> ## Content Index
> Fetch the complete content index at: https://www.altamiraequity.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# Amazon: The Miss Was Small — the $34.8 Billion Cash Hole Was Not
- URL: https://www.altamiraequity.com/amzn-2026-11-09/
- Published: 2026-11-09T18:06:20.000Z
- Updated: 2026-11-10T07:15:14.000Z
- Description: The miss itself was small: EPS $1.89 vs $1.97 and revenue a hair under the estimate. The $34.8 billion free-cash-flow hole, with capex at 182% of operating cash flow, is the story.
- Author: Altamira Capital
- Tags: Research, AMZN, #qc-needs-fix, Earnings

Revenue $201,944M, −0.14% vs estimate and +12.09% YoY · EPS $1.89 vs $1.97 (−4.06%) · capex $77,259M overwhelmed operating cash flow of $42,422M — free cash flow −$34,837M, and after-hours −11.22%

*Amazon.com, Inc. (AMZN, NASDAQ) · Q3 2026 (quarter ended Sep 30, 2026), reported Thu Oct 29, 2026 after the close. All figures are filed figures, traced to the sources at the end of this post.*

## The story in 30 seconds

- The miss itself was small: EPS $1.89 vs $1.97 expected (−4.06%) and revenue $201,944M vs $202,222M expected (−0.14%), with revenue still +12.09% YoY. The market's −11.22% after-hours move — against an eight-quarter average absolute move of ±6.23% — was not about the miss.
- The cash line was the story: operating cash flow of $42,422M against capex of $77,259M left free cash flow of −$34,837M for the quarter.
- The two numbers that would explain the spend — AWS growth and advertising growth — are not in the print, and the earnings-call transcript had still not posted when the underlying research closed, so this post argues only from what was filed.

Amazon has trained the market to forgive small misses. It beat earnings estimates in seven of the last eight quarters — twice by more than 70% — and revenue in all eight. So when the third quarter printed a 4.06% EPS miss and a 0.14% revenue miss, the after-hours fall of 11.22% looked, at first glance, like an overreaction to rounding. It was not. The market was reading the cash flow statement, where a single quarter of capital expenditure — $77,259M, the largest line in the entire filing — consumed the whole of operating cash flow and $34,837M more besides.

That reframes the question for this print. It is not whether Amazon grew — revenue rose 12.09% year over year to $201,944M, gross margin widened 238 basis points to 53.17%, and operating margin widened 423 basis points to 13.90%, which on derived prior-year figures implies operating income growth of roughly 61%. The question is what the growth cost, and whether the two engines that are supposed to pay for it — AWS and advertising — are accelerating fast enough to justify a capex line running at 182% of operating cash flow. On the evidence available, that question is currently unanswerable: the print carries no AWS or advertising split, and the earnings-call transcript had still not posted, leaving guidance unstated.

**In this edition:** (1) the print in one picture — where a $201,944M revenue quarter actually went; (2) what to make of it — including the quality-of-earnings line; (3) the long view — five years of margin repair against five years of free-cash-flow erosion; (4) the Key Issue — whether AWS and advertising can outgrow a capex bill that just consumed 182% of operating cash flow.

## At a glance — Q3 2026 scoreboard

This scoreboard is the quarter on one screen: what was filed, what consensus expected, and how the market reacted.

![Amazon.com, Inc.: eight quarters of EPS and revenue surprises against the next-session reaction. Source: FMP earnings and daily closes.](https://www.altamiraequity.com/content/images/2026/11/AMZN-2026-11-10-surprise_history.png)

Amazon.com, Inc.: eight quarters of EPS and revenue surprises against the next-session reaction. Source: FMP earnings and daily closes.

