AWS at 20: From a Million-Dollar Dream to a $128 Billion AI Powerhouse
In 2006, when Amazon Web Services was just a side project inside an e-commerce giant, no one—not even most Amazon employees—could see what was coming. A product manager named Matt Garman sat down for lunch with a colleague from another division. His friend asked, “How’s that AWS thing going? I heard it sounds interesting.”
Garman’s answer was bold: “I think this could be a billion-dollar business for Amazon.”
His lunch mate warned him about how audacious that goal seemed. Two decades later, AWS didn’t just hit that billion-dollar mark. It obliterated it. In 2025, AWS generated $128.7 billion in revenue. Last year alone, it contributed $45.6 billion in profit to Amazon’s bottom line. That’s profit, not just revenue—the lifeblood that funds everything else Amazon does.
And now, AWS is turning 20. But it’s not slowing down. It’s going all in on AI.
The Birth of a Cloud Giant: How It All Started
Amazon didn’t set out to build a multibillion-dollar cloud business. It started with a practical problem: Amazon’s own engineers needed a better way to manage the infrastructure behind its e-commerce operations. The company had developed internal tools for computing, storage, and networking. Someone had the idea: “What if we sold this to other companies?”
In 2006, AWS launched its Simple Storage Service (S3). That date—March 13—is now considered AWS’s official birthday. It was a radical concept. Instead of buying servers, managing data centers, and hiring teams of sysadmins, any organization could rent computing power and storage from Amazon. You could build an entire online presence without touching a single piece of hardware.
Wall Street analysts were baffled. One famously complained, “I have yet to see how these investments are producing any profit.” Back then, if AWS collected $100 in a single day, it was a big deal. The idea that it could ever become a billion-dollar business seemed laughable to outsiders.
The Skepticism That Almost Killed Cloud Computing
Even as AWS gained traction, detractors had a recurring meme: “Cloud will quickly become a commodity. Everything will normalize out.” The fear was that once everyone offered the same basic storage and compute services, margins would collapse, and AWS would be just another utility provider.
Garman, now CEO of AWS, reflects on that period: “When we started to get a little traction, there was this kind of meme about how AWS would quickly become a commodity and everything would kind of normalize out. And our team has shown incredible invention to prove that that’s not true.”
That invention is the key. AWS didn’t just succeed because it was first. It succeeded because it kept inventing. It added databases, machine learning, analytics, IoT, security tools, and eventually, AI services. Every time the market thought cloud was a done deal, AWS launched something new that changed the game.
The Competition That Followed
AWS’s success didn’t go unnoticed. Two other tech giants saw the opportunity and built their own cloud platforms: Microsoft Azure and Google Cloud. What started as a single-player market became the most competitive battleground in tech history.
Today, the three cloud providers spend billions annually on data centers, chips, and AI research. They compete on price, performance, features, and—most importantly—developer mindshare. But AWS remains the leader by a wide margin, controlling roughly one-third of the global cloud market.
That dominance is no accident. AWS’s early-mover advantage gave it a decade-long head start in building relationships, tools, and ecosystems. But the real secret sauce is something else: AWS never stopped listening to customers.
How AWS Turned Skeptics Into Evangelists
At first, large enterprises were wary of cloud computing. Ceding control of critical infrastructure to a third party felt risky. But AWS won them over by proving itself reliable, secure, and scalable. Today, organizations of all sizes—from startups to Fortune 500s—run their core operations on AWS.
The platform now offers more than 200 services. You can spin up a virtual server in minutes, store petabytes of data, run machine learning models, host websites, stream video, and even build quantum computing experiments. It’s a far cry from that first S3 storage bucket.
The key lesson for SaaS leaders? Don’t just build a product. Build a platform that makes your customers’ lives easier. AWS did it by removing the pain of infrastructure management. You can do it by removing the pain of whatever your customers struggle with.
Where AWS Goes Next: The AI-First Future
Now, at 20 years old, AWS is making a massive bet on artificial intelligence. The company has launched Amazon Bedrock (a managed service for building generative AI apps), Amazon Q (an AI-powered assistant for developers and business users), and its own custom AI chips (Trainium and Inferentia).
This isn’t just a side project. AWS is integrating AI into nearly every service it offers. From S3 to Lambda to SageMaker, AI is becoming the default layer. Garman has made it clear: AI is the next wave of cloud innovation, and AWS intends to lead it.
What does this mean for B2B companies? If you’re not already thinking about how AI fits into your product, you’re already behind. AWS is betting that every business will need AI capabilities embedded in their workflow. The same way every company eventually needed a website, every company will soon need AI.
What SaaS and Tech Leaders Can Learn From AWS’s 20-Year Journey
There are five takeaways from AWS’s growth story that apply to any SaaS or tech company.
1. Start With a Problem, Not a Venture
AWS began as an internal solution to a real operational challenge. It wasn’t developed in a boardroom. It came from engineers trying to make their own lives easier. The best products are born from genuine pain.
2. Ignore the Noise
Wall Street analysts didn’t believe in cloud computing. Neither did many of Amazon’s own employees. But the team kept building. If you have conviction in your vision, don’t let early skepticism distract you.
3. Keep Inventing—Even When You’re Winning
AWS could have rested on its laurels after crushing the $1 billion mark. Instead, it kept adding services, improving performance, and lowering prices. Complacency is the enemy of long-term success.
4. Think in Decades, Not Quarters
AWS didn’t become a profit powerhouse overnight. It took years of investment before the returns materialized. Optimize for long-term value, not short-term quarterly hits.
5. Make AI Your Core, Not an Add-On
AWS’s current bet is that AI will reshape every aspect of cloud computing. Whether you’re building a CRM, a marketing platform, or an analytics tool, AI should be embedded from the ground up—not bolted on as an afterthought.
The Road Ahead: What’s Next for AWS?
With $128.7 billion in annual revenue and a 20-year track record of innovation, AWS is far from a mature business. The cloud market is still growing, and AI represents perhaps the biggest opportunity since the internet itself.
Garman, who took over as CEO in June 2024, is now leading the company into its third decade. His message is clear: AWS will keep inventing.
“Our team has shown incredible invention to prove that [the commodity theory] is not true,” he says. That same spirit will define the next 20 years.
For B2B leaders, the lesson is simple. The companies that win are the ones that solve real problems, ignore the doubters, and keep innovating—even when they’re already on top.
AWS didn’t just change Amazon. It changed how the entire world builds software. And it’s just getting started.
Matt Garman photo credit: Josh Edelson for AWS