3 ms·
Coinbase scaling: Uber's demand with Facebook's support
In light of what just happened with coinbase, I though we should talk startup scaling..
PG tweeted in 2014 that coinbase crossed 1 million wallet.
<https://twitter.com/paulg/status/439098584378929152>
that's almost 3 years ago..why do you think they are still struggling to scale?
- phone support is still from 9-5
- email support is a blackhole
- so many bugs
- so many angry users on twitter, reddit and forums..
The company has so much to unleash and could conquer the world, but its not catching up..I don’t know what’s happening behind the scene but I think as a start maybe have a 24/7 call center? reply to emails faster?
I cannot help but to think of Travis Kalanic and what he did at uber..I still remember when they first launched in Boston, there weren’t many drivers in 2012..You had to wait 25 minutes for a ride, but they quickly scaled….by 2013 everyone in boston/cambridge was using uber… the wait time was reduced to 10 minutes, now its 2017 the wait time is less than 4 minutes, sometimes even seconds..
why cannot coinbase scale?!
Please no mean comments or trolling. I'm just curious to hear what others think..
- adjkant 9y agoScaling means very different things for Uber and Coinbase. This is not to make excuses, but handling so much instantaneous trading is a lot harder than a server responding to a ride requests and searching for the closest driver. By multiple magnitudes. Coinbase is not without its flaws, and they could be doing better, but it's a much harder problem than Uber faced. For Uber it wasn't even a problem technically, it was just a network effect.
- maram 9y ago>much harder problem than uber Respectfully disagree. We simply cannot compare the difficulty level of Uber and Coinbase to each other. But for the record, they faced many technical challenges..accurately matching you with a nearby driver and predicting ETA required a lot of technical work.. Uber's success wasn't only a matter of network effect. Uber is known for efficiency and what personally fascinates me is that the bigger they grow the more efficient they become. Uber out-mathed Google on ETAs..and to scale they hired hardcore mathematicians,rocket scientist, a computational neuroscientist and a nuclear physicist. Their support from the first day I used it never took more than 24 hours.
- adjkant 9y agoAgain, technical challenges generally and scale are not the same thing. Also I'm sorry but I don't think those challenges are nearly that hard. ETA's are not that hard to solve, nor am I convinced that they beat Google on ETA's. In fact most Uber ETA's change consistently along the ride and are never accurate for me (before I switched to Lyft). For matching drivers, all you have to do is look at all the X closest drivers available (an easy query given that drivers don't stray more than 50-100 miles from their home, usually much less), plot the routes using a routing API for each, and pick that with the lowest estimate. The pool algorithm gets a bit harder but is more or less just a preference for other close pool drivers when they are in a state where more riders are allowed to be added. Uber's app is copied regularly to a workable point by competitors. It's not that complicated and scaling is relatively simple given that it's more or less just upping the number of servers. Their support is good, but again, is not really scaling related. Coinbase never offered good support. Scaling isn't the issue, a lack of caring for support is. ---------------------------------------- The point to all of this being that it's pretty useless to compare scaling of Uber and Coinbase. Coinbase needs to improve, but that can be done just by looking at them alone with enough experience in the needed relevant areas. Uber is not applicable here.