Dear Investors.
Zee here. Most people know Amazon as the place a parcel comes from.
Last quarter, the parcels were the smaller half of the story.
Amazon reported results on Thursday, 30 July 2026, and the numbers were the strongest the company has posted in years. Revenue crossed $200 billion in a single quarter for the first time. Operating profit rose 43%. The cloud division grew at its fastest rate in four and a half years. The stock jumped roughly 10% in after hours trading.
And in the same set of accounts, free cash flow was negative.
That combination is the whole story. Amazon is now two things at once: a highly profitable business, and a business spending faster than it earns in order to build something bigger. Whether that is exciting or worrying depends entirely on how long you plan to hold it.
WHAT DOES AMAZON ACTUALLY SELL?
Before the numbers, a quick reminder of how the company makes money. Amazon is not one business. It is roughly five, sitting under one roof.
1. Online retail. The part everyone sees. Amazon buys products, holds them in warehouses, and sells them. This produces enormous revenue and thin margins, the same economics as any retailer.
2. Third party marketplace. Other sellers list on Amazon and pay a commission, plus fees for storage and delivery. Amazon takes a cut without owning the inventory. Higher margin, less risk.
3. Subscriptions. Prime memberships, plus video, music, and audiobooks. Predictable recurring revenue, and it makes customers shop more often.
4. Advertising. Sellers pay to appear higher in search results on Amazon’s own site. This is quietly one of the most profitable things the company does, because the traffic already exists.
5. Amazon Web Services (AWS). Renting out computing power, storage, and now artificial intelligence infrastructure to other companies. AWS is around a fifth of revenue but the majority of profit.
Retail brings people in. AWS and advertising pay the bills. That is the model in one line.
1. REVENUE CROSSED USD$200 BILLION IN A SINGLE QUARTER FOR THE FIRST TIME
Quarterly revenue came in at $200.6 billion, up 20% from $167.7 billion a year earlier. Operating income rose 43% to $27.5 billion.
The detail that matters more than the headline: profit grew roughly twice as fast as revenue. That is operating leverage. It means the extra dollar of sales is costing Amazon less to produce than it used to, which is what you want to see in a business that has already spent a decade building out its warehouses and delivery network.
North American sales rose 16% to $116.2 billion. International sales rose 15% to $42.2 billion. Both segments grew, which has not always been the case.
One caution on the next quarter. Amazon guided for revenue of $197 billion to $202 billion for Q3, below what analysts were hoping for. Management explained that Prime Day was moved into June this year instead of July, which pulled sales into Q2 and leaves a gap in Q3.
Strip that timing effect out and the company said growth would have been meaningfully higher. This is a calendar artefact, not a demand problem, but it is the kind of thing that moves a share price for a week.
2. AWS IS THE ENGINE, AND IT JUST SPED UP
AWS revenue rose about 37% to $42.2 billion, its fastest growth in 18 quarters. Analysts had expected around 31%.
Three numbers underneath that are worth more than the growth rate itself:
Operating margin of 39.4%. For every dollar of cloud revenue, roughly forty cents is operating profit. AWS produced $16.6 billion of operating income in the quarter, up from $10.2 billion a year ago.
A backlog of around $496 billion. This is contracted future revenue that customers have already committed to. Backlog is one of the most useful things a long term investor can look at, because it tells you what has been sold but not yet delivered.
AI and custom chip businesses each above a $25 billion annual run rate. Amazon designs its own chips (the Trainium and Graviton families) rather than buying everything from Nvidia. That gives it some control over cost.
CEO Andy Jassy told investors he now believes AWS can eventually become a business measured in the hundreds of billions of dollars annually, and possibly far beyond that.
Treat management ambition as ambition, not forecast. But the backlog gives that claim more support than it would have had a year ago.
3. THE HEADLINE PROFIT IS NOT WHAT IT LOOKS LIKE
This is the one most people will miss.
Reported net income was $62.6 billion, or $5.75 per share, against $18.2 billion a year earlier. Read quickly, that looks like profit more than tripled.
It did not. Around $53.4 billion of that was non operating pre tax income, largely a revaluation gain on Amazon’s stake in the AI company Anthropic. In plain terms, an investment Amazon holds was marked up in value on paper. No cash changed hands. Nothing was sold.
Strip it out and adjusted earnings were about $1.97 per share, against roughly $1.82 expected. Still a beat. Still a good quarter. But a very different number from $5.75.
The lesson generalizes well beyond Amazon. When a reported profit figure jumps by a multiple in one quarter, the first question is always the same: did the operating business do this, or did an accounting entry do this? Paper gains can reverse just as quickly as they appear.
4. THE SPENDING IS ENORMOUS, AND FREE CASH FLOW HAS GONE NEGATIVE
Amazon raised its 2026 capital spending plan to roughly $220 billion, up from about $200 billion, blaming higher memory chip prices. Capital expenditure in the quarter alone was $54.2 billion, against $32.1 billion a year earlier.
The consequence shows up in cash. Free cash flow, which is what remains after a company pays its operating costs and builds its data centres and warehouses, swung to an outflow of about $7.6 billion over the trailing twelve months, from an inflow of $18.2 billion the year before. Amazon has also announced plans to raise around $25 billion through a bond sale.
Two ways to read this, and both are legitimate.
The optimistic reading: this is what AWS looked like in its early years. Heavy upfront spending on infrastructure that later became one of the most profitable businesses in the world. Jassy has said that even at $220 billion, Amazon still cannot serve all the demand it sees, and that demand already visible for 2028 is unusually strong.
The cautious reading: capital spending is a promise, and demand is a forecast. Data centres depreciate. If AI demand slows before the assets are paid off, the depreciation stays and the revenue does not.
What tipped the market toward the optimistic reading this time was simple. Cloud revenue accelerated faster than the spending did. That is the test to keep applying each quarter.
5. THE BUSINESSES NOBODY PRICES IN YET: ADS AND SATELLITES
Advertising grew 26% to $19.8 billion in the quarter. That is a business roughly half the size of AWS, growing quickly, attached to almost no incremental cost. Amazon knows what you searched for and what you bought. That data is why the ads work.
Amazon Leo, the satellite internet business formerly called Project Kuiper, is the long shot. Around 396 satellites are in orbit, well short of the 1,618 the FCC licence originally required by 30 July 2026, and Amazon has asked for an extension. In April the company agreed to acquire satellite operator Globalstar for about $11.6 billion, a deal expected to close in 2027. In late July it filed with the FCC for a separate network of more than 5,100 satellites to beam service directly to phones, targeted for early 2028 if approved.
Leo is years from mattering to the income statement and is competing with a rival that started far earlier and is far larger. For now, treat it as an option, not an asset. It costs money today and may be worth something later.
Bottomline
Three questions are worth carrying into the next set of results:
Is AWS revenue still growing faster than capital spending? That is the single ratio that decides whether the buildout was disciplined or expensive.
When does free cash flow turn positive again? Negative free cash flow is acceptable while assets are being built. It is not acceptable indefinitely.
How much of reported profit came from the operating business? Paper gains flatter earnings on the way up and hurt them on the way down.
Disclaimer:
All information here is for educational purposes only. This is not financial advice. Please do your own research and speak with a licensed advisor before making any investment decisions. Past performance is not indicative of future returns. How we invest may not suit your investment goals and risk management profile.