Amazon Fell 7% on a $77B AI Spending Quarter: the Halal Investor's Read (2026)
The third act of big technology's earnings week arrived after the close on Thursday 29 October 2026, and it rhymed with the second. Amazon reported its fastest cloud growth in nearly five years, its highest operating margin ever, and revenue up 12% to 201.9 billion dollars — and its shares still fell about 7.3% in after-hours trading. The reason was the same one that knocked Alphabet down the night before: the market is no longer grading artificial intelligence ambition on trust. It is grading the bill. Amazon spent 77.3 billion dollars on property and equipment in a single quarter, more than double a year earlier, reaffirmed a full-year capital spending plan of about 220 billion dollars, and reported trailing free cash flow of negative 42.4 billion dollars against a positive 14.8 billion a year ago. This article reports what Amazon actually printed, why a record-margin quarter still sold off, how the story connects to Alphabet's fall and Microsoft's rise earlier in the week, and how a halal investor should read it: through the screen first, the price second.
Updated October 2026 · Reviewed by the Canadian Halal Investor team
What happened overnight
Thursday 29 October 2026 was already a heavy day before Amazon reported. The technology sell-off that began with Alphabet's earnings miss on Wednesday carried into the regular session: the Nasdaq Composite fell 1.38% to 26,184.72, the S&P 500 slipped 0.3% to 7,720.39, and the Dow Jones Industrial Average fell 132.54 points, or 0.25%, to 52,342.56. Alphabet's slide continued, with reporting describing a drop of around 12% on Thursday, its worst day on record, after the Google parent missed expectations and posted negative free cash flow. Caterpillar tumbled more than 9%, its steepest daily decline since March 2020, after construction equipment sales fell slightly short of expectations despite stronger overall results. The market, in short, was already in a mood to punish any print that did not clear a high bar — and Amazon's print, good as much of it was, did not quite clear it.
After the close, Amazon reported net sales of 201.9 billion dollars, up 12% from 180.2 billion a year earlier, against roughly 202.3 billion expected, and diluted earnings per share of 1.89 dollars against 1.96 expected. Net income slipped to 20.7 billion dollars from 21.2 billion a year earlier — but that comparison flatters the year-ago quarter, which included 9.5 billion dollars of pre-tax gains on Amazon's investments in Anthropic. Strip out that one-off, and the operating picture improved sharply: operating income climbed 61% to 28.1 billion dollars from 17.4 billion, a margin of 13.9% that the company describes as its highest ever. The shares had already fallen 4.5% in regular trading on Thursday before the results, closing at 243.79 dollars, and then fell a further 7.3% in after-hours trading to about 225.99 dollars. The two-step fall matters for how the story reads: this was not a single headline reaction, but a market that was already de-risking technology names and then found a second reason to sell.
Guidance added to the caution rather than relieving it. For the holiday quarter, Amazon projected net sales of 240 billion to 247 billion dollars, a midpoint roughly in line with the roughly 244.5 billion analysts expected, and operating income of 28.5 billion to 33.5 billion dollars against an average expectation of about 33 billion — the top of the range, in other words, barely reaches what analysts had pencilled in. Chief executive Andy Jassy told the earnings call that the company's major business segments are performing well, and said he believes AWS will grow at levels above 40% next year and even in 2028. The market heard the spending plan more loudly than the growth promise, and that is the pattern of this whole week: growth is assumed, funding is questioned.
AWS won the quarter and still lost the night
If any single segment deserved a celebration, it was Amazon Web Services. AWS revenue grew 39% to 46.0 billion dollars, beating the roughly 45.3 billion expected, its sixth consecutive quarter of accelerating growth and its fastest in nearly five years. AWS operating income rose to 17.8 billion dollars from 11.4 billion a year earlier. On the call, Jassy pointed to rising demand for AI computing as the driver, and the segment's acceleration fits the same demand story that lifted Microsoft's Azure 48% and Alphabet's Google Cloud 103% earlier in the week. All three cloud businesses are growing fast; the AI demand story, on the revenue side, is not in doubt.
