[alistproducer2]

All the speculation around the turmoil over seems to be missing one key component. I think a bigger driving force then this sexual harassment and CEO antics is the waymo lawsuit. Uber itself has acknowledged that without self-driving capabilities the company will most likely not find profitability. From reporting I've read, it appears the waymo lawsuit has a high probability of success.

If the waymo lawsuit essentially forces Uber to start its self driving program back from scratch then the path to self-driving viability has essentially been scrapped. To me, this is a better explanation of the recent spat of high-level exits from the company.

[vannevar]

Uber doesn't just need self-driving cars, they need driverless cars. That's ten years away, and there's no way Uber can sustain its losses for that long. It's an investment Ponzi scheme, and it's losing steam. When it collapses, it's going to take a big chunk of the tech private equity market with it.

[alistproducer2]

This is my assessment as well. People at the c-suite level had to be aware that viability of the company is highly reliant on the magical thinking that they could get to market with driver-less tech before the VC money runs out. The (very real) prospect of complete restart of that research program has likely freed them of the fantasy.

I'm not totally discounting the PR issues as a partial cause of the exits. As a c-level employee, you (more than most hires) have to answer for rotten company culture. I also, however, think both issue (lawsuit and PR issues) make it untenable for the current batch of executives to stay on.

[alrs]

I was in an elevator once, and overheard this conversation:

"Yeah, computers don't do that yet. Don't worry, you just wait for awhile, and eventually they do. That's how technology works."

I saw red.

[__abc]

Why did you see red?

[alrs]

Because it shows a complete disregard for what engineers do, and reduces our work to something akin to fermentation.

[chopin]

That shows disregard to what yeast does, though...

[zelias]

Probably because that kind of statement doesn't pass all the tests

[amygdyl]

I have to say, I am so stealing that line.

I know exactly who will repeat it, seriously, deadpan as if Q.E.D. shown, if I just encourage them a little...

There is is something about that which smacks of borrowing the internet for a day, for a show and tell....

[ExactoKnight]

Was this actually said by someone on Uber's C-suite?

[mabbo]

Ponzi schemes pay out large returns in the short-term, making you think it's a good idea to put more money into them. Uber has done one better, and paid no one anything in terms of returns, yet still collected billions in investment.

[nerfhammer]

"Ponzi scheme" apparently is just invective that can refer to any sort of market failure now, not necessarily the actual Ponzi scheme type investment fraud such as that which Charles Ponzi carried out.

In this particular example, remember that Charles Ponzi's books were totally fake; outright fraud. That does not characterize other phenomena such as other things like speculative bubbles or simple investment gambles that don't pay off. In contrast the Bernie Madoff scam is a recent example of an actual Ponzi scheme.

[vannevar]

Well, earlier investors are seeing paper returns. But other than consumers who are getting subsidized rides, yeah, no real payout.

[dopamean]

There were a few funds of funds that saw big paper returns with Madoff as well.

[LaikaF]

The subsidized ride thing is pretty crazy too. Uber offered me the ability to pay them a fee, then get a number of flat rate rides. I got a ten pack, used it to go across my city (a 25 dollar ride normally, even with pool) for 2 bucks. I think they lost a grand total of 200 bucks on me alone, and I haven't touched the app since then. Switched back to Lyft when they offered a saner version of the same thing.

[jedberg]

> they need driverless cars. That's ten years away

Maybe in the US. But Uber is everywhere. Given their penchant for "bending" the law, I wouldn't be surprised if they start deploying driverless cars in places with more lax regulation, which will 1) help their bottom line, 2) give them more training data, and 3) prove it works to make regulatory issues easier in the US.

That's of course if they can keep their tech, which seems less likely given the waymo suit.

[thrillgore]

I guarantee that if the claims have validity Waymo and Uber will settle and a licensing deal will be agreed upon.

[gtirloni]

Is Waymo required to make a licensing deal?

What would it be worth in $ at this point? Does Uber have deep pockets for _that_ much money at this point?

[tedivm]

I don't believe they are required to make a deal. However, their parent company is a big investor in Uber, so they have some incentive not to drive the Uber into the ground.

[gingerbread-man]

This could be the next IBM PC license if Waymo owns the tech that all driverless cars need to operate. They don't need to make much money off Uber, they need to get their tech out there into production so they can refine it before their competitors can get to market.

[gingerbread-man]

It seems more likely that GV would get additional equity in exchange for the license.

[api]

As someone trying to raise right now, I am very concerned that this is going to implode the entire industry. An Uber collapse will set off a panic as people wonder who else is swimming naked. It will also tank the return on any VC fund of which it is a part, leading to a whole round of terrible VC returns.

[tomblomfield]

I think the swing in VC herd-mentality happened about 18-36 months ago.

There was a palpable shift from "prioritise growth above all else" to "let's think about about unit economics".

[vbtemp]

FYI this comment has been quoted by The Atlantic... https://www.theatlantic.com/technology/archive/2017/03/uber/...

[sjg007]

That's amazing.. I also think it is quite funny when reddit posts become news.

[ChuckMcM]

A clear business model that works is the best defense here.

[api]

Yes and no. The problem is that panics affect the entire market and have cascading follow-on effects. Not only does this affect fund raising but it also affects the entire tech economy, and thus the ability to convert new customers if your customers have anything to do with the tech economy.

We are fine if we can't raise more, so I am not panicked... just concerned.

[thearn4]

> they need driverless cars. That's ten years away

Honestly even that seems way overly optimistic in my opinion, if we're talking about R&D + regulatory acceptance & general adoption.

[msabalau]

Is your theory that absent subsidies no one would prefer an Uber over a taxi? Because if not, presumably Uber could just raise prices and hunker down for 10 years. (Assuming they survive lawsuits and assorted revelations) Is their another clock that prevents them not just from not just growing, but existing?

[usefulcat]

Not a clock, but potential competitors. The barrier to entry is simply too low. For example, look how quickly Fasten/Fare/etc appeared in Austin after Uber left.

As a product, ride sharing is not quite a commodity, but it's not far off either. As a consumer, I care more about what the product does (a better taxi service) and how much it costs than exactly what company is providing it. Absent artificial constraints (e.g. taxi licenses) it's never going to be a high margin business.

So yes, they could hunker down and "exist", but that doesn't seem like a viable strategy to the get the kind of projected growth their current valuation depends on.

It's hard to see what inherent long-term advantages they have. Their core business is already a commodity. Even if they somehow get usable driverless tech better/faster/cheaper than anyone else, at some point that will become a commodity too. When that happens, there won't be much left on which to compete beyond price.

[yourexwasteman]

You bring up some good points. But I think it's important to not forget that Uber has been around 2009, and wasn't spending their entire time trying to achieve driverless vehicles.

Don't doubt for a second that there is a ton of engineering that went on in Uber's app that allows it to handle scale. For example, I was in Austin a few weeks ago and used RideAustin. Same experience as Uber. Then I read on the news that the weekend of SXSW, the entire app was basically crushed from the volume of ride requests and couldn't service anybody.

And besides, who wouldn't use Uber if it cost the same amount as a taxi? For me, the advantages of Uber has always been the quality of service. The drivers are nicer, the cars or cleaner, the drivers don't demand tips nor insist at the last minute that their credit card machine doesn't work at the last minute.

And most importantly of all, I would argue that Uber's inherent long-term advantage is their brand. Anywhere in the world that has Uber, I get the same service. I can go to Honolulu and Uber works the same. I just open the app, punch my destination in, look for my driver and go. Same if I was in Mexico City. Same if I was in Delhi.

However, the same is not true for taxis. In certain places (notably South East Asia as a tourist), you agree on a certain fixed fare before getting into the taxi. In the United States they expect tipping, whereas other countries may not. They might insist their credit card machines aren't working in Toronto. There's no accountability of the driver, no GPS trip details in the app...the list goes on.

I think that kind of brand and user experience is highly valuable.

[untog]

> Anywhere in the world tha

In my experience that's not true at all. In NYC it has to be a qualified black car driver. In LA it can be anyone.

[vannevar]

I personally think they could get away with raising prices, but then the drivers would become their customers, and profit margins for Uber as a dispatcher would be driven very thin---it's just not that hard to run a GPS-driven dispatch service. The only thing that makes an Uber an Uber is a window sticker and an app on the driver's phone. How long before there's an uber-Uber that just sends your request to all the dispatch services and the first one that shows up gets your fare?

[ProblemFactory]

> How long before there's an uber-Uber that just sends your request to all the dispatch services and the first one that shows up gets your fare?

There already is an app for this, with over a billion existing users: Google Maps.

If you search for directions in London, there's a "ride hire" tab next to public transport. It gives you price estimates for Uber, London official taxis, and Hailo - all booked with a button from Google Maps, never having to open the Uber/Taxi/Hailo app.

[__jal]

Whether or not humbling themselves into a mere taxi firm with unsubsidized rates would be competitive, I doubt the investors would allow that.

There's an awful lot of other people's money in Uber. There's an Ayn Rand joke here, but I'll skip it.

[sf_rob]

Ya, the cynic in me thinks the subsidy is to bleed Lyft not retain users until driverless cars are a thing.

[mr_luc]

Yeah - I dislike Uber's tactics as much as the next guy, and the Waymo lawsuit is nuclear.

But they're spending money like crazy because they still feel like it's getting them "upwind" in ways that will be lasting. New markets, and market dominance in existing markets.

If they scaled back their immense outflows, and used profitable markets to subsidize less profitable ones, I'd bet they could be profitable.

[ryanmarsh]

Do you have any data to back this up?

[bedhead]

While I wouldn't literally describe Uber as a ponzi scheme, I think you are right in the figurative sense. It usually helps to make that distinction. But yeah, it's hard to not see Uber running into a massive liquidity crisis at some point in the next deaced. The company is only viable to the extent the private capital markets allow it to be. I don't think it has any hope at all of accessing public markets.

[yeukhon]

What's the difference between driverless and self-driving car? The ultimate goal of self-driving car is driverless. Current state of the art does require a driver to be with the car.

[dogecoinbase]

What's the difference between driverless and self-driving car?

Whether you have to pay someone.

[cosmie]

And years of regulatory red tape. Moving from current -> self-driving legal and regulatory framework will be one battle. Moving from that to a driverless model will be much larger.

And those updates have to percolate through federal, state, county, and city level regulations. Making it an inherently slow process.

[yeukhon]

I agree, and I think that's also good for driver and traffic safety. As I mentioned a while back, in case of traffic light going dark, officer has to stepped in to regulate the traffic. Current self-driving car is not capable of handling that without any faults. Among cars, they need to negotiate who to move first on a all-away-stop sign. Self-driving cars need to be able to get data from government relating traffic condition. Currently we all rely on GPS data to determine whether there's congestion or not, as well as user's input (Waze) but that's far from enough to be 100% driverless.

Dataset availability is really really poor today. We can move with great confidence on known conditions, but so many edge cases remain to be solved.

[yeukhon]

But the end goal is that as a driver of my own car, I can just sit in the passenger seat. As a passenger of a taxi, similarly, I expect no driver.

The ultimate goal of self-driving car is to be driverless. But, if you want to produce a car without driver seat & steering wheel, then yes, from that perspective, driverseat-less car is different from self-driving car.

Otherwise, same.

[zild3d]

It's not about the seat, it's about whether or not a driver needs to be in that seat. Lower levels of autonomy need a driver to be able to step in. That is self-driving, but not driverless.

[yeukhon]

As I said three times already, self-driving car ultimately doesn't need a driver to be on the steering wheel. Why is it so hard to understand that? That's the whole idea of self-driving car.

[at-fates-hands]

They said the same thing about Groupon and somehow they're managing to keep their heads above water.

I haven't seen them in the news at all in the last year or so.

[vannevar]

Groupon went public, and their stock went from $26/share to $4. If the same thing happens to Uber in a down round, that's pretty much the definition of collapse.

[danielweber]

Without saying whether I think such a thing is likely, Uber compensates significantly with stock, and they cannot retain talent if the trust in the cashout goes away.

[user5994461]

They give ISO which are worthless and might fire back for a lot of employees.

They won't be able to retain talent when the talent will figure out they can't get the stock that was wrongly promised.

[teekert]

Maybe I'm being naive but how can a company that replaces a whole layer of management and overhead of traditional Taxi companies by an App not reach profitability? Is maintaining an app that expensive? I can't imagine.

[emodendroket]

The overhead is not as large as you're imagining. http://www.nakedcapitalism.com/2016/11/can-uber-ever-deliver...

> By contrast, in the hundred years since the first motorized taxi, there has been no evidence of significant scale economies in the urban car service industry. That explains why successful operators never expanded to other cities and why there was no natural tendency towards concentration in individual markets. Drivers, vehicles and fuel account for 85% of urban car service costs. None of these costs decline significantly as companies grow. As the P&L data above demonstrates, Uber has not discovered a magical new way to drive down unit costs.

[dagw]

That explains why successful operators never expanded to other cities

Sweden's largest taxi company currently operates in 50+ cities in Sweden, Norway and Denmark. So "never" is probably the wrong word.

[cynicalkane]

Sweden famously deregulated their taxi industry. American taxi regulations are awful--I can't compare them to other countries, but they're terrible in an absolute sense, and they vary widely by city. In New York, they even change depending on where you are and what color the taxi is (yellow ones and green ones have different rules). So all these city-by-city variations become a big diseconomy of scale.

[1024core]

The reason for the US' horrible state of affairs: corruption. Ever since regulation of the taxi industry started, the 'players' in each city decided to use this artificial constraint to enrich themselves. Medallions (the right to drive a cab in a city) were doled out to the politically connected, who, in turn, rented them out to the actual workers (drivers) for a tidy sum. It was a great scam.

If there's one positive thing about Uber, it's that they brought an end to this medallion monopoly.

[emodendroket]

Come on; that's too simple. Unregulated gypsy cabs present a real risk to people.

[emodendroket]

Having to comply with tons of different rules on a per-state or even per-municipality basis is characteristic of the entire US system, though.

[dagw]

American taxi regulations are awful

But that is a quirk of the current American taxi system, not something inherent to the industry.

[ExactoKnight]

It's actually pretty widespread. If you think they're bad here, taxis are even more entrenched in Europe.

[jknoepfler]

One imagines the cost of complying with labor regulation in those countries makes for better economies of scale. I agree that "seldom in the U.S." is more precise, though.

[jdavis703]

The population of these countries are comparable to a Metropolitan area in the U.S.

[emodendroket]

It's basically correct in the US case.

[metaphorm]

robot driven electric powered vehicles that operate 24/7 (minus maintenance and recharge time) are a major way to gain efficiencies in drivers/vehicles/fuel. the nature of an automated ride hailing service, also, gains efficiencies in densely populated areas. it doesn't quite reach economy-of-scale level stuff the way a manufacturing business does, but there are definite efficiencies that Uber (and their competitors) are aggressively pursuing.

[emodendroket]

The parent comment on which this thread is based is arguing that Uber's elimination of top-heavy management in traditional taxi companies is so great that they should be able to profit even if they never achieve self-driving cars. Clearly if they can entirely eliminate the cost of labor it complicates the question.

[013a]

That seems like a strange conclusion considering Uber doesn't even pay for the vehicles and fuel.

UberPool is one pointed example of scale economics working. Another would be Uber's brand encouraging more riders and the ability to scale into smaller markets that taxis normally couldn't service.

[emodendroket]

> That seems like a strange conclusion considering Uber doesn't even pay for the vehicles and fuel.

Well, they do, indirectly, in that that is part of the money the driver is paid. Yes, one way they've improved the numbers is squeezing their drivers more, but they can't do that forever. They're already encountering a lot of pushback.

I question the claim that Uber is serving a bunch of small markets that traditional livery services do not. Do you have an example?

[cat199]

Not OP, but I can say from personal observation/experience that while not 'small', Uber has increased taxi use in the general sense in Texas cities that are so car-centric and-or sprawl-like that traditional taxi services were never really popular (Houston, Austin, Dallas, etc) - having more drivers available in an ad-hoc manner has made the dispatch time much more reasonable (minutes vs a hit-or-miss scheduling ordeal of ~1h) and so using taxis (e.g. Uber) as a viable means of transportation, especially for things like social events, is much more common than it was before. I suspect the same is likely in similar places where 'everyone has a car' and the geography is very spread out.

[emodendroket]

That's the kind of place I live but there are multiple livery services operating here too.

[mrgordon]

For example, Boca Raton in Florida is a wealthy area that had no taxi service until Uber. You literally had to get a limousine to go somewhere. Now its simple and cheap to get an Uber.

[emodendroket]

When you say a "limousine" are you not just describing a livery service where you call a dispatcher and they send a car?

[darpa_escapee]

> the ability to scale into smaller markets that taxis normally couldn't service.

Go on.

[yesiamyourdad]

Mountain towns in Colorado, for example. Uber and Lyft operate in the major mountain towns where there is no taxi service at all.

[sulam]

Simple economics argues you are wrong about the costs aspect of things here. Do you think drivers are simply doing public service? Perhaps they have special gas pumps that are 30% cheaper?

[crispyambulance]

Traditional taxi companies are now increasingly employing 3rd-party apps which function much like Uber's. In many US cities, the "Curb" app will hail you a regular taxi cab and allow you to pay for the whole thing through the app (including tip, even if you just happen to hail the cab without opening the app).

Uber's long-game, and the reason they've been funded so handsomely, is the autonomous taxi plan.

Its a pity that taxi companies were so slow to adapt to Uber by providing convenient ways to summon and pay for a cab. All that people really want is to get from point A to point B without a hassle.

[pjmlp]

> Its a pity that taxi companies were so slow to adapt to Uber by providing convenient ways to summon and pay for a cab.

Have you ever picked a taxi in Nordic countries?

Those guys already had Uber like apps and all possible ways of paying for a ride, before Uber was a thing.

[crispyambulance]

Never, but not surprised. How is Uber doing in Nordic countries?

[Nekorosu]

Talking about Sweden Uber is just another taxi company here. It's not possible to provide taxi service without having a license. Uber has to comply. Still it's more convenient to use their app because it has better UX than everything else on Swedish market.

[gtirloni]

Which isn't hard to replicate or improve upon. Is Uber fighting the local government in Sweden to have less regulation too? It seems that's what they are doing here (Brazil).

[martinp]

Uber remains illegal in Norway as all taxi drivers need a permit granted by the government. The subject is under heavy debate though.

[pjmlp]

Not sure, my last trip there has been around 2012.

[debaserab2]

I have yet to see one of those app's come even close to replicating the experience that uber brings. It's one thing to have an app, it's another thing to have a well designed app with consistent policies that are followed the same by every driver.

Any time I've had to use an uber clone apps (from a traditional taxi service, not lyft) they haven't instilled a lot of confidence in me to how reliable they are.

[erroneousfunk]

> consistent policies that are followed the same by every driver.

This was the most amazing thing about Uber, to me. I live in Boston and used to take cabs all the time pre-Uber (still take the occasional cab from the airport). Cab drivers are rude, frequently claim the credit card machine is down (despite violating city laws saying they're literally not allowed to drive with a broken CC reader), take the long way / detours to run up the meter, air conditioning or heating might be broken, the cars are dirty, no real recourse for bad service. Even if they have a GPS, drivers refuse to "waste time" by typing in your destination address and require turn-by-turn directions. I'd call a cab and it was hit or miss whether they'd show up or not. I even had a cab driver threaten to call the police when I didn't have cash after he tried to claim that Boston was "cash only" on Saturday nights -- and he was driving me to my HOME in the South End, so it should have been pretty clear I wasn't a tourist unfamiliar with local laws.

Even with the most well-designed and reliable app in the world, cab companies can't compete with Uber -- it's the service and consistency that really made it take off.

[debaserab2]

Yep, agreed. I've literally been hijacked and taken to an ATM before by a cab driver who clearly had a VISA sticker on the window of his car. While uber seems pretty evil these days, I just can't risk the unreliability of traditional cab services.

[crispyambulance]

Although I admit the UX of Curb is not as seamless/slick as Uber's, it is certainly usable enough and given time the kinks will be worked out.

IMHO it was a small price to pay for deleting my Uber account.

[erroneousfunk]

Do you think cab drivers have gotten better since Uber came out? I only take them from the airport occasionally now, but the biggest thing about pre-Uber taxis was the service issues.

[gtirloni]

As I've learned while participating in these discussions is that all of this is HIGHLY localized. The issues you have with taxis and/or Uber will not make sense in other places with different regulations, better apps, etc.

Almost every regular/licensed taxi I take here, the driver has Uber/Lyft/99Taxi app running simultaneously so they can reach more clients. In that case, it's the same quality of service as before (which is just fine where I live, so Uber/Lyft aren't disrupting anything here).

[cstejerean]

How much more time exactly? Curb used to be called Magic Taxi if J remember correctly and it's been around for more than 4 years.

[darkmouth]

True. Escaping the law, sexual harassment of employees and low wage to drivers aside, it will still make me feel very inconvenienced to see Uber go out of business. Not even Lyft comes close to how good and fast the Uber app is. I was sold the first time I used it. I just hope some other company steps in to fill in the void.

[ExactoKnight]

eCab in Vancouver is getting pretty close. The app's technology is all but reliable, but where the problems still exist are when demand is high and taxis (by the very nature of being limited) are unable to meet it.

[galdosdi]

I honestly never had a problem with the UI of

1) Dial a 10 digit number

2) Someone says "Hello, XYZ car service"

3) You say "I need a car at the corner of Sample street and 123rd road"

4) They say "5 minutes" and hang up

It takes the same or less time as using an app like Uber with fewer things that can go wrong.

[mcgrath_sh]

That was effective in cities with good (or any) taxi service. In Pittsburgh, Uber and Lyft have utterly changed people's ability to get around the city.

One Saturday night we wanted to go out to a place 10 minutes away. We called a taxi cab around 7:30/8:00. We were told that it would be between 2.5 and 3 hours for a cab, if one came. We were considered "out of the way" when we were a 10 minute ride from downtown/the main bar area in the city. Forget about getting a cab back from those bars. We used jitney cabs all the time to get home. Yellow cabs wouldn't bother or were too infrequent.

[crispyambulance]

I remember jitney's in Pgh !! In the 70's I used to ride in jitney cars with my mother to get back from the grocery store. In some cities, "jitneys" are called "hacks"-- unlicensed cab drivers driving their own cars. Perfectly safe if they're neighborhood folks that you know.

There used to be a similar/related practice called "slugging" as well, it seems to have faded as Uber rose up.

[galdosdi]

Great point -- ridesharing apps have genuinely contributed something important in areas that didn't use to have much taxi service. My point is limited specifically to big cities that already had decent service -- in those places, there wasn't a significant improvement brought by Uber and friends.

[planteen]

5 minutes would be fine. But sometimes it was more like 45 minutes to an hour. You wonder if a cab is ever going to show up. That's why I started using the apps - you know the driver can get in touch with you and you can see where the car is.

[crispyambulance]

Yeah, but you have to guess at which cab company to call and then dig up the number.

With the Curb app, your request goes out to multiple cab companies and you usually get a cab that is very close.

[watter]

For me this can be problematic in my second language, and near-impossible in my third, fourth, and fifth languages. I travel a lot and it is the only time I ever use Uber.

[fs111]

> Maybe I'm being naive but how can a company that replaces a whole layer of management and overhead of traditional Taxi companies by an App not reach profitability?

They employ more than 2000 Engineers, many of them in the Bay Area. Those cost easily more than any manager in a Taxi corporation..

[Arizhel]

Yep. Amazing how you need 2 thousand engineers just to make an app (which is already made) and run some servers (which are probably hosted on AWS anyway).

[Balgair]

Well, the greyball program took some real coding and testing out to implement as it was reported. From the point of view of 'if you see one cockroach scurry on the floor, your walls infested with them' then we can make an educated assumption that there are more 'secret' projects they may be working on to further obstruct justice (highly dependent on location). The question is then: How many coders work on the core app, how many worked on the greyball section, how many worked on the self-driving cars (auto-automobile is maybe a better term for these vehicles, but not by much), how many does that leave left? Take that answer, assume that the greyball is average the number of coders on a 'secret' project (very very dubious to do this, but what other data do you have?), and then you have an estimate on the number of other 'secret' projects they have going on and what their abilities are as compared to your cadre of coders. Then try and steal them if they are doing better than your folks, now is a good time to invest in their coders before the crash and rush for them happens.

[sosborn]

> just to make an app (which is already made) and run some servers

I assume they want to do more than let the product stagnate, not to mention the resources they are putting into driverless vehicles.

[Arizhel]

You don't need thousands of engineers to keep a mobile app up-to-date, and the driverless vehicle thing is a boondoggle as there's already other companies doing that work, which are much better at it (Google, GM, etc.), and don't need tons of VC for such a risky endeavor.

[sosborn]

> there's already other companies doing that work

In your opinion, is work not worth doing if those mammoth companies are pursuing it?

Also, you say it is a risky endeavor. Would you argue that the driverless vehicle is not an inevitability? I would certainly argue that betting on Uber to come out on top is risky, but that doesn't mean there isn't money to be made.

[iamatworknow]

I wonder if they were just overzealous in hiring and don't actually need all those employees, but now can't let them go because of the impression that would give to the industry/equity holders. Large staff cuts usually do not instill confidence in investors, even if that staff literally isn't required.

[GhostVII]

Their backend consists of hundreds of microservices that are frequently breaking, so it's more complex than just an app and a basic backend on AWS. Also they have been doing a lot of massive refractors in the last couple years, so they need lots of people for that.

[passiveincomelg]

How much of that is self-imposed complexity from chasing the hype of the week?

Also:

1. Software is all about automation and increasing efficiency!

2. Let's hire thousands of programmers!!1!12!

Empire building much?

[GhostVII]

I would imagine most of the complexity is from handling a massive number of requests from around the world, and needing all of their data to be backed up and secure. They have their critical data backed up to like 10 different servers, when your dealing with scale like they are things get complex really quick.

[dasimon]

The problem is not competition from traditional taxi companies (they do indeed replace taxi companies' management structure) but other ride sharing companies, ie Lyft. Uber has only managed to stay ahead of Lyft by undercutting Lyft's prices, incurring severe losses in the hope that as soon as they have self-driving cars they will be able to become profitable by eliminating one of their main expenses, the drivers themselves, while maintaining lower prices than Lyft.

Without self-driving cars, that strategy isn't sustainable, and so eventually Uber will find itself unable to maintain its advantage over Lyft.

[CPLX]

> Uber has only managed to stay ahead of Lyft by undercutting Lyft's prices, incurring severe losses in the hope that as soon as they have self-driving cars they will be able to become profitable by eliminating one of their main expenses

I think someone should pause to note that, if true, this is one of the dumbest long term business strategies in the history of high finance.

This idea, apparently, is a bet on a technology that not only doesn't exist, but is extremely highly regulated, that the company has literally no demonstrated core competencies in, hasn't been even successfully prototyped, that represents the hardest most complex use case of the technology, and is obviously years away at best, but yet it justifies a policy of losing billions of dollars in the present just to get market share when the costs of switching brands are literally so non-existant that a typical customer often does it several times in a single evening out.

Maybe it's my old age and having lived through the first dotcom crash, but it seems to me that even when you feel like the only person who sees that the underlying business logic is nonsensical magical thinking, it's still quite possible you're correct.

[tyingq]

You aren't alone in that thought. Being privately held the financials are opaque enough that no outsider knows for sure how much of the spend is subsidizing the low prices, versus being used for expansion and R&D.

The leaked data from Naked Capitalism is really the only data outsiders have at their disposal: http://www.nakedcapitalism.com/2016/11/can-uber-ever-deliver... The use of EBITAR (vs EBITDA) makes it difficult to draw real conclusions, though the fact that they use EBITAR at all is suggestive on its own.

[TuringNYC]

However, something as epic as self-driving cars are one of those things that come once every two generations. SDCs, if achieved, have the potential to change the entire American way of life and massively disrupt society. Start-ups love to disrupt and this would be the mother of all disruptions.

So, for that reason alone, I think firms are willing to bet a sliver of their portfolio. If I was a family office I certainly would do so.

[CPLX]

Right. But it's the second half of the business strategy that I outlined above that's key to the discussion, ie the idea that paying billions of dollars for market share in this field makes sense. That's literally crazy, the network effect and lock-in at scale is really modest at best.

Sure the more cars you have the better the service can be, but it's trivially easy for a competitor to come along at any time and attack your most profitable market segment in a given city, and trivially easy for any customer to switch as easily as clicking on a different app and glancing at the estimated time and price. That's going to be true forever, this isn't a market that will ever have a defensible monopoly position.

Investing some amount to hedge in self-driving cars could conceivably be defensible, but looking at what's going on it feels like that's more of an rationalization for their present behavior.

They've been lighting money on fire subsidizing rides for a few years, and have hunted around for a plausible excuse. One is the "pool" functions, as that has a slightly more plausible network effect story, and the other is self-driving cars.

Both appear to be post-hoc rationalizations designed to provide some plausible story for why they need to borrow another couple billion dollars.

[TheOtherHobbes]

I wouldn't say it's trivial for a competitor to take over.

The Uber app is still best-in-class, and all my friends now say "Get an uber" instead of "Get a cab" because the UX is so much smoother.

Generally it's extremely difficult to dislodge an incumbent from a market slot. Newcomers can only compete on price, UX, and brand recognition. Ideally all three need to be significantly better than the incumbent to have a hope of taking over. Without all three the best a newcomer can hope for is a small slice of the pie.

So market share definitely has value in the abstract. Unfortunately in Uber's case it has negative economic value because of the costs/subsidies.

Then again I suppose it's possible Uber has always been a cunning plot to take VC money and spend it on subsidised transport. If so, it's definitely been a success - for now, at least.

[ExactoKnight]

What you are saying is right about Uber's current business. But that is not true for self-driving cars.

The infrastructure costs of deploying a fleet of cars to compete with Uber if they gain market share will be massive, all but assuring they will hold a monopoly position for years unless a decentralized competitor could actually become reliable quickly (I doubt it).

[fmihaila]

Once you have a self-driving fleet sure, you can start printing money, but it makes no sense to sacrifice revenue today by starting an unsustainable price war long before the tech is ready. All this does is shorten your runway, and you don't even know how long a runway you will need, given that reliable, fully autonomous driving is so hard.

[CPLX]

> Once you have a self-driving fleet sure, you can start printing money

Why?

Are drivers who make $10-15 an hour so ludicrously expensive that saving that money fundamentally changes the business?

Are the carrying costs and maintenance costs and depreciation costs of self-driving cars likely to be lower or higher than the cost of a 2017 Toyota Camry? How about the regulatory and insurance costs?

Is there likely to be some magic secret that allows one company to dominate self-driving cars, or will it resemble the historical markets for transportation devices, whereby there are dozens of companies who make different offerings of similar technology, and a network of component suppliers and hardware and software companies that contribute?

I guess it captures a unique place in our imaginations, but this topic has an unusually severe infestation of magical thinking for some reason.

[danielweber]

Assuming self-driving cars are just about to work, why is Uber significantly better positioned to profit off of them than other people?

[danielweber]

> if true, this is one of the dumbest long term business strategies in the history of high finance.

You speak of the first dot-com crash, but there was a lot of stupidness back then. One of my favorites is furniture.com http://www.ecommerce-digest.com/early-dot-com-failure-case-s...

[Systembolaget]

It's déjà vu all over again. Remember "eyeballs" from 1999.

Research into "autonomous driving" began 1987 with the Prometheus programme consortium, then came C2C and C2X.

We have "autonomous driving" the day an automotive CEO is happily blindfolded on the back seat, alone, chauffeured a random journey through Seoul on a morning commute in monsoon season or a scooter-mania evening on Friday in Milan.

[Systembolaget]

So far, Uber et al. have operated on the principle of pitching easily disposable worker bees against each other, at the mercy of an opaque rating system, with zero rights to appeal, burdening them with all costs of doing business (car, insurance, maintenance, etc.) while providing nothing more than an app with server-farm back end - the classic founder/VC/underwriter/IPO-seller benefit narrative.

[sulam]

Analysts who have looked at it from the outside suggest Uber may at best be worth 1/3 its last financing round valuation. You are not alone.

[ProblemFactory]

> Without self-driving cars, that strategy isn't sustainable, and so eventually Uber will find itself unable to maintain its advantage over Lyft.

I see the "Uber is a bet on self-driving cars" coming up in every discussion about their business plan. But even if that's true, why is there an assumption that Uber would have a monopoly on self-driving cars?

Google, every car manufacturer, and a whole series of startups and universities are working on self-driving technology. If for example Google perfects it first, they will sell licenses to car manufacturers, who in turn will sell cars to Uber, Lyft, and every taxi company in every city of the world. Even if Uber develops the technology first and keeps it to themselves - then others cannot be too far behind, the potential payoff is just too great.

In the end the situation is just the same as now.

[nomnombunty]

I totally agree with you. I don't think Uber would have a monopoly on self driving cars. In terms of the tech, they are definitely not ahead of e.g. Google. According to this TechCrunch article [1] their self driving cars require human intervention every minute.

[1]: https://techcrunch.com/2017/03/16/uber-recode-leaked-autonom...

[samfisher83]

They are subsidizing the rides. They are charging you less than what they pay the driver. That is why they are losing money.

[danielweber]

That's not what private leaks show. For example, in the first half of 2015, they got $3.6 Billion from customers and paid $3 Billion to drivers.

[anilshanbhag]

I disagree with the other comments. I think it's because the rides are currently cheaper than they need to be. From the airport to my place, it costs 50$ via normal taxi. Via uber pool I pay $10-12. Most of the times it's only me in the car. There you have it.

[pyvpx]

because they subsidize every ride. significantly. traditional taxis are actually profitable with what they charge. uber is not.

[hueving]

>because they subsidize every ride. significantly

I don't believe this is true in all markets. IIRC the heavy subsidizing behavior causing severe losses is in new markets like China.

[mandevil]

Nope, worse than that. Uber did have its butt kicked in China and surrendered last summer, having it's China operations purchased by Didi (the company that applied the foot to Uber's posterior). But the problem was everywhere: Uber lost 800 million in 3Q 2016 ex-China, 2.2 billion for the year. "Even in the U.S., Uber's home market, the company continues to lose money." They did manage to turn a profit in the US in 1Q2016 but lost 100 million in 2Q and the losses "increased" in 3Q, looking only at the US. Source: https://www.bloomberg.com/news/articles/2016-12-20/uber-s-lo...

[danielweber]

That doesn't show they are losing money by subsidizing rides.

From your source:

> Net revenue—the amount of money Uber generates after it pays its drivers—was $1.7 billion in the third quarter,

That's what it takes to be a real business. If you are a grocery store that buys bread for $1 and sells it for $2, you potentially have a real business even if you only move 20K loaves a month and need 30K to cover all expenses.

People keep on alleging that Uber is somehow buying bread for $1 and selling it for less than $1. There was stupidness like this in the first dot-com bubble. You aren't a business if you are doing this.

This distinction really matters. Since Uber is making money on each ride, there is a real business here.

[macNchz]

I don't claim to know what's going on behind the scenes, but even years after launching in NYC, both Uber and Lyft have continued to send me notifications with steep discounts, free rides, etc. I've used both apps since they launched here and a major percentage of my rides have been discounted below 'sticker price'.

[logicallee]

Not even a dating app would be profitable if you just spent VC money on paying for (subsidizing) users' dates ;) It would be very popular though.

Hey - brb!

[ExactoKnight]

Uber has massive costs as it scales, but unlike Amazon that has fixed costs, Uber's problem is that it has massive VARIABLE COSTS in three areas: driver recruitment, local advertising and local regulation.

The work Uber needs to do to onboard a driver in Alberta, Canada is completely different from Ontario, Canada (e.g. drivers needing a commercial class 4 vs consumer class 5 license).

As a result, they are facing unique regulations that is making it difficult to automate the work that is being done in each local market.

[cauterized]

By keeping their prices below costs in order to undercut the competition.

[JustSomeNobody]

Poor management. If a company is not successful it is because they have a product nobody wants or they have bad management.

[bmelton]

Uber may indeed have poor management, but I believe you paint with far too broad a brush in this scenario.

MOST companies fail. 75% of venture backed companies fail quickly. Only about 50% of all businesses are capable of lasting 5 years, and only about 35% are able to last 10 years.

We have all seen companies that were offering products you wanted, but that closed for one reason or another, and it is not in every case that you can attribute it to bad management.

[Arizhel]

Companies can fold while offering products that are in-demand. The problem is that you have to have enough demand, and sell at a price that allows you to be profitable at that demand level. So it's quite possible for a company to have a product that's in-demand, such as glittery pink mechanical computer keyboards (there's got to be a few people out there who want them after all), but not be able to sell them at a price high enough to keep the company afloat. So glittery pink mechanical keyboards surely have some demand, but it's not going to be high, and the manufacturing cost of such things is going to be high, so between the low volume and high cost, that means you need to charge boutique prices. If you can't get enough people to buy them, then the company goes under.

But this is still attributable to "bad management". Good management would not even try to sell a low-volume product where they cannot price it high enough to make up for the limited demand and volume, and would instead come up with a better business plan in the first place.

[iamatworknow]

You don't even need to make up your keyboard example -- just look at Pebble.

[Arizhel]

Very good point. By most accounts a great product, but not great enough to get either enough volume, or fetch a high enough price.

[JustSomeNobody]

What were those reasons?

[danesparza]

Taxi's don't have an app to maintain. Uber does everything that Taxi's do, but also has an app (and associated server resources and operations) to maintain.

Your assessment of "replaces a whole layer of management and overhead" isn't entirely true as far as I know. If anything, they have more overhead.

[officelineback]

I believe you're wrong and here's why (maybe the previous poster's way of explaining it wasn't good).

1) Economies of scale. Airbnb has management and overhead but it's spread across thousands of cities. Cab co.s need a full structure for each city in which they operate. 2) Automation. Many of the manual jobs in local cab companies like dispatch and scheduling are automated. You don't call a human when you need a ride, you push a button on an app. So that's also less overhead.

They are also very clever in how they treat their drivers, so labor cost is also low, and of course there is no maintenance on cars (drivers change their own oil, buy their own tires, etc).

So I believe Uber could probably make money just by replacing cabs, but like others have said, they have to compete with other VC money receiving ventures that are wiling to burn money to come out on top.

For now, we the customers are benefiting greatly from the situation.

[vmateixeira]

You forgot: not paying local taxes and being able to operate illegally in some countries (without public transport insurances and legal permits). Which actually also happens to be the case of Airbnb.

[ExactoKnight]

Naked Capitalism made a good argument that Uber cannot experience economies of scale with its costs the way Amazon could when it scaled. Why? Because Amazon's fixed costs were... well.. fixed.

Unlike Amazon, Uber faces massive variable costs in driver recruitment/onboarding, legal regulations and local advertising.

Seriously. The rules needed to onboard a driver in Alberta, Canada are completely different from New South Wales, Australia. You can't automate driver recruitment/onboarding in either of these cities because these markets have passed actual legal regulations to allow ridesharing that require unique, but idfferent things (e.g. like market-specific commercial drivers licenses, background checks... or filling out registration forms made by the jurisdiction).

[kristiandupont]

Would you also argue that AirBnB does everything hotels do, as well as a website?

[ncphillips]

I'm not sure of their business model, but my understanding is that AirBnB doesn't set the price of the rentals or pay the home-owners, they take a cut of whatever price the homeowner decides to charge.

As I see it, Uber employs drivers, whereas AirBnB facilitates BnB owners.

[kylehotchkiss]

Imagine Uber built by a team of 10 people making $75,000k each in the midwest. That would not be the worst profit share / stock payout.

[paulcole]

This classic HN-trope simply isn't possible in this case. The app, while perhaps technically impressive, is the least of Uber's problems:

1. Recruiting drivers and passengers

2. Legal issues

3. Marketing and branding

4. Support

[Fricken]

Waymo's case isn't as strong as it looks.

1. Levandowski was critical to the development of the Lidar technology he allegedly stole

2. Google has never before enforced IP theft by former employees, and makes it clear to engineers that it doesn't do that. This is the first time.

3. Levandowski was selling technology he developed at Google to other companies, while working at Google, and Google looked the other way.

Waymo obviously has good reasons to be pissed off, but they've already set a precedent of permissiveness towards Levandowski's antics, which are not news to people familiar with the matter. All these years Goolge has been going around with a 'kick me' sign taped to it's back, and big surprise, Levandowski kicked them. They've already let him get away with this kind of shit.

I can't actually speak to the 14,000 documents he allegedly downloaded, so it's possible there's a silver bullet hidden in there somewhere. We'll see what sort of defense Uber trots out likely sometime over the next few weeks.

[swang]

Maybe I'm missing it here, but your points (mostly 1&2) seem to make Google's case _stronger_. I haven't heard about the point that Google let Levandowski sell tech to other companies.

Also Google did their own internal investigation which led to them strongly believing that the theft of important tech and secrets happened.

> The company decided to perform a forensic investigation of Levandowski’s former company computer after a Waymo employee was inadvertently copied on an email from a lidar supplier with the subject line “Otto Files.” The email was being sent to a list of people that Waymo believes were working with Uber. Attached to the email were drawings of Otto’s lidar circuit board.[0]

Seems pretty straight forward

[0]: http://www.recode.net/2017/2/23/14717432/waymo-otto-uber-ant...

[enraged_camel]

>>Waymo obviously has good reasons to be pissed off, but they've already set a precedent of permissiveness towards Levandowski's antics

Doesn't matter. It is Google's prerogative whether to go after people who steal their IP. Just because they have not done so in the past does not weaken their case. In fact, it strengthens it, because now they can make the argument that the reason they made an exception in Levandowski's case is because of the humongous potential impact of the technology that was stolen.

IP is not like a trademark where if you don't enforce it you lose it.

[brotoss]

Yea I was pretty much blown away by the difference between the Bloomberg feature in last week's Businessweek and the blog post that was submitted here earlier. Paints Levandowski in a MUCH better light with sources and high level interviews.

[Fricken]

Not enough people read that feature, it was the best write-up on the lawsuit to date, from a reporter who has a relationship with Levandowski and other Googlers going back years:

https://www.bloomberg.com/news/features/2017-03-16/fury-road...

[dantiberian]

Author of https://danielcompton.net/2017/03/14/uber-bombshell here.

I thought the Bloomberg article was pretty favourable towards Uber and Levandowski too, especially some of their reporting on the lawsuit without context:

> Levandowski defended Uber’s lidar technology as “clean”—that is, not the product of stolen design documents—and told the company’s engineers that he’d downloaded the files to work from home.

The major cache of documents that he downloaded were from an SVN system that he apparently had never connected to before (based on search logs). After he downloaded them and copied them onto the memory card, he wiped the laptop.

> A handful of Google employees soon followed him to the new company.

They fail to mention that those Google employees followed him out the door with confidential docs.

In my opinion, the Bloomberg piece is trying to blunt the revelation [1] that Levandowski was consulting/meeting with Uber just weeks after leaving Google and forming Otto by comparing it with his earlier behaviour at Google. They do a good job of that, but didn't ask the hard questions of why he would consult for a direct competitor, especially when he had his own startup where he had the freedom to do whatever he wants.

[1]: http://www.recode.net/2017/3/14/14923056/uber-google-waymo-s...

[Dwolb]

There's a way out of this that doesn't involve self-driving vehicles if Uber's willing to give up some control: 3rd party integrations.

If they allowed other ride-share/delivery services to integrate seamlessly into the Uber app/driver onboarding process, they don't need to build self-driving vehicles. They just level up from a driver dispatch service to a general dispatch service.

It'd be pretty tough for any one car manufacturer to go their own on a ride-sharing/delivery platform as long as Uber does everything they can to keep competition on both their inputs and outputs thriving.

[zelos]

Self-driving cars as in with no driver at all, even as a backup? Surely being allowed to run a fleet of self-driving taxis is 10+ years away from being legal?

[ethbro]

Running an unregulated taxi service was also 10+ years away from being legal...

[vkou]

If you're willing to pay the fines, and shoulder the legal risk, every illegal service is 0 years away from being legal.

[danielweber]

They're not a taxi service. This might seem like splitting hairs, but it's important. The car-for-hire industry just got good enough that it ate the taxi companies' lunch.

[vkou]

Assassinat.io is not an assassination service.

It's just a social network, whose premium users get to sponsor surprise meet-ups between its other users.

If it walks like a duck...

[danielweber]

No, seriously, it isn't a taxi service.

A taxi service can pick you up from the side of the road when you hold up your hand and whistle.

A car-for-hire service is where you call a company and ask for them to give you a ride from place A to place B.

It should be obvious from those two sentences why taxis are more regulated.

Now, with increasing technology (primarily universal smartphones), the need for taxis cruising the streets at random looking for hails has dwindled significantly. This still doesn't mean car-for-hire services are taxi companies.

It might be how you understand "taxi" but that's not relevant, any more than if a robotic factory was doing work with a lot fewer people and a critic can't understand why minimum wage laws don't apply to the assembly machines. In his mind "factory" is something that uses labor to produce goods so there must be some scam going on. But he just doesn't understand.

[ethbro]

How is pressing a button on my phone to signify I would like a ride different than holding up my hand and whistling?

[erikpukinskis]

It's probably already legal in some jurisdiction somewhere.

It doesn't need to be legal everywhere, it just needs to become legal in a way that scales with your ability to deploy the tech.

[bernardlunn]

If Uber really needs self driving cars to become a viable business, that seems like a hail mary pass. Agree, this not just about culture probs.

[beager]

At their proposed scale, yes. There's another version of reality where Uber can massively scale back and become profitable, but that screws over a ton of investors, and causes an almost complete turnover. At that point, Uber is a much smaller, and really unsexy company.

[metaphorm]

from our perspective (very focused on technology) it seems that way but I have my doubts. Jones was hired on as a marketing and PR oriented executive. I don't think his role had much of anything to do with technology or dealing with the lawsuit. My opinion is that he got out of there because he didn't sign on to be doing constant spin and damage control over a rancid company culture.

[nradov]

Such lawsuits rarely destroy the defendants. Uber has been working on autonomous vehicles for several years now so they probably have some patents of their own. Most likely outcome is they will pay Alphabet (Waymo) a few hundred million $ and agree to cross-license all of their patents.

[throwaway2048]

I dont see a reason that waymo has to license anything to a potential competitor that they could very easily put out of business instead, especially at such a low amount.

[nradov]

The risk in such cases is always that the defendant will use their own patent library to countersue. So the plaintiff has an incentive to settle.

[danielweber]

It's a lot different if Uber is found, as is alleged, to have induced the employee to take the secrets and bring them to Uber.

[smallgovt]

I think we should be asking whether this exodus is actually happening before why.

Uber is a large company with hundreds of "execs" at this level.

If we assume there are 100 "senior execs", even with a healthy 5-year retention, we'd expect 20 execs to leave every year. Sure, we'd expect churn to ramp over time, but 1 senior exec leaving every month would not be out of the ordinary.

I guess it's not really surprising that some people (and therefore reporters) want to believe this story. Uber is the villain and the villain must die. But, it doesn't make it true.

[nomnombunty]

There are probably hundreds of senior execs, but Jeff Jones was the second in command... not just any senior execs

[samfisher83]

Google ventures is an investor in uber. They invested 250 mil when it was worth 3.6 billion. That works out to about 7%. Lets assume their stock has been diluted somewhat to 5% however the company is now worth 60 billion. That is is 2.75 billion paper gain. I think it makes sense they just work together and settle the issue because even if google wins the only people who really win are the the lawyers, because you will be hurting your investment. Also google track record with getting R&D projects to market is not too great.

[alistproducer2]

Interesting point. Cheeky counter point: Google having done the research knows better than most that Uber's future profitability dependence on autonomous vehicles is doomed and is trying to recoup it's $250 million while Uber's still flush with other VC's money.

[Tepix]

I think they have more to gain here by winning the lawsuit even if they have to write off their $250mil investment.

[ExactoKnight]

According to Uber, the engineer who was the target of the lawsuit developed a self-driving motorcycle way before he ever worked at Google.

So while Google/Waymo probably have a case, it's not black and white. A lot of techniques developed over years of research was probably stolen from Google, but the truth is Google does not have a monopoly on using LIDAR sensors to train self-driving cars.

[dreamcompiler]

> Google does not have a monopoly on using LIDAR sensors to train self-driving cars.

True. But they do have a LIDAR that's an order of magnitude cheaper than the competition -- which makes LIDAR practical as a use-phase sensor (not just a training sensor), as well as a patent on "zoomable" LIDAR with Levandowski listed as one of the inventors.

[ExactoKnight]

Interesting, but I thought the best LIDAR sensors were being built primarily by 3rd party vendors? These vendors are working very hard to make a powerful under thousand dollar LIDAR sensor possible, so I wouldn't say that LIDAR sensor pricing is Uber's achilles heel.

[yeukhon]

I think it's the (1) internal politics, people do get freakin' tired of yelling and backstabbing in the end, (2) public blackslash and the down image working there. Associate oneself with this company at an executive position / senior management position is going to hurt people's profile down the road.

[elmar]

Flying cars of "Uber elevate" that is the path Travis is counting own to propel Uber to the next level.

[CSMastermind]

Absolutely. I wouldn't be suprised if most of the recent bad PR Uber has been getting is actually being driven by Google ahead of the lawsuit to hurt Uber's reputation and distract them from preperation.

[zaphar]

I would be surprised. That would be way out of character for Google.