4 ms·

This. I never got how Uber was still unprofitable for so long. Sure you have a bunch of Silicon Valley salaries to pay but realistically speaking, one could bui

This. I never got how Uber was still unprofitable for so long. Sure you have a bunch of Silicon Valley salaries to pay but realistically speaking, one could build an Uber/Lyft for under $1M and basically can operate with two or so devs fixing issues. Yet, they are blowing hundreds of millions of dollars yearly all over the place while shorting drivers who can’t make a living wage and are basically trading vehicle expenses down the road for payment today.

ozim 6y ago
I think they are blowing hundreds of millions of dollars yearly on marketing/sales people not on developers.
ta988 6y ago
I'm not sure you realize that a product at that scale can't run with two people. You need lawyers, service support, testers, advertisers... People that hacl the other company to sabotage their launch (if you are not aware, uber did that).
- agloeregrets 6y ago
  I was referring to Silicon valley salaries, obviously you would need ~ 50-100 people or such. At current scale of Uber, if they charged just a penny (instead of the 20-30% they currently do) They would make 16 million dollars which could pay 100 people an average of 100K per year, they could spend $4M on infrastructure and still make $2M in profit despite being not much more than a payment processor and app provider (like square or shopify) They bill more than 200X that on average.
n4r9 6y ago
I subscribe to an email newsletter called BIG by Matt Stoller. It's all about US monopolies and he's very interested in exactly what you describe. His response is that both ride-hailing and delivery apps are taking a gamble on establishing a monopoly in several years time, at which point the profits will pay off the current losses.
- andrepd 6y ago
  This is evident. We're at such a crazy level of capitalism/neoliberalism that it's a regular occurrence to see large businesses operating at a loss for 5 or 10 years.
  - twelvechairs 6y ago
    Being able to run at a major loss for 5-10 years is great for 'hard tech'. For a simple phone app though its a different story.
mindvirus 6y ago
Which of those two devs will be responsible for ensuring taxes are properly charged and keeping up to date with the tax code? Which will maintain parking and map zones? Who will work on routing and matching? Who's working on fraud? On safety? Who is working on payments, both from riders and paying drivers? Who is generating required tax reports for drivers? Who's doing driver onboarding? Which will ensure that drivers have the proper paperwork to operate in their market (insurance etc)? And who is working on Android and iOS apps, both for drivers and riders? Who's working on the backend?
- agloeregrets 6y ago
  My numbers were talking app expenses, obviously a company of Uber's scale would need 100+ employees. I broke down below that if they worked at their current scale and charged just a penny, they could pay their employees an average of $100K per year and still make millions in profit after spending millions on Infrastructure. Uber bills on average more than 100-200 times that and spends insane numbers on unprofitable projects like self driving cars. I've read the whole book, a set of co-ops could easily spend as little as a million dollars, retain devs for work on it, bill a tiny service fee of a penny per ride (plus card processing) and then have local co-ops operate the other half of everything above at a reasonable scale whereas drivers get paid the rest other than management costs. Uber is intentionally using automation to overswamp driver markets to keep prices low and driver pay low to boost ride numbers and eliminate competition.
  - aerosmile 6y ago
    Nope, your words were clear - you were flat out wrong, and are now trying to save face. I don't have a tendency to call people out, but blindly applying the "two devs in a garage" axiom is not only wrong when it comes to labor marketplaces, but it also unnecessarily spreads discontent to the point that we have to stop this silliness. I don't care one bit about your recalculated math from now on - for me, you're just another armchair quarterback who speaks before he thinks.
    - ryandrake 6y ago
killbot5000 6y ago
The Uber business model is: Undercut competition. Use borrowed money to sell rides under cost. Wait for all competitors to die. Jack up prices. AFAIK they’re still living on borrowed money. It reminds me of the airline deregulation of the 90s. New airlines would use borrowed money to sell seats at below cost to attract customers, driving established carriers out of business. The funny thing is that this smells a lot like “the tragedy of the commons”. Everyone wants to use this amazing infrastructure for flying, but no one wants to pay for it. New firms undermine the stability of the system by charging less than cost in order to starve established competitors whose business model is focused on being profitable.
- tmoertel 6y ago
  > The Uber business model is: Undercut competition. Use borrowed money to sell rides under cost. Wait for all competitors to die. Jack up prices. The interesting thing is that there is more than one startup using this model. The competition isn't the other startup; it's the other startup's investors. I wonder how long those investors will keep on pumping money into the scheme, hoping their guy will be the last standing. Or will they ride the sunk cost fallacy all the way down?
- missedthecue 6y ago
  That's not their business model. Uber knows very well how low the barrier to entry is in ride hailing. It's not like they're building railroads here. It's an app.
objclxt 6y ago
> one could build an Uber/Lyft for under $1M and basically can operate with two or so devs fixing issues This is fundamentally untrue, although you hear it a lot here. I have to wonder whether the people who think this have actually worked on a system at the scale Uber operates at. Uber is an incredibly complicated system. The trick is it's presented to the user as a very simplistic one, so people overlook what's actually going on behind the scenes. Sure, like many well-funded tech companies with a large engineering staff there's a fair amount of fat you could probably cut away (and indeed, Uber have - they've done engineering layoffs in the past). But just to sustain the app in all the territories they currently operate in you're talking hundreds of engineers, not "two or so devs fixing issues". That's how large the problem surface area is.
MrMan 6y ago
This is my analysis too. The app should be open sourced and then subsequent new market entrants enjoy the benefit, lowering costs. I don't think running servers to handle a few transactions per minute per zip code is too expensive

ozim 6y ago

I think they are blowing hundreds of millions of dollars yearly on marketing/sales people not on developers.

