Hacker News new | comments | show | ask | jobs | submit login

I see a lot of comments on here about the efficiency of Lyft/Uber's and lots of comments about their overspending.

The new book by Reid Hoffman, Blitzscaling, was an eye-opener for me as to how these companies operate and what seems as incredibly inefficient now will all be but forgotten tomorrow when they have complete monopolies over all transportation... The nature of technology makes it winner take all almost always, it's only a matter of time until they can offer other forms of transportation to people. Perhaps people someday won't even own any cars at all especially if it's cheaper to call an autonomous one from your device whenever you need it.

These are interesting times we are living in and these business models are completely counterintuitive to traditional norms of taking small risks and that much is obvious from seeing the business models of Uber/Lyft, WeWork, Airbnb, etc.






Why would ridesharing be winner take all? There's almost no network effect (beyond splitting the cost, but Venmo has largely solved that) and the drivers all drive for all of them.

In NYC I think there's at least 4 ridesharing companies and I usually choose which one based on which one has given me a promo.


There are plenty of network effects.

To start a new ridesharing company, you have to get drivers and riders. You'd have to offer them a ton of incentives to start driving/riding with your service, when the existing options are well established.

For drivers, Lyft and Uber have incentives (lower fees) if you do the majority of rides with their service, which incentivizes you to only drive with one service.

When's the last time you downloaded a new app on your phone? For me it's been weeks, maybe months. For so many people, it's easier to just not try your new rideshare service.


This is just not correct.

To start a new ridesharing company all you have to do is attack and carve off the most profitable element of the big company's product mix.

All you need is to undercut them in certain high profitability routes, like airport runs, or luxury vehicles for executives, or anything, and you have a toehold.

There are some positive network effects, but there are also negative network effects. Having more users doesn't mean you always win as the big company.

For example if you have no customers all you have to do is be outside of a stadium at the end of a show, or waiting by an airport during a busy landing period, and have rudimentary abilities to let people know you exist, and you'll get defectors.


Not sure why you're saying the op is not correct. Nothing you said discredits that.

The reality is that both are true.

There are multiple positive network effects at work, as it's evidenced by the growth of these companies.

However the barrier to entry is super low, so competition can be quite high. Like you said, just by subsidizing rides/drivers you will start gaining some market share.

In the end it's all gonna be about who has the most cash and whose efficiency is the highest. Yeah you could hire a promoter and give out flyers outside a venue, but if you check the data, the ROI is abysmal.


Without a large network, the time for a car to arrive is high and unpredictable.

This is bad for customers.


Not necessarily.

It's pretty easy to have a network that's not very big that always has someone at the airport, or always is available for trips that stay within a specific neighborhood, or can always be arranged with a little more advance notice, or whatever.

And the relevant part of pointing this out is that as long as that's true, which is forever, then these large companies will never be able to extract monopoly pricing.

As soon as they start gouging customers people will immediately flood in to siphon off the lowest hanging fruit or easiest segments of their customer base.

That dynamic will never go away, it's not at all clear what the point of all this buying of market share is, given that it will never be truly defendable.


Yes. You can overprovision in a constrained area and you can also let people schedule ahead of time. That's pretty much every black car service which often are not very big companies. But they tend to be relatively expensive. (Which is fine for some uses; I use one to travel to and from the airport.)

I'd make the argument that because there are low costs of switching (as has been mentioned elsewhere), these are scale effects (like in most industries), rather than network effects.

Remember that you're almost certainly not a trendsetter or first customer. You are never the person a startup goes after first. There's a whole group of people who behave very differently from you who do download new apps on their phones all the time.

There are entire chapters dedicated to talking about each group of adopters, and you saying that no one is in the "early adopters" group is simply not true.


From an economic standpoint if the winner is substantially more effective, the cost of producing the product is high, and the marginal costs of servicing additional users are small the system will end up as winner takes all. This is common in the software, sports and entertainment industries.

