Prediction: Oracle Will Be Worth More Than Meta by 2028. Here's Why.

Oracle reported a remaining performance obligation (RPO) of $553 billion (up 325% YoY), with cloud revenue +44% to $8.9B and overall revenue +22% to $17.2B, signaling strong backlog conversion. Management sees revenue accelerating from a projected 17% this fiscal year to +32% next fiscal year and +46% in fiscal 2028, targeting $185B in fiscal 2029. The article models valuation scenarios where Oracle could reach ~$2.8T by 2028 (15x sales) versus Meta at ~$2.7T (7.8x sales) if Oracle’s faster growth persists. Key risks to monitor: execution on large AI/cloud contracts, backlog realization timing, and revenue concentration from major customers.

Analysis

Oracle’s business is transitioning from software licensing to being an industrial-scale capacity aggregator for AI compute — that changes the margin and cash-conversion profile in ways the market underappreciates. Contracted, multi-year capacity deals convert capex-heavy supply into annuity-like revenue, giving Oracle optionality to layer higher-margin services (private interconnect, managed stacks, GPU leasing) on top of raw rack sales, and allowing it to capture a larger share of downstream TAM than a pure software play would.

Second-order winners include GPU and subsystem suppliers (NVIDIA and their board/PSU partners) and colo/real-estate owners because long-term commitments will prioritize capacity reservation over spot pricing; losers are mid-cycle spot-market brokers and any software vendor that relied on elastic, on-demand pricing to arbitrage cost. Hyperscalers face a choice: secure committed capacity (higher near-term cash spend) or accelerate in-house fabs/architectures — that bifurcation will widen margins between capacity sellers and self-sufficient cloud giants.

Key risks are behavioral and technical: a meaningful improvement in model efficiency or an oversupply surge in accelerators could materially compress demand within 12–24 months, while large customer cancellations or sovereign restrictions on data localization could strand capacity. Near-term catalysts to watch are quarterly contract conversion cadence, disclosed multi-year anchor customers, and GPU supply dynamics; these will move valuation multiples faster than revenue beats alone.

Key Decisions for Investors

• Pair trade (12–36 months): Long ORCL equity (size 3–5% of risk budget) / Short META equity equal notional (size matched). Thesis: capture valuation re-rate if contracted capacity converts to high-margin services; target 40–60% gross upside on the pair if Oracle rerates to a premium multiple while Meta reverts to sector multiple. Tactical stop: 15% adverse move on pair; take-profits at 30% and 50%.
• Options play (6–12 months): Buy NVDA 9–12 month calls (delta 0.30–0.40) to express continued GPU tightness; finance ~50% by selling 1–3 month covered calls. Risk: high IV; reward: asymmetric upside to further supply tightness or upside surprise from enterprise demand. Position size: 1–2% of portfolio.
• Selective overweight (12–24 months): Add INTC via equity or cheap LEAPS (conservative notional), sized as a hedge to NVDA convexity. Rationale: benefits from server CPU refresh cycles and potential ASP stabilization if Intel wins more OEM server sockets. Risk manage with 20% stop-loss and cap exposure to 2–3% of portfolio.
• Event-driven watchlist: Put alerts on announcements of large contract cancellations, model-efficiency breakthroughs (>20% reduction in compute/unit) or a sustained decline in GPU spot rental prices; any such event should trigger re-evaluation and potential quick de-risking of capacity-exposed positions.