Drivers care that lots of people use Uber and you care that lots of drivers use Uber. So there's definitely a network effect. It's just a lot weaker than something like Facebook because I don't care how many Uber drivers there are in Memphis if I'm in Salt Lake City. It's even weaker than that because the geographic area is more like "within 10 minutes of me". You can see a competitor relatively easily getting a toe hold by starting service at a specific time for a specific area. For example, the popular bar street on Friday/Saturday nights.

The problem comes with your "a dollar less" caveat. You star EpicEng's ride share and undercut Uber/Lyft by a dollar. They respond by matching your price and giving drivers more incentives so they don't switch. Now what do you do?

It's not network effects so much as the strength of the two-sided market they've created. Installing an app and registering is a lot of friction. On either side you'll probably bother doing so for no more than one or two services, which in most situations will cover most of whatever market exists.

But in general the answer is either having a disruptive strategy or else a huge amount of money to woo existing market participants. Most of us don't have either. That's the barrier to entry.

Today, VC money subsidizing rides is the advantage Uber and Lyft have.