It is very hard for a new player to break that.
Since its origins as a garage bookstore in 1995, Amazon has grown into something closer to an infrastructure than a retailer — a layered ecosystem of commerce, cloud computing, and culture that shapes how much of the Western world shops, streams, and stores data. Its dominance is not merely the product of size, but of timing, patience, and architecture: early market entry, shareholders willing to absorb losses, and platform mechanics that make departure feel costly. Now, as antitrust courts and artificial intelligence both press against its edges, the question is whether Amazon's model represents the natural endpoint of digital commerce — or simply its longest chapter so far.
- Amazon surpassed Walmart as the world's largest company by annual sales, yet its most striking feature is the near-total absence of serious Western e-commerce rivals.
- Traditional retailers who tried to match Amazon's loss-tolerant growth strategy faced collapsing stock prices — a structural disadvantage that allowed Amazon to outpace the field before competitors could adapt.
- The 2000 pivot to a third-party marketplace and the 2005 launch of Prime created self-reinforcing loops: more sellers drew more buyers, and free shipping made leaving feel irrational.
- The FTC and California have filed antitrust suits alleging Amazon punishes sellers who offer lower prices elsewhere, effectively trapping shoppers and freezing out competitors — trials are set for 2027.
- The most credible disruption may not come from a rival retailer at all — generative AI shopping interfaces are beginning to route purchases without ever touching Amazon's marketplace.
A household's monthly shopping list — vitamins, tape, mango chutney — is also a map of modern commerce. The items arrive via Amazon. So do the evening's streaming shows, the books on the Kindle, and the websites running quietly on Amazon Web Services. This layering is not coincidence. It is design.
Amazon began in 1995 as an online bookstore in a rented garage. It has since surpassed Walmart as the world's largest company by annual sales, and in Western e-commerce it faces remarkably few serious challengers. Walmart, Tesco, Zalando, Temu, and eBay each occupy corners of the market. GameStop recently bid $55.5 billion to acquire eBay in hopes of building a stronger rival. None of it has meaningfully threatened Amazon's position.
The reasons are structural. Amazon arrived early with a clear vision, captured market share aggressively, and — crucially — operated at a loss for years while reinvesting every early profit into growth. Its shareholders waited. Traditional retailers attempting the same strategy would have faced collapsing stock prices and revolts from investors. Harvard's David Yoffie notes this patience alone "constrained the competition." Today, AWS — Amazon's cloud division — generates the profits that subsidize its lower-margin retail arm and fund experiments in healthcare, devices, and original content.
Two decisions proved decisive. In 2000, Amazon opened its platform to third-party sellers, creating a network effect: more sellers attracted more customers, who attracted more sellers. Harvard's Sunil Gupta observes that breaking such a loop is "very hard for a new player." Then came Prime in 2005 — an annual membership offering fast, free delivery that proved, in the words of communications professor Emily West, "very sticky." Over time, Prime bundled streaming, grocery discounts, and original content, making cancellation feel like a genuine loss. Amazon had become an ecosystem, not a website.
Yet regulators allege a darker dimension to this dominance. The FTC and California have both filed antitrust suits, with trials scheduled for early 2027, alleging that Amazon penalizes sellers who offer lower prices elsewhere — suppressing their search visibility or removing their Buy Box placement. The effect, if proven, would mean competitors cannot undercut Amazon because sellers cannot afford to try. Amazon denies the allegations.
Some advocates call for breaking the company apart to "oxygenate the market." Most experts agree a well-funded rival could replicate Amazon's model given time. But the next disruption may not come from a retailer at all. Generative AI interfaces are beginning to embed shopping directly into conversation — allowing purchases without ever visiting Amazon's marketplace. The company that seemed impossible to rival may ultimately be challenged not by a competitor, but by a different way of buying altogether.
A household's monthly shopping list—vitamins, repair tape, mango chutney—tells a story about the shape of modern commerce. These items arrived via Amazon. So did the streaming shows watched that evening, the books read on a Kindle device, and the websites browsed, many of them running on Amazon Web Services, the company's cloud-computing engine. This layering of services is not accidental. It is the architecture of dominance.
Amazon, which started in 1995 as an online bookstore in a rented garage, has become something far larger. Earlier this year it surpassed Walmart to claim the title of the world's largest company by annual sales. Yet in the Western world, it faces remarkably few serious challengers in e-commerce. Walmart and Target have built online operations. Tesco leads in UK grocery delivery. Zalando dominates clothing retail in Germany. Temu and Shein offer ultra-cheap goods from China. eBay operates in a different lane, focused on auctions and second-hand items. GameStop recently made a $55.5 billion takeover bid for eBay, hoping to create a stronger Amazon competitor. None of this has dented Amazon's commanding position. The company towers over all of them.
The reasons for this dominance are structural and deliberate. Amazon arrived early to online retail with a clear vision of how the internet could remake shopping through speed and convenience. It captured market share faster than rivals could follow. More importantly, its shareholders tolerated something traditional retailers could not: years of operating at a loss, selling products below cost, then aggressively reinvesting early profits back into growth. Amazon has never paid shareholders a dividend. This strategy, as Harvard Business School professor David Yoffie notes, "constrained the competition." A traditional company pursuing the same path would have faced a collapsing stock price and furious shareholders. Amazon's shareholders waited.
Today, the company's most profitable business—AWS, its cloud-computing division—generates the cash that allows Amazon to sustain its lower-margin retail operation and fund experimental ventures. The company has positioned itself as a technology firm, not merely a retailer. Algorithms, automation, and data drive its efficiency. It experiments boldly, entering cloud computing, consumer devices, private-label products, original content, and healthcare, abandoning ventures that fail without hesitation.
Two pivotal moves cemented its position. In 2000, Amazon transformed from an online retailer into an online platform, allowing third-party sellers to offer goods through its store. This created what economists call a network effect: more sellers meant more products, which kept customers from leaving, which attracted more sellers. As Sunil Gupta, another Harvard professor, explains, "It is very hard for a new player to break that." The second move was Amazon Prime, launched in the US in 2005 and the UK in 2007. For an annual fee, members received free and fast delivery. The service proved "very sticky," in the words of Emily West, a communications professor at the University of Massachusetts Amherst. Once you have free shipping, why search anywhere else? Prime itself generates modest profits, but what it bundles has expanded dramatically—streaming libraries, original content, discounts at Whole Foods—making the membership increasingly difficult to cancel. Amazon is no longer a website that sells products. It is an ecosystem of reinforcing businesses.
Yet there may be another reason Amazon lacks rivals: behavior that regulators allege violates competition law. The Federal Trade Commission and the state of California have filed separate antitrust lawsuits against Amazon, both set for trial in early 2027. California recently released evidence supporting allegations that Amazon penalizes sellers—lowering their product visibility in search results or removing their "Buy Box"—if it discovers they offer lower prices elsewhere. The effect is to trap shoppers on Amazon's platform. Competitors cannot undercut prices because sellers fear losing Amazon sales. Amazon denies the allegations and is defending the lawsuits vigorously.
Some advocates have proposed breaking Amazon into separate companies. Stacy Mitchell, co-executive director of the Institute for Local Self-Reliance, argues this would "oxygenate the market." With sufficient capital and time, most experts agree a company could build an Amazon replica. Walmart, with deep pockets, is clearly borrowing from Amazon's playbook. Yet the next threat to Amazon's dominance may not come from a traditional retailer at all. Generative AI interfaces like ChatGPT are beginning to embed shopping directly into their platforms, allowing users to purchase without leaving the site. This development could fundamentally alter how people buy online, shifting power away from Amazon's centralized marketplace. The company that seemed impossible to rival may face disruption not from a competitor, but from a different way of shopping altogether.
Citas Notables
Amazon is not an undisputed monopolist in e-commerce, but it is the dominant firm. And the scope of what it sells is unparalleled.— Annabelle Gawer, director of the Centre of Digital Economy at the University of Surrey
Amazon is not just a website that sells products. It's an ecosystem of multiple businesses that are reinforcing each other, which makes it very hard to compete with.— Annabelle Gawer
La Conversación del Hearth Otra perspectiva de la historia
Why couldn't Walmart or Target simply replicate what Amazon did? They had the capital, the infrastructure, the customer base.
They could have, in theory. But Amazon's shareholders tolerated losses for years—something traditional retailers couldn't do without destroying their stock price. By the time Walmart and Target tried to catch up, Amazon had already built a network effect that was nearly impossible to break.
The network effect—more sellers, more products, more customers. But couldn't a new competitor offer better terms to sellers to lure them away?
In principle, yes. But Amazon's Prime membership and its alleged practices—penalizing sellers who list cheaper prices elsewhere—trap both sellers and buyers. A competitor offering lower fees can't overcome the fear of losing Amazon sales.
So it's not just that Amazon is bigger. It's that the system itself prevents competition.
Exactly. And that's why the antitrust cases matter. If Amazon is using its dominance to suppress competition, breaking it up might actually create space for rivals to emerge.
What about the AI threat you mentioned? How does ChatGPT shopping change the game?
It removes the middleman entirely. Instead of going to Amazon's website, you ask an AI to buy something, and it happens. Suddenly Amazon's platform advantage—the ecosystem, the Prime lock-in—becomes less relevant. The power shifts to whoever controls the AI interface.
So Amazon's dominance might not be overthrown by a better retailer, but by a completely different way of shopping.
That's the possibility. The company that seemed unbeatable might be disrupted not by competition, but by the market itself changing shape.