What happened when the CEO of a Boston burrito chain became an Uber driver
boston.com
boston.com
He drives his Jeep, falls to a 4.3 rating and is about to get released from the service. He starts driving his Tesla, finds that passengers start treating him as an equal and give him a perfect 5.0 every time.
Somehow that's not at all surprising and disappointing at the same time.
A Camry or something is totally fine; doesn't have to be a Tesla, but should be something relatively quiet and comfortable.
If you can find the niche of uber customers who want to go offroad though, maybe you could make a fortune.
You don't get to justify providing poor service by accusing your dissatisfied customers of being soft.
Understatement of the millenium.
If someone has a problem with a ride, the only thing they really care about is letting Uber know that the ride was terrible (I bet the rating distribution clusters heavily around 1, 4 and 5 stars).
Looking at passenger rating histories should also play a role in how these are interpreted (if someone is an asshole and rates tons of drivers poorly even if they're well liked, that should be discounted)
By doing something like this, you have a way of weeding out bad drivers, rewarding great drivers, and leaving everyone in the middle alone.
Of course, this means you can't really reward drivers who are amiable and make you feel comfortable. But I feel as though answering "no" to the above question implies that your ride was a at least a "4."
I think three tiers of experience make more sense anyway. Your ride was either Bad, Good, or Exceptional. It's difficult to gauge what a 2/5 or 3/5 means.
That gives a very accurate picture of the driver. It's cool that sometimes my sidecar driver gives me candy and tells jokes, but that's not why I use the service.
I expect a driver to get me from A to B, and notice whether I'm interested in talking (in which case talk and be friendly) or not interested in talking (in which case don't talk please).
A half-pint of water is nice but hardly necessary - it won't convert a horrible ride into a great one, and its absence doesn't make a great ride a bad one.
So much is outside the control of the driver - a late night drive with no traffic is generally more pleasant than a sluggish rush-hour drive, but it's not the driver's fault.
I once took a longer Uber Black ride to pick up my wife a few towns away, who was quite sick. Once I mentioned this to the driver, he drove noticeably more 'efficiently', offered me his phone to call to check on her once mine had died, and waited outside the building briefly to make sure everything was ok.
Also, I've got a bit of prior race driving experience, and this guy clearly grokked car control. I wouldn't be surprised if he'd spent a good bit of time on a track. He also, without drawing any attention to it in any way, adjusted the suspension settings from 'firm' when we were going around on- and off-ramps to comfort on the highway, to keep the ride comfortable and level.
I gave him a five-star rating and sent Uber customer support an email raving about him. Wish I could have done more.
I think it this case you did it right with a letter to customer service.
When the "expectation" is a 4/5 star rating, then the average is generally 4.6-4.7 stars and there's no way to distinguish "normal, good service" from truly "above and beyond".
Additionally, if the "above and beyond" just gains a 5 star review, that doesn't really give the feedback needed for extraordinary service.
In my experience with cab drivers if they don't offer unsolicited conversations about religion or politics that's an exceptional experience. When I say conversations I mean they just drone on and on one sided about how great God is and how Obama is ruining the country.
If there 4.5 is the minimum average rating, everyone not rating 5 should be asked what was wrong. Or .. fix the rating so that not everything needs 5 out of 5.
It's still goofy, though.
It's a bit more nuanced than this, but the fundamental dilemma is:
• An overly negative review essentially dooms someone to never working again.
• An overly positive one leads to sticky questions from the next person to hire them (whom you probably know socially).
Clarke's solution is to write ... very closely ... an accurate but difficult to parse recommendation. As I recall, the essay ends with him noting that a household servant he'd dismissed some years before (pursuing a "flim" career) had since returned, to the pleasure of both parties.
Sadly I can't find a copy online -- seems that at a $1.78 purchase price the friction of commerce is excessive here for a 50 year old essay.
A syndicated content payment system (not micropayments) is another option.
This way a small sample size doesn't distort the rating too much, and predictions can be customized for every individual, and it's somewhat resistant to fake reviews, if they can be picked up on by the computer. It also gives you an incentive to rate and what you actually think.
It's even worse to have a points scale where not awarding the first driver the same as the second moves him a few percentage points closer to losing his livelihood.
