Outrageously Good Customer Service

There are not many strong incentives for individuals to provide great customer service. There may be small financial rewards that accrue if customers routinely tell an employee’s supervisor what a great job they did; but if someone owns the business, the rewards are greater because positive word of mouth will generate new customers. Not surprisingly, many reports of great customer service (one instance of which we’ve blogged about before) are associated with small business owners.

For most employees, the only real benefit of great service is the good feeling that comes with making another person happy and the pride of doing a job well. Add to that the chance to be written up in Freakonomics.com.

In the last week, I have had two shockingly positive experiences with major airlines. The first was with American Airlines — a connecting flight through LaGuardia as I headed home to Chicago. (Note to self: never, ever try to connect through LaGuardia.) I hadn’t realized it, but my arriving flight let me off in one terminal, while my departing flight left from a different terminal. To get from one to the other, I had to make my way across no small distance via a makeshift sidewalk.

The walk, along with a slight departure delay, got me to the ticket counter too late to check in for my original flight. The machine could do no better than to put me on standby on the next flight. I dashed through security to reach the gate from which my original flight was scheduled to depart. There were people everywhere – literally dozens of standbys who would not be accommodated because every seat was taken.

Halfheartedly, I approached the counter and said that I had a seat on this plane, but it was probably long gone. The woman behind the gate, Carlene Boyd, replied, “Is your name Steve Levitt?” I said, “Yes.” She said, “I thought you might show up. So I saved your seat until the last second. Here it is, feel free to board.” I didn’t think to ask her why she thought I would show up so late. But that one simple act was enough to make me loyal to American Airlines until the end of time.

That is, until United Airlines did one better yesterday. Once again, I was returning home via LaGuardia. Because of weather, all the flights were delayed two hours or more. I arrived an hour early, which meant at least a three hour wait. Because of an earlier cancelled flight, the person at the counter told me there was no way I could catch one of the earlier (but also delayed) flights.

As I sat down to a dinner of fast food Chinese, my cell phone rang. The caller was a United Airlines employee named Michael. (Sorry, Michael, I can’t recall your last name.) He said, “I see that you’re at the airport and your flight is delayed a few hours. A seat opened up on an earlier flight, so I grabbed it for you in case you wanted it. It leaves in forty minutes, so you’ll have to hurry.”

When I met up with Michael to get the boarding pass, I asked him what his job title was. It turns out his job is to watch out for the interests of frequent flyers and make sure things go as smoothly as possible for them. Who knew United had people specifically employed to handle that job? I asked how he got my cell phone number, and he said it hadn’t been easy — he’d had to make four calls before anyone would give it to him. That is great customer service.

I have no illusions as to why American and United are nice to me: I travel way too much and they are the major airlines serving Chicago. I am a good customer. Still, compared to all the other things that airlines can do — serve warm nuts, show good movies, give a few inches of legroom — I would trade it all for a few more instances in which the airline does something out of the ordinary to get me home faster to see my wife and kids.

Finally, one last example of amazing customer service that has nothing to do with being a regular customer: This spring I was visiting my best friend from college, a fantasy baseball co-owner and high school principal extraordinaire named Matt Spengler, at his new house outside of Boston. For dinner, we ordered take-out from Bertucci’s in Needham, MA. Somehow, the order was miscommunicated over the phone — we’d ordered two pizzas, and when we arrived to pick them up, they’d made us one pizza, half of it covered with the first pizza’s topping and the other half with the second pizza’s topping.

The manager was incredibly apologetic. “No problem,” I told him. “But I need you to make me a second pizza exactly like this first one.”

Matt and I sat down for a beer at the bar. Within ten minutes, there were three pizzas: the original one, and two more. I thought that there had been another mix-up (three pizzas instead of two), but the guy was so nice that I wasn’t going to say anything. “How much do I owe you ?” I asked, since so far I had only paid for the first pizza. “Nothing,” he said. “It was our mistake.” Matt and I argued with him for about five minutes, but he refused to take any more money.

My guess is that Matt will order enough extra pizza from this restaurant over the next ten years to pay them back for their kindness twenty-fold. If that is the case, this kind of service makes good business sense. But if it makes good business sense, why is it still so rare?


#18: Ah, I see. Thanks for the clarification (regarding a question I asked in comment #12).

But now a new question arises: according to this hypothesis, only the owner has an incentive to be nice; no one else does. But is the increased probability that customers deal directly with the owner in small businesses vs. large ones sufficient to explain why small businesses are nicer than large ones? It seems not.

That is, it seems that, even in a small business, your chance of dealing directly with the owner is low, and niceness stems not from the owner but rather from the employees, just as in a large business.


What value is it to John Q. Retail employee

None. All Levitt's examples are either employees paid to be extra nice to a few customers or a pizza manager/owner who is paid not to be John Q. Retail Employee.

What Levitt is really saying, nay yearning for, is a time (the 1950s) when employees were paid a living wage in order to be loyal to that employee. Americans, including me, prefer cheap and six sigma quality. Not good service.

Alas, the Kenmore washing machine salesman at Sears is gone. Perhaps for the best.


I'm not too convinced about the airline examples. Both airlines had to fill their planes anyway, and may just as well have given the last seat to someone else, which would have been equally good customer service. Handing out seats for a capacity-constrained plane is ultimately a zero-sum game: the fact that you got a seat simply means that somebody else didn't.


Mmmm, Bertucci's.


Once a Schwann's truck (a company that deliver ice cream and other groceries to your home) was blocking my driveway while making a delivery to a neighbor. I sat in my car for a minute and scowled at the driver until he moved.

The next week when the truck returned for another delivery, the driver came up to my door and handed me a free half gallon of ice cream along with a tip for making a delicious shake. I felt awful about scowling, but the gesture went a long way to making me not hate Schwann's anymore. (I can't say I've bought anything, but that's a matter of affordability.)


I completely agree with #20. The turnover in retail is extremely high, with most employees being relatively unskilled or young. They aren't paid well (although I think the labor market for retail workers is pretty fair), they aren't invested in the company, and they aren't invested in the outcomes of individual transactions with customers. Negative customer service experiences are always quickly pointed out, but how often are good experiences reported and rewarded? And how about the middle of the curve, all the customer transactions that took place as normal with no fuss, and perceived as neutral? Not to mention the frazzled, impatient, easily-irritated general public trying to get their pizza ordered or gallon of milk home in time to take Johnny to soccer practice.

This brings to my mind the NetPromoter concept (http://www.netpromoter.com/), which measures customer loyalty and satisfaction in terms of whether people would recommend your business to another person. I've actually conducted these surveys for my company. Check out the link for more info on the methodology.



"I hadn't realized it, but my arriving flight let me off in one terminal, while my departing flight left from a different terminal. To get from one to the other, I had to make my way across no small distance via a makeshift sidewalk."

Excuse me if I'm mistaken... but isn't that the definition of a connecting flight? I've only flown maybe 100 times in my life, but 9 times out of 10 my connecting flight is in a different terminal. I'm getting ready to fly to DC in a month, and when I read this story I checked my flight information, and suprise suprise, my connections are all in different terminals.


I'll tell you why it is -- a generation of business book authors have convinced CEOs and management types that everything should have "metrics" and that if you can't measure something on a spreadsheet, it is worthless to your business.

There is no way to measure these kinds of acts of goodwill in any objective way (you could run customer satisfaction surveys or followup with specific customers, but you're never going to be able to measure the way you can graph how many calls per hour a specific customer service rep handles -- regardless of whether those calls were handled in a way that pissed the customer off or left them feeling valued).

As you say, a small business owner can see with his own eyes the repeat business and ongoing relationships that result from good service. I certainly do in my business. I've fired some customers before but I've never lost a single one to a competitor, despite being far from inexpensive for my line of work. Because they know I'm good to my word, that I will solve whatever problems come up, and I will never, ever fail them or embarrass them to their superiors by not coming through. People are happy to write big checks if you make their job easier.

But trying to explain to someone in accounting why you have an employee whose sole job is to solve problems and give things away is like arguing with a brick wall -- they only see a cost with no directly measurable benefit. Some places get away by sneaking such efforts into "marketing", because "customer service" is seen as a cost to be cut at every opportunity.



i don't know if you're familiar with Karma Kitchen but basically, every Sat night in Berkeley, CA, there are no charges for what you eat. It's up to you if you want to donate. You can read about their excellent customer service here: http://www.charityfocus.org/blog/view.php?id=1517


My mother taught me to say "thank you" when it's warranted. I frequently receive good customer service, and always offer to commend such service to the customer service rep's supervisor. Both rep and supervisor frequently have thankless jobs, most often receiving complaints, and rarely receiving positive feedback.

On the other hand, I view poor customer service as a competition, and, with a smile on my face, will metaphorically shake the hand of my perceived adversary, gradually tightening my grip until I get what I feel I deserve. Results have ranged from having an airline put my forgotten luggage in a taxi for a 90-mile delivery to having a major appliance manufacturer give me a free part (which would have cost over $200) plus $300 apologetic credit toward their merchandise. The key is knowing how, and with whom, to negotiate; rather than blowing a gasket.


As a manager, what I always taught my employees is that we are all customers - treat each one the way you want to be treated when you are their customer.

Not only did my employees receive accolades from customers often, they reported back that it made them less stressful when out shopping or whatever because it was easier to put themselves in the place of the clerk (or whoever).


"At MetWest there are no final exams, no letter grades, and no bells herding students into crowded hallways.

Students spend only three days a week in classrooms. The other two days they are interns at hospitals, government agencies, schools and community organizations. "

I'm sorry to say i don't find anything
extraordinaire about Matt Spengler other than the altruistic fact of being a part of a school for the disadvantaged.

However, the school's performance is rather horrible I feel. My first problem is that if the school doesn't have any tests then how can it possibly measure or develop academic potential in anyone???? Without tests students aren't learning about abstract skills.

The only reason these kids work because they can't afford the expenses to study, so in that way it might be a good thing for them to work part-time and do projects. But there is nothing exceptional about this! An exceptional school(for poor) will be something like which takes disadvantaged students and makes them have a career which changes their income level significantly, or in other words, make them smart in a way that we could say that the schooling was statistically significant.

It would have cases where it takes ordinary poor kids and motivates them enough to become someone like Roland Fryer (ok this is a bit of a extreme example)!



Regarding airlines filling up planes: I'd say the marginal cost is pretty close to zero if they can get you on an earlier flight, so I can see a large incentive for them to fill every plane to capacity. The other folks on standby will likely not be lost revenues because they've already purchased their tickets.

As far as holding your seat until the last possible moment - I guess that actually could be a negative for their short term revenue, so that's is more purely a customer experience effort which is an attempt to solely create loyalty.


I know it's easy and petty, but your good pizza experience brings up my not-so-good pizza experience-

Readers Digest version. North Beach Pizza in SF. The opposite treatment.

Didn't deliver the pie ordered to my hotel. I called at the 1.5 hour mark, asked if they could still deliver the pie.

Was told, "No problem. 45 minutes." Like the first time.

"Will this one get delivered?"


"Will it be free?"

"I can't authorize that. The manager is not in."

The discussion went on for 15 minutes. You get the picture.

Needless to say, I said, "Don't bother."

I even emailed them at their customer service address. Never got an answer.

I get that a hotel delivery is not a repeat customer and therefore carries almost no incentive. I just wish there were a way to wrench that satisfaction out of the phone.

And I know there are others out there that, even though traveling, want to support local small businesses. Wouldn't good service spread the news beyond their locale?

Unfortunately my small monetary vote won't count for much in this case.


Vincent Clement

Claywinter: You should have told the manager at the hotel about your pizza experience. He could do any number of things including calling up the pizza store and telling the manager that he will not be recommending the pizza store to hotel guests. In many areas, a group of hotels is often owned by the same person or company - all those hotels could cease to send business to the pizza store.

"I can't authorize that. The manager is not in."

That is the sign of a poorly managed business. It says that the employer does not trust it's employees to empower them to make time-critical decision and that it does not trust them to not abuse their powers.


Americans, including me, prefer cheap and six sigma quality. Not good service.

To borrow from a management theory, isn't service in many sectors a sort of hygiene factor? E.g. as long as its above a certain threshold (which is certainly varying), its not much of a factor but when it drops below that level, it's a pretty significant matter? And certainly expectation theory has got a significant role here.

I had my own surprisingly positive experience on the airlines yesterday. Flying back from Denver on Frontier (which I've called "Air Greyhound" for years given my previous terrible experiences with the airline), we had exceptional treatment from the gate to the competence and professionalism of the pilot and flight crew. Serious kudos to the Denver to Omaha pilot and crew on the 6/5/07 5:25 flight.

In my case, expectation theory certainly played a role. I expected a mediocre experience and would have been pleased with normal competence, given previous experiences with Frontier. Having really strong service shocked me. Omaha's furniture retailer, Nebraska Furniture Mart (part of Warren Buffet's Berkshire Hathaway) has this same issue. Customers walking into the Mart expect an experience commensurate with the greatest furniture retailer in the US. Anything short of that expectation is a real disappointment, while some tiny retailer who barely performs above marginal expectations is well received.

While I agree that most Americans discount the importance of service, I'd suggest there's still a moving, minimum threshold factor that has to be considered. Push service delivery too low and risk being made obsolete by competitors who don't cut corners as aggressively.


Tom Kelly

I believe that outrageously good customer service (OGCS) is correlated with certain business models, not business size.

A pizza business is an example of an enterprise that is highly correlated for OGCS. In most cases, the customers are local, hard to get, easy to keep, and well connected to each other and other potential customers. And the value of any one transaction is trivial compared to total revenue. There is a rapid cycle rate from one purchase to the next and demand is highly elastic.

Local customers mean that "what goes around, comes around". Faced with a limited pool of prospects, every one must be given optimal service in order to build reputation.

Hard to get means that the marginal cost of providing OGCS is less that the marginal cost of acquiring a replacement customer if you don't provide OGCS.

Easy to keep means that providing OGCS actually works- provide it and customers stay.

Well connected, which is a close cousin of local, means that the benefits of (not) providing OGCS are transmitted beyond the receiving customer.

And the value of an individual transaction relative to all revenue is important because a business model with many small transactions can afford to sacrifice profit considerations on any given transaction to provide OGCS.

The rapid cycle rate encourages OGCS because the payoff for providing OGCS comes quickly. A recipient of OGCS from a pizza store will likely reorder within 2 weeks.

The highly elastic demand allows the market to expand if OGCS is provided. I was involved in a "service war" with a competitor once where we both ended up with far higher sales as the market demanded far more of our products after they were sold with an incredible level of service by both of us.

A business where you would not expect OGCS would be the opposite of this- nonlocal customers who are easy to get, hard to keep, not well connected (though with the internet that's hard to do), and with a high transaction value relative to overall revenue. a long cycle time, and inelastic demand.

My last experience with a business like this was an airport jewelry store which was my last chance to redeem myself from a perceived indiscretion before flying home. It was a small business and the owner/operator seemed well aware of the sort of situation that had me needing to buy jewelry at that moment. He seemed more interested in extortion than service.



I think that good customer services is sometimes correlated to business size. But a more important factor, in my mind, is the amount of competition the business faces. Take the book business, for example. Small independent bookstores face an inordinate amount of competition from the chains like Barnes & Noble and Borders. Their margins are very small and they often struggle to keep afloat. At the bookstore where I worked, the owners (after 30 years owning the business) were barely making $12/hr.

Because of this serious competition and the fact that the chains could offer cheaper books (because of publisher discount structures) and a wider variety (because they can afford more rent), the main defining feature of independent bookstores is their customer service, their knowledge of their inventory, their capacity to thoughtful and intelligently recommend books to customers.

Right now, the airline industry is having a lot of problems, facing stiff competition amidst rapidly increasing expenses (fuel). Sounds like United is doing its best to make a stressful situation (any connecting flights can be stressful) better for its repeat customers. With all our tech link-ups now, for every bad experience, the average person will tell about 20 people about it (unless they have a blog such as this), whereas they might tell one or two about a positive experience. Sounds like United it trying to minimize that negative publicity.



I wonder whether the airlines have a list of people who are opinion-leaders, and whom they (although not publicly) treat with priority. I wouldn't be surprised if they had, and I am sure that if they have such a list, then the name Levitt is on it.

A blog post about an appaling flight experience published on Freakonomics blog must be pretty bad for the airline. On the contrary, a positive blog entry is like a free lunch for them.

If the airlines are at least a bit clever, they must have such a list.

And then, of course, a heretic thought... Is Steven Levitt rational enough to realize that posting such blog entries will likely enlist him to even more preferred-customers lists? ;-) If so, does he write these stories with this purpose in mind? ;-)

(end of the hyperbole)


i think its all about short-term vs long term - I would expect established firms to have bad customer service and young firms entering the market to have good customer service relatively - but den i dont hav data to prove that - Old firms probably feels that their business will run forever and therefore wont have to improve on the margin - new firms on the other hand is struggling to make a big name and therefore would have long-term objectives