eBay Didn't Lose to Amazon in 2008. It Lost to Skype in 2005.

In September 2005, eBay — the company that had taught the world to buy a stranger's old camera at 11pm — wrote a check for an internet phone company. The price was about $2.6 billion in cash and stock, with another $1.5 billion in earn-outs dangling behind it.² The logic, briefly, was that buyers and sellers would want to talk before they traded. They didn't. Four years later, eBay sold off most of Skype at a valuation lower than what it had paid.⁴ But the money was never the real loss. The real loss was the two years of corporate attention spent integrating a product that had nothing to do with the marketplace — the years Amazon spent building Prime.

The official story is that Amazon raced past eBay in the early 2000s, a fast clean overtaking. It didn't happen that way. Amazon's enterprise value did not outgrow eBay's until 2008 — and in many countries it took years longer.⁷ eBay didn't get knocked off the throne in a battle. It got up, wandered away, and came back to find the chair occupied.

Here is the thesis a smart friend can repeat at dinner: eBay didn't lose the crown to a better competitor in one stroke; it surrendered it slowly, between 2004 and 2008, through a compounding sequence of capital-allocation failures — while its own revenue kept rising the entire time. The decline was real. The collapse is a fiction.

The revenue kept growing — which is exactly why nobody panicked

The most dangerous kind of decline is the kind that doesn't show up in the top line. Look at eBay's own SEC filings: net revenue went $748.8M, $1.214B, $2.165B, $3.271B, $4.552B, $5.970B from 2001 through 2006.¹ That is not a company in crisis. That is a chart that goes up and to the right every single year — the kind of chart that buys management the benefit of the doubt for any acquisition it wants to make. eBay was still considered the world's largest e-commerce site as late as 2001, and it kept compounding through 2006. There was no quarter where the alarm bells could have rung, because the numbers never gave anyone a reason to ring them.

But eBay's revenue is a trap for the analyst, because it isn't the same number Amazon reports. eBay's reported revenue is its cut — transaction fees and commissions on goods other people sold. Amazon's reported revenue includes the full price of everything it sold first-party. So when Amazon's net sales hit $6.92B in 2004 versus eBay's $3.27B⁵, the headline 'Amazon is twice as big' was almost meaningless: it compared a toll-taker's tolls against a retailer's gross receipts. The two companies recognized revenue in fundamentally different ways. The real contest was never visible in a side-by-side revenue line — which is precisely why the early-2000s-takeover story is so wrong, and so durable.

Where the capital actually went

Strategy is where you point your money and your attention. In February 2005, Amazon announced Prime in the same breath as its 2004 results — net sales up 31%, the whole machine being reinvested into faster shipping and a customer-loyalty flywheel.⁵ By the next year, Amazon's operating cash flow had grown 29% to $733M, funding more of the same.⁶ Amazon was pouring its capital into the one thing a marketplace can't fake: the experience of the package arriving fast and free. eBay, in the same window, pointed its capital at Skype.

Watch the mechanism, not just the deal. eBay had actually made one brilliant acquisition — PayPal, for roughly $1.5 billion in 2002, which it sensibly let keep its own brand while quietly retiring its in-house Billpoint service.⁹ That was capital allocation done right: buy the rail your own transactions ride on. Skype was the opposite move three years later — buying a business adjacent to nothing eBay did, on a theory of buyer-seller voice chat that the market never validated. The deal closed in October 2005, eBay wrote down $1.4 billion of it in 2007, settled the earn-outs for $530M, and in 2009 sold 65% of Skype at a valuation of just $2.75 billion — below what it had paid to begin with.³⁴ The dollars hurt. The two years of executive attention hurt more, because they were spent on a phone app instead of on shipping speed, while the competitor that was about to take the crown spent those same two years on exactly that.

And then eBay broke its own auctions

If the Skype era was the distraction, September 2008 was the self-inflicted wound. eBay's identity was the auction — the live drama of bidding, the thing it had taught the world. In 2003, auctions were 96% of active listings.⁸ By 2015, they were 7.2%. The collapse wasn't because buyers suddenly hated bidding; the peer-reviewed data points at seller incentives. When eBay introduced permanent 'Good Till Canceled' fixed-price listings in September 2008, the auction share of listings fell from 65.4% to 33.4% in four months.⁸ eBay changed the rules so that posting a fixed price was the easy, default, economically rational thing for a seller to do — and in doing so it quietly dismantled the one experience that made eBay eBay, converting itself into a worse, less distinctive version of the catalog model Amazon already dominated.

The honest objection: wasn't Amazon just going to win anyway?

The fair counter is that none of this was decisive — that a logistics juggernaut willing to lose money for a decade to build fulfillment was always going to beat a fee-taking marketplace, Skype or no Skype. There's real force to that. Amazon's first-party retail-plus-logistics model is structurally a different and arguably stronger machine than a commission marketplace, and Prime created a loyalty loop eBay had no equivalent for. So the counterfactual where eBay skips Skype and still loses is entirely plausible.

But notice what the steelman concedes: that the contest was close enough to matter for years. Amazon's enterprise value didn't pass eBay's until 2008⁷ — three years after the Skype deal, the same year eBay broke its auctions. The window where eBay could have pressed its advantages was not theoretical; it was open, and it stayed open longer than anyone now remembers. eBay had the leading marketplace, it owned PayPal, and it had a global brand. It did not need to out-build Amazon's warehouses. It needed to defend the two things that were genuinely its own — the trusted payment rail and the irreplaceable auction — and pour its capital into them. Instead it bought a phone company and then deprecated its own format. The decline was not destiny. It was a sequence of choices, each one survivable, that compounded.

That is the real lesson of how eBay lost the e-commerce crown, and it's stranger than the legend. There was no single catastrophic quarter, no dramatic collapse, no moment the music stopped. Revenue rose the whole way down. The crown didn't get seized — it got misplaced, $2.6 billion and one rule change at a time, by a company that was too busy looking elsewhere to notice it had set the thing down. The most expensive way to lose is the one that never shows up on the income statement until it's already over.