From page 29, average customer acquisition cost has risen by ~33% in just the last two years... they make the point on the next slide that having customer acquisition costs exceed a customer's lifetime value can't really succeed for long.
I think this, specifically, is what will spell the doom of Uber and Lyft, and potentially many of the food delivery companies -- they rely on insane growth curves to generate new investment, and customer acquisition is just so unbelievably competitive/expensive, it's an arms race that (IMO) has to collapse at some point.
Car sharing isn't going anywhere. It's too convenient. The prices will rise, the number of drivers and passengers will fall and the companies will shrink, but they'll still be around. I have trouble seeing suddenly seeing a day when there's no car share service because they all finally went belly up.
I don't think car sharing will go extinct either, but be careful with your analyses. Uber and Lyft are precariously balanced on a system of (substantial) low-income demand, high-income demand, and subsidy. If prices rise, low-income demand plummets, ending subsidies. High-income pricing could rise significantly higher than you might expect, at which point you no longer have "ride sharing" so much as you have private car services with app hailing instead of phone hailing, and, like NYC in the 1990s, so expensive that most people will only use it to get to the airport.
There's going to be a point where taking a scooter becomes more attractive for low income people (it's already half the price of a lyft), and then there will be a point where owning your own $400 scooter or $100 bike will make more sense rather than relying on the rental (which is normalizing the idea of riding an electric scooter/bike through a city). I wonder if these companies have dug their own grave.
Owning a bike isn't really that good for commuting. I use Citibike to go to work and pretty much anywhere else they have Citibike docks. People always ask why I don't just buy a bike. I had a bike. It got stolen. If a Citibike gets stolen, that's not my problem. Moreover, I don't have to try to take it on the train, or carry it into a building, or find a place to park it outside where it won't get stolen or vandalized. I've seen people in my neighborhood with giant bolt cutters going after people's bikes. Then there's always flat tires, bent rims, brake pads, etc. maintenance cost. Why bother with that. Cars are really the same way. What's the point of owning a car? So you can drive out to Walmart somewhere and buy ten bags of stuff every two weeks? Just order it and have it delivered to your door instead.
But those shared bikes are miserable to ride. The frames are heavy and not stiff enough, no suspension, no pedal clips, etc. They're fine to go a few blocks but for anything longer it's an unpleasant experience.
It's not hard to make your bike less attractive to steal. Big fat D lock. Shitty looking seat. Locking nuts for your wheels. Parking in a rack amongst more attractive bikes. Taking your bike into your apartment. Not giving a fuck about what your coworkers think about your bike parked in the loading dock.
If you are commuting, I wouldn't pay more than $100 for your bike anyway. Unless you are biking 100 miles a ride, you'd be hard pressed to notice a huge difference between a scratched up steel frame trek you found on craiglist and tuned up vs. some carbon fiber precision instrument imported from Italy. My maintenance costs are taking it to the bike shop once a year for a tune up - $60 and they give me new pads. I haven't had a flat since I bought gatorskins, and even though I suck at changing them, an inner tube is cheap and takes less than 30 mins to change. Those dollar or two rides on bike share really add up if you start relying on it heavily.
I get that it sucks hauling in a bike on a packed subway, but in the u.s. that's only really a concern for commuters in nyc. Even L.A. is a bike utopia once you get comfortable with traffic grazing your elbow and cutting you off all the time; it requires thick skin, and planning your routes a little differently, but you will always be faster than a car in traffic due to filtering.
If you live/work in a larger building you will probably have dedicated indoor bike parking, which solves the problem of keeping your bike safe at least at those two endpoints.
To your point, I want to know what share of Uber/Lyft customers are converts from public transportation versus taxis (maybe these categories aren’t very distinct). There are so many times when I’ve missed the bus ($2.50) and opted for a $5 shared ride that will deliver me in 10 minutes (same as bus) instead of waiting 15-30 minutes on the next bus. It leads me to wonder how many people almost never took taxis before, but now accept a 2x markup to reclaim their time when public transportation doesn’t fit their schedule constraints.
I never really used public transit or cabs, but I make pretty heavy use of ridesharing. It's insanely convenient when I want to have a few drinks out somewhere, and I also almost exclusively use it to get to / from concerts and sporting events.
It seems pretty reasonable to assume that equilibrium looks like taxis and private cars with app hailing. Personally, it wouldn't affect my usage much; I just use these services when I'm traveling because they're better experiences for the most part than taking a cab.
But, anecdotally, a fair number of urbanites regularly use Uber/Lyft instead of public transit or just because they don't want to drive. Presumably, significantly increasing prices would lose casual users who are using these services because they're a bit more convenient.