ta988 6y ago

I'm not sure you realize that a product at that scale can't run with two people. You need lawyers, service support, testers, advertisers... People that hacl the other company to sabotage their launch (if you are not aware, uber did that).

agloeregrets 6y ago
I was referring to Silicon valley salaries, obviously you would need ~ 50-100 people or such. At current scale of Uber, if they charged just a penny (instead of the 20-30% they currently do) They would make 16 million dollars which could pay 100 people an average of 100K per year, they could spend $4M on infrastructure and still make $2M in profit despite being not much more than a payment processor and app provider (like square or shopify) They bill more than 200X that on average.

n4r9 6y ago

I subscribe to an email newsletter called BIG by Matt Stoller. It's all about US monopolies and he's very interested in exactly what you describe. His response is that both ride-hailing and delivery apps are taking a gamble on establishing a monopoly in several years time, at which point the profits will pay off the current losses.

andrepd 6y ago
This is evident. We're at such a crazy level of capitalism/neoliberalism that it's a regular occurrence to see large businesses operating at a loss for 5 or 10 years.
- twelvechairs 6y ago
  Being able to run at a major loss for 5-10 years is great for 'hard tech'. For a simple phone app though its a different story.

mindvirus 6y ago

Which of those two devs will be responsible for ensuring taxes are properly charged and keeping up to date with the tax code? Which will maintain parking and map zones? Who will work on routing and matching? Who's working on fraud? On safety? Who is working on payments, both from riders and paying drivers? Who is generating required tax reports for drivers? Who's doing driver onboarding? Which will ensure that drivers have the proper paperwork to operate in their market (insurance etc)? And who is working on Android and iOS apps, both for drivers and riders? Who's working on the backend?

agloeregrets 6y ago
My numbers were talking app expenses, obviously a company of Uber's scale would need 100+ employees. I broke down below that if they worked at their current scale and charged just a penny, they could pay their employees an average of $100K per year and still make millions in profit after spending millions on Infrastructure. Uber bills on average more than 100-200 times that and spends insane numbers on unprofitable projects like self driving cars. I've read the whole book, a set of co-ops could easily spend as little as a million dollars, retain devs for work on it, bill a tiny service fee of a penny per ride (plus card processing) and then have local co-ops operate the other half of everything above at a reasonable scale whereas drivers get paid the rest other than management costs. Uber is intentionally using automation to overswamp driver markets to keep prices low and driver pay low to boost ride numbers and eliminate competition.
- aerosmile 6y ago
  Nope, your words were clear - you were flat out wrong, and are now trying to save face. I don't have a tendency to call people out, but blindly applying the "two devs in a garage" axiom is not only wrong when it comes to labor marketplaces, but it also unnecessarily spreads discontent to the point that we have to stop this silliness. I don't care one bit about your recalculated math from now on - for me, you're just another armchair quarterback who speaks before he thinks.
  - ryandrake 6y ago

killbot5000 6y ago

The Uber business model is: Undercut competition. Use borrowed money to sell rides under cost. Wait for all competitors to die. Jack up prices. AFAIK they’re still living on borrowed money. It reminds me of the airline deregulation of the 90s. New airlines would use borrowed money to sell seats at below cost to attract customers, driving established carriers out of business. The funny thing is that this smells a lot like “the tragedy of the commons”. Everyone wants to use this amazing infrastructure for flying, but no one wants to pay for it. New firms undermine the stability of the system by charging less than cost in order to starve established competitors whose business model is focused on being profitable.

tmoertel 6y ago
> The Uber business model is: Undercut competition. Use borrowed money to sell rides under cost. Wait for all competitors to die. Jack up prices. The interesting thing is that there is more than one startup using this model. The competition isn't the other startup; it's the other startup's investors. I wonder how long those investors will keep on pumping money into the scheme, hoping their guy will be the last standing. Or will they ride the sunk cost fallacy all the way down?
missedthecue 6y ago
That's not their business model. Uber knows very well how low the barrier to entry is in ride hailing. It's not like they're building railroads here. It's an app.

objclxt 6y ago

> one could build an Uber/Lyft for under $1M and basically can operate with two or so devs fixing issues This is fundamentally untrue, although you hear it a lot here. I have to wonder whether the people who think this have actually worked on a system at the scale Uber operates at. Uber is an incredibly complicated system. The trick is it's presented to the user as a very simplistic one, so people overlook what's actually going on behind the scenes. Sure, like many well-funded tech companies with a large engineering staff there's a fair amount of fat you could probably cut away (and indeed, Uber have - they've done engineering layoffs in the past). But just to sustain the app in all the territories they currently operate in you're talking hundreds of engineers, not "two or so devs fixing issues". That's how large the problem surface area is.

MrMan 6y ago

This is my analysis too. The app should be open sourced and then subsequent new market entrants enjoy the benefit, lowering costs. I don't think running servers to handle a few transactions per minute per zip code is too expensive