Page 2

By Jeffrey Rothfeder

Can Information Technology Save the Airlines?


Page 1

: Introduction">

It was an annoying incident that served as an apt metaphor for much of what's wrong with the airline industry: On Christmas Day, Comair Inc., a Cincinnati-based regional subsidiary of Delta Air Lines, grounded about 1,100 flights and shut down for four days, after a computer couldn't keep up with a large number of passenger scheduling changes caused by an ice storm that had hit the Midwest a few days before.

It was an expensive glitch—one that cost the carrier about $20 million in revenue and almost wiped out Comair's operating profit in the third quarter of 2004. In the wake of the episode, Comair's president, Randy Rademacher, a 20-year veteran of the company, resigned.

"Comair knew there was a chance there would be a problem," says John Kasarda, a professor of management at the University of North Carolina's Kenan-Flagler Business School, who specializes in airlines.

"They said they had planned on updating their computer system, but other cost pressures—such as meeting payroll—were too great. So they just hoped that their system, patched together, would work. It didn't. Now they're playing catch-up."

The same could be said of all the major airlines. Technology, at one time a source of pride for the biggest carriers, has become the bane of their existence: Old computer systems often become unhinged, or prove themselves unprepared to handle the needs of the big airlines, while innovation has become the purview of their younger, more inventive competitors.

Thanks primarily to this technology shortfall, the major airlines' costs of operations cannot be controlled in any meaningful way, exposing serious flaws in their once-pioneering business model. The combination of the two has produced the greatest amount of red ink, and simultaneous airline bankruptcies, in the history of the industry.

In just the past three years, the Big Three carriers—American Airlines, Delta Air Lines and United Airlines—posted combined operating losses of nearly $15 billion, while United and US Airways are still in bankruptcy.

Now, airline executives concede, nothing short of a radical overhaul of their business model, including the adoption of new technologies, can turn them into highfliers again.

"At this point, if we look at the way airlines typically operate as a model for how to fix our industry, we're setting the bar far too low," says Steve Jarvis, staff vice president of e-commerce and distribution at Alaska Airlines, the nation's ninth largest carrier.

Next Page: Did Sept. 11 cause the decline of the major carriers?

Page 2

: Pointing Fingers">

Pointing Fingers
The conventional wisdom is that the decline of the major carriers began after the terrorist attacks of Sept. 11, 2001. Not so.

As early as the third quarter of 2000, following a long-term boom in airline profitability, the large carriers' average cost per seat-mile climbed above revenue per seat-mile (two key measures of carrier performance) for the first time in nearly a decade—a sure sign of the beginning of their current decline, according to an analysis by consultants Booz Allen Hamilton.

Part of the problem back then was rising fuel costs, and that's still a concern today. More intractable, however, were the fundamental technological disruptions that confronted the big airlines.

The business model maintained by large carriers, such as American, Delta, Northwest Airlines, British Airways and Alaska Airlines—called hub-and-spoke, or network, airlines, because they route flights to smaller destinations through central airports—is essentially designed to take anyone from anywhere to everywhere, seamlessly.

At its peak, in the decade or so after airline deregulation in 1978, this remarkably reliable approach produced lower ticket prices, greater productivity and, in most years for most airlines, consistent profits.

It also produced a suicidal addiction to yield-management systems, which had been created by the techno whiz-kids at the airlines to adjust dynamically, minute to minute, the pricing of seats as reservations came in and as the flight date got closer. Because of these computer programs, the airlines were able to charge a business flyer, often a late booker, $1,000, while a tourist sitting next to him was paying $250.

Before long, this technology became a primary profit driver for the carriers, thereby elevating the lucrative business traveler, a by-product of yield management, to the status of Most Important Passenger, responsible for as much as 60 percent of revenues and more than 100 percent of earnings.

That profitable arrangement ended with the arrival of the Internet, or more specifically, buying tickets on the Web. The emergence of sites such as Travelocity and Expedia in the late 1990s made the shadowy and perplexing world of flight ticketing transparent.

Travelers, from families visiting granny to businesspeople, could view fares from multiple carriers in an instant and, not surprisingly, usually chose their least-expensive option. In other words, seats became a commodity, and control over pricing shifted in key ways from the carrier to the customer.

As a result, prices began to fall for all types of passengers—just as the nation's lengthy economic expansion, fueled in part by the Internet, came to a halt. That, in turn, meant corporate belt-tightening and fewer business travelers.