I think it's important to remember that a big part of the value prop of these services is that when you open the app to hail a car, you're pretty much assured of getting that car within 15 minutes or so. Change the current equilibrium much and that benefit might go away. I could hail a cab by phone before Uber, but, in Chicago, where we have pretty reasonable cab service (certainly light years better than SFBA's), it would often take multiple attempts and involve a 30+ minute wait. You could reliably get a private car in NYC, but, as I recall from all the times I worked in NYC, you weren't getting that car in 15 minutes.
Fair point. Around where I live, they seem pretty scarce the few times I've looked at the app out of curiosity. I don't know how long a typical pickup would take. But there's certainly some density level of drivers/passengers below which the market doesn't work. In fact, any percentage decrease presumably decreases service everywhere to some degree and shrinks the areas where Uber/Lyft can effectively operate at all.
I do use private cars but pretty much to get back and forth to the airport.
You’re overestimating how good the cab systems are outside of Chicago and the coasts. There is a lot of value in just having a reliable hailing system with a review system and tracking of where your pickup is, even if there is a 30 min delay.
The state of the art of cabs in smaller cities is getting hung up on, ghosted, or massively delayed when given a different ETA if you don’t book the cab hours in advance.
I think I'm doing the opposite, and saying that on-demand hailing of cabs without subsidy was and, when subsidies collapse, likely will be a lot worse than people think they'll be.
Yes, this is absolutely key for me. Effectively never needing to worry about being able to get a ride to and from wherever I need at any time removes a huge amount of stress and overhead from my life. The minute that stops being reliable, I have to start planning my travel and movements around transportation methods that ARE reliable (e.g. rental cars, or driving my own car). Once I've rented a car or driven my own car, the chances that I will use an Uber or Lyft for any part of that trip drop to near zero.
I for one have already switched from Uber to Jump (which Uber now owns anyway) for casual use where I’m not in a hurry. A bicycle is more pleasant to use, I don’t have to wait for pickup, it’s way cheaper, and I get a little exercise. The electric motor helps me not get sweaty.
During rush-ish hour it’s also faster than a car due to lane splitting and such.
The math goes like this: $15 for a car that takes 15min + 5min of waiting, or $3 for a bike that takes 10min + some pedaling? Is easy decision
This is downtown SF so we’re talking lots of traffic and short distances.
I think it's important to point out though the number of cities where biking is impractical (and can even be unsafe). In many parts of Europe, this is probably a good model. In San Francisco, this could work. In Seattle, where now live, I've seen an uptick in rented bikes (though plenty of regular bikers too) -- but it really depends on what type of commute you're making. In New York, it depends on area -- but if you're really concerned about time/money you'll take the subway. But in Atlanta or Los Angeles, this isn't going to be a solution for most people that use Uber now.
I think your point downthread about how this could impact the truly "on-demand" nature of the services is apt. As you said, the value prop is you get a guaranteed car in about 15 minutes (and often it is much less), whereas the old systems required calling, being routed to a nearby car service and having someone dispatched or a on-the-road driver routed to you.
In heavily urban areas, the average user might not see much of a difference (with few exceptions for weather, getting a cab in Manhattan pre and post Uber was basically the same and I've frequently canceled an Uber when I was able to hail a cab faster), but if you expand that just a smudge, say Brooklyn, which doesn't have yellow cabs unless they happen to be on the way back from a drop-off -- (and don't get me started on the uselessness of the green cabs) and as such was one of Uber's first really strong markets (I signed up for Uber the month it launched in NYC, back when it was livery only -- and used it almost exclusively in Brooklyn). The subsidized price (even though it has only decreased for drivers) has allowed cars to circle neighborhoods in Brooklyn -- but if that goes away and prices increase, drivers aren't necessarily going to sit or drive around each of those neighborhoods and will instead migrate to the most dense parts of town.
This is even worse for cities with historically little or no cab ecosystems (Atlanta -- and Atlanta is at an even worse advantage because of the terrible stage of its public transit, which only exists in the city proper), where Uber/Lyft effectively have become the cab system. I certainly don't think that market will leave those cities, but it costs rise (which seems likely), that will impact demand and will also in turn, impact how long it takes someone to get a ride.
(That said, I do not expect us to return to a pre-Uber in Atlanta world where it would cost $100 to get a cab from the airport to Dunwoody)
I do think the combination of increased demand (or at least awareness/ease of access to call a car) and logistical improvements (to me, this has always been Uber's greatest strength) will prevent things from turning into NYC in the 90s in most larger cities. But it is an open question in smaller cities that didn't have that infrastructure to begin with or are much more spread out.
Car sharing? There is no sharing involved, all transactions are done with money. They're taxi dispatch services that live in a grey area of employee/contractor.
The question is: will Uber and Lyft always be the entities leading car-ordering apps?
Hypothetically, Uber and Lyft stock prices could collapse, Google and Apple buy them, respectively, and then they could become features in Google and Apple Maps, as opposed to standalone companies and apps.
Sure, there are additional logistics, city management, back-end processing pieces that are not trivial. But the customer app experience certainly can be replicated as companies like Via have done, and Google and Apple could afford the transition costs of taking over all the behind-the-scenes logistics.