Everywhere else, the quarter was merely solid — and in a week when merely solid is punished, that was enough to sell. Online store sales grew just 2% to 68.5 billion dollars, the slowest growth since early 2023, when Amazon was navigating the post-pandemic decline in online retail. Advertising revenue rose 11% to 19.64 billion dollars, below the roughly 20.82 billion expected, and the slowest growth since Amazon began reporting advertising separately in 2020. Part of the e-commerce softness is timing: Amazon moved its big summer sale from July to June this year, pulling a chunk of sales into the earlier quarter. But timing explains a quarter, not a trend, and the market read the two slowest-growth figures in years as evidence that the consumer side of Amazon is maturing while the cloud side carries the growth — and the cloud side is precisely where the spending is concentrated.
The contrast with the other two cloud giants sharpens the point. Microsoft beat on revenue and earnings, grew Azure 48%, and still generated positive free cash flow of 18.6 billion dollars — and rose. Alphabet grew cloud 103% but missed on revenue and earnings, spent 51.4 billion dollars, and posted negative free cash flow of 9.07 billion dollars — and fell about 10% in premarket trading, as covered in yesterday's article. Amazon now completes the set: the fastest AWS growth in nearly five years, a record operating margin, and still a 7% after-hours fall, because the market's question this week is not "is the cloud growing?" but "what is the growth costing, and who is paying for it?" Canaccord, defending the stock after the drop, called the sell-off a "spend-scare overreaction, not a thesis break," pointing to genuine AWS demand rather than excess capacity. That may prove right — but it is a judgment about the future returns on today's spending, and the market spent Thursday pricing the spending itself.
The 77 billion dollar bill: capital spending, cash flow and borrowing
The numbers behind the sell-off deserve to be stated plainly, because they are large enough to lose meaning in headlines. Amazon purchased 77.3 billion dollars of property and equipment in the third quarter alone, more than double the 35.1 billion of a year earlier, and well above the roughly 58.7 billion analysts had expected. Over the trailing twelve months, operating cash flow grew 29% to 168.3 billion dollars — a genuinely strong figure — but net capital spending over the same period reached 210.7 billion dollars. Subtract the second from the first and you get the headline that moved the stock: free cash flow of negative 42.4 billion dollars, against a positive 14.8 billion a year ago. The company reaffirmed plans to spend about 220 billion dollars in 2026, mostly on AI chips and data centres, with a significant increase expected in 2027. The buildout, in other words, is not a one-quarter event; it is a multi-year programme whose bill is arriving faster than its cash returns.
None of this means Amazon is short of cash in any ordinary sense. A company generating 168.3 billion dollars of operating cash flow over twelve months is not in distress, and capital spending on data centres and processors buys long-lived assets whose returns arrive over years, not quarters. The question the market is asking is narrower and harder: whether the returns on this particular vintage of spending will exceed its cost, and how the gap between spending and operating cash flow is funded in the meantime. That funding question is where an earnings story becomes a balance-sheet story. Amazon's interest expense in the quarter was 1.7 billion dollars, up from 538 million a year earlier, as the company has borrowed more — a tripling that tells its own story about how part of the gap is being closed. Borrowing to build productive assets is normal corporate finance, not a red flag by itself. But every dollar borrowed adds to interest-bearing debt, and every dollar of interest paid is a cost that the new assets must eventually cover.
There is also a mechanical point that matters for screened investors, and it is easy to miss in the noise of a 7% fall. The debt screen measures interest-bearing debt against market capitalisation, so a falling share price raises the ratio even if debt does not change at all: the numerator stays put while the denominator shrinks. A 7% fall does not move Amazon anywhere near its ceiling, as the next section shows, but the direction is worth understanding, because it means price falls and debt ratios move in the same direction at exactly the moment investors are most nervous. The right response is not to panic about the ratio — it is to re-check it at the next screening update, when both the new debt figure and the new market capitalisation are in the data, rather than guessing at it from a single night's price move.
For Canadian readers, one more translation is needed. All of Amazon's figures are in US dollars, so a Canadian investor's return in Canadian dollars also reflects the exchange rate — a currency effect, not a compliance effect, and one that cuts both ways over time. And the holiday quarter now becomes the next test of the whole thesis: if AWS growth holds above 40%, as Jassy suggested it could next year, and if the consumer business steadies into the holiday season, the spending story gets easier to tell. If the holiday quarter's operating income lands near the bottom of the 28.5 to 33.5 billion dollar guided range, the market will ask the funding question louder. Either way, the answer will come from the next print, not from Thursday night's price.