Because service like Lyft depends on more than just the software to provide the end service so the product is not substantially better. For the most part the drivers are also working for the competition, the time to get to the destination, and the overall cost will be similar. At some level the network effect is part of both the barrier to entry, and the quality/value of the product but it is not clear if this is enough to end up on top of a winner take all market.


The future of ridesharing is uber pool/lyft line in order to increase efficiency. The network effect is huge.

Have you ever, like, actually used the shared ride options?

It can be handy when you're not in a hurry, but as long as people are involved, and they aren't always where they say they are going to be, there's a pretty hard upper limit to how efficient things can get.


A shared car is great when you have little money, few other options of transportation, and the route allows to share it with other people.

I did it a few times. But usually when I need a ride, it's because I'm in a hurry, and paying $5 more is no object. Else mass transit suffices. This only works in urban areas, but it's where the most profitable ride-hailing markets seem to be anyway.


Ride sharing is generally supply constrained. The dirty secret is that drivers are considered internally as the real customers for ride sharing cos - they literally bring them all the money.

There is a network effect. Imagine if you made a ridesharing app with only one driver. No one would use the app due to high wait times. That's an edge case but it gets the point across: too few drivers and no one will use your app, and if no one uses the app no drivers will enlist in your service.

Of course your app with one driver would get usage. Your comment contains its internal contradiction right there in plain sight.

If "no one would use the app due to high wait times" then your driver would always be available and ready to go. So someone would be happy to use it whenever the dominant app is having trouble keeping up with demand. Which will always happen from time to time as long as demand has occasional peak periods and supply has some practical constraints, which a fleet of cars always will.


But drivers aren't locked in to a single service, and there's little to no cost for drivers to be on both Uber and MyStartup, so the network effect is diminshed.

You should read more into this, but even though switching seems easy in theory, the truth is different. Going through another driver application process / training takes time, and even then using two apps creates extra annoyance. You would be surprised how lazy People are even if there was a way to grow their earnings through switching.

You're right that people often make lazy decisions, but every single Uber I've taken in the last year had a Lyft sticker as well and vice versa.

Which doesn't necessarily indicate that they would drive 50/50 for both. It might be closer to 80/20 easily.

Users already have to make arrangements with traditional car services for early AM airport runs because there is no Uber of Lyft supply in most areas (outside of NYC, SF, LA).

You need to reread what a network effect is. It's exactly what goes on in a ridehailing platform.

I'm not really sure that's true at all.

If all of my friends are using Uber and I'm using Lyft, I get the same amount of value out of Lyft whether anyone else is using it or not.

Now, there is some network effect with new drivers signing up for a more dominant platform, making it harder to use others. Since drivers can (and usually do, in my experience) sign up for multiple platforms. That effect is greatly diminished.


Im not talking about a onesided network effect. Im talking about indirect network effect concerning a two-sided market.

If more of your friends use Uber than Lyft, that suggests that there are more customers on Uber.

This means Uber is more valuable for the drivers as a platform.

This means more drivers will driver for Uber.

This means Uber has better availability.

This means Uber is more valuable to customers.


The switching costs for ride-share are far lower than Reid's examples.

AirBnB can make recommendations based on past experience. Uber/Lyft are one-time transactional commodities. There's no loyalty incentive.


I'm not sure WeWork is a great example either. Sure, they're the hip kid in the co-working space these days but there are tons of other providers of co-working and other office facilities around the world, both chains and one-offs.

AirBnB is a reasonable example because there are fairly strong network effects (as with e.g. eBay--especially in the old flea market days).

There are definitely some network effects with Uber and Lyft but they're mostly local and most people do most of their taxiing locally. There's some benefit to brand and there's some benefit to amortizing the back-end but taxis/ride-hailing is mostly a local business.


> The nature of technology makes it winner take all almost always

Citation? I am having trouble thinking of a major market where there's some company with 100% market share. Software has hudnreds of multi-billion dollar companies. Autos has hundreds? Until your scope gets very narrow, there's tons of competition.


> complete monopolies over all transportation

No way I'm going to pay Uber for the right to use my bike.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: