Amazon’s latest earnings report is a study in contradictions, a story of extraordinary ambition colliding with Wall Street’s impatience. The company reported that Amazon Web Services, its cloud-computing division, grew 39 percent in the third quarter to $46 billion in revenue, the fastest pace in nearly five years. That acceleration is a direct result of the artificial intelligence boom, as startups, enterprises, and even governments rush to rent computing power and run machine-learning models on AWS. CEO Andy Jassy went even further, predicting that AWS would grow by more than 40 percent in both 2027 and 2028, a rate that would make the cloud unit one of the fastest-growing large businesses in corporate history. The division is now generating revenue at a $184 billion annual pace, putting it within striking distance of Amazon’s third-party seller services business, the fees the company collects from independent merchants, which has long been the second-largest source of Amazon’s sales. In other words, the long-promised AI payoff is finally beginning to show up in the numbers. And yet, the stock fell sharply after the announcement. Why? Because even with those blockbuster cloud figures, Amazon’s overall earnings came in below analyst expectations, and the company’s spending plans have become so enormous that they are starting to spook even the most patient investors. The tension is classic Amazon. For decades, the company has traded short-term profits for long-term dominance, building warehouses, logistics networks, and devices that often lost money before eventually reshaping entire industries. Jassy is asking shareholders to do it again, this time for AI, satellites, and data centers. The market, meanwhile, is anchored to the present, and the present includes higher interest costs, negative free cash flow, and a stock that has already priced in a great deal of success. This is the core drama of the report: Amazon sees the future clearly, but it needs investors to see it too.
Diving into the numbers, the mixed picture becomes clearer. Amazon reported earnings of $1.89 per share for the quarter, missing the $1.97 that analysts had expected. The shortfall wasn’t because the core business is suddenly struggling. Rather, higher interest costs on Amazon’s growing debt, along with other losses, ate into the bottom line. Operating income, which measures profit from the company’s actual operations before interest and taxes, was a different story. It came in at $28.1 billion, above the high end of Amazon’s own guidance and a sign that the underlying businesses are running efficiently. The operating margin hit 13.9 percent, the highest in Amazon’s history, meaning the company is squeezing more profit out of every dollar of sales than ever before. Revenue reached $201.9 billion, up 12 percent from a year earlier. That was just below analyst forecasts, but at the top of the range Amazon had provided, suggesting that the company saw its own performance as a success. Investors, however, were less forgiving. Shares fell more than 7 percent in after-hours trading, adding to a 4 percent decline during the regular session. For a company of Amazon’s size, these are astonishing numbers. Generating more than $200 billion in a single quarter is something only a handful of companies in the world can do. But the market’s reaction shows how high the bar has become. With AI investments consuming cash at a historic pace, every miss, even a small one, is amplified. There is also a sense that Amazon’s story has become harder to tell in a simple headline. The company is no longer just an online retailer. It is a cloud giant, an advertising network, a logistics empire, a media company, and soon a satellite internet provider. That complexity makes it harder for investors to model, and harder for analysts to compare to peers. The stock’s drop after this report was not a judgment on Amazon’s execution. It was a reflection of the enormous expectations that now surround the company and its AI ambitions.
Perhaps the most important number in the entire report wasn’t in the headline revenue or profit figures at all. On the earnings call, Jassy revealed that AWS has a backlog of $578 billion in contracted future revenue, more than double what it was a year earlier. That means customers have already signed deals that will generate revenue for years to come, giving Amazon unusual visibility into its cloud growth. It is the kind of number that should reassure investors that the company’s massive spending is not a shot in the dark but a response to real demand. The flip side is that Amazon is spending enormous sums to build the data centers, chips, and infrastructure needed to meet that demand. The company plans to spend roughly $220 billion on capital projects this year, and Jassy said Amazon expects to spend “meaningfully more” in 2027. That is a staggering amount of money, even for a company with Amazon’s scale. To put it in context, $220 billion is more than the annual GDP of many countries. Jassy acknowledged the dynamic in characteristically Amazonian fashion, invoking the company’s legendary willingness to be “misunderstood for long periods of time.” He argued that the current investments are laying the groundwork for two massive businesses. In the earnings release, Jassy noted that Amazon’s retail business is gaining market share while improving its margins. Retail sales, including online stores, physical stores, and third-party seller services, totaled nearly $500 billion over the past 12 months. Combined with AWS’s $184 billion annual run rate, Jassy said, “There has never been a trillion dollar revenue business, and we may have two of them in the next 5-10 years.” He later clarified that the two businesses are AWS and Amazon’s stores business. It’s a vision of the future that few companies could credibly articulate, and perhaps only Amazon could back it with actual numbers. The company’s history suggests that it has earned the benefit of the doubt. But the scale of the bet is also unprecedented, and the margin for error is shrinking. If AI demand slows, or if the investments take longer to pay off than expected, the patience of investors could wear thin.
The quarter’s segment results show why Jassy is so optimistic, but also why the near-term picture is a bit messy. Advertising, one of Amazon’s most profitable businesses, grew 11 percent to $19.6 billion, or 12 percent excluding currency changes. That was a significant slowdown from the 22 percent growth rate it posted a year earlier. Amazon blamed the timing of Prime Day, which moved into the second quarter this year, along with a few other factors, and said it expects ad growth to return to the high teens in the fourth quarter. Online stores grew just 2 percent to $68.5 billion, also held back by the Prime Day shift. Without that timing change, Amazon said overall revenue growth would have been about 4 percentage points higher. Third-party seller services, the fees Amazon charges merchants, grew 9 percent to $46.5 billion. That puts AWS within about $500 million of that business, meaning the cloud unit is on the verge of becoming Amazon’s second-largest source of revenue. It is a remarkable moment, one that underscores the transformation of Amazon from a retailer that happens to have a cloud business to a technology infrastructure company that also happens to run a massive marketplace. Subscription services, which include Prime memberships, grew 11 percent to $14 billion. Physical stores, mostly Whole Foods, grew 2 percent to $5.7 billion. None of these are bad numbers, but they are not the kind of explosive growth that AWS is delivering. The story of this quarter is really a tale of two Amazons: a mature retail and advertising machine that is growing steadily, and a cloud and AI business that is booming. The challenge for Jassy is to keep both engines running while the company spends heavily to fuel the faster one. It is not an easy balance. The retail business generates steady profits and cash flow, but it cannot match the growth rates of AI. The cloud business is growing spectacularly, but it requires relentless investment. Jassy’s job is to convince the market that the two are complementary, not contradictory.
Looking ahead, Amazon’s guidance for the holiday quarter was solid but not enough to calm investors. The company expects fourth-quarter revenue of $240 billion to $247 billion, representing growth of 12 to 16 percent, and operating income of $28.5 billion to $33.5 billion. Wall Street had expected operating income of about $33.2 billion, near the top of that range, so the midpoint of Amazon’s forecast was a little light. CFO Brian Olsavsky explained that costs related to Amazon Leo, the company’s satellite internet service, are weighing on the fourth-quarter profit forecast as the service prepares to launch later this year. That is a reminder that Amazon is not just investing in AI and data centers; it is also building in space. The financial strain is visible in Amazon’s cash flow. Over the past 12 months, free cash flow, the cash left over after capital spending, was negative $42.4 billion, compared with positive $14.8 billion a year earlier. The swing is mostly due to AI investments. In other words, Amazon is spending far more than it is bringing in, and it is betting that the spending will pay off. The company still expects to spend about $220 billion on capital projects this year, a figure that would have been unthinkable for any company just a few years ago. For investors, the question is whether Amazon’s future growth will justify the enormous outlays. Jassy’s answer is that the backlog and revenue acceleration prove the bets are working. But the market wants proof in the form of profits, not promises. There is also a broader macroeconomic uncertainty. Interest rates, consumer spending, and global competition all factor into Amazon’s performance. The company’s debt load has grown, and higher interest costs are already eating into earnings. Amazon Leo, meanwhile, is a long-term bet that could take years to become profitable, if it ever does. All of these factors make the short-term picture murkier, even as the long-term story remains compelling.
At its core, this earnings report is about the changing nature of Amazon and the broader tech industry. The company that built its reputation on shipping boxes and undercutting retailers is now a massive infrastructure provider for the AI era, a digital advertiser, a subscription platform, and, soon, a satellite internet operator. The AWS boom is real, and the numbers are getting harder to ignore. But the costs of that boom are also real, and they are showing up in everything from interest expenses to cash flow. Amazon has been here before. In the late 1990s and early 2000s, it was criticized for spending on warehouses and logistics when it could have shown higher profits. That spending eventually created the retail empire that generates hundreds of billions of dollars in sales. Now, the company is making a similar bet on AI, and Jassy is asking investors to be patient a little longer. The risk is that the AI boom could cool, or that Amazon’s huge capital expenditures could become a burden rather than an advantage. But if the AWS backlog is any indication, the demand is already locked in. For the average person, the practical impact of all this is likely to be a world shaped by Amazon’s AI services, its satellite internet, and its ever-expanding retail network. The stock may wobble in the short term, as it did after this report, but the company’s trajectory is clear. Amazon is not just trying to be a part of the AI revolution. It is trying to build the infrastructure underneath it, and it is willing to be misunderstood, and unprofitable in the traditional sense, for as long as it takes. That is both the source of Amazon’s greatness and the reason it makes investors nervous. The company has earned the right to be believed, but even believers need a little reassurance. This report offered plenty of evidence that Amazon’s bets are paying off, but it also made clear that the price of admission is rising. For Jassy and Amazon, the next few years will be the ultimate test of whether patience really is a virtue.