- **Revenue:** $201,944M — vs $202,222M est · −0.14% surprise · +12.09% YoY · +0.67% QoQ
- **EPS (diluted):** $1.89 — vs $1.97 est · −4.06% surprise · −3.08% YoY · estimate range $1.83–$2.19 (30 analysts)
- **Guidance:** not stated — the transcript had not posted at the research close, so no guidance can be reported, and none is invented here
- **Reaction:** −11.22% after-hours, to $226.03 from a $254.60 previous close · eight-quarter average absolute move ±6.23%
- **Gross margin:** 53.17% · +238 bp YoY · gross profit $107,371M on cost of revenue of $94,573M
- **Operating margin:** 13.90% · +423 bp YoY · operating income $28,076M
- **Net margin:** 10.24% · −152 bp YoY · net income $20,685M
- **Cash generation:** operating cash flow $42,422M · capex $77,259M · free cash flow −$34,837M
- **Balance sheet:** cash $55,783M · cash + short-term investments $170,009M · total debt $271,949M

**The quarter in one line:** a small miss on both headline lines, a large expansion in operating margin — and a capex bill so large it turned the quarter's free cash flow $34,837M negative, which is the only number the market actually punished.

\[CHART: flow.png — Q3 2026 income flow (filed). Caption: Revenue $201,944M became $107,371M of gross profit, $28,076M of operating income and $20,685M of net income — the operating business had an excellent quarter. Source: company filings\]

## What to make of this print

Seven readings of the quarter — each is a finding plus what it implies.

- **The miss was small; the reaction was not — and the gap between them is the information.** EPS missed by 4.06% and revenue by 0.14%, yet the stock fell 11.22% after hours, roughly 1.8 times its eight-quarter average absolute move of ±6.23%. Amazon had beaten EPS in seven of the prior eight quarters and revenue in all eight, so some of the move is a broken streak being repriced. Most of it, on the filed evidence, is the cash line: this is the first print in the set where capex alone exceeded the entire quarter's operating cash flow by $34,837M.
- **Quality of earnings — operational at the operating line, cash-negative below it.** Diluted EPS fell 3.08% YoY while the diluted share count rose 0.38% sequentially to 10,929M — slight dilution, and no buyback was disclosed to flatter the line; stock-based compensation was $5,406M, 2.68% of revenue. The operating engine itself improved: operating margin expanded 423 bp YoY. What fell was conversion: free cash flow of −$34,837M against $20,685M of net income. Label: operationally earned, deliberately unconverted — profit was real, and it was spent, in advance, on capacity whose return is not yet in the print.
- **Margin expansion is the verified half of the story.** Gross margin of 53.17% (+238 bp YoY) and operating margin of 13.90% (+423 bp YoY, +21 bp QoQ) are the consolidated signature of a mix shifting toward advertising and services, and of logistics automation compounding. On derived prior-year figures, operating income grew roughly 61% YoY. The caveat is depth: the print carries no North America segment margin, so the engine is confirmed in aggregate, not in segment detail.
- **Capex at 182% of operating cash flow is a choice, not an accident — and choices need a disclosed return.** Capex of $77,259M in one quarter was funded without strain: cash and short-term investments stand at $170,009M. Solvency is not the issue. The issue is sequencing: the spend is verified, the AWS and advertising returns that are supposed to justify it are not — neither segment figure is in the print, and the transcript that would carry them has not posted.
- **The balance sheet is why this is strain, not breakage.** Total debt of $271,949M looks large until it is set against $170,009M of cash and short-term investments and the company's long record of operating cash generation. Amazon can fund several more quarters of this capex pace from liquidity and operating flow; the question the Key Issue weighs is whether it should have to.
- **The streak mattered, and it is over.** The eight-quarter history shows EPS surprises of +215.93%, +70.55%, −1.02%, +24.20%, +28.24%, +16.06%, +24.83% and +25.44% before this quarter's −4.06%, with revenue beating in every one. Part of Amazon's market value was a reliability premium. A single small miss does not erase the record, but it transfers the burden of proof: the next print must re-establish the pattern while also showing the capex curve bending.
- **The quarter's demand context arrived before Amazon reported.** Both of Amazon's cloud rivals had already printed accelerating cloud growth this earnings season (covered in the Key Issue below). Amazon reported into a market that had seen the demand — which made an unverified AWS line, next to a $77,259M capex bill, the hardest possible combination to defend on the night.

