New quarterly report expected Apr 29, 2026. Report will refresh after results.
A
Amazon.com, Inc.
AMZN · Consumer Cyclical
Q4 FY2025
Earning ended Dec 31, 2025Report generated Mar 7, 2026
The Toll Booth Empire Temporarily Blind to Its Own Cash
The market is debating whether Amazon is expensive on FCF — but FCF is being voluntarily suppressed, and the more consequential question is whether the advertising segment, not AWS, is the next decade's fastest-compounding margin engine, because it monetizes the highest-value signal in all of marketing without building a single data center.
Flowly Score
Apr 16 Close
-20%$249.70
Fair Value
$200.00
65
Strongat $249.70
Business Strengthstrong
Three structurally distinct moat types — logistics scale, enterprise switching costs, and two-sided network effects — run through a single self-reinforcing flywheel, with the most defensible and fastest-growing segments (AWS, advertising) compounding their advantages faster than the retail core can dilute them. The advertising business monetizing purchase intent at the transaction layer may be the most underappreciated moat in the entire portfolio.
Financial Resiliencemoderate
Operating cash flow quality is unimpeachable — it ran strongly positive even in the year reported earnings went negative, which is the fingerprint of a genuinely cash-generative business. But the voluntary FCF near-collapse due to the AI capex wave means the company is funding its future from its present, and the Piotroski and Altman scores reflect a balance sheet that is carrying real leverage against a temporarily suppressed free cash flow line.
Growth & Trajectorystrong
AWS re-accelerating to its fastest growth rate in thirteen quarters — at a scale where adding sequential revenue rivals the entire annual revenue of many S&P 500 companies — is the most important data point in this report, and it suggests the AI infrastructure cycle is pulling forward demand rather than cannibalizing existing workloads. Advertising at similar growth rates, retail margins expanding structurally, and the mix shift toward high-return segments all point to a business whose earnings power is compounding well above its revenue growth rate.
Valuationweak
The FCF yield is effectively a rounding error and the EV/FCF multiple is uninvestable on its face — but both of these are artifacts of voluntarily suppressed free cash flow, not structural impairment, which makes the P/E a more honest anchor for this moment in the capex cycle. The real valuation question is the timing and magnitude of the FCF normalization: if the capex build matures on the schedule management implies, the current multiple compresses rapidly on forward earnings; if it doesn't, the premium has nowhere to go.
Riskmoderate
The single most dangerous scenario is that the AI infrastructure arms race is structurally permanent rather than cyclical — if AWS must sustain capex at roughly double D&A not as a build phase but as the ongoing cost of competitive parity against Azure's OpenAI integration, then normalized FCF never materializes at the level implied by current ROIC trends and the entire investment thesis rests on a foundation that cannot be verified until it's too late to matter. The antitrust and regulatory overhang targeting the marketplace flywheel adds a second tail risk that is genuinely difficult to handicap.
Amazon is one of the rare businesses where the reported financials actively mislead you about the underlying quality. The retail segment creates the optical illusion of a thin-margin, capital-intensive conglomerate, while the actual profit engine — a cloud infrastructure monopoly with enterprise switching costs so deep they span multi-year migrations, fused with an advertising platform that captures purchase intent at the moment of transaction — quietly compounds behind it. The P/E multiple, sitting well below its five-year average despite accelerating AWS growth and expanding operating margins, suggests the market is still anchoring to the retail optics rather than pricing the mix-shift that is already underway. That's the quality-price interaction worth paying attention to. This business is heading toward a smaller but structurally more valuable version of itself: fewer SKUs Amazon sells directly, more rent Amazon extracts from the commerce and cloud ecosystems it controls. Every dollar of revenue that shifts from online stores to advertising or AWS drops to the bottom line at multiples of the retail margin, which means earnings can grow substantially faster than revenue for years without any new product or market. The AWS backlog, growing at a rate that dwarfs the actual revenue run rate, is the most credible forward indicator in the dataset — that is not speculative demand, it is signed contracts. The single most specific risk that would break the thesis is a capex cycle that refuses to plateau. If the gap between capital expenditure and depreciation is not a temporary build phase but the permanent structural cost of competing in AI infrastructure — because Azure's OpenAI integration or Google's Gemini architecture proves natively superior for the workloads that matter most — then normalized free cash flow never arrives at the levels the current multiple requires. A business that must outspend its depreciation indefinitely to maintain competitive position is not compounding for shareholders; it is running a technological treadmill, and the current price leaves almost no room for that outcome.