May 2001, Volume 69, Number 3

In the conservative world of banking and finance, it takes a bold and passionate leader like Richard Fairbank, MBA ’81, to change an industry.

By MIKE McNAMEE

Case Discussion: Rich irbank, MBA '81 The first-year MBA marketing classes completed a case on Capital One Financial Corporation in February 2003, then heard founder Rich Fairbank's comments on the case and the founding of his company. Video File, 30:00 minutes

HAVE YOU EVER TRIED TO EXPLAIN a credit card to a child—why the supermarket will give you valuable items like Froot Loops, Lunchables, and pudding cups, and all you do is swipe a piece of plastic through a machine? “It’s like, the bank gives the money to the store, then the bank sends me a bill, and ... well, it’s kinda complicated.”

If you think that’s tough, try to see your credit card the way Richard D. Fairbank does. To the founder and CEO of Capital One Financial Corp., that slab of plastic is the key to an information machine—a machine fueled by data on who you are, what sort of people you live among, whether you’ll carry a balance or avoid finance charges at all costs. This data machine may offer the closest thing ever invented to perpetual motion, because the more you use the card, the more data it produces. The machine takes your bytes, combines them with transactions by millions of other cardholders—and before long, it can pinpoint what you’re likely to buy next, whether you’ll respond to a pitch for long-distance telephone service, and how likely you are to sign up for Internet banking.

Fairbank, MBA ’81, built that data machine—and revolutionized the credit industry. Before Fairbank, all credit cards cost the same: 19.8% annual percentage rate, $20 fee. Before Fairbank, no one had ever heard of teaser rates, “superprime” cards charging just 9.9% interest, or credit cards printed with your kids’ pictures. Since Fairbank—to be precise, since October 1991 when he rolled out the first cards to break the plastic price barrier—credit cards have become hotly competitive, customized products with thousands of combinations of rates, fees, credit lines, rewards, and services.

Fairbank’s insight—that credit cards offer digital finger-prints of consumer behavior—has two results of which he’s particularly proud. Before 1991, half of America couldn’t get a credit card at all, because banks turned thumbs down on anyone who didn’t fit their one-price-fits-all approach. Fairbank realized that the data machine could design cards to make those screened-out borrowers profitable while maintaining Capital One’s low default rates. Millions of Americans now can rent cars, book airline tickets, and shop online as members of the credit card economy. “We’ve democratized credit,” Fairbank boasts.

The other achievement is Capital One itself. Based in Falls Church, Va., the company is now the country’s ninth largest issuer of MasterCard and Visa cards, with 34 million customers and $29.5 billion in cardholder balances at the end of 2000. Fairbank and his partner, Nigel W. Morris, built the company to embody their data machine: Every aspect of Capital One’s business, from hiring methods to how to answer the phones, is subjected to the same intensive “information-based strategy” that it uses to design and market credit cards. It must work: Earnings per share have grown at a head-spinning 31 percent average rate for the last seven years. And in each of the last three years, Capital One has ranked as one of Fortune magazine’s best places to work—even as it was quadrupling its payroll from 5,000 employees to almost 20,000.

Even after six years as Capital One’s chief executive, Fairbank still acts like he can’t believe what he’s done. Telling his company’s story to a class of budding entrepreneurs at the University of Virginia’s Darden Graduate School of Business Administration one day last January, he joked: “If Rich Fairbank can start a successful company, there’s hope for everybody in this room.” Nothing in his career prepared him to be CEO of a Fortune 500 company—after all, he had never worked for a large company until one grew up around him.

But friends and associates see Fairbank’s traits in many aspects of his company. There’s his tenacity: “Rich had to overcome a lot of resistance in banking, an industry where new ideas are quite rightly met with suspicion,” says William Barnett, associate professor of strategic management and organizational behavior at the GSB, who’s used the Capital One case in his classes. There’s also his dedication: “He eats, sleeps, and breathes Capital One and the information-based strategy,” says James V. Kimsey, cofounder of America Online and a director at Capital One.