**Takeaway:** the operating business had an excellent quarter and the cash flow statement had a historic one; until segment disclosure shows what the $77,259M is buying, both statements are true and neither settles the argument.

## How Amazon makes money

This section shows where revenue actually comes from — the mix decides which growth rates matter.

![Amazon.com, Inc.: how the latest quarter's revenue became operating income and free cash flow. Source: FMP income and cash-flow statements.](https://www.altamiraequity.com/content/images/2026/11/AMZN-2026-11-10-income_flow.png)

Amazon.com, Inc.: how the latest quarter's revenue became operating income and free cash flow. Source: FMP income and cash-flow statements.

**Sector primer — hyperscale cloud and platform retail.** Amazon competes in two linked industries. In cloud infrastructure, hyperscalers sell computing, storage and now AI capacity from vast data-centre fleets; economics turn on utilisation, the depreciation life of accelerators, and whether AI demand compounds faster than capacity is added — the capex cycle is the industry. In platform retail, the marketplace takes a margin on other sellers' volume, layers high-margin advertising and subscriptions over first-party sales, and uses logistics density as the moat; consolidated margin therefore rises mechanically as the services mix grows. The named peers for the cloud read-across used in this post are Microsoft (Azure) and Alphabet (Google Cloud); on the retail side, Walmart and Costco are the physical-scale comparators.

Amazon's company profile structures the business in three segments — North America, International, and Amazon Web Services (AWS) — spanning merchandise and content sold directly, third-party seller services, devices (Kindle, Fire, Ring, Blink, eero, Echo), media production, and the cloud business. This quarter's available set, however, contains **no segment split**: no segment revenue or segment margin for Q3 2026 is stated in this post, and that absence is flagged as its central disclosure gap. The post argues only from the consolidated filed lines: revenue of $201,944M at a 53.17% gross margin, with the mix effect visible indirectly in that margin's 238 bp YoY expansion.

\[CHART: operating-margin.png — Q3 2026 margins (filed). Caption: A 53.17% gross margin at Amazon's scale is the services mix made visible; the 13.90% operating margin is up 423 bp YoY. Source: company filings\]

## The long view

One quarter is weather; five years are climate. Amazon's long view is a tale of two repairs: margins rebuilt year after year, free cash flow sacrificed to the build-out that made the margins possible.

![Amazon.com, Inc.: the long view — annual revenue, margins and returns on capital. Source: FMP annual statements.](https://www.altamiraequity.com/content/images/2026/11/AMZN-2026-11-10-long_view.png)

Amazon.com, Inc.: the long view — annual revenue, margins and returns on capital. Source: FMP annual statements.

### Revenue growth

Growth is the first quality test, because a company compounding revenue in double digits can fund experiments a stagnant one cannot. Filed fiscal revenue rose from $513,983M (FY2022) to $574,785M (FY2023), $637,959M (FY2024) and $716,924M (FY2025) — a five-year compound growth rate of 13.18%, with the most recent three-year rate at 11.73%, a mild deceleration. This quarter's +12.09% YoY sits between the two windows: no re-acceleration, no fade. **Judgment: growth passes — durable double-digit compounding at a scale no peer matches in absolute dollars, with the acceleration question now belonging entirely to AWS, the segment this print does not disclose.**

![Amazon.com, Inc.: quarterly revenue and year-on-year growth. Source: FMP income statements.](https://www.altamiraequity.com/content/images/2026/11/AMZN-2026-11-10-revenue_trend.png)

Amazon.com, Inc.: quarterly revenue and year-on-year growth. Source: FMP income statements.

\[CHART: revenue-growth.png — Amazon revenue, filed fiscal years. Caption: The top line compounded steadily through the margin-repair years — growth was never the problem this print poses. Source: company filings\]

### Operating margin

Margin tells you whether scale is converting into operating leverage or being competed away. The filed annual record is a repair story: operating margin of 2.38% (FY2022), 6.41% (FY2023), 10.75% (FY2024) and 11.16% (FY2025), reaching 13.90% this quarter (+423 bp YoY). **Judgment: margin passes outright — the strongest verified trend in this post.**

### Cash — free cash flow margin and conversion

Cash is where strategy becomes arithmetic. Over five years, Amazon's operating cash flow has run far ahead of its reported profit — earnings are thoroughly cash-backed at the operating line — while free cash flow has compounded negatively over the same five-year window, because capex has grown faster than the cash it serves. This quarter is the pattern at full stretch: operating cash flow of $42,422M, capex of $77,259M, free cash flow of −$34,837M. **Judgment: cash fails on the only definition readers can spend — free cash flow — while passing on operating cash quality; the entire debate is which of those two lines the next two years will vindicate.**

\[CHART: fcf.png — Q3 2026 cash: capex overwhelmed operating cash flow (filed). Caption: Capex of $77,259M was 182% of operating cash flow — the quarter in one chart. Source: company filings\]

### Returns on capital

Returns on capital ask whether the build-out earns its keep. Amazon's five-year average return on invested capital sits in the high single digits — respectable for a retailer, inadequate for the capital intensity Amazon has chosen, and weighed down by a capital base swollen by exactly the capex supercycle this quarter extends — while earnings per share have compounded rapidly off the depressed FY2022 base, a filed net loss year. **Judgment: returns on capital are the long view's open wound, improvable only if AWS monetises the capacity now being installed.**

## Beyond the print — the business this quarter

The earnings release is the scoreboard; this section is what changed in the business itself during and around the quarter, each item traced to the sources at the end.

- **The capex line is itself the quarter's biggest business development (filed).** $77,259M of capital expenditure in a single quarter — 38.3% of revenue — is a strategic statement no press release needed to make: Amazon is building capacity ahead of disclosed demand. Depreciation and amortisation of $20,949M in the quarter shows the installed base already compounding.
- **Equity compensation remains a real cost of the model (filed).** Stock-based compensation of $5,406M — 2.68% of revenue — alongside a diluted share count up 0.38% sequentially. No buyback or dividend is disclosed in the print: capital return this quarter was zero by design, with every available dollar routed to capacity.
- **The cloud race's external scoreboard moved first (verified in this publication's own earlier editions).** Microsoft's quarter, featured here on Nov 2, disclosed Azure revenue of $33,170M growing 48.2%; Alphabet's quarter, featured Oct 30, showed Google Cloud revenue up 103% to $30.7B with a $561B backlog. Amazon reported into a market that had already seen both rivals accelerate and spend.
- **The disclosure gap itself is a fact.** When the underlying research closed, the earnings-call transcript had still not posted, guidance was unstated, and AWS growth, advertising growth and the capex path were all unverified. For a company whose investment case now rests on two segment numbers, the absence of both is this post's central limitation.

## What management said — filing-only this edition

This section normally reports what management guided and claimed, in their own words. This edition cannot — and says so rather than filling the space.

**No verbatim management commentary is available for Q3 2026 in the traced set.** The earnings-call transcript had not posted when the underlying research closed its definitive edition, so no quotes are reconstructed from summaries and no guidance is inferred: guidance is **not stated**, and this section is filing-only. The filed facts on record this quarter are the statements themselves: revenue of $201,944M, operating income of $28,076M, capex of $77,259M, and a balance sheet carrying $170,009M of cash and short-term investments.

