Here's My Top Reason to Buy Amazon Stock Right Now

AWS generated 60% of Amazon's Q2 operating income despite representing about 21% of revenue, underscoring its outsized profitability. AWS revenue grew 37%, versus 16% growth in North American commerce and 15% internationally, while its backlog reached $496 billion. Amazon is expanding data-center capacity amid AI computing demand, though CEO Andy Jassy said available capacity is unlikely to meet demand in 2026 and potentially 2027; the article argues this supports a bullish multi-year outlook for AMZN.

Analysis

The investable question is not whether AWS demand is strong, but whether incremental cloud profit can outrun the depreciation, power commitments, and lease liabilities required to relieve capacity constraints. If utilization remains tight, AWS should gain operating leverage as scarce AI capacity is allocated toward higher-value workloads; if capacity arrives ahead of monetization, consolidated free-cash-flow conversion could disappoint despite strong reported revenue. The relevant earnings catalyst is therefore AWS margin and capex guidance, not headline backlog growth.

Capacity scarcity also creates a competitive vulnerability: enterprise customers unable to secure AWS capacity can design around Azure or Google Cloud, making supply constraints a potential share-loss mechanism rather than purely a pricing opportunity. This favors infrastructure bottleneck suppliers such as VRT and ETN over hyperscalers in the next 6-18 months, since their economics improve regardless of which cloud captures the workload. NVDA remains exposed to the same capex cycle, but the marginal spend mix should increasingly rotate toward networking, power distribution, cooling, and data-center construction as accelerator availability normalizes.

Consensus appears to be treating AWS growth as unambiguously multiple-accretive while underweighting the duration mismatch between multiyear infrastructure investment and customer contract conversion. Near-term sentiment is already constructive, limiting the value of chasing a routine bullish article. A more attractive entry emerges on any post-earnings selloff driven by elevated capex, provided AWS growth, operating margin, and remaining-performance-obligation conversion remain intact; the thesis is falsified by sequential AWS deceleration combined with another upward revision to infrastructure spend or evidence of material workload diversion to MSFT/GOOGL.

Key Decisions for Investors

• Maintain a 6-12 month long AMZN bias, but add only on a capex-driven drawdown rather than on current narrative momentum; target a 2:1 upside/downside profile based on AWS margin expansion versus consolidated FCF pressure.
• Structure a relative 6-12 month trade: long AMZN / short WMT in equal beta-adjusted dollars. This isolates AWS and advertising optionality against a primarily consumer-discretionary retail margin exposure; exit if AWS growth decelerates materially for two consecutive quarters.
• Overweight VRT and ETN versus NVDA for the next 6-18 months as AI buildout spending broadens from chips into power and thermal infrastructure. Reduce if hyperscaler capex guidance falls or data-center project lead times normalize sharply.
• Set an earnings watch item for AWS revenue growth, segment operating margin, total capex, depreciation growth, and disclosed capacity availability. Do not increase AMZN exposure if capex rises faster than cloud monetization without a corresponding improvement in contract conversion or pricing.