I found some estimates online that a taxi driver might drive about 150 miles per day. Let's say 125 since this was sort of a short day. The IRS mileage allowance is 56 cents per mile - this is supposed to be an "all-in" number that includes all of the costs of owning and operating a vehicle. So that's $70/day for vehicle costs which the operator will have to pay - fuel, repairs, tires, whatever.
So we'll knock that $150 down to $80. So he's earning around $12/hour, before taxes, doing one of the more dangerous professions (cab driver is more dangerous than police officer, for example), and he's doing it completely uninsured - if he cracks up that Tesla during an Uber trip his insurer is not going to pay out on it. If the Tesla is stolen during an Uber trip his insurer is not going to pay out on it. If he has one accident, ever, he'll lose more than he could ever earn from a lifetime of working for Uber.
> 1. Never, ever try to call Uber with problems… because they don’t even have phones and there is very little if any reason to talk to them.
And Uber won't even pick up the phone to talk to drivers.
While cab driving is one of the more dangerous occupations, I'd avoid that comparison. Police officer barely rates as occupation of above-average danger in most ratings of fields.
Further, this is what the IRS allows you to DEDUCT. So his taxable income is reduced by the 56 cents per mile - to the extent that his costs are less than 56 cents per mile, he is making more due to the tax-advantaged nature.
Also, at the end of the first year he can elect to either use per-mile deduction, or, the MACRS method, which may or may not give him an advantage (but he has to choose and use one or the other for subsequent tax years).
But remember to factor in depreciation and wear and tear on the car, in addition to all the gas and repair costs. If you're driving an old Toyota Yaris it'll be much less than 56 cents per mile. If it's a brand-new Mercedes S-Class you've financed, it's probably going to be more.
Better money for them?? Yes Safer for the traveling and waling public? Not so much.
Drivers using Uber to fill free time or excess capacity is a radical shift away from how Uber started, with professional drivers using it to make a living. A friend of mine does this with Lyft. Imagine how many cars are on the road right now with only one driver and the rest of the seats empty. Lots of potential new drivers...
I use both Uber and Lyft frequently and, for me, Lyft has been always cheaper for the same trips.
I have no problem with that at all given the fact that Uber's image and product both rely on having good customer experiences.
If Uber were more concerned with providing a living to its employees, then they likely wouldn't last very long as a profitable business. $21/hour is a fair wage (a living wage in most locations, in fact), and that's all they really need to provide to be good corporate citizens.
Not to mention the "people are interchangeable" concept actually dates back to at least the Industrial Revolution, and is not only not limited to startups, but is probably less prevalent in startups than in manufacturing or corporate jobs in general. So I'm not sure what you're complaining about.
I am not saying that this is a black-and-white thing. There are certainly nuances here, and in a purely economic sense, humans are often interchangable. The problem starts when this attitude uncompromisingly becomes part of how companies (read: people) do business. A purely utilitarian, "amoral", selfish attitude to the world has bad ramifications if widely adopted.
Keep in mind that giving the boot to everyone below a 4.3 average rating means that if only one out of three customers conclude that your service was perfect, you will have your contract cancelled with no warning. It's easy enough to say "just have perfect performance, or otherwise just get another job" when you have a 5-year STEM education, grew up in a happy family and don't suffer from any serious condition or illness.
But not everyone has it this easy. If this is the future of employment, there need to be simpler options for people who can't perform significantly better than "average" most of the time. You might be safe in your software development job since you can't easily be replaced, so you boss won't fire you if, say, you go through a really messy breakup and perform at 50% capacity for six months. Everyone in an uncompromising performance-measured job will be canned under such circumstances. And having a tough time is something that all people experience once in a while.
As long as everyone could have their basic needs met when unemployed: Sure, make whatever labor structure is the most efficient. Otherwise, there needs to be a certain degree of compromise.
Let me rephrase the question. Do I have a moral obligation to pay said employee more than the current value he is providing to me? Now, most economists would say, "No," though there may be an argument about the present value of what the employee can provide me with in the future, perhaps in terms of special skills or increased loyalty.
Now, let me rephrase one last time. What if the cost of supporting this employee is make or break for my business? What if it is not just the under performing employee, but several others who would lose their jobs? What then is my moral obligation?
It's great to be loyal to employees, if you can. If you put your loyalty to your employees above the quality of the product you're providing, you may find you have no ability to pay any of your employees. Or in the case of Uber, that people stop hiring drivers.
The problem here is that you're asking economists moral questions.
I seriously doubt that. After having gotten to know some actual economists (you know, academics; not the ones who pretend to know about economics on TV) I have learned that most of them are actually quite different from the "hooray free markets all the way" stereotype.
In fact, I would bet that most of them will say that you do have a moral obligation, and many (perhaps most) will say that it also makes economic sense to help an under-performing employee through a bad period. Real economists don't emphasize the short term.
It's not that Uber is paying $21 an hour. It's that they're allowing the driver to go on and be matched with people who want that service. It might average out to $21 an hour for some set of hours (I imagine it's higher on the weekends and lower at low peak times). The fact you can back out dollars per hour from someone providing services on a marketplace doesn't really change things.
This whole line of argument ignores reality. The normal, real action is actually to pay the employee less than the value he is currently providing you.
That means that at the moment, there is less competition in the business than there could be (competitors could be nicer places to work, and hence could get by for example with paying less to workers, and be cheaper to customers as a result), or there are simply too much workers available compared to the amount of work available.
It's probably some mix of those. I've also noticed that in low skilled jobs, there can be many companies with very different cultures, and people pick the one where they fit in. They might all pay roughly the same and they don't differentiate that much with their service quality, because there isn't that much to differentiate with. Despite all the talk about excellence, the customers are not willing to pay much for any cosmetic extras.
If Uber were convinced that it was terribly immoral to fire someone, they would just put greater effort into making sure they got the best people in the first place.
So your policies favor a society with 5% of the people unemployed 100% of the time, instead of 100% of the people unemployed 5% of the time.
Job security is of questionable social good because it places a nonsensical emphasis on the harm of getting fired, as opposed to the harm of not getting hired in the first place.
I didn't communicate my point well.
My point wasn't that "it's OK to treat people like crap because: PROFIT!!" I'm the last person to claim that profit justifies everything.
My point was: If Uber doesn't impose serious quality control on its drivers, then even the best drivers will fail to make a living wage, because the perceived value of the Uber service (by customers) will drop with every bad experience.
Uber is creating a market for its service by being better than a taxi. Better in all ways. It takes serious differentiation to convince people to change their current habits.
>Otherwise, there needs to be a certain degree of compromise.
If you're talking about non-customer-facing jobs, or even basic retail jobs where the expectations aren't high, then absolutely.
If you're a giant corporation and someone has been working with you for years and they go through a bad spot, then sure.
If you're trying to create a "new kind of taxi" and part of your business plan is that everyone has a superb experience, then you have no choice but to fire people who can't live up to your goal.
The ex-CEO had a rating of 4.3 (the cutoff mentioned in the article was 4.5), but they didn't "immediately fire him;" he had a chance to bring his rating up, and he did. It sounds like there's some compromise already built into the system. Not to mention the minimum 40 ratings before they require a 4.5.
At the end of the day, it's a business with specific goals. Some businesses can employ people who only rate a 3.0 and still meet their goals; some can't.
>you go through a really messy breakup and perform at 50% capacity for six months.
Two months in my case (happened once; was going to leave the job in 2-3 months regardless, but it's the only job I've ever had that they weren't begging me to stay). But you are right that, with my awesome specialized software skill-set, I was able to pick up a new job in just a couple months, and at this point I can practically write my own ticket.
If your hypothetical break-up survivor is pissing off customers, though, Uber has no choice but to fire him or her if they want to meet their goals. How else are they going to keep the fleet-wide ratings at or above 4.5 stars, which is likely their goal?
Honestly I do wonder where people in, say, the bottom percentile of job or interpersonal skills (or both) can get work. I'd fully support a "living wage for everyone" concept, at least if it can be shown to work at some scale, to sidestep this problem. Or a government "works" program that guaranteed a job to everyone who wanted one (in conjunction with a disability system, of course, that could provide support to people who didn't have the mobility or brain function to perform even menial labor).
But how the system works when it's working well (i.e., prior to the 1970s [1]) is that there are more jobs than employees, and so the jobs have to make due with who they can find, and/or the jobs have to be attractive enough to get the applicants. It's how the programming market is right now: I know that most developers have really basic skills compared to mine (probably only 1 in 20 professional developers that I encounter are even in the same ballpark), but even the least productive developers get great salaries and benefits.
And how the system needs to change is that workers need to own the companies they work for. Then their personal motivations line up with that of the company, and employee/owners who have a bad breakup maybe spend six months collecting a lower salary and working part time until they get back on their feet. Or maybe because of their closer support network (employees who own a piece of the action probably will stick around longer -- and care more about their coworkers) they don't take six months to get over a break-up.
I agree that things need to be fixed. I don't agree that a company should be required (morally or otherwise) to shoot itself in the foot to achieve social justice.
[1] A long watch, but totally worth it if these questions are of interest to you: https://www.youtube.com/watch?v=_-KqeU8nzn4
But these people are real, most of them are still within the "normal" and their problems are more complex than missing interpersonal skills. I would not want to live in a world that treats "under-performing" (implication: "performing" is our raison d'être) people as second-rate citizens.
But again, our views on the economics of it could probably be reconciled even if our perspectives are different.
I would like to help them, but honestly charity is probably the least helpful of the options (if the only one available); I'm sure that no one likes to accept charity.
I believe that everyone wants to feel useful, meaning that accepting charity actually hurts their self esteem. Which is better than starving, of course. But it seems like there should be a better way.
>I would not want to live in a world that treats "under-performing" (implication: "performing" is our raison d'être) people as second-rate citizens.
Agreed, with the caveat that some jobs still need to have a minimum performance level.
We've been talking about driving people around, where the performance level is "only" critical to the survival of the company. But if you take the "ignore under-performance" concept to its logical conclusion, then you couldn't fire surgeons who kill all their patients.
Yes, that's a far more extreme example, but the point is some jobs really require that you are able to do them well. You can't ignore job performance for social ends, or you will end up harming society. There are many reasons that Communism failed: One is that people who know they can't be fired don't have as much motivation to perform well. An even better motivation to perform well would be to own the business you are working for (just look at the differences between neighborhoods that are primarily owned versus primarily rentals to see a very real example of this). But we can't flip the ownership equation overnight, and in the mean time we don't want incompetent surgeons killing people.
What you can do is create jobs where being a top performer isn't critical. And ensure those jobs pay a living wage -- even at 30 hours per week, ideally. A 30-hour work week might be the right answer, regardless. [1] It worked for Kellogg until the 80s, after all.
[1] http://www.alternet.org/labor/when-america-came-close-establ...
Rental works for nice neighbourhoods in Germany. But Germans tend to rent long term. That makes a difference.
Well, in a competitive market, something has to fall to the cost of the inputs required to produce it.
Whether it should be Uber's profit margins or driver's wages is probably a value judgment.
Naturally this is predicated on your ability to earn enough to take some time off, and it sounds as though he could (for a certain lifestyle).
On amazon for example 4,5 is considered positive and 1,2,3 negative. You really want to keep 90%+ positive. So when someone comes along and decides to give a 3 because they thought the transaction was good but nothing out of the ordinary it hurts you.
The lack of nuance doesn't feel great either. For instance when you look though the bad reviews on trip advisor for a good hotel they are often people that have enlarged expectations of what the hotel could reasonably do for them. In some cases the hotel even going above and beyond couldn't please them.
The insurance companies already spend a lot of effort trying to quantify those risks.
According to Consumer Reports (see linked article), the median car costs about $9k per year for the first 5 years (including the sticker price). That comes out to $0.75 per mile--not an insignificant amount. Driving 30 miles per hour, that's $22.50 in costs.
This sounds like a poor deal to me.
To actually calculate the costs of driving for Uber, we should only be looking at the variable, or the per-mile cost, which as far as I can tell should be:
- Gas - Maintenance - Loss in resale value
Automobile reliability has, generally, advanced worlds from where it was in the 1970s. It used to be that getting 100,000 miles was a significant accomplishment, now it's pretty much expected, and having a car run 200,000 miles isn't uncommon. I'd had a vehicle I ended up selling to a friend which went 280k before it was finally cash-for-clunkered. In that time, it went through a couple of radiators, a couple of exhaust manifold repairs, and a transmission (protip: keep 'em lubed and watch for oil leaks). And a few sets of brake pads, possibly a clutch. But all told, pretty remarkable.
In the 2000s you saw some significant improvements in safety (ubiquitous airbags, ABS, and traction control), but not a whole lot else.
That said: doing a lot of city driving in traffic is fairly high-risk and high-wear on a car.
1) Assumes that you only use your car for Uber/Lyft driving and that it is entirely valueless otherwise. If you were going to own a car anyway, you were going to eat a bunch of that cost to begin with. The extra costs are in the incremental mileage that you drive.
2) Assumes you are buying a brand new car, and uses averages for first 5 years. If you just start with a 1-year old car, that drops your average around 25% (from ~$9K to ~$7K).
3) Averages across a number of makes and models, including a number of luxury cars (pretty sure most people looking to earn $35/hour aren't in the market for Porches, Land Rovers and Mercedes Benz', all of which top out the scale being used in the consumer reports article). For example, a Toyota Camry would cost about $5,700/year to own for years 2-5 (http://autos.yahoo.com/toyota/camry/2013/l/cost.html), and around $7K/year for all first 5 years.
4) As stated earlier, assumes that a Lyft driver is just going 12,000 miles/year and that the rate $0.75/hour stays constant regardless of how many hours you drive in a year.
5) As far as I can tell, you can write off your expenses on your taxes, recouping part of the value as well (standard cost at 56 1/3 miles per gallon: http://www.irs.gov/publications/p463/ch04.html#en_US_2013_pu...).
Not quite sure how to evaluate, and it's very possible that I may have estimated on the lower end. But, the methodology used in that blog post linked is pretty visibly flawed in a number of ways. Now I've already given this more thought than I initially planned to, but you're right that I shouldn't have off-handedly thrown out $5/hour without investigation. That said, it bothers me that the blogger you referenced put out an entire article (that was referenced on HN) that used such poor reasoning and assumptions using misleading data. All of which to come up with a bolded "$3.45/hour!!!" conclusion. /rant
I'd hardly call Uber immoral for this. It's up to each person to square up with their insurance company.
When an Uber driver has an accident, the next thing that will happen is that their insurance denies the claim, and you have to claim against your uninsured motorist policy. Good luck to anyone who doesn't have such a policy because they don't own a car!
Hey, uninsured motorist rates just went up! I wish they didn't do that, and instead Uber fares and pizza prices went up because I don't use them. They have no business imposing externalities on me.
Indeed. If someone drives in such a way as to be uninsured, shame on them.
Still not the company's fault for the individual's lack of responsibility.
No. If they know, or more importantly from a legal liability standpoint SHOULD have known, then it's on them.
We haven't seen the test case that will decide the legal liability/insurability of P2P driver services, but it's clear that Uber and others are operating in a somewhat legal grey area and are not bothered by it. Indeed, you could argue that they are forcing the issue of regulation and insurability with their business model.
But I'd stop short of saying that they have NO responsibility. It's clear that Uber is hoping to delay any reckoning on that question until after their dominance is a fait accompli.
And take it with a grain of salt, since it's internal PR, but:
http://blog.lyft.com/post/75739276230/introducing-the-p2p-ri...
You could very easily argue that not requiring drivers to prove they've got commercial vehicle insurance is a lack of responsibility on Uber's part.
Uber's business is connecting drivers to passengers, and it takes a commission for doing so that is in some cases equal to or greater than what taxi companies collect for roughly the same service. The difference is that taxi companies are required to have insurance, and Uber arbitrages insurance costs and books it as "revenue."
Worst part is regulators seem to be letting them get away with it in many states, Not all.
For those in the hunt for a CEO, take note here. Curiosity, exploration, not afraid to get dirty, a desire to understand and learn. Well done.
This kind of article title is really starting to piss me off. Especially for this particular article where nothing "happened". It's just a QA with a new driver.