I've also never heard of anyone using Uber, though it is an option and I'm sure a few do.
Didi and Kuaidi are by-far the leaders, completely entrenched. I think first-mover advantage will be the case here. The big thing these services are now competing on however is not hailing a ride, it is providing financial services through the app. They're no longer ride-hailing apps, they're finance apps with the option to hail.
[Perhaps a big difference with Uber compared to the US, is the use of it by taxis. Almost all taxis are on Didi/Kuaidi (indeed, these were taxi-only services at the start) and taxis are inexpensive.]
Uber has _already_ generated non-zero revenues in China. I know because I have paid them money myself.
You are right that services like didi dache are head-on competitors.
You are wrong about calling a preferred driver. Sure, I do this when I want to pre-book an out-of-town journey or trip to the great wall.
But, if I want a car in 5 minutes, from a random part of town, there's no way I can achieve that reliably by calling a guy, even he has a 1st degree connection who happens to be 5 minutes away from my current location.
While you may be in the tiny-to-irrelevant minority of people in Beijing that use Uber and find it useful (probably mostly foreigners?), Beijing is not a normal Chinese city and in the scale of the Chinese market foreigners are statistically insignificant. I remain deeply skeptical "a car in five minutes" is reliably available in Beijing from Uber in many locations outside of the center of the city.
However, car-hailing apps make more efficient use of passengers' time, drivers' time, and cars themselves, compared with passengers calling a guy. Taxis make good use of drivers and cars, but waste passengers' time as they are hard to hail in many places in large cities.
Car-hailing apps are here to stay and, in the long run, some company will make money. Whether any single company will make monopoly-type rents due to being the only game in town, though, ...
But Uber drivers provide their own vehicles, and can choose rides as desired.
The main service I am aware of is dididache or 嘀嘀打车 in Chinese which was apparently launched in 2012 (same as Uber).
My observation is that a lot of people just call their preferred driver, too. In the event that the driver is busy, they generally forward the business to a trusted friend. It's pretty hard to compete with free, personally reputation-audited referrals and 24 hour service.
Further, only certain cities have functional traffic: Chengdu being a prime example. Many Chinese cities' road networks have been so oversubscribed by the millions of new cars that have appeared in the last half decade or so that calling someone to pick you up is a physically impossible model during large portions of the day.
Yet another source of competition are e-bike or e-trike taxis, which have a few hundred dollar entry cost, hang around on many city corners, are totally unregulated and generally faster and cheaper (5-15元 = $1-3) than regular taxis (due to lack of road space).
I am skeptical Uber can generate any profits whatsoever in this market.