What the Shariah screen says about Amazon
Here is the question this article exists to answer: does a 7% fall change whether Amazon is halal to hold? On this site's October 2026 screening, the answer is no — and the reasons are worth stating precisely, because they show what the screen measures and what it does not.
The screen has two gates, applied in order, as documented on our methodology page. Gate one is business activity: is the company's core business permissible? Amazon's business — online retail, cloud computing and advertising — passes, and our Amazon screener explains the question investors most often ask, about the co-branded credit card: the business being screened is the retail, cloud and advertising operation, a permissible commercial activity, not a lending business, and the card is a partnership product rather than Amazon's core activity. Nothing in Thursday's earnings report changes that business mix — AWS grew fastest, but retail and advertising are still the same businesses they were — so gate one is untouched by the sell-off.
Gate two applies financial ratios, and here the October 2026 figures leave Amazon far from its ceilings. Amazon carries interest-bearing debt of 3.03% of market capitalisation and interest income of 0.61% of revenue. The ceilings providers use are roughly 33% for debt and cash and 5% for non-compliant income — though providers differ, with some using a 30% debt ceiling rather than about 33%, a disagreement this site reports rather than resolves. Against either ceiling, Amazon passes by a wide margin: its debt ratio would need to rise roughly tenfold to approach the ceiling. A single quarter's borrowing, even one in which interest expense tripled, does not close a gap of that size. What it does do is move the direction of the debt ratio, which is why the ratio deserves a re-check at the next screening update rather than an assumption that it stands still — and why the falling share price, which shrinks the market-capitalisation denominator, is part of that re-check, not a separate worry.
The contrast with this week's earlier stories sharpens the point. Celestica surged 11.9% on AI earnings on Tuesday, yet fails this site's screen at gate one because of its Aerospace and Defense business, whatever its ratios say. Broadcom's 63 billion dollar AI deal raised the same funding question as Amazon's quarter: how much of the AI buildout is debt-funded, and what does that do to balance sheets over time. Amazon is the mirror image of Celestica: it passes the screen, and the market's verdict is about cash conversion and funding, not compliance. There is also a practical footnote for dividend-focused readers: Amazon pays no dividend, so there is nothing to purify for dividend holders, as the screener explains — the 0.61% interest income figure is what the 5% income screen measures at the company level, and it clears comfortably.
For Canadian holders, two practical notes complete the picture. First, registered accounts do not change any of this: a TFSA or RRSP changes the tax treatment of what sits inside it, not the compliance of the holding, as our halal RRSP guide explains. Second, for investors who want to verify the result themselves rather than take this article's word for it, the screener hub lists every screened company with its ratios, and the screener roundup compares the independent apps that run the same gates with their own data. A screen result is only as good as its inputs, and the inputs here — debt, market capitalisation, interest income — are all due for a fresh look when the next update lands.
An earnings-day checklist for a halal investor
Earnings season produces more commentary than almost any other market event, and most of it is noise for a screened investor. A short checklist, run in order, filters it. None of these steps is a buy or sell call; together they are a decision procedure for any earnings day, including this one.
- Separate the print from the price. Amazon's sales, AWS growth, capital spending and free cash flow are facts; the 7.3% after-hours fall is the market's reaction to them. Read the print first, the price second, and never let the size of the move substitute for the contents of the report.
- Check the screen before the chart. Run the ticker through the screener hub or one of the apps in the screener roundup before drawing any conclusion from a price move. A falling price is not a failed screen, and a rising price is not a passed one — Celestica this week proved the second half of that sentence.
- Ask how the growth is funded. The question Thursday's verdict turned on was not whether AWS is growing — it grew 39% — but whether growth is funded from operations. Trailing operating cash flow of 168.3 billion dollars against capital spending of 210.7 billion leaves a gap, and the interest expense tripling to 1.7 billion dollars shows part of that gap is being borrowed. The debt gate is where that story eventually shows up.
- Watch the ratio inputs, not just the ratio. New borrowing, market capitalisation and interest income all move between screening updates. A company far from its ceilings today can drift toward them over several quarters; a re-check at each update catches drift that a single earnings reaction cannot.
- Leave the interest-bearing shelf alone. Nothing in Amazon's report makes an interest-bearing product permissible, and nothing in a technology sell-off changes that rule either. Earnings volatility is precisely when the discipline of staying inside screened, asset-backed holdings matters most, because the temptation to reach for yield or leverage is strongest when prices move fast.
