$AMZN Stock: Why I’m Bullish on Amazon’s Three-Engine Machine

Amazon is one of the rare companies where the pieces become more valuable because they sit beside one another. The store creates purchase intent. Prime creates loyalty and frequency. The marketplace attracts selection. Fulfillment turns scale into speed. Advertising monetizes the shopping traffic. AWS turns the company’s infrastructure obsession into a high-margin technology platform.

That is why I am bullish on $AMZN. I do not see one giant retailer with a cloud side business. I see three powerful engines—commerce, advertising, and AWS—feeding the same compounding machine.

The latest numbers make the thesis difficult to dismiss. In the second quarter of 2026, Amazon’s net sales increased 20% year over year to $200.6 billion. Operating income increased 43% to $27.5 billion. AWS sales increased 37% to $42.2 billion, advertising services revenue increased 26% to $19.8 billion, and both the North America and International segments produced operating profit.

Those are elite numbers at Amazon’s scale.

But this is not a blind bull case. Amazon is spending so aggressively on artificial-intelligence infrastructure that trailing-twelve-month free cash flow fell to an outflow of $7.6 billion. The question is no longer whether Amazon can grow. The question is whether today’s extraordinary capital spending creates enough future revenue and profit to justify the bill.

My view is yes—but investors need to watch the receipts.

Amazon owns a commerce network, a purchase-intent advertising platform, and a cloud infrastructure business that reinforce one another, while its willingness to spend heavily on AI could widen the moat if management converts that spending into durable cash flow.

The market often analyzes Amazon as a collection of separate businesses. I think that misses the real advantage. A merchant can sell through Amazon, store inventory in its network, buy ads to reach shoppers, use its checkout and logistics tools, stream content to Prime members, and run the rest of the company on AWS.

Every additional service makes the ecosystem harder to leave. That does not eliminate competition, but it gives Amazon several ways to win the same customer relationship.

AWS is the clearest reason I am bullish. Q2 sales increased 37% to $42.2 billion, the fastest growth rate in 18 quarters. AWS operating income reached $16.6 billion, up from $10.2 billion one year earlier. The segment generated roughly 61% of Amazon’s consolidated operating income even though it represented about 21% of sales.

That is the earnings engine.

The acceleration matters because the cloud debate has changed. Investors spent years asking whether AWS was losing momentum as Microsoft Azure and Google Cloud grew quickly. Amazon’s latest results show AWS is not standing still. Core cloud workloads are growing, AI workloads are growing, and management says each side helps the other.

AI models need accelerators, but they also need CPUs, storage, databases, networking, security, monitoring, and tools that connect models to company data. Amazon can sell the entire stack. It has NVIDIA capacity, its own Trainium AI chips, Graviton CPUs, Bedrock for model access and inference, SageMaker for development, and AgentCore for deploying agents.

Management said both the AWS AI business and the chips business exceeded $25 billion annual revenue run rates in Q2, with each growing at triple-digit percentages. Those figures are company-defined run rates rather than separately reported accounting segments, so I would not treat them like audited annual revenue. They still show that Amazon is building meaningful businesses around the AI infrastructure wave instead of merely attaching “AI” to an earnings call.

The backlog is also substantial. Amazon’s Q2 10-Q reported approximately $496 billion of performance obligations, primarily related to AWS long-term customer commitments, with a weighted-average remaining life of 6.4 years. That is not the same as guaranteed near-term revenue, and contracts can contain conditions. It does provide a long runway of contracted demand that the company can work to fulfill.

Advertising may be the most underappreciated part of Amazon.

Q2 advertising services revenue reached $19.8 billion, up 26% year over year. That makes the quarterly business almost half the size of AWS revenue, even though Amazon does not disclose a separate advertising operating profit figure.

The strategic advantage is simple: people visit Amazon to research and buy products. That purchase intent is extremely valuable to advertisers. A sponsored product shown beside a relevant search result can be measured against an actual sale, not only a click or impression.

Amazon is also expanding beyond sponsored search. Prime Video, Thursday Night Football, the NBA, WNBA, and NASCAR add streaming inventory. The company says advertisers using multi-sport campaigns get broader unduplicated reach, while viewers who watch several sports also spend more on Amazon. The same Prime relationship can support entertainment, advertising, membership retention, and commerce.

AI can make that engine more efficient. Amazon said shoppers who clicked a Sponsored Prompt converted to a sale 48% more often and spent 21% more on average than shoppers who did not. That is a company-reported comparison, not proof that every advertiser will receive the same result. It points to the opportunity: conversational shopping can create a new surface where product discovery and advertising happen together.

The risk is that Amazon pushes too many ads and damages the customer experience. Search results must remain useful. Prime Video must still feel worth paying for. Regulators may also scrutinize how Amazon uses marketplace data and ranks products. Advertising is a powerful engine only if trust survives.

Amazon’s retail business is no longer just a low-margin volume story.

North America sales increased 16% to $116.2 billion in Q2, while segment operating income increased to $9.1 billion from $7.5 billion. International sales increased 15% to $42.2 billion, and operating income increased to $1.7 billion from $1.5 billion.

The revenue mix is getting stronger too. Online stores generated $70.4 billion, third-party seller services generated $46.8 billion, subscriptions generated $13.7 billion, and advertising generated $19.8 billion. Amazon earns from its own retail sales, seller fees, fulfillment, Prime, and ads around the same transaction ecosystem.

Speed strengthens the flywheel. Amazon said it delivered more than 40% more items the same day or overnight during the first half of 2026 than one year earlier. Faster delivery improves the customer experience, encourages more frequent orders, attracts more sellers, and gives Amazon more volume over which to spread its logistics investment.

Grocery, pharmacy, and supply-chain services can deepen that habit. The company reported that monthly active perishables customers grew more than 50% from the start of the year, while new Amazon Pharmacy customers more than doubled in the first half. It also opened its supply-chain capabilities to outside businesses through Amazon Supply Chain Services.

This is why I think the word “retailer” undersells Amazon. The company increasingly provides infrastructure for commerce: discovery, payments, fulfillment, delivery, subscriptions, advertising, and cloud technology.

Each Amazon engine is strong on its own. The bull case becomes more powerful when they work together.

More selection and faster delivery make Prime more useful. More Prime members create more shopping traffic and streaming attention. More traffic attracts sellers and advertisers. More sellers expand selection. More advertising and service revenue can improve the economics of commerce. AWS supports Amazon’s own technology needs while selling the same infrastructure discipline to outside customers.

That flywheel creates data, but the moat is not data alone. It is Amazon’s ability to turn information into physical and digital execution. The company can recommend a product, serve an ad, process the order, move the inventory, deliver the package, host the seller’s software, and keep the customer inside Prime.

Very few competitors can match that full chain.

Amazon’s AI shopping tools could make it stronger. Management said more than 350 million customers used Alexa for Shopping during the last 12 months and that Q2 interactions increased more than fivefold year over year. If conversational shopping helps customers compare products and automate purchases, Amazon can reduce friction while creating another advertising surface.

The bear case is that AI agents could bypass Amazon and send shoppers directly to brands or lower-priced marketplaces. The bull case is that Amazon’s selection, reviews, logistics, payments, and delivery promise make it one of the best places for an agent to complete a transaction. I believe the physical network gives Amazon an advantage that a chatbot alone cannot copy.

The strongest argument against my thesis is cash flow.

Trailing-twelve-month operating cash flow increased 33% to $161.4 billion, but purchases of property and equipment, net of proceeds and incentives, reached $169.0 billion. That pushed Amazon’s reported free cash flow to a $7.6 billion outflow, down from an $18.2 billion inflow one year earlier.

The spending increase is primarily tied to AI. During the first six months of 2026, AWS net additions to property and equipment reached $90.1 billion, compared with $36.5 billion in the same period of 2025.

I do not automatically treat high capital spending as bad. If Amazon builds scarce compute capacity that customers use for years, today’s spending can create tomorrow’s revenue, operating leverage, and strategic control. Custom chips can also lower Amazon’s costs and give customers more price-performance choices.

But capital intensity can destroy value when demand is overestimated, hardware becomes obsolete quickly, electricity and data-center constraints delay utilization, or competitors force prices down. A $25 billion AI run rate sounds impressive; it must eventually be judged against the far larger infrastructure bill.

This is the checkpoint that matters most. I want AWS growth, backlog conversion, and operating income to prove that the capex is productive. I also want operating cash flow to keep rising so free cash flow can recover as the investment cycle matures.

Amazon reported Q2 net income of $62.6 billion, but that number included $53.4 billion of non-operating pre-tax other income, primarily related to its Anthropic investments.

That gain is economically meaningful, but it is not recurring operating profit. Investors should separate it from the performance of the store, advertising, and AWS. The cleaner business signal is the 43% increase in operating income to $27.5 billion.

This distinction matters because a headline earnings number can make valuation look cheaper than the underlying business really is. My bull case does not depend on repeating investment gains. It depends on Amazon compounding operating income and eventually converting more of it into free cash flow.

The first risk is AI overbuilding. Amazon may spend ahead of demand, earn weak returns on data centers, or face rapid depreciation as chips improve.

The second risk is competition. Microsoft and Google are formidable cloud competitors. Walmart, fast-growing international marketplaces, direct-to-consumer brands, and specialized retailers keep pressure on commerce. Meta, Google, TikTok, and streaming platforms compete for advertising budgets.

The third risk is regulation. Amazon faces antitrust, marketplace, labor, privacy, and tax scrutiny across several countries. Rules that limit self-preferencing, data use, acquisitions, or marketplace practices could weaken parts of the flywheel.

The fourth risk is retail execution. Fast delivery is expensive. Wages, fuel, tariffs, inventory mistakes, and weak consumer demand can pressure margins. International profitability has improved, but it remains thin relative to the revenue base.

The fifth risk is valuation and expectations. A great business can still produce a poor return when investors pay too much or assume every growth engine will remain perfect. I am bullish on Amazon, not bullish at any price.

For Q3, Amazon expects net sales of $197 billion to $202 billion, representing 9% to 12% year-over-year growth, and operating income of $22.5 billion to $26.5 billion. Prime Day timing makes the reported growth comparison less clean, so I will focus on segment trends rather than one consolidated percentage.

I want AWS growth to remain strong while operating income proves the infrastructure is being used profitably. I want advertising to keep growing faster than the company without degrading shopping quality. I want North America margins to hold and International operating profit to expand.

Most importantly, I want the gap between operating cash flow and capital spending to begin moving in the right direction. Amazon does not need to stop investing. It needs to show that the return on those investments can outrun the cost.

If those checkpoints hold, I believe Amazon can compound through three connected engines: AWS sells the digital infrastructure, advertising monetizes intent, and commerce owns the customer habit and physical delivery layer.

That is why I am bullish on $AMZN. Amazon is not simply the everything store. It is becoming the infrastructure underneath how companies compute, advertise, sell, and deliver—and the businesses that become infrastructure are difficult to replace.

This article is education and research only. It is not personalized investment advice, a recommendation to buy or sell $AMZN, or a promise of performance. Investing involves risk, including loss of principal.