In the fall of 2008, Louis C.K. was a guest on Late Night with Conan O’Brien and delivered a soon-to-be-viral rant called “Everything’s Amazing and Nobody’s Happy.” It was about how we live in a time of mind-blowing technological achievement, and all we do is complain about it. His main source of amazement was—again, this was seven years ago—airplane Wi-Fi. He recounted his experience with it, how incredible it was to watch YouTube while soaring above the clouds, and how the network broke down minutes after passengers started using it. “The guy next to me says, ‘This is bulls---,” Louis tells O’Brien. “Like, how quickly the world owes him something he knew existed only 10 seconds ago!”

It’s a clip Michael Small knows well. “Oh sure,” he says. “That’s huge around here.” Small is the chief executive officer of Gogo, the largest in-flight Internet provider in the U.S. You might think an old comedy bit about in-flight Wi-Fi would be charmingly quaint; that most of the kinks would have been worked out by now and service would be fast and reliable. But you don’t think that. If you’ve flown for work on a major U.S. airline over the past five years, you’ve probably used Gogo, and “fast and reliable” are probably not how you’d describe it. More like “hell-sent and extortionate.”

Photographer: Matthew Avignone/Bloomberg

Since pioneering the in-flight Internet business, Gogo has dominated, commanding about 80 percent of the market. And as often happens with near monopolies, Gogo has become a name people love to hate. “So, Gogo is officially a joke at this point, right?” is the title of a well-commented-on thread on the road warrior site FlyerTalk. “They’ve got a monopoly, and they just don’t care,” says pharmaceutical executive and frequent flyer Keith Lockwood. “Once you have it, it’s hard not to have it.”

Gogo hasn’t done itself any favors. Steadily increasing fees and deteriorating data speeds have further annoyed already cranky flyers. “The service is so unreliable at this point that I don’t get a good enough ROI to spend $60 a month to maybe be able to download my e-mails,” says health-care executive and former Gogo user Manuel Hernandez.

For years, customer perceptions that Gogo is basically Comcast at 35,000 feet didn’t hurt the company’s bottom line. Users were literally a captive audience, and if they didn’t like the service, too bad, read a book. But for the first time since that Louis C.K. rant, Gogo has some serious competition. At least two companies—ViaSat and Global Eagle Entertainment (GEE)—are encroaching on its airspace, winning business by offering faster, cheaper connections that use satellites instead of cell towers. Gogo is launching its own satellite system that should come online by the end of the year. “We’re going to create a great new future in aviation,” Small says. “And as long as we keep making progress, the customers are going to hang with us.”

 

In the late 1990s, Boeing began building a satellite network called Connexion that would provide Internet access on planes. The technology worked; people who tried it loved it; but Wi-Fi, even home Wi-Fi, was new and there wasn’t enough demand. Flights were still mostly downtime for business travelers—a few precious hours of unreachability. So the service muddled along, there for the taking but mostly unwanted, like a seat-pocket copy of SkyMall. Then came the airline industry collapse following Sept. 11. Boeing shut Connexion down in 2006.

Gogo, which started that same year, had much better timing. It’s spent almost $1 billion developing onboard equipment and a network of transmission towers across North America. Back then, travelers in business class who needed to work used laptops or occasionally BlackBerrys or Palm Treos. A year later the iPhone arrived, and data-hungry smartphones soon became more or less a human appendage.

Courtesy Gogo

By 2008 it was clear that any airline worth its wings needed to offer some kind of in-flight connectivity, fast. Gogo had the cell towers and the FAA-approved onboard antennas and servers to make it happen. “Airlines didn’t want to do this organically,” says industry analyst George Hamlin. “Here comes somebody with a system in place, and you had to have it or people would leave.”

The first airline to sign up was American, which added the service to its transcontinental routes. With no competition to speak of, Gogo quickly expanded to Delta, United, Virgin America, Alaska Air, and Air Canada. Today, the company provides service on more than 2,000 commercial aircraft. It employs almost 900 people and had revenue of $409 million in 2014, up almost 25 percent from the previous year.

But demand for in-flight wireless has far outpaced capacity. Gogo’s CEO isn’t ignorant of customer dissatisfaction. “One of the reasons we get a bad rap out there sometimes is people compare what we do in the sky to the ground and just wonder why isn’t it the same,” Small says.

What Gogo does in the sky is, indeed, different from what wireless companies do on terra firma. It uses an air-to-ground system that functions similarly to traditional cell service, but its radio towers point up, not down. Gogo’s towers are anywhere from 50 to 200 feet tall and can be located in rather remote locations, such as atop peaks in the Rocky Mountains or deep in the Alaskan tundra. The tower signal is received by a device on the plane’s belly that looks a bit like those antennas you used to see on stretch limos. The signal is routed to an onboard server about the size of an old-fashioned tower PC and then continues to the cabin.

Gogo has to design much of its hardware. “The scale is low,” Small says. “There are only 40,000 planes in the whole world.” That means there isn’t a constellation of Lucents and Huaweis churning out new and improved airborne wireless equipment. “There is nothing off-the-shelf. Everything is custom,” says Anand Chari, Gogo’s chief technology officer. “People have made either hundreds, or at most a few thousand, of those, and that’s it.”

Gogo can provide a plane with as much as 10 megabits per second of connectivity, which is about half the average download speed on Verizon’s 4G network. Only one-third of Gogo-equipped planes, however, have the hardware to reach even that speed. The rest top out at 3 Mbps. And the signal is shared among all the passengers, so the more people using it, the more bogged down the service gets.

To balance this demand-speed trade-off, the company has focused on what consultants and B-school professors call price optimization. From its Chicago headquarters, Gogo is constantly analyzing usage—how many passengers are logged on and how much data they’re consuming—to come up with dynamic pricing that acts as a kind of capacity regulator. For example, Gogo charges more on a flight from New York to San Francisco (typically about $33) than it does on a flight from Detroit to Miami ($10). JFK-SFO is full of business travelers who want to be connected much more than passengers on the other flight, who are probably on their way to the beach and just want their kids to shut up and watch a movie. Prices also differ depending on the day of the week—Mondays and Thursdays cost more: Fly from New York to San Francisco on those days, and it will cost you $40 (you can save money on Saturdays, when the price drops to $28).

The practice is like the surge pricing used by Uber. It may make perfect logical sense—varying the price of a scarce resource according to demand—but it hardly wins the hearts and minds of noneconomists. More to the point, people who use Uber usually have options: a yellow taxi, the subway, walking. If you’re on a Delta or American flight, it’s Gogo or no go. “The airlines have plenty of choices,” says Andrew De Gasperi, an analyst at Macquarie Group. “It’s just that the passenger, who is the one who uses it, does not.”

Gogo differs from Uber in another way. While the taxi app’s surge prices tamp down demand, thus preventing the service from becoming overloaded, they also encourage idle drivers to hit the streets and increase capacity. When Gogo charges more, capacity doesn’t improve. “They’re participating in something we like to call ‘incremental value capture’ without also offering a better service,” says Frances Frei, a professor at Harvard Business School. “If I’m going to raise your rates, I also have to give you a better value proposition.”

Gogo has gotten away with these pricing games, because its main demographic, business travelers, are so price-insensitive. For them, staying connected is a necessity, not a luxury, and they tend to be on expense accounts, so it’s someone else’s dime. Also, these passengers are usually not bandwidth hogs. “The business traveler is working on a PowerPoint or a spreadsheet locally and sending e-mails,” says analyst Tim Farrar of TMF Associates, a consulting and research firm that focuses on mobile data services. “He’s not using it as intensively as a leisure traveler who wants to surf the Internet.”

Only 7 percent of passengers on an average flight use Gogo, according to the company. That’s still too many if you’re a user with a corporate American Express card and some Google Docs to work on. In a perverse logic, these customers would actually be better served if Gogo cost more, driving out the teenagers in Row 37 posting selfies on Instagram.

Earlier this year the company raised the price of its all-airline monthly pass from $45 to $60. Many customers complained, but so far it’s working. In Gogo’s latest quarterly financial filing, average revenue per aircraft was up 13 percent from 2014, “driven primarily by connectivity service price increases,” the report said. Farrar says: “They’ve found that there really isn’t much limit to what people are willing to pay if they have to get work done on the plane.”

While Gogo manages demand, it’s working on supply, too. Toward the end of this year, the company plans to roll out its 2Ku satellite-based service. The network will reach about 70 Mbps per plane at the outset and possibly be as fast as 100 Mbps in the near future. It will also be available over oceans and wherever else Gogo can’t put towers, which means the company will be able to go after intercontinental routes.

Gogo’s two main competitors, ViaSat and GEE, use satellites exclusively for customers such as JetBlue and Southwest. Both services have faster connections and lower prices, yet neither has dented Gogo’s dominance. That’s mainly because Gogo did a good job early on of locking up airlines into decade-long contracts. Gogo will install and maintain the equipment across a fleet for 10 years and share the Wi-Fi fees with the airline. As soon as the contracts are signed, hardware lock-in takes hold. Gogo (and ViaSat and GEE) equipment is proprietary, so switching providers means switching servers and antennas and everything else. “Once something is installed on the plane, it’s very hard to change it out,” Macquarie’s De Gasperi says. “For someone to install something and then decide a few years down the line, ‘You know what? I don’t really like it anymore,’ that would involve probably days or a week of that aircraft not flying. Aircraft lose money when they’re on the ground.”

ViaSat and GEE haven’t only been different from Gogo technologically but also in how they sell their services. Gogo charges passengers directly and then cuts a check to the airline for about 20 percent of that revenue. With ViaSat and GEE, the airline writes them a check for their service, the same way it may pay a caterer for food or a fuel company for tanker trucks of Jet A. Once the in-flight Internet is up and running, it’s the airline’s decision what to charge, if anything.

ViaSat CEO Mark Dankberg thinks the best way to sidestep customer rage is to eliminate the transaction altogether. The company’s basic service, which is capable of streaming video, is free for JetBlue customers—the fee’s already been baked into the fare (for heavier uses, a premium tier is available for $9 an hour). For Dankberg, the key isn’t dickering with prices, but eliminating the friction required for customers to get online. “It almost doesn’t matter what you charge,” he says. “Just people having to use their credit card is what the big issue is.” He’s proud to point out that more than 40 percent of JetBlue passengers use his service, compared with Gogo’s single-digit take rate.

What it comes down to is whether an airline views Wi-Fi as a source of revenue or as a perk. It’s not a purely philosophical choice. “I think there are some airlines that can’t afford to invest significant amounts, and therefore they like getting a check every month from Gogo,” Farrar says. “Remember,” he adds, “this is an industry that can’t afford to give too many bags of peanuts away or full cans of Coke.”

Contracts may be long-term and the hardware locked in, but as the wise man once said, nothing is forever in in-flight connectivity. ViaSat just signed a 10-aircraft deal with Virgin America for Hawaii flights. And there’s the all-too-imaginable scenario of some other technology coming along and upending Gogo and the rest of the industry. “If you truly believe your position in the market is impenetrable, then treat your customers like s--- and gouge them,” Harvard’s Frei says. “But I’ve seen very few organizations that get away with customers hating them.”

(Corrects Gogo's annual revenue in the ninth paragraph. Also corrects which airline was first to use Gogo, in the ninth and final paragraphs.)