- Write down what you will watch, then stop watching. For Amazon, that is the holiday-quarter print against the 240 to 247 billion dollar sales guide and the 28.5 to 33.5 billion dollar operating income range, whether AWS growth holds near 40%, and the next screening update's debt ratio. Note those markers, note what would actually change your plan — a screen result changing — and ignore the hourly commentary in between.
That last step is the one most investors skip, and it is the one earnings days most reward. A portfolio reviewed on a short list of markers, and otherwise left alone, will beat a portfolio reviewed on every after-hours tick. The screen does the compliance work; the calendar does the patience work.
What to watch next
The rest of the week and the weekend give the framework several live tests. Apple reports after the close on Monday 2 November, the last mega-cap print of this earnings round, and it will deserve the same treatment as Amazon's: separate fact from expectation, check the screen before the price, and decide on the business rather than the move. This morning, before the open, ExxonMobil and Chevron both beat quarterly profit expectations on soaring refining margins, with Exxon shares up about 2% and Chevron up about 1% in premarket trading — a reminder that earnings season is not only a technology story, and that energy names, which many screened investors exclude at gate one or hold only through screened funds, play by different drivers entirely. The Federal Reserve's December meeting on 8 to 9 December is the next macro marker, with markets pricing a high chance of a further increase; whatever data arrives between now and then will be parsed for what it says about that meeting, and none of that intraday noise changes a screen result by itself.
For the running list of every trending story, see the market news page, and for the split verdict that set this week's pattern, see Alphabet fell 10% as Microsoft beat: the AI spending test through a halal lens. Earnings season will come again next quarter, and the checklist above will still apply, unchanged, because the screen it serves does not move with any single print — it moves only when a business or its ratios do.
FAQ
What did Amazon report for the third quarter of 2026?
Amazon reported net sales of 201.9 billion dollars, up 12 percent from a year earlier, against about 202.3 billion expected, and diluted earnings per share of 1.89 dollars against 1.96 expected. AWS revenue grew 39 percent to 46.0 billion dollars, operating income rose 61 percent to 28.1 billion dollars, and capital spending was 77.3 billion dollars in the quarter.
Why did Amazon fall if AWS grew 39 percent and margins hit a record?
The market focused on the bill rather than the growth. Amazon spent 77.3 billion dollars on property and equipment in one quarter, reaffirmed a full-year capital spending plan of about 220 billion dollars, and reported trailing free cash flow of negative 42.4 billion dollars against a positive 14.8 billion a year ago. Online store sales grew only 2 percent and advertising grew 11 percent, its slowest since separate reporting began, so the rest of the business did not offset the spending concern.
How does Amazon's quarter compare with Alphabet's and Microsoft's?
All three cloud businesses grew fast, but the market rewarded funding growth from operations. Microsoft beat expectations and generated positive free cash flow of 18.6 billion dollars, and rose. Alphabet missed expectations, spent 51.4 billion dollars and posted negative free cash flow of 9.07 billion dollars, and fell. Amazon beat on AWS and margin but missed slightly on sales and earnings while spending 77.3 billion dollars, and fell about 7.3 percent in after-hours trading.
Does a 7 percent fall change whether Amazon passes the Shariah screen?
No. A price move does not change a company's business activity, and as of October 2026 Amazon passes this site's screen with interest-bearing debt of 3.03 percent of market capitalisation and interest income of 0.61 percent of revenue, far below the roughly 33 percent and 5 percent ceilings. Rising borrowing and a lower share price both deserve a re-check at the next screening update, but one quarter's move does not change the result.
Why does rising borrowing matter for a halal investor if Amazon is far from the debt ceiling?
Because the screen measures interest-bearing debt against market capitalisation, borrowing moves the numerator while a falling share price shrinks the denominator, so the ratio can drift upward from both sides at once. Amazon's interest expense tripled to 1.7 billion dollars in the quarter, which shows more of the AI buildout is being funded with debt. The gap to the ceiling is still very large, but the direction of travel is what the next screening update should confirm.
Is anything in this article a fatwa or a buy or sell call?
No. Everything here is facts and screening methodology only, with no fatwas and no personalised buy or sell calls. Rulings on your own situation belong with a qualified scholar, and investment decisions belong with advice suited to your circumstances.