|
|
|
|
|
|
by Shindi
1268 days ago
|
|
|
People think that they can build Lyft in a weekend with simple GPS + queue system in reality it's a ton more complicated. There are 1. Dynamic pricing. Both drivers AND riders have different price elasticities depending on surge, etc. I keep hearing stories of "rider paid $100 but driver was only paid $30". It's almost always the case that a ride like that had surge pricing for a rider but not for the driver. In some scenarios drivers will get paid a higher fraction of the amount riders paid to incentivize more drivers. 2. Bonuses, streaks, and other incentives. Lyft/Uber are famously commoditized. So how do you incentivize drivers to driver for Lyft more? Using bonuses and streaks. There are whole teams that handle this. 3. National scale. Each taxi operator is roughly localized. Lyft is a national service. Uber is international. That's going to come with a ton more complexity. 4. Self driving. Ah yes, everyone loves to hate on self driving bc of broken promises. However, if you look at recent news of Waymo/Cruise in SF, you'll see it's having it's 0->1 moment. Uber/Lyft have to handle this in their business because ride share is actually THE path to commercialization for Waymo/Cruise. Taxi cab companies were never going to support self driving, and for the above reasons Waymo/Cruise are not going to build their own ride share infra. |
|
|
They're trying to obfuscate and gamify the economics enough to disguise that they simply don't work out long-term for many drivers.
There are probably other models, many of which require much less complexity. Imagine a scheme with drivers on flat salaries, rather than trying to tweak exactly how much you have to pay for the 10 minute drive to compensate for 45 minutes of travel to and from the trip endpoints and sitting idle waiting for a summon.
3 is an extremely marginal feature. I need a Lyft in Chicago when I visit once or twice a year; I could equally use Happy Shiny Chicago Rideshare Ltd, and the fact they have no service in my hometown is rather immaterial. Maybe they could go for a co-op sort of concept-- regional operators who contribute towards shared marketing and development budgets.