Not saying this will happen, but it's an illustration of how car-sharing survives but Uber and Lyft do not.
I mean, even in this scenario the company still exists, just with a different brand. They'll still have the same code, the same infrastructure, the same users.
If Facebook buys Oculus, that doesn't mean Oculus stops existing. It just means Oculus is now owned by Facebook.
Correct, the main difference (in my mind, open to other arguments) is how financial resources are allocated.
Right now, Uber and Lyft are burning a ton of money on customer acquisition with the goals of gaining enough scale to be sustainable, independent companies.
Relating this back to the great-grandparent post, there's an alternative: Google and Apple could use their existing brand recognition and software reach to maintain wide-scale car-ordering apps without having to spend as much (certainly still some) and with bigger balance sheets to take any short term financial hits. Then they could let the supply (drivers and the cost of rides) and demand (how many customers will pay for non-VC subsidized cars) play itself out over time, instead of the Uber/Lyft model of trying to juice both sides of the marketplace with excess spend.
In a strictly legal and technical sense yes, but IMHO a company going down the toilet that is sold for parts at a loss hasn’t really survived in any meaningful sense.
I can totally see a situation where a new entrant comes in at a point where the system stabilizes, and can be quickly profitable, if only slightly, but survive, while Uber and Lyft are contending with extreme pressure of what is essentially soft debt. Do any of Uber or Lyft's investors want those companies to by in a super competitive market where the profits are thin? Maybe that pressure causes them to make wild swings that are much more risky. How rationally will investors approach their investment?
I think, for me, the idea is that these companies aren't really gunning to make profits immediately and all are really in the play to get to automated taxis.
I'd assume automation of the vehicles plummets labor costs and then subsidizing the prices due to increased rider volume would be permissible.
That may be in the 10-15 year range, but I think they'll float along until something like that occurs. It's just a matter of time.
By 'Car sharing' I think you mean taxis. Ride sharing is a tiny fraction of their business.
To be fair, one could make the same argument about Pets.com. People will always need dog food, it will always be inconvenient to get it from the shop. But here we are 20 years on and I'd say online delivery of pet supplies across the world is virtually non-existent.
What makes you think it’s non existent? All the pet owners I know order their supplies online. It’s no different from any other form of online shopping, which is ubiquitous.
I could see Uber and Lyft being still around, with one or both of them having gone through a bankruptcy. Alternately, with one or both having their stock price crater.
Just as the on-demand ride service was disrupted by ride-sharing, I see ride-sharing ready to be disrupted by an on-damand ride service aggregator.
If Uber and Lyft are like arilines, what happens when a travelocity equivalent enters the game? Does all that market position that was bought through VC money matter if all of a sudden many (or most!) your fares are being viewed side by side with every competitor that feels like being in that market?
What matters more to a consumer of these services, the past good behavior of the companies they've used, or the fact that it's a dollar cheaper and picks you up a minute earlier, or in a nicer car? Even if they are reticent to use an unknown company, everyone knows Uber and Lyft, and the damage done to the VC funded low-fare strategy just by showing those two head to head would be enormous. And what happens when entry is that easy and localities start forming co-op groups to provide the benefits of Uber or Lyft but funnel all profits back to the drivers?
All I can say is that I'm grateful for the VC backers of Uber and Lyft for their generous donations.
IIRC this is how Uber started. They catered for some kind of "bespoke" customer that was more than happy to pay a lot. Another relevant question is what Uber will look like in developing countries where keeping the prices low is necessary even for the middle / upper middle class.
While margins in on-demand are usually pretty slim, companies can get pretty creative about opening up revenue streams: ad-serving, corp partnerships, enterprise pricing, licensing out tech).
None of these are a silver bullet, but they open up new growth that isn't tied to the standard hockey stick growth curve people use to get funding.
That's actually an interesting point. Uber Eats and the various scooter companies are just the very beginning of these companies trying to branch out into new areas of revenue. Whether they can branch out into profitability... I guess we'll see!
my understanding is that driver acquisition (and retention) is much more costly for uber and lyft (at least on a unit basis, and probably overall).
in late majority markets like ride-hailing, (1) awareness is not really a problem on either side of the marketplace as most folks have at least heard of uber and lyft, and (2) inertia is mainly due to uncertainty and risk aversion (for both sides of the market).
so then, the primary friction on the user side is downloading the app and then entering a credit card.
on the driver side, it's considering whether to buy/replace a car, entering personal info and answering intrusive questions, getting a background check, getting your car inspected, and being interviewed and trained. more costly and risky than signing up to be a user.
I think this, specifically, is what will spell the doom of Uber and Lyft, and potentially many of the food delivery companies -- they rely on insane growth curves to generate new investment, and customer acquisition is just so unbelievably competitive/expensive, it's an arms race that (IMO) has to collapse at some point.