Palantir Has an $11.2 Billion Revenue Backlog, a $10 Billion Army Deal, and a 61% Growth Forecast. Is It a Buy for 2026?

Palantir reported 2025 revenue of $4.5B, up 56% year-over-year with customer count rising to 954, and ended 2025 with $11.2B in remaining performance obligations (up 105% y/y). The company secured a landmark U.S. Army contract worth up to $10B over the next decade and management guides for 61% revenue growth in 2026. Despite these growth metrics and cross-sell/ontology-driven moat, the shares trade at steep multiples (P/S ~87, P/E ~241), prompting valuation caution even as the author views Palantir as a compelling long-term AI-themed buy.

Analysis

Winners will include firms that sit directly upstream of decisioning compute and system integrators that embed Palantir’s ontologies into mission workflows; that should increase demand for GPUs, inference racks and professional services rather than generic cloud storage alone. Conversely, vendors whose value is primarily commoditized storage or query engines face a margin squeeze — the second-order effect is higher attach rates for compute and integration spend versus pure-seat SaaS budgets.

Key risks cluster around cadence and convertibility of the long-tail backlog: a multi-year RPO is only valuable if conversion and expansion rates stay elevated, and any visible slippage (quarterly customer churn uptick, delayed government contract milestones, protest awards) will compress multiples quickly. Near-term catalysts that can move the stock materially are quarterly revenue/ARR beats, high-profile contract delivery milestones, and any regulatory or DOD audit findings; time horizons are asymmetric — days around prints and contract announcements, months for backlog conversion, and years for moat durability.

Trade implementation should exploit both binary upside and valuation complacency: use LEAPs to capture multi-year optionality while monetizing near-term premiums, and express conviction through relative-value pairs that isolate enterprise analytics share shifts versus pure data-infra vendors. Size asymmetrically—concentrated if you own conviction on backlog conversion, modest if the thesis depends on continued broad AI capex.

The contrarian angle is that market pricing embeds sustained hyper-growth and near-zero execution risk; that’s a fragile position. If you’re constructive, own time (long-dated calls) rather than spot equity, hedge delivery risk with short-dated puts or pair trades, and be prepared to cut exposure if expansion rates or cross-sell cadence fall by more than ~300-400bps year-over-year.

Key Decisions for Investors

• Buy PLTR 24–36 month LEAP calls (size 1–2% NAV) financed by selling 6–12 month 30–40% OTM calls; objective: capture multi-year conversion of RPO into cash flow. Risk/reward: downside limited to net premium, upside 2–4x+ if secular adoption persists; trim if 6-month revenue growth decelerates >300bps.
• Pair trade — Long PLTR equity (or LEAPs) / Short SNOW equal-dollar (size 1% NAV each) over a 6–18 month horizon to express ontology-driven share gains vs. pure data-storage incumbency. Target: relative outperformance of 20–40% or absolute spread tightening; stop-loss if SNOW outperforms PLTR by >15%.
• Buy NVDA 6–12 month call options (size 0.5–1% NAV) to play increased inference/GPU demand from enterprise adoption of decisioning platforms. Risk/reward: limited premium loss vs outsized upside if enterprise deployment accelerates; reduce position if NVDA’s data-center guidance softens next quarter.
• Protective hedge: buy 3-month PLTR puts sized to 30% of long exposure ahead of key contract or earnings windows (size variable). Purpose: limit drawdown from delivery/protest/regulatory shocks; cost is insurance premium — acceptable if it prevents >25% equity drawdown on execution misses.