Amazon, Alphabet, and Microsoft: 2 Cloud Giants I Am Buying in September and 1 I Am Avoiding

The article favors Alphabet and Amazon over Microsoft as AI-driven cloud demand accelerates: Google Cloud revenue grew 82% year over year in Q2 with a 36% operating margin, AWS grew 37%, and Azure grew 43%. Alphabet plans roughly $200 billion of data-center spending this year, while AWS and Google Cloud are described as showing stronger growth acceleration and companywide operating-income momentum than Microsoft. The author argues Microsoft could underperform its cloud peers unless Azure’s growth meaningfully reaccelerates.

Analysis

The actionable signal is not reported cloud growth in isolation, but whether incremental AI capacity earns above its cost of capital. GOOG has the clearest potential for multiple expansion if cloud profitability remains durable while capex converts into contracted backlog; its valuation is also less dependent on cloud than MSFT's, creating an asymmetric upside if the segment becomes a larger share of consolidated earnings. AMZN offers operating leverage through AWS, but retail/logistics and consumer demand can obscure that benefit, making it a less pure expression of the thesis.

MSFT's relative risk is a mismatch between AI infrastructure spending and monetization timing: sustained capex with stable, rather than accelerating, Azure growth would pressure free-cash-flow expectations and its premium earnings multiple over the next 1-3 quarters. That said, a simple growth-rate comparison may be misleading; Microsoft can prioritize capacity for higher-value enterprise commitments, while Copilot and bundled software monetization may appear outside reported cloud metrics. The bearish relative case is falsified by renewed Azure acceleration, rising commercial remaining-performance-obligation growth, or evidence that AI revenue is lifting gross margin rather than merely consuming capacity.

Consensus may be too quick to extrapolate capacity scarcity into lasting share gains. Over 6-18 months, hyperscaler capex convergence can shift the bottleneck from compute availability to power, networking, and enterprise implementation; this would compress GPU rental pricing and favor operators with proprietary chips, power procurement, and software distribution. Watch quarterly capex guidance versus cloud operating-margin progression: revenue growth without margin conversion is not a durable equity catalyst.

Key Decisions for Investors

• Initiate a 3-6 month relative-value position: long GOOG / short MSFT, dollar neutral, only after confirming next-quarter cloud revenue and operating-margin trends. Target 10-15% relative return; exit if Azure growth reaccelerates by more than 300bps or GOOG cloud margin declines materially despite revenue growth.
• Maintain AMZN as a smaller satellite long versus GOOG rather than a full-size cloud bet. Add only if AWS growth translates into consolidated operating-income upside while North American retail margins hold; retail demand weakness or incremental fulfillment investment can offset AWS upside.
• Do not short MSFT outright on this signal. Use the GOOG/MSFT pair or buy 3-6 month MSFT put spreads funded by selling farther-out puts, limiting exposure to a broad AI multiple expansion that could lift all hyperscalers.
• Set an earnings alert around capex-to-monetization: reduce hyperscaler longs if capex guidance rises while cloud operating margins contract for two consecutive quarters. That outcome would indicate industry overbuild and raises the case for a sector-level de-rating rather than share-specific winners.