

Defense contractors compete for budgets, but not always for the same work

Palantir’s government business places it alongside traditional defense contractors such as Lockheed Martin, Northrop Grumman, RTX, General Dynamics, and L3Harris Technologies, as well as newer defense-technology companies. The comparison is imperfect. Traditional primes manufacture aircraft, missiles, sensors, communications systems, and other hardware. Palantir primarily sells software that integrates information and supports planning, intelligence, logistics, targeting, and operational decisions.

The distinction can make Palantir complementary to hardware vendors. Modern defense systems generate enormous volumes of data, and governments need software to combine those feeds with logistics, intelligence, and command systems. Yet budget competition is real. Procurement authorities must allocate finite resources, and large programs can be delayed, restructured, challenged, or canceled.

Palantir’s government opportunity therefore depends on more than total defense spending. It depends on the share of budgets devoted to software, data integration, autonomy, artificial intelligence, and decision support; the company’s ability to win classified and unclassified contracts; and the durability of its relationships across agencies and allied governments.

Enterprise application vendors may bring AI into existing workflows

Salesforce, ServiceNow, SAP, Oracle, Workday, and other enterprise software companies already sit inside critical workflows. Each is adding AI agents and automation. Their advantage is distribution: customers may prefer to buy new capabilities from a vendor already embedded in sales, finance, human resources, supply-chain, or service operations.

Palantir’s counterargument is that large organizations do not operate within one application. They use dozens or hundreds of systems, and many of the highest-value decisions require data that crosses those boundaries. The ontology is intended to represent the organization across software silos rather than remain confined to one application.

The outcome will probably vary by use case. A customer may use a native agent inside a customer-relationship-management system for routine tasks and Palantir for cross-functional planning. The market does not have to be winner-take-all. Palantir’s valuation, however, implicitly assumes that the company wins a substantial portion of the highest-value workloads.

Open-source software lowers some barriers while raising the value of governance

Open-weight models and open-source tools can reduce the cost of experimentation. A company can deploy a capable model without paying a premium to one proprietary provider. That weakens scarcity at the model layer and may pressure model pricing.

It does not solve the entire enterprise problem. A model still needs approved data, identity controls, auditability, evaluation, monitoring, and rules governing what actions it may take. Models can hallucinate, expose sensitive information, or make decisions that conflict with policy. The more capable AI becomes, the more important those controls may become.

Palantir’s neutral-platform argument is strongest when customers want to switch among proprietary, open-weight, and internally developed models while preserving the same data and permission architecture. The company has highlighted work with multiple model ecosystems and, together with Nvidia, has promoted secure deployments using Nvidia’s open models for U.S. government and enterprise settings. The partnership supports the claim that Palantir does not depend on one model vendor.

The competitive danger is that cloud and data platforms can build similar governance layers. Palantir’s defense is accumulated implementation experience in environments where permissions, provenance, and operational consequences matter. Whether that experience remains a durable moat will be tested as rivals improve.

From Government Data Platform to Commercial AI Operating Layer

Palantir’s current growth is easier to understand in the context of its product evolution. The company’s original reputation centered on government intelligence and defense work. Gotham helped users integrate information from multiple sources, analyze relationships, and support operational decisions. That history gave Palantir credibility in complex, security-sensitive environments, but it also contributed to the perception that the company was a specialized government contractor.

Foundry extended the approach to commercial organizations. It provided a common data and operations environment for industries such as manufacturing, health care, energy, aviation, and financial services. Apollo addressed software delivery across public clouds, private infrastructure, edge systems, and disconnected environments. AIP added a layer for integrating large language models and other AI systems with customer data and workflows.

The common element is the ontology. Palantir describes it as the operational representation of an organization: the objects, relationships, actions, permissions, and logic through which data becomes usable. A factory, for example, may have machines, components, suppliers, work orders, maintenance schedules, quality records, employees, and delivery commitments. The ontology links those elements so software can answer operational questions and, within approved limits, recommend or execute actions.

This architecture helps explain why management emphasizes deployment rather than model development. The company does not need to create the most powerful general-purpose model. It needs to make models useful within real organizations. The customer must be able to ask a question, trace the answer to authorized data, understand the assumptions, and decide whether the system may take action.

AIP arrived when enterprises were struggling to convert enthusiasm about generative AI into production systems. Many companies could build demonstrations, but fewer could connect models to sensitive data and mission-critical workflows. Palantir’s existing platforms gave it a starting point: data integration, access controls, deployment infrastructure, and relationships with customers that already trusted it with important operations.

The commercial acceleration since 2023 suggests that this positioning resonated. Palantir’s U.S. commercial growth moved from a much lower base into triple digits, and management has repeatedly pointed to bootcamps as a major customer-acquisition and expansion mechanism. The second-quarter 2026 results showed the strategy at its most powerful: rapid top-line growth, large contracts, rising remaining deal value, and expanding margins.

The historical narrative still requires caution. A strong sequence of quarters does not prove that every deployment will scale or that competitors cannot close the gap. It does show that Palantir found a repeatable way to sell enterprise AI at a time when many vendors were still discussing prototypes.

Revenue Quality, Customer Economics, and the Meaning of Large Deals

Revenue growth receives the most attention, but the quality of that growth determines its durability. Palantir’s second-quarter results contain several favorable signals: large contract counts, rising remaining deal value, strong cash flow, and expansion in both commercial and government segments. They also contain reasons to interpret the numbers carefully.

Large contracts indicate organizational expansion

Palantir closed 220 deals worth at least $1 million during the quarter. Within that group, 98 were worth at least $5 million and 73 were worth at least $10 million. The eight-figure category is important because it is difficult to explain as a minor experiment. A contract of that size usually reflects multiple users, workflows, business units, or a mission with substantial operational value.

That does not mean each contract is recognized immediately as revenue. Revenue recognition depends on contract terms and the delivery of products and services. Nor does it mean the full contract value is guaranteed. Palantir states that TCV and RDV can include customer options and contracts that may be terminated for convenience.

The better interpretation is that contract size provides evidence of adoption depth. Customers appear to be moving beyond isolated pilots into broader commitments. That supports the operating-system thesis, but investors should still look for conversion into recognized revenue, renewal, expansion, and cash.

Remaining deal value provides visibility, not certainty

U.S. commercial RDV increased 124% year over year to $6.238 billion. The measure represents the remaining value of contracts, including certain assumptions about options and termination rights. It can grow because the company signs more business, signs longer contracts, or expands existing agreements.

RDV is broader than remaining performance obligations under accounting rules. RPO generally reflects contracted revenue that has not yet been recognized and is subject to more specific accounting criteria. Palantir uses RDV to offer an operational view of its contract base, but the company explicitly warns that the measure may not predict future revenue as reliably as investors sometimes assume.

The growth is still meaningful. A rapidly expanding contract base reduces dependence on finding an entirely new customer for every dollar of future revenue. It indicates that customers are making commitments that extend beyond the current quarter. The key is conversion: the value becomes economically important when it turns into delivered software, recognized revenue, cash collection, and renewal.

Cash flow supports the quality of reported earnings

Cash generation is one of the strongest features of the quarter. Operating cash flow was $1.216 billion, or approximately 63% of revenue. Adjusted free cash flow was $1.220 billion, also about 63%. The small difference reflects the company’s limited capital expenditure relative to revenue.

That cash profile distinguishes Palantir from growth companies that report adjusted profits while consuming cash. It also supports the argument that the business can scale without enormous physical investment. Customers fund contracts; Palantir spends primarily on people, software development, sales, infrastructure, and deployment rather than factories or data centers.

Cash flow is not identical to economic profit. Stock-based compensation is added back in the operating cash-flow statement because it is a non-cash expense in the period, even though it transfers value to employees and can dilute shareholders. Working-capital timing can also affect quarterly cash generation. Nevertheless, sustained cash flow at this level provides flexibility and reduces financing risk.

Customer concentration and contract timing can create volatility

Large government and commercial contracts can make quarterly results uneven. A contract may begin, expand, or reach a revenue-recognition milestone in one period. Procurement schedules can move. Customers may exercise options later than expected. That means a single quarter’s acceleration should not be extrapolated mechanically.

Palantir’s broadening U.S. commercial business reduces some concentration risk, but the company still serves large organizations and public agencies whose purchasing decisions can be material. Investors should monitor customer count, average revenue per customer, large-deal concentration, renewal, and the relationship between bookings-style metrics and recognized revenue.

The operating model must keep scaling

Palantir’s margins imply that revenue is growing much faster than the cost required to support it. That is the core promise of enterprise software. Once the platform is developed, an additional deployment should require less incremental expense than the revenue it produces.

The model becomes less attractive if customization and on-site support rise in proportion to revenue. Palantir employs forward-deployed engineers and other personnel who work closely with customers. That proximity is a competitive strength because it helps solve difficult problems. It can also constrain scalability if every customer needs a unique labor-intensive engagement.

The second-quarter numbers favor the scalable interpretation. Adjusted operating margin rose while revenue nearly doubled, suggesting that implementation and support costs did not keep pace with sales. The durability of that leverage will matter more than any one-quarter margin record.

GAAP Profit, Stock-Based Compensation, and Dilution

Palantir’s adjusted results are exceptional, but a complete assessment must include statutory accounting. The company reported $912 million of GAAP operating income, a 47% operating margin, and $1.062 billion of net income attributable to common shareholders. Those figures show that profitability is not merely the product of an adjusted presentation.

The difference between GAAP and adjusted operating income was driven mainly by stock-based compensation and related payroll taxes. Stock-based compensation was $265 million in the quarter, equal to approximately 13.7% of revenue. The expense helps Palantir recruit and retain employees, particularly in a competitive technology labor market. It also represents compensation paid through equity rather than cash.

Why stock-based compensation matters even when cash flow is strong

Stock-based compensation does not require an immediate cash payment, so it is added back when calculating operating cash flow. That improves cash generation, but the economic cost does not disappear. New shares or equity awards can dilute existing owners, and employees receive part of the company’s value.

The appropriate question is not whether stock-based compensation should be ignored or treated as cash. It is whether the company creates enough incremental value to offset dilution. Rapid revenue growth and rising GAAP profit can make a given level of equity compensation more manageable. If compensation grows faster than the business or share count expands materially, per-share economics weaken.

Palantir’s progress is visible in the income statement. The company now produces substantial GAAP profit after recognizing equity compensation. That is a stronger position than a software company that can show profitability only by excluding a large recurring expense. Still, adjusted margin should not be used without disclosing what it excludes.

Net income benefited from more than operating profit

Net income exceeded operating income. Palantir’s large cash and securities balance generates interest and investment income, and the tax line can also affect the result. That means net margin should not be interpreted as purely the profitability of selling software.

The distinction matters for valuation. Operating income shows the economics of the core business before interest and taxes. Net income includes returns on the balance sheet and other items. Both are legitimate, but they answer different questions.

Per-share results depend on the diluted share count

Palantir reported diluted earnings per share of $0.41 on both a GAAP and adjusted basis for the quarter. The diluted share count incorporates the effect of potentially dilutive equity awards. Investors evaluating long-term value should focus not only on total profit but also on profit and cash flow per share.

A company can increase aggregate earnings while delivering less impressive per-share growth if the share count rises substantially. Palantir’s ability to expand revenue, margins, and cash flow faster than dilution is therefore a central part of the investment case.

International Adoption Is the Largest Strategic Gap

The United States generated approximately $1.573 billion of Palantir’s $1.935 billion quarterly revenue, or about 81%. The implied international contribution was roughly $362 million. That concentration reflects extraordinary U.S. strength, but it also highlights a gap between Palantir’s global ambition and its current geographic mix.

