Think AWS Is Losing To Azure and Google Cloud? You Need To Hear This Quote From Amazon CEO Andy Jassy

Amazon reported AWS revenue growth of 24% in 2025 — its fastest in 13 quarters — on a $142 billion annualized run rate, adding $21.2 billion of AWS revenue while AWS operating income rose to $45.6 billion (versus $13.9 billion for Google Cloud). Consolidated quarterly revenue rose 14% to $213.4 billion and operating income increased 18% to $25 billion, but management guided to $200 billion of capital expenditures mostly for AWS and AI workloads; with $139.5 billion of operating cash flow in 2025 that capex target will likely drive negative free cash flow in 2026 and triggered a post-earnings stock selloff. Management defended AWS leadership against faster-percentage growth at Google Cloud and Azure, but investors remain cautious until the spending demonstrably improves returns.

Analysis

Market structure: AWS remains the incumbent cash engine (24% growth on a ~$142B run rate, $45.6B operating income) which preserves pricing power vs. Google Cloud and Azure despite share momentum to competitors. Winners: GPU/AI supply chain (NVDA), AWS in-house silicon ecosystem (ARM/Graviton beneficiaries, pressuring INTC), and utility/energy suppliers in AWS regions; losers: wholesale colocation REITs (EQIX/DLR) and incumbent x86 CPU vendors. Increased capacity guidance ($200B capex) signals a near-term supply push that should compress spot cloud pricing and raise hardware demand for 12–36 months while pressuring free cash flow in 2026.

Risk assessment: Tail risks include a macro slowdown that shaves enterprise cloud spend (-10–20% scenario), execution failure on data center builds, or adverse export/antitrust actions within 6–24 months that could restrict AI chip supply/use. Immediate (days) risk is volatility around guidance updates; short-term (3–12 months) is FCF and margin compression; long-term (1–3 years) is market-share shifts and potential regulatory breakup. Hidden dependency: AWS’s ROI hinges on workload migration to its Trainium/Graviton stack — if customers resist, capex becomes stranded.

Trade implications: Favor asymmetric exposures: long NVDA 6–12 month call spreads to capture GPU demand, selectively accumulate AMZN on >15% post-earnings weakness with 12–36 month horizon, and hedge by shorting colocation REITs (EQIX/DLR) via put spreads 6–12 months. Use relative-value pairs (long MSFT vs short AMZN) over 6–12 months to express preference for cash-generative cloud growth without heavy new-capex risk. Options: sell premium into H1 2026 earnings and buy 9–18 month LEAPs on AMZN/NVDA to capture structural AI upside while limiting time decay.

Contrarian angle: Consensus fears of capex dilution may be overstated — Amazon’s past capex cycles (warehouse, prior cloud buildouts) produced multi-year moat gains; if AWS converts even 20–30% of new capacity into differentiated AI services, unit economics and market share could re-accelerate within 18–36 months. Market likely overprices short-term FCF pain while underpricing the option value of proprietary chips (Trainium/Graviton) and integrated stack. Watch for 1) signs of enterprise migration to Graviton (revenue mix shifts) and 2) incremental pricing moves from Google/Microsoft as triggers to reassess exposure.

Key Decisions for Investors

• Establish a 2–3% portfolio long in AMZN, scaling in on any >15% post-earnings decline over the next 4–8 weeks; target 30–40% upside over 12–36 months, place stop-loss at -20% or if AWS operating margin falls >200bps YoY.
• Deploy 1.5–3% notional into NVDA via 6–12 month call spreads (buy 1–2 strikes OTM, sell nearer-term OTM) to capture sustained GPU demand from hyperscalers; take profits at >25–35% gain or if NVDA revenue guidance misses by >5% sequentially.
• Open a 1–2% short via put spreads on colocation REITs (e.g., EQIX, DLR) with 6–12 month expiries to express negative view on outsourced data-center demand; add size if AWS announces new region/mega-cluster builds in next 3–9 months.
• Implement a pair trade: long MSFT (1–2%) and short AMZN (1–2%) over 6–12 months to prefer cloud growth with lower capex risk; unwind if MSFT guidance decelerates >300bps or AMZN demonstrates consistent FCF recovery over two consecutive quarters.