I'll research AMZN thoroughly before writing the preview. Let me start by gathering recent earnings materials, guidance, and news.Let me read the most recent earnings call transcript and release, plus get stock price context.Let me get more recent news context and analyst estimates.Let me search for analyst consensus estimates for the upcoming quarter.I now have enough to write a well-grounded preview. Here it is:
Amazon heads into this print as the last of the mega-cap hyperscalers to report, and it does so into a jittery tape. The stock closed at ~$230 on July 29, down roughly 16% from its late-May high near $274, and has slid over the past week — pressured after Alphabet raised its 2026 capex outlook to $195–205B, which reignited investor anxiety about AI infrastructure spending and returns. The debate going into the quarter is less about whether the business is healthy (it clearly is) and more about whether the payoff on ~$200B of annual capex justifies the depressed free cash flow, rising debt, and a "prove-it" market mood.
Notably, hyperscalers as a group now trade at a forward P/E below 18 vs. the S&P 500 above 20 — the market is not paying up for AI capex, so a beat alone may not be enough; the narrative on ROI and the Q3/capex guide will likely drive the reaction.
| Metric | Q2 2026 Consensus | Company Guidance (given 4/29) | Q2'25 Actual |
|---|---|---|---|
| Total revenue | ~$196.7B | $194B–$199B | — |
| EPS | ~$1.82 | — | $1.68 |
| Operating income | ~$21–22B | $20B–$24B | $19.17B |
| AWS revenue | ~$40.5B | — | — |
| AWS operating margin | ~33.8% | — | — |
| North America revenue | ~$113.8B | — | — |
The consensus $1.82 EPS implies ~8% YoY growth off last year's $1.68. Several sell-side desks (e.g., TheStreet/TipRanks aggregation) have flagged that the bar on AWS may be set too low, arguing for a potential upside surprise on cloud.
One important nuance: management guided assuming Prime Day falls in Q2 this year for the U.S. and most large geographies, whereas in 2025 Prime Day was entirely in Q3. That shift flatters Q2 comps (and pressures Q3 comps) — worth keeping in mind when judging the "beat."
Amazon set the bar high last quarter: - Revenue $181.5B, +17% (+15% FX-neutral); operating income $23.9B at a record 13.1% margin. - AWS $37.6B, +28% YoY — the fastest growth in 15 quarters, now a ~$150B annualized run-rate; AWS operating income $14.2B. - Advertising $17.2B, +22%. - North America op margin 7.9%, International 3.6%. - AWS backlog of $364B — and that figure excluded the >$100B Anthropic deal signed after quarter-end.
This is the single most important number. AWS re-accelerated to 28% in Q1 and management framed AI as a "once-in-a-lifetime" demand wave, noting a strong feedback loop where AI spend pulls through "core" (non-AI) consumption (post-training, RL, agentic tool usage). Watch: the AWS growth rate, AWS operating margin (consensus ~33.8%, off a Q1 that was flattered by efficiency gains), and any update to the backlog. A sequential deceleration would be poorly received; a hold or acceleration would validate the bulls.
Amazon guided to ~$200B of 2026 capex in February and reaffirmed it in April ("plan is largely the same"). Q1 cash capex was $43.2B. After Alphabet's hike, several analysts now expect Amazon to lift its capex guide, driven by AI, custom silicon, Leo satellites, and — critically — surging memory/DRAM prices. Watch how management frames the ROI: Jassy's pitch is that capex funds long-lived assets (30+ yr data centers, 5–6 yr chips) with attractive ROIC "a couple years after being in service," but that free cash flow is "challenged" in high-growth years. A big capex number without a compelling monetization story is the key downside risk to the stock reaction.
A genuine bright spot and increasingly a valuation driver. Q1 disclosures: chips business at a >$20B run-rate (~$50B if sold standalone), ~40% QoQ growth, and >$225B in Trainium revenue commitments, with Trainium2 "largely sold out" and Trainium3 "nearly fully subscribed." Recent signals — hiring a Google TPU veteran for Trainium software and continued large AI-lab commitments (Anthropic, OpenAI) — suggest more disclosure likely. Any hard revenue figures or margin commentary here could be a positive catalyst.
The overhang. Long-term debt jumped 81% to ~$119B between December and March, and Moody's has warned that "unprecedented" AI spending is pushing even cash-rich giants like Amazon toward debt and off-balance-sheet financing. Amazon still has enormous liquidity (~$255B current assets as of April), but FCF is being squeezed by the capex ramp. Watch the FCF trajectory and any commentary on financing.
Fundamentals point to another likely beat — AWS is re-accelerating, ads are compounding at ~20%+, retail margins are structurally improving, and the custom-silicon story is bigger than most appreciated. The risk is not the print; it's the guide and the tone on spending. With the stock already down ~16% from its highs and AI-capex fatigue running high across the group (see the SK Hynix and Alphabet reactions), the reaction likely hinges on three things: (1) whether AWS holds/accelerates near 28%, (2) whether — and by how much — the ~$200B capex guide moves higher, and (3) whether management can convincingly tie that spend to monetization (backlog, Trainium commitments, margins). A clean AWS beat with a disciplined capex message could break the stock out of its recent funk; a capex hike without a monetization story likely gets sold, given the current "prove-the-ROI" mood.
Key figures reflect Q1 2026 reported results, management's Q2 guidance issued April 29, and Street consensus/news as of late July 2026. All forward metrics are estimates and subject to change.