Amazon Just Surpassed Walmart as the World’s Largest Company. It Took Thirty-Two Years.

In 1994, Jeff Bezos set up Amazon in the garage of his house in Bellevue, Washington. The company wasn’t even called Amazon yet — the first name was “Cadabra.” That same year, Walmart was already, by far, the largest retail chain in the United States: national presence, a distribution network nobody could match, satellite-connected inventory systems, and the trust of millions of consumers that Amazon could only dream of.

Thirty-two years later, just a couple of months ago, Amazon surpassed Walmart as the company with the highest revenues in the world — ending Walmart’s thirteen-year consecutive run at the top of the Fortune 500 ranking.

It wasn’t a coincidence or a sudden stroke of luck. In 2010, Walmart still posted revenues more than twelve times higher than Amazon’s. The difference wasn’t the size of the bet, but where they placed it. Walmart treated the internet as an additional channel on top of a business that already worked perfectly fine without it. Amazon built the entire business around the internet, from day one, with nothing to dismantle first.

This isn’t an isolated case or a nice story to tell at a motivational talk. The World Bank studied this with data from thousands of companies across more than a hundred developing countries, and found something very consistent: when a general purpose technology arrives in an industry, small companies capture disproportionately larger productivity gains than large companies. The benefit shows up in high- and low-tech businesses, exporters and non-exporters, of any size — but small companies gain the most. Not because they’re smarter. Because they don’t carry the structure, inherited processes, and approval committees that slow a large company down fifteen steps before it can try anything new.

Artificial intelligence is creating exactly the same dynamic right now. A team of ten people with AI truly integrated into how it researches, builds, sells, and serves clients can do today what just three years ago required fifty people. Meanwhile, the large company keeps trying to fit AI inside workflows, committees, and hierarchies designed for a world without it — exactly like Walmart treated the internet as a separate channel while Amazon was building everything around it.

More recent research on cloud adoption shows the same repeated pattern: small and new companies are, systematically, the ones that benefit most from each new wave of technology infrastructure — precisely because they have nothing to dismantle before starting.

I’m not saying that being large is a death sentence. Walmart, after all, did eventually react: in 2014 it brought in a CEO with a digital mindset, acquired an e-commerce company, and rebuilt its operation until its thousands of stores became part of a logistics network that now competes with Amazon head-to-head on several fronts. But it took nearly two decades to realize and course-correct. That’s the window we’re talking about.

For a small team, a startup, or a division inside a larger company that operates almost like a startup, Amazon’s lesson is the most hopeful in this entire series: the advantage of having nothing to dismantle is real, it’s documented in the economic evidence, and today it’s available to anyone willing to build around AI instead of bolting it onto what already exists. But that window isn’t going to stay open forever. Every year that passes, the large players learn a little faster how to dismantle their own central shaft.

If artificial intelligence is repeating the same pattern as the internet in the ’90s, does your company today look more like Walmart in 1994, or like Amazon in a garage?