Palantir (PLTR) Q3 2026 Earnings Call Transcript

Palantir reported Q3 2026 revenue of $2.313 billion, up 96% year over year and 20% sequentially, with U.S. commercial revenue rising 148% to $984 million and surpassing U.S. government revenue for the first time. The company raised full-year revenue guidance to $8.530–$8.534 billion, implying approximately 91% growth, and lifted adjusted free-cash-flow guidance to $4.8–$5.0 billion. Adjusted operating margin was 62%, and management attributed accelerating commercial demand to adoption of its AI platform and Sovereign AI offering.

Analysis

The important read-through is not simply AI adoption; it is a potential shift in bargaining power from model providers to the enterprise workflow layer. If customers can use open-weight models and switch models while retaining their data, logic, and operating processes, differentiation may accrue to deployment and integration platforms rather than frontier-model access alone. That supports Palantir’s position, but management’s customer examples are not independent evidence of repeatable economics across the installed base.

Near term, results raise the bar for execution: faster pilot-to-contract conversion can pull forward revenue, but shorter initial contract terms make expansion and renewal—not signed deal value—the key proof point. TCV, RDV, and customer anecdotes should not be treated as equivalent to durable recognized revenue. RPO being flat sequentially is a reason to monitor conversion, not by itself evidence of deterioration given the stated government-contract structures.

Over 6–18 months, model efficiency is a two-sided read-through. Smaller, fine-tuned models may reduce cost per task, while lower unit costs and edge deployments can expand total inference demand; net demand for NVIDIA compute is uncertain. Anthropic may benefit as a model supplier even if application-layer economics accrue elsewhere. AIG and L3Harris illustrate possible customer productivity benefits, but the reported examples do not establish sector-wide savings or vendor displacement. The principal downside is a capacity or implementation bottleneck: demand that cannot be staffed or converted could disappoint against elevated expectations. Verify contract duration, expansion rates, revenue conversion, and hiring capacity before extrapolating the quarter.

Key Decisions for Investors

• PLTR: Favor a staged long entry after the initial earnings volatility rather than chasing a gap. Add only if the next 1–2 quarters show continued U.S. commercial growth alongside conversion of deal value into revenue; cut or reassess on a material sequential slowdown in commercial bookings, weaker retention, or guidance that fails to keep pace with demand claims. No price target is warranted without valuation and post-earnings price data.
• For a relative-value expression, consider long PLTR versus a diversified enterprise-software basket only if the view is specifically that workflow deployment captures more AI economics than standalone software seats. Keep the position modest: the basket may not isolate that exposure, and PLTR’s execution and valuation sensitivity could dominate the thesis.
• NVDA: Do not read smaller task-specific models as automatically bearish for compute. Track total deployment, edge-system volumes, and customer inference workloads; efficiency-driven cost reductions could expand use, but a decline in aggregate GPU demand would falsify the positive read-through.
• Monitor realized revenue against TCV/RDV, renewal and expansion rates, initial contract duration, and implementation capacity. If reported deal momentum remains strong but revenue conversion or customer expansion weakens, treat the growth narrative as less durable; government award timing is a separate catalyst and risk.