International adoption can be slower for several reasons. Data-sovereignty requirements vary across countries. Governments may prefer domestic suppliers for sensitive workloads. European procurement can be fragmented. Privacy rules and public attitudes toward surveillance technology can complicate deployment. Large organizations outside the United States may also have different technology budgets, purchasing processes, and risk tolerance.

Palantir has responded by emphasizing sovereign AI: systems that allow governments and enterprises to deploy models within jurisdiction-specific security, data, and infrastructure requirements. Its architecture can run across public cloud, private cloud, on-premises systems, edge environments, and disconnected networks. That flexibility is valuable when customers cannot send sensitive data to a standard commercial service.

Partnerships can expand distribution. Palantir and Nvidia have promoted combined offerings that connect Nvidia models and accelerated computing with Palantir’s ontology and operational platform. Such arrangements may help governments and regulated industries deploy AI while preserving control over data and infrastructure.

The international challenge remains substantive. Reuters reported that France’s domestic technology company ChapsVision was positioned to replace Palantir in an intelligence-related role, illustrating the political pressure to develop national alternatives. Public authorities may value Palantir’s capability and still decide that sovereignty requires a local provider.

For the long-term thesis, two outcomes would both be constructive. Palantir could accelerate internationally and prove that its sovereign architecture overcomes local resistance. Or it could continue expanding so deeply in the United States that domestic demand supports years of growth. The valuation would be easier to defend if both occurred.

Government Contracts Bring Durable Demand and Unusual Risks

The 90% increase in U.S. government revenue demonstrates that Palantir’s original market is not being displaced by the commercial AI story. Defense agencies and intelligence organizations are investing in data integration, software-defined operations, autonomy, logistics, and AI-assisted decision support. Geopolitical competition increases the urgency of those capabilities.

Government business can be durable because missions persist across economic cycles. A manufacturer may delay a discretionary technology project during a recession; a defense agency still has national-security responsibilities. Multi-year programs can create visibility, and successful systems become difficult to replace when they are integrated into operations.

The same market carries risks that ordinary commercial software companies face less often.

• Procurement risk: Awards can be protested, delayed, reduced, or canceled. Budget authorization does not guarantee immediate revenue.
• Political risk: Changes in administration, congressional priorities, or allied policy can alter program demand.
• Reputational risk: Work involving immigration, intelligence, policing, or military operations can attract criticism from employees, activists, customers, and the public.
• Legal and regulatory risk: Privacy, civil-liberties, procurement, and data-handling disputes can produce litigation or review.
• Security risk: A breach, model failure, or unauthorized access in a sensitive deployment could cause severe operational and reputational damage.
• International sovereignty risk: Foreign governments may seek domestic alternatives even when Palantir’s product is technically competitive.

These risks do not negate the demand. They explain why government software expertise can be both a moat and a source of controversy. Palantir has spent years operating in environments where failure has consequences beyond ordinary business inefficiency. That history strengthens its commercial pitch to regulated industries, but it also keeps the company exposed to public-policy debates.

What the Earnings Commentary Got Right—and Where It Went Too Far

The market commentary that accompanied the results correctly identified several central features of the quarter. Palantir produced an unusual combination of growth, margin expansion, contract momentum, and guidance. U.S. commercial and U.S. government revenue both accelerated at scale. The company’s cash conversion was strong. Open-model adoption does not automatically undermine Palantir because its platform is designed to connect different models to the same governed operational layer.

The description of two demand engines is also useful. Commercial customers and government agencies buy for different reasons, follow different procurement cycles, and respond to different budgets. Strength in both reduces dependence on one market.

Several statements need qualification.

Calling the Rule of 40 score of 155 nearly four times the “goal” is arithmetically correct, but it can imply more comparability than exists. The standard benchmark usually concerns recurring software businesses at lower growth rates. Palantir’s score is exceptional, yet the growth component will normalize as the revenue base expands. The metric is a snapshot, not a permanent state.

Describing adjusted operating margin as the share of revenue that becomes operating income is incomplete without noting the excluded stock-based compensation and payroll taxes. Palantir still produced a 47% GAAP operating margin, which is outstanding, but the accounting distinction matters.

Saying the full-year guide exceeded $8.2 billion overstates the official range. The company guided to $8.150 billion to $8.158 billion. The increase from prior guidance was approximately $498 million at the midpoint, which remains a very large one-quarter raise.

The argument that open-source models necessarily make Palantir more valuable is plausible rather than proven. Model commoditization can increase demand for governance and integration, but it can also reduce barriers for competitors. Palantir must continue demonstrating that its ontology, deployment model, and customer relationships create differentiation that cloud and data-platform rivals cannot easily reproduce.

Finally, the assertion that momentum will not fade is a forecast. The quarter supports confidence, but no company can guarantee continued acceleration. Customer budgets, procurement timing, competition, technology change, and difficult comparisons remain relevant.

A balanced reading therefore reaches a strong but narrower conclusion: Palantir delivered one of the best combinations of growth and profitability among large public software companies in the quarter. The results materially strengthened the business case. They did not make the future certain or the stock inexpensive.

How to Evaluate Palantir After the Q2 2026 Earnings Report

The most useful framework is not a binary question about whether Palantir is a good or bad company. The business quality and the equity valuation are separate variables. Readers can track a small set of evidence that determines whether the company is growing into the expectations embedded in its market value.

Evidence that would strengthen the thesis

• Sustained U.S. commercial growth: Continued expansion well above the software industry would support the view that AIP is becoming a standard operating layer rather than a temporary AI experiment.
• Conversion of RDV into revenue: Large remaining deal value should produce recognized revenue and cash over time. A widening gap would require explanation.
• Stable or rising GAAP margins: Operating leverage after stock-based compensation would show that scalability is not dependent on adjusted exclusions.
• Per-share cash-flow growth: Cash generation should outpace dilution.
• International acceleration: Stronger growth outside the United States would broaden the opportunity and reduce geographic concentration.
• Repeatable industry deployments: Evidence that one customer’s successful workflow can be adapted across an industry would increase the addressable market and reduce implementation cost.
• Model neutrality in practice: Customers using multiple proprietary and open models through Palantir would validate the neutral-layer strategy.
• Government contract breadth: Growth across defense, intelligence, logistics, health, and civilian agencies would reduce reliance on a small number of programs.

Evidence that would weaken the thesis

• A sharp commercial deceleration: Slower growth is inevitable eventually, but a sudden fall before the revenue base becomes much larger would challenge the current valuation.
• Lower contract conversion: RDV or TCV that rises without corresponding revenue and cash could signal optimistic assumptions or delayed deployments.
• Margin compression: A need for substantially more implementation labor, sales spending, or infrastructure would weaken the operating-leverage story.
• Persistent international weakness: Failure to expand abroad would make the long-term opportunity more dependent on unusually deep U.S. penetration.
• Pricing resistance: Customer evidence that Palantir’s cost outweighs operational gains would create room for internal builds and lower-priced competitors.
• Platform convergence: If cloud, data, and application vendors offer comparable governance and deployment with lower friction, Palantir’s differentiation could narrow.
• Greater dilution: Rising share count without faster per-share profit and cash-flow growth would reduce the benefit to shareholders.
• Government disruption: Major protests, contract losses, legal restrictions, or procurement reversals could weaken one of the two growth engines.

Three broad operating scenarios

Exceptional execution: U.S. commercial growth remains very high, government revenue continues expanding, international adoption improves, and margins stay elevated. In this scenario, Palantir grows into a meaningful portion of its valuation through rapid increases in revenue and cash flow.

Strong business, normalizing growth: Revenue continues rising at an impressive rate but decelerates as the base expands. Margins remain strong, yet valuation multiples compress because the market no longer pays an extraordinary premium. The company succeeds operationally, while shareholder returns depend heavily on the entry price.

