Editorials

When a software company's operating income rises 21% while revenue climbs just 19%, something worth studying is taking place. Palantir Technologies reported second-quarter 2026 results on August 4, and the earnings reveal an enterprise that is not merely growing, but improving its unit economics at an accelerating clip.

The fundamentals speak plainly. Revenue in Q2 reached $1.935 billion, up from $1.633 billion in Q1. Gross profit expanded to $1.639 billion with a sustained 85% gross margin, a figure that places Palantir alongside the most profitable software franchises in operation. Operating income jumped to $912 million from $754 million the quarter prior. Diluted earnings per share rose to $0.41 from $0.34.

This is not rocket science, but it is worth attention. Companies that grow revenue while simultaneously expanding operating margins are compressing their cost base relative to sales, which is the opposite of what most growing firms do.

Operating income matters, but operating cash flow is the true test. In Q2 2026, Palantir generated $1.216 billion in operating cash flow. That single quarter's cash generation exceeded the company's operating profit by $304 million, a spread that reflects strong receivables collection and favorable working capital timing.

To contextualize this: annual operating cash flow is becoming a meaningful multiple of net income, signaling that the business converts earnings into actual cash with exceptional efficiency. For an enterprise that spent $192.5 million on R&D in Q2 alone, this cash generation capacity matters tremendously. Heavy reinvestment in the platform (Foundry, AIP, Gotham, Apollo) is not strangling cash returns.

Year-over-year, operating cash flow in Q2 2026 was roughly 84% higher than the comparable period in 2025, according to the key metrics digest. That kind of cash acceleration outpaces revenue growth and certainly outpaces typical software industry norms.

Gross margin of 85% is not a ceiling; it looks more like a floor. When a software company maintaining two distinct customer bases—government (54% of 2025 revenue) and commercial (46%)—can achieve such margins across both segments while scaling from $3.7 billion in 2025 full-year revenue to a $7.7 billion annualized run rate (based on Q2 trailing quarter extrapolation), the underlying physics are sound.

The cost of revenue in Q2 was $297 million on $1.935 billion in sales. That 15% cost structure is a product of platform maturity and scale. Foundry, the foundational data operations system at Palantir's core, is increasingly deployed at existing customers as a common operating system across divisions and functions. This horizontal expansion within the base requires minimal incremental R&D per dollar of revenue.

The question investors face is whether margins remain sticky as the company pursues new verticals and geographies. The data so far suggests they do.

Operating metrics alone do not capture why this matters. Palantir's business model relies on deep, long-term customer relationships in sectors with high switching costs and high implementation friction: defense, intelligence, utilities, pharmaceuticals, supply chain management. The company serves 954 customers across these domains, with the top 20 customers generating trailing-twelve-month revenue of $93.9 million per customer, up from $64.6 million in 2024.

This progression reflects either deeper wallet penetration, higher-value customer wins, or both. Either way, it signals that the sales funnel is producing higher-quality opportunities and that existing customers are expanding their use of Palantir's platforms. The alternative explanation—that the company is simply raising prices—is belied by the fact that customer count is growing and gross margins are stable.

The government segment, Palantir's original market, remains a substantial revenue driver at roughly 54% of sales. The commercial segment is accelerating from a smaller base, with enterprise software adoption of AIP (the generative AI platform) creating new points of entry. Neither segment is mature; both are in active expansion.

Palantir trades at a P/E multiple of roughly 129x based on the most recent quarter's earnings run-rate and current market capitalization of approximately $389 billion. The enterprise value sits around $387 billion. On a price-to-sales basis, the stock sits at 63x trailing revenue. By historical and comparative standards, these are elevated multiples.

For perspective: revenue is growing year-over-year at 56%, and net income is surging at 252% on a year-over-year basis. EPS growth stands at 232% year-over-year. A firm expanding earnings at triple-digit percentages commands a different valuation regime than one growing at single-digit rates, but the multiples reflect significant expectations embedded in the stock price.

The company is not unprofitable. It is not burning cash. Its operating leverage is real and measurable. But the market has priced in substantial future growth and margin expansion. Any stumble in the top line, any slowdown in customer wins, or any compression in deals would be met with sharp revaluation.

Investors must ask: Is 56% revenue growth sustainable? Is 252% net income growth a feature or a temporary anomaly driven by leverage expansion? How much of the valuation assumes that both trajectories persist?

Bulls see a best-in-class software franchise with secular tailwinds (AI adoption, government IT modernization, enterprise data fragmentation) and improving unit economics. The data supports this view: margins are widening, customers are deepening relationships, and cash is flowing.

Bears see a stock priced for perfection, with valuation metrics that assume not just sustained high growth but acceleration into new markets and new use cases. They argue that the valuation leaves no room for error and that the company's addressable market, while large, is not infinite.

Both sides have merit. The company's fundamentals are genuinely strong. The question is whether strength is reflected fairly in a market cap that exceeds $389 billion, or whether investors are paying a premium for upside scenarios that have not yet materialized.

One fact is incontrovertible: Palantir is not a classic unprofitable growth story. It is a profitable, cash-generative business with measurable competitive advantages (embedded customer relationships, switching costs, proprietary data ontology). The premium valuation is not unjustifiable; it is simply worth scrutinizing.

Palantir's Q2 2026 results are genuinely impressive by the metrics that matter: revenue growth, margin stability, operating income leverage, and cash generation. The company is demonstrating what happens when a software platform achieves meaningful market penetration and begins to realize economics of scale.

But impressive results at a premium valuation are a test of conviction, not a gift. Investors who buy Palantir at current levels are betting that the company will sustain extraordinary growth rates for years to come, that its competitive moats are as wide as they appear, and that management can deploy capital wisely as the company scales. Those are reasonable bets, but they are bets nonetheless.

For current shareholders, the results validate the thesis. For prospective investors, the opportunity cost of capital deployed at these multiples is the critical variable. In a world where 10-year Treasury yields are measurable in real percentage points, the math on high-multiple growth stocks looks different than it did two years ago.

The company is excellent. Whether it is excellently priced is a separate question.