"Yield-management systems made it possible to have more fare types than seats on an airplane," says Henry Harteveldt, a vice president at Forrester Research Inc. "But the airlines got greedy, and their bad pricing strategies caught up with them when the Internet exposed the pricing structures."

With revenue falling, the airlines tried to cut costs—usually by demanding labor concessions or trimming airline perks—but their complex system of hub airports, flexible ticketing, multiple connecting flights to every possible location and huge employee pools made that difficult.

Meanwhile, the discount airlines, such as Southwest Airlines, JetBlue Airways and AirTran Airways, which generally offer nonstop, point-to-point flights, often from smaller airports, faced none of these constraints. The result: As the large carriers stumbled, the low-cost airlines entered their greatest period of growth. With a much simpler business model and newer operations, the low-cost carriers could bring to bear newer technology, such as the Web, internal networks and productivity enhancers, to handle many of the administrative, operational and customer-facing activities that the large airlines had to trust to aging legacy systems or manual processes.

As a result, the low-cost carriers have kept their costs down—freeing them from having to concentrate too much on increasing revenue, especially during difficult economic periods—while riding discounted ticket prices to profitability. In 2000, the discount carriers' cost per seat-mile was 7.2 cents lower than that of the hub-and-spoke airlines.

Most telling, 65 percent of that cost advantage was the result of savings in scheduling, operational processes and ticket distribution systems, according to Booz Allen Hamilton.

By the third quarter of 2004, after firing thousands of workers and extracting salary and benefit cuts from those employees who kept their jobs, the large carriers had trimmed the cost per seat-mile differential with low-cost airlines to 4.8 cents. But revenue per seat-mile for all the carriers was still well below 1998 levels, and the top seven hub-and-spoke airlines reported operating losses, on average, of 7.6 percent of sales. By contrast, the much healthier low-cost carriers produced an average profit margin of 3.6 percent.

United and US Airways are still in Chapter 11—the first time two major carriers have sought bankruptcy protection at the same time.

The large carriers, experts say, suffer from a common business problem: They know they have to cut costs, but they have not been self-critical enough—at least not until very recently—to attack their most stubborn expenses, because doing so would require a fundamental change in the way they operate.

The big airlines keep slashing at the obvious expenses—labor, a frill or two, an underused hub—but continue to sidestep the more pressing need to purge their systems of the overwhelming complexity that makes true cost-cutting so difficult. To do that would require tearing down old systems—hardware and software—and rethinking, like many industries have in the past couple of decades, the role of technology and automation to improve productivity and customer service.

British Airways is among the leaders in that effort, but even it has not yet landed safely.

"For a lot of airlines, technology has been something appended to their business—an add-on application that's little more than window dressing—when in fact technology is as central to the success of an airline as the lift underneath the airplane's wing," says Forrester's Harteveldt.

The most obvious potential cost savings are in ticket distribution and interaction with customers. Through Web ticket sales and check-in, as well as airport self-service kiosks for handling baggage and boarding questions, airlines can save as much as $15 a ticket and significantly cut employee expenses. Alaska Airlines (along with sister Horizon Air) claims to have been the first domestic network carrier to book a seat over the Internet, in 1995, as well as the industry pioneer in Web check-in and airport kiosks.

Since then, most carriers have followed suit, although usually without as much success, either because they haven't made Internet booking a priority, or because integrating Web check-in with their internal systems has been too difficult, and the resulting customer interface is sluggish, or not attractive enough to catch on widely with customers.

Page 3

: A Tech Strategy"> A Tech Strategy

Of all the mainline domestic carriers, Alaska has been the most bullish on technology—and results seem to indicate that its strategy is paying off. Alaska says that about a third of its revenue comes from its Web site, compared to more than 60 percent for discounters such as Southwest and JetBlue, but just 15 percent, or so, for large airlines such as Delta and American.

Thanks primarily to Web ticketing and stingier policies regarding travel agency commissions, Alaska, which is headquartered in its hub city of Seattle, has shaved off about $60 million in distribution and selling expenses since late 2001. Additional savings are expected from Alaska's reengineering of its computer architecture.

Alaska is also heavily promoting a technology to the lower 48 states that it developed in Juneau: The technology uses global positioning satellites to land in bad weather. In January, the Federal Aviation Administration gave Alaska the okay to use the system at California's Palm Springs International Airport, but the carrier hopes to spread the technology to other airports in the next couple of years. The airline claims that the GPS approach, which so far it alone is using in the U.S., would have allowed it to land 21 of 24 Palm Spring flights that were diverted, cancelled or delayed last year.

Thanks to such initiatives, Alaska has become the lowest cost U.S. network carrier (10 cents per seat-mile versus 14 cents for Delta, for example) and the most profitable, with an operating margin of 6.9 percent in the third quarter of 2004.

Alaska will never be a discount carrier, says e-commerce staff vice president Jarvis, because its hub activities are too costly and much of its business comes from companies such as Microsoft Corp., Boeing Co. and Hewlett-Packard Co., which require flexibility in the form of late bookings, first-class seats and the ability to change plans. "But philosophically we're trying to deploy a low-cost carrier's architecture—cost control, quick turns and operational simplicity—while we want to provide a high level of service and enjoy a revenue premium because of what we do," says Jarvis.

"Our model is companies in other industries, such as Target, Nordstrom, Starbucks and Costco, that are able to drive value at a higher revenue point."

Widespread computerization is now three decades old at some industries, and as many as four to five decades in others. Yet the network carriers' lack of automation is, unfortunately, their most salient characteristic, experts say.

What other industry, for instance, would allow its most important employees to carry around manuals, maps and related documentation, the way pilots do, to perform jobs that are generally routine but at times require split-second thinking?

With so many essential safety rules governing every flight, this paperbound system severely slows down the pace of a business that makes money only when the planes are flying.

The benefits of jettisoning this old-fashioned approach have been aptly demonstrated by JetBlue Airways. The five-year-old discount carrier has been so stingy that its cost per seat-mile is more than a penny below industry-standard Southwest.

As with so many aspects of its business, JetBlue, a start-up with no legacy technology to reinvent—in fact, 75 percent of its information systems is devoted to new projects—is in the enviable position of being able to reexamine every feature of its operations before committing to a system. As a result, the carrier made some radical cost-saving decisions, such as creating a virtual call center with hundreds of agents working out of their homes.

But it was the situation in the cockpit that needed the most attention, because what went on in the cockpit slowed down operations, and that cut into potential revenue. So the airline equipped pilots and first officers with laptops and access to the Web to retrieve manuals electronically, perform load calculations, submit airplane data and ease flight clearances and updates. This speeds turnaround time by as much as 15 or 20 minutes, says Todd Thompson, JetBlue's vice president of IT.

For the large carriers, it would take more than a paperless cockpit to get planes back in the air quickly, although that would be a good start, experts say.

The problem is that, in order to manage many connecting flights and not inconvenience passengers too greatly, these carriers generally stack arrivals and departures in peak periods. That results in long aircraft turnaround times—planes can sit for three or four hours on the tarmac—while passengers and baggage must be moved from one side of an airport to another, and traffic congestion builds up.

Productivity necessarily falls and labor costs rise as manpower expenses are driven by how long an aircraft sits idle at the gate.

To mitigate this, while not completely replacing their business model or adopting the rules of low-cost airlines, the network carriers need to shift to so-called rolling hubs, says Tom Hansson, a Booz Allen vice president who specializes in airlines. This approach involves eliminating rush-hour connecting flights in favor of a continuous all-day schedule of arrivals and departures designed to meet the needs of nonstop passengers rather than those of the connecting flyer; the latter, after all, provide about 45 percent lower revenue per seat-mile than do point-to-point travelers.

In other words, connections might be lengthier, but airline operations would be sped up, thus enabling the carriers to get more of their planes in the air with more profitable customers in the seats. To make this work, Hansson adds, the airlines would have to invest in technology to improve scheduling, compress baggage handling and speed up airport logistics.

"The airlines schedule flights to arrive and depart in peaks to provide effective passenger connections," says Hansson. "The result is low labor and aircraft utilization, and a system structured around the needs of the least profitable customers. By redesigning the network around the needs of the local passengers, making connections a by-product of the operation, it's possible to nearly halve turnaround times, increase aircraft utilization, reduce congestion, and significantly improve labor productivity."

Page 4

: Generating Revenue"> Generating Revenue

These kinds of suggestions are not new to the network carriers, but even as they struggle with how to keep their identity as full-service airlines, and achieve hefty cost cuts at the same time, they can't afford to neglect shrinking revenue.

Since 1978, airline revenue per seat-mile has declined an average of 1.8 percent annually. This reflects post-deregulation competition, marked by low-cost carrier fares as low as $200 for a round trip from New York City to Los Angeles, and by price wars, such as Delta's recent 50 percent fare reduction, which was matched by other large airlines. Making matters worse, most passengers have very little loyalty to individual airlines; they simply buy the ticket that meets their needs at the lowest cost. Or, as Forrester's Harteveldt put it in a report: "Airlines don't own their customers; they lease them flight to flight."

To illustrate this, in a 2002 study, Forrester found that 39 percent of United's U.S. passengers also flew on American Airlines, 28 percent flew Southwest and 21 percent traveled on US Airways. Numbers like these are particularly vexing to the airlines, because they discount the value of the large carriers' heavily promoted frequent-flyer programs, which hold out the inducement of free tickets for repeat business.

As it turns out, these marketing campaigns fail to engender significant customer loyalty and each year cost the airlines as much as 10 percent of their total passenger miles in redeemed awards.

In order to keep revenues from falling any further, and to give cost cuts a chance to have a positive impact on bottom-line earnings, the airlines are beginning to realize that they have to do a better job of using the huge storehouses of information they have collected about consumers from ticketing, Web sites, e-mail, frequent-flyer programs and credit cards. Most of this information has gone untouched, lost in the database systems that were layered one on top of another every time a new customer service application or an Internet ticketing program came online.

Airline executives say that improving their ability to alert customers to special deals based on their travel habits, or customizing the flying experience—that is, giving passengers meal or entertainment preferences without making them feel like it is an imposition, or calling passengers by their names, the way some high-quality hotels do—is a must if carriers want to build a loyal customer base.

But achieving this level of data mining requires replacing legacy systems with a much smaller, integral network, says British Airways CIO Paul Coby, and most carriers are unwilling to commit to such a wholesale investment in technology. Coby, who was named to his post at BA just days after the Sept. 11 attacks, has received strong support from the airline's management to do whatever he must to modernize the company and eliminate its wasteful technology.

Now, in the middle of what he calls a post-Sept. 11 "strategic fight back," Coby says that perhaps his most important accomplishment to date is overhauling the airline's internal network so that customer data is now available in a form that can be used to the advantage of both the business and customers at the marketing, sales and operational levels. Coby sees it as a competitive edge that's responsible, in part, for the company's recent stellar performance in a disastrous business environment. In fiscal 2004 (ended March), BA was one of the few standouts among network carriers, with operating income of $740 million, an operating margin of 5.4 percent and a sharp turnaround from a loss of $202 million two years earlier—and BA managed this even though revenue has slipped 8 percent in that period. "My view has been that IT revolutionizes the airline industry once a decade," says Coby.

"Our relationship with our customers is changing completely. We know what customers want by how they interact with us. And we are building simplified systems that allow us to respond to customer preferences."


"Few Survivors Predicted: Why Most Airlines Are Caught in a Tailspin"

Knowledge@Wharton, Jan. 26–Feb. 8, 2005


Flight for Survival: A New Business Model for the Airline Industry

By Tom Hansson, Jürgen Ringbeck and Markus Franke
Strategy & Business Magazine
Summer, 2003

Although they may be progressing too slowly for their own good, the large airlines are in fact addressing some of the basic flaws in their overly complex business models.

Delta and United, for example, are creating hybrid flight networks by trimming their hub-and-spoke operations where possible and shifting business to newly launched low-cost carriers such as Song and Ted. These new companies are still a very small part of the large carriers' operations; nonetheless, they have surprised many airline analysts by operating at costs per seat-mile on par with discount carriers.

And all of the large carriers are at least paying lip service to the importance of finding technological solutions to their most profound problems. Recently, for instance, the airlines have begun to turn up the heat on travel agents to use Internet-based booking for purchasing tickets rather than the large mainframe systems that have been the industry mainstay for years.

If successful, this lobbying effort could save the airlines as much $10 per ticket, experts say.

But experts argue that these are piecemeal changes and that nothing short of a radical departure from the past will save the airlines this time around. The consensus for the future of the U.S. airline industry is that it will be composed of one international carrier, possibly United; two large domestic carriers, perhaps Delta and Southwest; a smattering of regional airlines, which can also provide network services, such as Alaska Airlines; and a few low-cost carriers, with JetBlue leading them.

That's conjecture, of course. But what isn't in doubt is that the threat to the mainline carriers, despite their post Sept. 11 rhetoric, is primarily internal—linked more to their own operational deficiencies and lack of imagination than to any external problems they can conjure up.

"This isn't child's play," says UNC's Kasarda. "It's an industry that's not going to look the same ever again. Look, in a strong wind even turkeys can fly, and the airline industry has shown that in the past. But now there's no wind—and a lot of turkeys."

This article was originally published on 02-05-2005