Expectation reset: Commercial deployments take longer, competition increases, international adoption remains weak, or government contract timing turns unfavorable. Palantir still grows, but not fast enough to justify the post-earnings valuation. The stock can fall materially even without a collapse in the business.

These scenarios are not price forecasts. They illustrate why an excellent company and a high-risk valuation can coexist.

What Happens Next

Palantir guided third-quarter 2026 revenue to between $2.160 billion and $2.164 billion and adjusted operating income to between $1.292 billion and $1.296 billion. The midpoint implies approximately 11.7% sequential revenue growth from the second quarter.

The full-year revenue range implies a fourth quarter of roughly $2.424 billion when the reported first half and third-quarter midpoint are subtracted from the annual midpoint. That calculation is an editorial estimate, not company guidance for the fourth quarter. It would represent approximately 12.1% sequential growth from the third-quarter midpoint.

The next report will test several questions:

• Can U.S. commercial revenue sustain triple-digit year-over-year growth?
• Does U.S. government revenue remain near its second-quarter pace?
• Will international revenue accelerate?
• Does U.S. commercial RDV continue converting into recognized revenue?
• Can GAAP operating margin remain close to the second-quarter level?
• Does stock-based compensation decline as a percentage of revenue?
• Will management raise full-year guidance again, or has the latest range already captured most of the upside?

The market will also reassess valuation as analysts update earnings estimates. A higher denominator can reduce a forward multiple even if the share price stays elevated. Conversely, continued price gains can preserve or increase the premium despite stronger estimates.

Beyond quarterly numbers, the most important evidence will come from customer expansion. Palantir needs to show that AIP bootcamps continue turning into production deployments, that large contracts extend across organizations, and that model neutrality creates a durable role rather than a temporary integration opportunity.

Frequently Asked Questions

What did Palantir report for Q2 2026 revenue?

Palantir reported $1.935 billion of revenue for the second quarter of 2026, an increase of 93% from $1.004 billion in the same quarter a year earlier. Revenue rose approximately 19% from the first quarter of 2026.

How fast did Palantir’s U.S. commercial business grow?

U.S. commercial revenue reached $764 million, up 149% year over year and 28% sequentially. It was the fastest-growing major reported segment and the clearest evidence of enterprise demand for Palantir’s Artificial Intelligence Platform.

How much U.S. government revenue did Palantir generate?

U.S. government revenue was $809 million, an increase of 90% year over year and 18% sequentially. The $109 million figure included in some summaries of the report is incorrect.

What was Palantir’s operating margin?

Palantir reported a 47% GAAP operating margin and a 62% adjusted operating margin. The adjusted figure excludes stock-based compensation and related payroll taxes, so the two measures should not be used interchangeably.

What does a Rule of 40 score of 155 mean?

Palantir calculated the score by adding its 93% year-over-year revenue growth rate to its 62% adjusted operating margin. The result was 155. The conventional Rule of 40 says that a software company’s growth rate plus profitability margin should total at least 40, although definitions vary and Palantir’s score uses an adjusted margin.

How many large contracts did Palantir close?

The company closed 220 deals worth at least $1 million during the quarter. That total included 98 deals worth at least $5 million and 73 worth at least $10 million. The categories are nested rather than additive.

What is remaining deal value?

Remaining deal value is Palantir’s estimate of the value left in contracts that has not yet been recognized as revenue. U.S. commercial RDV was $6.238 billion, up 124% year over year. The measure can include customer options and terminable contracts, so it is not guaranteed future revenue.

What guidance did Palantir provide for 2026?

Palantir raised full-year revenue guidance to $8.150 billion to $8.158 billion. It guided U.S. commercial revenue to more than $3.424 billion, adjusted operating income to $4.889 billion to $4.897 billion, and adjusted free cash flow to $4.5 billion to $4.7 billion.

Why did Palantir stock rise after earnings?

The shares rose because revenue, earnings, margins, and guidance exceeded market expectations, while both commercial and government growth accelerated. At approximately 11:53 a.m. Eastern Time on August 4, the stock traded at $159.28, up about 26.8% from the previous close. A market move has multiple causes, but the earnings surprise was the dominant company-specific event.

Is Palantir profitable under GAAP?

Yes. Palantir reported $912 million of GAAP operating income and $1.062 billion of GAAP net income attributable to common shareholders for the quarter. The company also generated $1.216 billion of operating cash flow.

Why are open-source AI models relevant to Palantir?

Cheaper and more interchangeable models can reduce the value of model scarcity. Palantir argues that this increases the importance of the governed layer connecting models to customer data, permissions, and workflows. The opportunity is real, but cloud and data-platform competitors are developing similar capabilities.

What is the biggest risk after the quarter?

The largest immediate risk is the gap between excellent business performance and an exceptionally high valuation. Palantir must sustain rapid growth and strong margins for years. International weakness, competition, pricing resistance, contract timing, government procurement, and dilution could make that path more difficult.

Final Assessment

Palantir’s second-quarter 2026 report was exceptional by almost any large-cap software standard. Revenue rose 93% to $1.935 billion. U.S. commercial revenue increased 149%, U.S. government revenue rose 90%, and the company produced a 47% GAAP operating margin. Contract activity, remaining deal value, operating cash flow, and full-year guidance all strengthened the conclusion that customers are moving beyond AI demonstrations into larger operational deployments.

The most persuasive interpretation is that Palantir has built a valuable position above the model and infrastructure layers of artificial intelligence. It connects models to governed data and real workflows, while the hyperscalers finance much of the underlying computing buildout. Its government experience supplies credibility in complex, security-sensitive environments, and AIP bootcamps have created a faster route into commercial customers.

The skeptical interpretation is not that the quarter was weak. It is that the valuation already assumes an unusual amount of future success. At the post-earnings price measured on August 4, Palantir was valued at roughly 50 times the midpoint of current-year revenue guidance. Growth will inevitably slow from the second quarter’s rate, and the market may react sharply if that normalization arrives sooner than expected.

The report changed the balance of evidence in Palantir’s favor. It showed that commercial adoption is broadening, government demand remains powerful, and profitability can rise alongside growth. It did not remove the need for careful accounting distinctions, contract conversion, international expansion, and per-share analysis.

The next stage of the story will be determined by execution rather than narrative. Palantir must turn its $6.238 billion U.S. commercial remaining deal value into revenue and cash, preserve its deployment advantage as competitors improve, expand beyond the United States, and sustain GAAP profitability after stock-based compensation. If it does, the company can continue growing into a demanding valuation. If it merely performs well instead of extraordinarily well, the stock can still face an expectation reset.

Sources

• Palantir Technologies Q2 2026 earnings release filed with the SEC
• Palantir Q2 2026 shareholder letter
• Palantir Q2 2026 business update
• Palantir Technologies 2025 Form 10-K
• Palantir Technologies Q1 2026 Form 10-Q
• Palantir Technologies Q1 2026 earnings release
• Reuters report on Palantir’s Q2 2026 results and guidance
• Reuters market report on the August 4, 2026 post-earnings reaction
• Associated Press U.S. market report for August 4, 2026
• MarketWatch analysis of Palantir’s earnings beat and outlook
• Investor’s Business Daily post-earnings analyst and valuation coverage
• Palantir AIP Bootcamp product information
• Nvidia report on Palantir, secure AI, and open models for U.S. agencies
• Nvidia and Palantir enterprise AI partnership announcement
• Palantir sovereign AI architecture overview
• Bessemer Venture Partners explanation of the Rule of 40 and Rule of X
• Google Finance market data for Palantir Technologies

This article is provided for general informational purposes and does not constitute financial, investment, tax, or legal advice.

Affiliate disclosure: Businessfinance.news may earn compensation from qualifying actions completed through selected links on this website, at no additional cost to the reader. Affiliate relationships do not influence our editorial reporting, analysis, or conclusions.