Amazon actually didn't have much competition at a key point in it's growth, i.e. just after the dot com crash. Amazon survived, many did not and Amazon then had a few years to cement their dominance in the online shopping space.

Amazon is also very different from Uber in that it a) went public earlier and b) didn't need huge outside cash injections like Uber and c) Wasn't making a huge gross loss on sales like people claim Uber is.

I believe Uber has already lost more money than Amazon has in it's entire existence!

http://s1.ibtimes.com/sites/www.ibtimes.com/files/styles/emb...

Also a good point. Low margin doesn't have to mean unsuccessful. But it's taken 20 years for Amazon to get where it is. Uber's valuation seems pretty premature. Amazon lost money for years because it was building infrastructure. Uber is losing money because it is subsidizing its sales. Amazon may have been low margin, but I don't know that they ever took investor money and gave it directly to customers in the form of price subsidies.

Amazon also has crazy AWS profits to subsidize major investments. Uber lacks such a super profitable but tangential revenue stream.

And now with them having become an online-only warehouse club [1] with Prime, there is even less reason to switch away.

[1] https://en.wikipedia.org/wiki/Warehouse_club

Since Amazon seems to have shifted to generate profit instead of subsidizing prices, it went downhill (at least in Berlin). Amazon Logistics is a huge pain (several lost packages, Saturday and 8pm delivery to business) and the prices of most goods at Amazon is mediocre.

As a very good and Amazon customer for a very long time, I've reached the point to jump to something else should it materialize.

Amazon got kicked by Zalando for Fashion at least in Germany and despite heavy investments and promotions for fashion can't compete with Zalando.

If someone starts an electronics site on the same professional, data drive, aggressive level as Zalando, I'm sure Amazon is in trouble here in Germany.

Network effects don't accumulate to a natural monopoly in this case. Yes you need a fleet large enough to get to a sweet price / waiting time optimum, but each new car in your fleet has a diminishing marginal effect.

You could say exactly the same for airlines: there's network effect. Each new plane allows your to fly more rotations, have more frequent lines, open new routes... The reality is that most people take decisions based on price.

We know most people are pretty bad at valuing their time, and even so, waiting a minute more for a cheaper service is something people will do for sure. As long as you're below an acceptable waiting time for your car, you can compete on price.

I'm curious about how Uber will avoid being a commodity just like airlines are.

The problem is that in this case, Uber doesn't own the car network. It's trivially easy for a driver to shift between networks, or even work for multiple networks simultaneously. Uber not only has to talk a rider into giving them the fare, they also have to talk the driver into taking that fare. There's just nothing special about Uber for either party unless Uber can be the one that performs that matchmaking service at the lowest possible transaction cost. Welcome to Uber's low-margin world.