*Remainder box: guidance — not stated (transcript pending); margins — gross 53.17%, operating 13.90%, net 10.24%; capital return — none disclosed (no buyback, no dividend), stock-based compensation $5,406M, diluted shares +0.38% QoQ; Q&A colour — unavailable (no transcript). When the transcript posts, the three disclosures that matter are AWS growth with its AI revenue run-rate, advertising growth, and a capex path that brings free cash flow back toward positive.*

## The Key Issue — can AWS and advertising outgrow a capex bill that just consumed 182% of operating cash flow?

**Thesis:** Amazon's operating business just had one of its best margin quarters on record while its cash flow statement had one of its worst — and both facts trace to the same decision: build AI capacity ahead of proof. The spend is fundable ($170,009M of cash and short-term investments; modest leverage against earnings) and the demand backdrop is real (both cloud rivals accelerated this quarter). What is missing is Amazon's own evidence: no AWS growth rate, no advertising growth rate, no guidance. The question is not whether Amazon can afford the build — it is whether the build is buying growth the print cannot yet show.

**Evidence for (the spend is buying something real):** operating margin expanded 423 bp YoY to 13.90% with derived operating income growth of roughly 61% — the retail-and-mix engine is demonstrably working; revenue grew 12.09% YoY at $201,944M of scale; the company's long record of operating cash generation is prolific before capex; the balance sheet carries the cycle without strain; and the external demand signal is verified in this publication's own traced editions — Azure +48.2% and Google Cloud +103% this earnings season say enterprise AI demand is compounding, not pausing.

**Evidence against (the bill is arriving before the proof):** free cash flow of −$34,837M is the quarter's fact, not a framing, and the five-year free-cash-flow trend has compounded negatively, so this is a pattern reaching an extreme, not a one-quarter blip; AWS and advertising — the two engines the spend serves — are both unverified in the print (no segment split, transcript unposted, guidance unstated); the eight-quarter beat streak broke on both lines simultaneously; five-year returns on invested capital sit in the high single digits, so the existing capital base has not yet earned a premium return; and the market's −11.22% reaction, the largest in the recent set against a ±6.23% average, shows holders pricing the same gap.

**What to watch:** (1) The transcript when it posts — AWS growth and its AI run-rate, advertising growth, and any capex guide are the three disclosures that settle this question. (2) The Q4 print's cash line: a second consecutive quarter of capex overwhelming operating cash flow without AWS acceleration would deepen the strain materially. (3) *External datapoint:* Azure and Google Cloud growth in the next earnings round — the verified read-across for the AI demand Amazon is building into; a rival deceleration would undercut the build's premise from outside. (4) The depreciation trajectory: $20,949M this quarter is the leading edge of the cost the new capacity will carry whether or not revenue follows. (5) *Historical analogy:* Amazon has run this experiment before — the FY2022 overbuild year in the filed record shows an operating margin of 2.38% and a net loss of −$2,722M when capacity outran demand, followed by three years of margin repair to today's 13.90%. The analogy cuts both ways: Amazon has proven it can sweat an overbuild into record margins, and proven it can overbuild. Which precedent governs depends on the AWS number that has not posted yet. **The forward question:** when AWS finally discloses, will its growth rate be large enough to make $77,259M of quarterly capex look early — or merely expensive?

---

*Sources: Amazon Q3 2026 filed financial statements and consensus estimates via Financial Modeling Prep (scoreboard, income statement, margins, cash flow, balance sheet, market reaction, eight-quarter surprise history); filed fiscal-year statements FY2022–FY2025 for the long view; Microsoft and Alphabet editions of this publication (Oct 30 and Nov 2, 2026) for the cloud read-across. Gaps, stated rather than filled: no segment split in the print (AWS, North America and advertising unverified); guidance not stated and no verbatim call commentary (transcript unposted at the research close); no consensus price target was available from the estimate sources checked.*

*Disclosure: This post is analysis, not financial advice. The author may hold positions in companies discussed.*