No network for Uber

This post was co-written with Jess Peterson.

Uber, love it or hate it, is one of the most fun companies to watch in Silicon Valley. Semil Shah just launched a newsletter to cover everything Uber, and it’s easy to understand why. The company is a bad-behaving operational machine, with an unquenchable ambition.

Despite all the scrutiny, there’s a common misconception: Uber is a monopoly due to network effects.

The mistake comes from not understanding that there are two kinds of “network effects”. What we in Silicon Valley call a network effect is actually a demand-side economy of scale. This is one of a handful of advantages that leads to a long-term monopoly.

Simply, this kind of network effect is where each additional user within a system increases the value for every existing user. The service is more valuable to me when you’re here too. Classic examples are LinkedIn or Facebook: you can’t beat either by having a better product, even a much better product. The value comes from your friends and colleagues being there.

The other “network effect” is actually a supply-side economy of scale. And this is far less defensible. A great example is bank ATMs. I use Bank of America because their ATMs are everywhere. It makes no difference to me if I’m the only Bank of America customer in the world.

Supply-side economies of scale can be bought. If a new competitor wants to start a bank to rival Bank of America, they can build an equally-sized ATM network. It would be expensive, but possible. On the other hand, demand-side networks can’t be bought. Money can’t replace your cousin’s wedding photos on Instagram.

Uber has scale, not a demand-side network effect. If say, Google, wanted to compete against Uber– how would they do it? They’d pay drivers twice as much and pay them to wait for rides, even in under-utilized areas. Or skip the drivers and go full self-driving. Either way the playbook is simple: use dollars to compensate for lack of scale.

Further, Uber is not a marketplace (like Airbnb or eBay), because the “products” on its shelves are indistinguishable — diversity doesn’t matter. I don’t care if there are 10 drivers or 10,000 drivers as long as they come to my house in two minutes.

Uber is aggressive, it’s the dominant player and it has mindshare. Those are huge advantages and they’re expensive to beat. But there are no networks effects.