Where Will Oracle Be in 2 Years?

Oracle reported a $553B backlog (RPO) with RPO up 325% YoY while revenue rose 22% to $17.2B and cloud infrastructure revenue surged 84% YoY to $4.9B. The company plans to raise $50B in debt and equity this year, has already secured $30B via convertible preferreds and bonds, and signed $29B of customer-funded contracts last quarter. Management guided $67B revenue for fiscal 2026 and $90B for fiscal 2027 (a 34% increase), and analysts model EPS of $10.73 for fiscal 2028 which, at a 24.5x multiple, implies ~68% upside to $263. Key risks are higher interest expense from financing and concentration of backlog exposure to OpenAI, but the customer-funded infrastructure model supports accelerated revenue and earnings growth.

Analysis

The market is treating Oracle’s buildout as a binary execution call rather than a multi-year operating leverage story; that creates a volatility-asymmetry opportunity. If Oracle converts contracted demand into running instances at scale, gross margins on those incremental workloads should compress customer acquisition and RPO conversion timing into a tight 12–24-month earnings acceleration window that the stock may not be fully discounting today. Conversely, any multi-quarter slippage in rack activation, GPU supply or power provisioning will show up quickly in cash interest coverage and re-pricing of the equity and credit curves.

Competitive second-order effects are underappreciated: hyperscalers and GPU vendors win when scale deployments accelerate, but OEMs and colo players face margin pressure as customers pre-pay for hardware or bring custom appliances that bypass traditional refresh cycles. That shifts profit pools away from systems integrators and blade-server OEMs toward chip/IP owners and high-density co-location / PUE optimization specialists; watch procurement cadence from the top 20 AI customers for early signal of demand elasticity.

Key catalysts and tail risks are operational (rack turn-up cadence, power/connectivity constraints), counterparty concentration (one or two large customers can re-price deals or delay activations), and financing friction (higher rates or covenant pressures would force pacing changes). Time horizons compress: expect meaningfully visible outcomes within the next 2–6 quarters; full realization of scale economics is a 12–36 month story. That framing argues for asymmetric, time-limited exposure rather than full reliance on a single multi-year buy-and-hold thesis.

Key Decisions for Investors

• Long ORCL equity (2–4% portfolio): scale in over 6–10 weeks on pullbacks that widen intraday volatility; target a 12–18 month upside of 30–50% if rack-activation cadence meets management’s near-term signals, set a protective stop at -20–25% from entry to limit execution risk.
• Buy ORCL long-dated calls (18–30 months, 10–20% OTM): use LEAPS to capture asymmetric upside from re-rating while risking only premium; size 1%–2% notional and roll if activation KPIs (customer instance counts, revenue conversion) are met over two consecutive quarters.
• Pair trade: long ORCL / short INTC equal notional (6–12 month horizon): hedge broad AI demand risk while expressing a relative tilt toward GPU-centric deployments; take profits if the spread compresses by 20–30% or cut if the short leg outperforms by 15% (indicating broader cyclical move).
• Credit opportunism (1–3 year horizon): accumulate ORCL senior bonds or IG CDS on basis widening >75bps vs peers — receives carry and convexity to a successful scale-up; limit exposure size so that a 2-notch credit hit does not exceed 1–2% portfolio loss.