Amazon: From a Garage Bookstore to a $638 Billion Empire That Runs the Internet

How Jeff Bezos drove cross-country with a business plan on his laptop, launched an online bookstore from a Bellevue garage in 1994, accidentally invented the cloud computing industry, built AWS into a $107.6 billion powerhouse commanding 31% of the global cloud market, and created a $2.3 trillion company employing 1.55 million people across 200 countries.

The Road Trip That Started Everything

In the summer of 1994, Jeff Bezos was a 30-year-old senior vice president at hedge fund D.E. Shaw in New York, earning a substantial salary and on track to extraordinary wealth. One statistic changed everything: internet usage was growing at 2,300% per year. Bezos began drawing up a list of products that could be sold online, and books rose to the top. They were commoditised, easy to ship, and came in more varieties than any single physical store could ever stock.

He resigned from D.E. Shaw, and he and his wife MacKenzie drove cross-country from New York to Seattle. Bezos typed the business plan on his laptop in the passenger seat as MacKenzie drove. They chose Seattle for its proximity to a major book distributor and its large pool of technology talent. They rented a house in Bellevue, converted the garage into an office, and launched Amazon.com on July 16, 1995. In its first month, Amazon sold books to customers in all 50 US states and 45 countries. Bezos installed a bell that rang every time a sale was made. The bell had to be disconnected within weeks, because it never stopped.

From Books to Everything: Building the World's Largest Store

Amazon's expansion from books into other product categories began almost immediately. CDs and DVDs were added in 1998. Electronics, toys, and tools followed by 1999. Bezos's stated ambition, articulated publicly within Amazon's first few years, was to build not a bookstore but an everything store, a single destination where any product could be found and bought. Third-party marketplace sellers were integrated in 2000, allowing external merchants to list their products on Amazon's platform. This single decision was transformational: Amazon's selection could now grow infinitely without Amazon itself needing to own the inventory.

The dot-com crash of 2000-2001 nearly destroyed Amazon. The stock fell from $107 to $7. Competitors with lavish venture capital funding went bankrupt around it. Bezos cut costs ruthlessly, focused relentlessly on customer experience rather than marketing, and refused to retreat from long-term investment even when short-term losses were alarming. Amazon emerged from the crash with a stronger competitive position than before it began, having outlasted dozens of well-funded rivals through the discipline of thinking in decades rather than quarters.

"We are stubborn on vision. We are flexible on details." Jeff Bezos, Amazon founder

AWS: The Accident That Now Powers Half the Internet

Amazon Web Services was not conceived as a product for external customers. It began as an internal project to standardise Amazon's own computing infrastructure after years of each engineering team building incompatible systems. By 2004, Amazon's engineers had built standardised internal computing services they called infrastructure primitives. The obvious question followed: if Amazon needed these services to run its own business, wouldn't every other company need them too?

AWS launched publicly in March 2006 with Amazon S3 cloud storage and Amazon EC2 computing. Initial adoption was almost entirely from startups and developers who could now build businesses without buying physical servers. What nobody predicted was the speed at which enterprise businesses would follow. Netflix migrated its entire video streaming infrastructure to AWS between 2009 and 2016. The CIA signed a cloud contract with AWS in 2013 for $600 million, a signal that AWS had crossed from startup tool to critical national infrastructure. Today AWS powers approximately 31% of the global cloud infrastructure market.

Revenue: From $89 Billion to $638 Billion in a Decade

Milestones: Three Decades of Relentless Expansion

Amazon's Four Core Business Engines

Amazon Prime: The Loyalty Engine Worth Trillions

Amazon Prime launched in 2005 as free two-day shipping for $79 per year. It has become the most successful subscription programme in the history of retail, with an estimated 200 to 220 million members globally paying $139 annually in the United States. Prime members spend on average more than twice as much on Amazon each year as non-members, shop more frequently, and are dramatically less likely to comparison-shop with competitors. Prime Video, Prime Music, Prime Gaming, and Prime Reading are all bundled in, making cancellation increasingly irrational from a consumer value perspective.

The Prime flywheel drives Amazon's entire business model. More Prime members create more shopping volume, which justifies more fulfilment investment, which enables faster delivery speeds, which attracts more Prime members. Amazon Advertising benefits because advertisers pay premium rates to reach Prime members who are demonstrably higher-intent buyers. AWS benefits indirectly because Amazon's scale gives it engineering talent and purchasing power that reinforces cloud dominance. Everything connects back to Prime.

Amazon and AI: The Next Frontier

Amazon's AI strategy operates at three levels. At the infrastructure level, AWS offers AI computing through custom silicon including Trainium chips for training and Inferentia chips for inference, designed to undercut Nvidia's GPU pricing for cloud workloads. At the platform level, AWS Bedrock provides access to dozens of third-party AI models, and Amazon Q is an enterprise AI assistant being deployed to corporate clients. At the consumer level, Alexa is being rebuilt on a large language model foundation, with billions being invested in a next-generation version capable of complex multi-step reasoning.

According to Reuters Technology, Amazon's $4 billion investment in Anthropic, the AI safety company behind the Claude large language model, gives AWS exclusive cloud provider status for Anthropic's most advanced models, positioning Amazon as a key infrastructure partner for the most commercially important AI systems currently in development.