Most of all, there’s his intensity and focus. “When he talks, you feel the fire that he feels,” says David S. Berry, director of research for Keefe Bruyette & Woods, an investment firm specializing in the financial industry. “When he listens, you get the feeling that he’s absorbing everything.” Indeed, Fairbank’s willingness to chew over ideas for hours on end is legendary among his colleagues—and sometimes, they confess, taxing, especially if they’ve been sitting for hours in the “beaten-to-death” armchairs where Fairbank prefers to hold meetings. “He’ll give you all the time you need. He tends to give people more time than they deserve,” says Paul P. Paquin, vice president of investor relations. The toughest job in the company, adds another exec, is “keeping Rich’s schedule.”

To Fairbank, 50, it’s just second nature to throw himself into ventures. For each of his four oldest children—he and his wife, Chris, have six, the youngest being infant twins—he’s immersed himself in the child’s favorite activity. He didn’t just coach his kids’ soccer or hockey teams; he took up competitive adult-league soccer and hockey himself to connect with his children’s experiences. It’s risky: “I took up golf five years ago with my son, and now that he’s 16 he’s way eclipsed me,” Fairbank says. “But by learning side-by-side with them, I share their passions.”

Fairbank credits these traits to the influence of his father. William M. Fairbank, a professor of physics at Stanford University, instilled a deep respect for empirical data and taught Rich persistence. An experiment that Professor Fairbank proposed in 1960 to test flaws in Einstein’s theories is scheduled to be launched on the space shuttle in 2002, 13 years after the physicist’s death.

Rich’s first venture after he got a Stanford economics degree in 1972 was entrepreneurial: He founded a swim school to try out his own methods of training champion swimmers. And when the Stanford Business School application asked for his proudest achievement, Fairbank listed “suing Stanford.” As general manager of the Ladera Recreation District in the late 1970s, he had taken the University to court to win eminent domain over playing fields the district leased from Stanford.

While at the GSB, Fairbank knew he ultimately wanted to found a company—“but how do you start a company when you have no experience, no money, and no business ideas?” So he set his sights on management consulting, figuring he’d encounter business problems that would inspire ideas. Hired by Strategic Planning Associates (which has since merged and become Mercer Management Consulting), he discovered just one flaw in his plan: The corporate behemoths that hire consultants don’t tend to spin off ideas that a lone entrepreneur can develop.

After a stint analyzing oil companies in Canada, Fairbank moved to northern Virginia to build a financial services specialty for SPA. In 1987, he and Morris, an Englishman eight years his junior, landed a contract to study the credit card operations of a large New York bank. “We’d been there just one day,” Fairbank recalls, “and it struck us like an epiphany.” Instead of rejecting half your potential customers and charging the rest the exact same price, why not use the mountains of data that credit cards produce to design cards with prices and terms to satisfy all sorts of customers?

Morris and Fairbank promptly pitched their vision to their client. “What if two customers got together at a party and discovered they were paying different rates?” the client replied. “It’ll be a PR disaster.” Undeterred, the two consultants spent the next year pitching to the CEOs or credit card chiefs of 20 of the 25 biggest card issuers. None bit. Finally, Fairbank and Morris lowered their sights—and found a partner almost in their own backyard. Signet Bank, based in Richmond, Va., was a regional bank hoping to expand its modest card operation. In October 1988, Fairbank and Morris joined Signet.

Their first challenge: Getting data. Signet’s card chief was a 30-year veteran who taught employees his own strict credit standards, and the result was the industry’s lowest charge-off rates: Only 2.6 percent of balances went unpaid. “His gut was the best in the business—but we needed a credit model based on data,” Fairbank says. “We told him, ‘We’ve got to know more about the customers you’re rejecting—take everybody.’” Amazingly, the bank agreed.

Charge-offs rose to 5.9 percent while Fairbank’s crew tested dozens of card permutations. Signet, meanwhile, was being dragged down by troubled real estate loans, and bank executives were wondering when this strange new operation would deliver results. “Rich kept us all motivated—the way he shared his dream kept us excited even when we hit all kinds of resistance,” recalls Cathryne Doss, who was the third person hired in Fairbank’s new marketing analysts department and is now a Capital One vice president. Finally, late in 1991, one idea showed promise: Offer potential customers a low initial interest rate if they consolidate debts from other accounts. The new card instantly built up big balances that brought in hefty interest payments once the three-month “teaser” rate expired. The teaser-rate balance-transfer card was born—and Signet mobilized 100 employees in a week just to mail checks to pay new customers’ old debts.

The idea worked because Signet’s data machine broke an ancient rule of credit: The people who want to borrow a lot of money are the ones you least want to lend to. Combing through the thousands of accounts Signet had opened almost at random, Fairbank’s team found ways to identify good credit risks who were also heavy chargers. Using their demographic and financial profile, the bank could search for similar prospects and hit them with heavy promotions—raking in new customers, profits, and new data to fuel the information turbine.

By 1994, the fast-growing credit card operation overshadowed Signet’s banking activities. Wall Street analysts complained that the company’s muddy image was holding down the stock price. So in November 1994, Signet spun off Capital One, with a market cap of $1.1 billion and Fairbank as chairman and CEO and Morris as president and COO. (At the end of 2000, the market cap was $13 billion.)

Once on their own, Fairbank and Morris applied their information-based strategy to every corner of the business. Take customer retention—a backwater for most card companies. By 1997, thanks to Capital One and its imitators, partygoers were talking about credit cards—and the one who paid the higher rate would be on the phone the next Monday claiming he’d switch to another bank unless he got a better deal. To help figure out which card-hoppers were serious, Fairbank ordered a test. When a card-hopper called, the customer service agent’s computer randomly ordered one of three actions: Match the claimed offer, split the difference in rates or fees, or just say no. “We called a lot of bluffs, because many people didn’t really have a new offer,” Fairbank says. Thousands of accounts were upgraded or closed— at random—and the company gathered data on who left, who stayed, and how they behaved. Now, when a card-hopper calls, computers can calculate the odds that this particular customer actually does have a better offer—and the lifetime profits that would be lost if she leaves. Within hundredths of a second, the operator gets a script spelling out the precise terms he can offer—or orders to bid the card-hopper a cheery farewell.

In Capital One’s lexicon, that experiment was a “scientific test.” While it involved thousands of accounts, it was just one of 14,000 tests that Capital One’s busy analysts and marketers conducted in 1997. In 2000, the company carried out 45,000 tests. Anyone in the company can propose a test, and if the results are promising, Capital One rushes the new product or approach into use immediately. “We don’t hesitate, because our testing has already told us what will work,” Fairbank says.

If all this profiling and predicting gives you the creeps, you’re not alone. Capital One executives insist that the company strictly safe-guards all the data it keeps on customers, does not share account numbers or credit data, and allows cardholders to block new solicitations. After all, information is the company’s greatest asset. And while other card issuers have run into scandals over misleading terms or unfair fees, “Capital One is the one company I know of that hasn’t been subject to a lawsuit over fees,” says analyst Berry. Executives say that Fairbank has personally vetoed card offers that could mislead customers.

Testing isn’t limited to products: The employees at Capital One undergo extensive psychological and personality tests, including one 4½ hour block of tests administered to 1,600 associates who are picked at random. The results are compared to performance measures to figure out which factors predict success in particular jobs. Hiring for call-center and sales jobs is almost totally test based and takes just five hours—down from 20 days. Potential hires for professional jobs must analyze business school cases and undergo batteries of interviews with executives from all parts of the operation.

EVER SINCE HIS OPERATION HIT 500 PEOPLE, Fairbank has fretted about corporate stodginess. So in 1997, he proposed a plan to boost entrepreneurship: All employees get stock options, and any manager could trade in other compensation for an increased options stake. To lead the way, Fairbank and Morris swapped all their compensation—salaries in 1997 of $676,250 and $497,083, respectively—for options that vest only if the company meets steep goals. To collect on their 1999 grant, for example, they’ve got to boost Capital One’s split-adjusted stock price from $56 in mid-1999 to $100 by June 2002. If they succeed, Fairbank can reap $49.2 million and Morris $32.8 million. But they have a ways to go—in mid-February, the stock was trading at $61. With their incentive plans, “they’ve got everybody in the company pulling together for the shareholders,” says Kimsey, the director.

“Our goal has always been to operate like a consulting firm—lots of very smart people throwing out ideas, then testing them against reality,” Fairbank says. To get those smart people, top executives spend 20 percent of their time recruiting. “At most companies, people spend 2 percent of their time recruiting and 75 percent managing their recruiting mistakes,” Fairbank says. By making the rounds of top business schools—including Fairbank’s annual visit to the GSB—Capital One’s execs “make sure we’re always offering challenges and an environment that appeals to the best and the brightest,” