The E-Commerce Revolution: Selling Everything to Everyone

Zusammenfassung

This article traces the history of electronic commerce — from Jeff Bezos driving across the country to start a bookstore in his garage, through eBay’s peer-to-peer marketplace, the invention of online payment, and Amazon’s transformation from retailer to infrastructure. It is the story of how the internet removed the physical constraints from commerce — distance, store hours, shelf space — and what happened to the industries that depended on those constraints for their survival.

The Spreadsheet in the Car

In 1994, Jeff Bezos was a vice president at the hedge fund D.E. Shaw in New York, tracking the growth of the internet. He encountered a statistic that stopped him: web usage was growing at 2,300% per year. He made a list of products that could be sold effectively online and concluded that books were optimal: the catalog was vast (more titles existed than any physical store could stock), they were standard commodities (a book was identical regardless of which store sold it), and they were easy to ship.

Bezos quit his job, drove across the country with his wife MacKenzie, and wrote the business plan for Amazon.com in the car. They incorporated in Washington state in July 1994 and launched the website in July 1995 from a garage in Bellevue.

The early Amazon was a genuine bookstore: Bezos and a small team purchased books from distributors, stored them in the garage, and mailed them. It immediately discovered a property of online retail that would define the industry: the long tail. A physical bookstore could stock perhaps 50,000 titles. Amazon’s catalog contained 1 million. Customers who couldn’t find obscure titles locally could find them on Amazon. The demand for obscure titles, aggregated across millions of customers, exceeded the demand for bestsellers.

Amazon was not profitable. Bezos invested every dollar of revenue in growth — expanding the catalog, building warehouse infrastructure, hiring engineers. He argued to his board, and to Wall Street, that in a winner-take-all internet market, market share was worth more than profit margin. The argument was contested. For several years, it was not clear he was right.

eBay: The Peer-to-Peer Marketplace

Pierre Omidyar was a software engineer in San Jose who launched AuctionWeb on Labor Day weekend 1995 — later renamed eBay. The premise was the opposite of Amazon’s: instead of a company selling goods to customers, customers would sell to each other, with eBay providing the platform and taking a percentage of each transaction.

The first item sold on AuctionWeb was Omidyar’s broken laser pointer, purchased for $14.83 by a collector of broken laser pointers. When Omidyar emailed the buyer to confirm he understood the item was broken, the buyer replied that he collected broken laser pointers. Omidyar realized he had built something unusual: a marketplace with sufficient scale to connect sellers of highly specific items with the rare buyers who wanted them.

eBay’s growth relied on a genuine network effect: each additional seller made the platform more useful for buyers, and vice versa. By 1997, the platform was processing $95 million in transactions annually. By 1999, $2.8 billion. Unlike Amazon, eBay was profitable almost immediately — it held no inventory and bore no shipping costs.

The feedback system — buyers and sellers rating each other — was eBay’s most important innovation. Online commerce required trust between strangers who could not inspect goods or verify identities. eBay’s reputation system created a proxy for trust: a seller with 10,000 positive ratings was genuinely safer to buy from than an anonymous unknown. The system was imperfect and gameable, but it was sufficient.

PayPal: The Payment Infrastructure

Online commerce required online payment. Credit cards worked, but the process was cumbersome for individuals — few people could accept credit card payments without a merchant account, which required a bank relationship, a fee structure, and a waiting period.

Confinity, founded by Peter Thiel and Max Levchin in 1998, built a payment system that allowed money to be sent via email. X.com, founded by Elon Musk in 1999, built an online bank. The two companies merged in 2000; the merged entity kept the name PayPal after the Confinity product proved more popular.

PayPal’s adoption was driven by eBay: sellers and buyers needed a way to complete transactions quickly and safely. PayPal spread virally — each transaction recruited both parties to the platform. By 2002, PayPal was the payment method for 70% of eBay auctions. eBay acquired PayPal for $1.5 billion in 2002.

The “PayPal Mafia”

PayPal’s early employees — forced out or bought out after the eBay acquisition — went on to found or fund a remarkable proportion of Silicon Valley’s subsequent generation of companies. Elon Musk founded Tesla and SpaceX. Peter Thiel founded Palantir and invested in Facebook. Reid Hoffman co-founded LinkedIn. Steve Chen, Chad Hurley, and Jawed Karim founded YouTube. David Sacks founded Yammer. Jeremy Stoppelman founded Yelp. The density of subsequent company formation from a single workplace suggests something about the culture PayPal created — an unusually high tolerance for ambitious, unconventional ideas combined with the experience of building a financial system under intense competitive and regulatory pressure.

Amazon’s Transformation: From Retailer to Infrastructure

By the early 2000s, Amazon had expanded from books to music, videos, electronics, toys, and — eventually — everything. Its 2000 partnership with Toys “R” Us, in which the toy retailer paid Amazon to be its exclusive online toy seller, illustrated the period’s logic: established retailers needed Amazon’s customer reach; Amazon needed the inventory.

The relationship with Toys “R” Us ended in litigation when Amazon began allowing other toy sellers on its platform. This pattern — partnerships that became subordination, distribution that became dependence — characterized Amazon’s relationship with the retail industry through the 2000s and 2010s.

Amazon’s deeper transformation was in infrastructure. The development of Amazon Web Services (2006) — selling its internal computing infrastructure as a service — is told in The Cloud Computing Era. But equally transformative was the Fulfillment by Amazon (FBA) program (2006): third-party sellers could ship their inventory to Amazon warehouses, and Amazon would pick, pack, and ship orders. Amazon became not merely a retailer but the logistics infrastructure for millions of small sellers.

By 2020, third-party sellers accounted for more than 50% of units sold on Amazon. Amazon collected fees from these sellers — listing fees, fulfillment fees, advertising fees — without bearing the cost of the goods. The marketplace was more profitable than the retail business.

Dead End: The Stores That Couldn’t Adapt

The costs of e-commerce’s rise were concentrated in physical retail. Companies that depended on foot traffic, physical browsing, or the friction of distance found their advantages systematically eliminated.

Borders bookstores, second only to Barnes & Noble in U.S. book retail, partnered with Amazon in 2001 to run its online bookstore — outsourcing its digital future to its most dangerous competitor. It filed for bankruptcy in 2011 and liquidated. Tower Records, which had defined the music retail experience for a generation, was killed by digital downloads and filed for bankruptcy in 2006. Circuit City, unable to match Best Buy’s pricing or Amazon’s selection, filed for bankruptcy in 2008 and liquidated in 2009.

The Innovator’s Dilemma in Retail

Established retailers faced a version of Clayton Christensen’s Innovator’s Dilemma: investing in e-commerce cannibalized their profitable physical store business in the short term, while not investing guaranteed long-term irrelevance. Most chose to protect existing margins and invested too little, too late. The exceptions — Best Buy, Target, Walmart — survived by accepting margin compression and building genuine omnichannel capabilities. Walmart’s acquisition of Jet.com in 2016 for $3.3 billion was an expensive lesson that infrastructure, not just website investment, was required to compete with Amazon’s logistics network.

For the payment systems that enabled online commerce, see The Connected World. For the cloud infrastructure Amazon eventually sold to the world, see The Cloud Computing Era.