Amazon earnings recap: AWS comes through big as Q1 tops expectations

Amazon beat Q1 expectations with revenue of $181.5B versus $177.23B expected and EPS of $2.78 versus $1.62, while AWS sales jumped 28% to $37.6B, its fastest growth in 15 quarters. However, free cash flow fell 95% as capex reached $43.2B and management gave no fresh capex guidance, leaving AI spend and returns a key investor concern. Q2 guidance was solid on sales ($194B-$199B) but below consensus on operating income ($20B-$24B vs $22.86B).

Analysis

AMZN’s print is less about top-line surprise than about proving the AI capex flywheel is still working: accelerating AWS growth alongside a near-term FCF squeeze implies the market is being asked to underwrite a longer payback period. That is usually fine in the first phase of an infrastructure cycle, but it becomes fragile if incremental AWS margin stops expanding over the next 1-2 quarters, because the valuation support is increasingly tied to eventual operating leverage rather than current cash generation.

The more important second-order effect is competitive pressure on the AI stack. Amazon is not just monetizing AWS demand; it is moving up the value chain with custom silicon and model distribution, which puts pressure on Nvidia at the margin and raises the stakes for Microsoft’s Azure AI narrative. If Trainium becomes a credible external rack product, the market may start valuing Amazon less like a cloud retailer and more like a vertically integrated AI infrastructure vendor, which would be a multiple-positive over 6-12 months.

Near term, the stock likely trades on whether investors believe the capex curve is still ahead of revenue or whether the spend is finally translating into visible revenue acceleration. The key risk is that guidance stays conservative while fuel, logistics, and memory/storage constraints keep suppressing cash conversion, creating a “great quarter, poor FCF” setup that can cap multiple expansion even if the shares gap up initially. Conversely, a sustained 30%+ AWS growth cadence for two quarters would force the Street to re-rate the whole complex.

The contrarian angle: consensus is already leaning too hard on the idea that every AI dollar spent is self-funding. The market may be underestimating how much of Amazon’s near-term strength is being pulled forward by internal AI productivity and Anthropic marks, which are not perfectly repeatable operating drivers. That means the risk/reward is better expressed relative to peers than outright naked longs.

Key Decisions for Investors

• Stay long AMZN on any post-earnings consolidation, but size for a 1-2 quarter FCF drawdown narrative; risk/reward is favorable if AWS holds >25% growth and the Street lifts 2026 revenue assumptions.
• Pair trade: long AMZN / short MSFT over the next 1-3 months to express stronger relative AI infrastructure momentum and custom silicon optionality versus Azure’s more mature base.
• Add a tactical short NVDA hedge against AMZN upside into the next 4-8 weeks; if Trainium rack sales gain credibility, marginal share shift away from merchant GPUs could pressure sentiment even if fundamentals remain strong.
• Consider AMZN Jan-2027 call spreads rather than outright equity for a 6-12 month view; the setup is a multiple re-rating story, but capex/F CF volatility argues for defined downside.
• Watch for any 2Q AWS margin commentary; if q/q margin expansion disappoints, fade the initial rally and reduce longs, because the capex-return debate will likely re-open quickly.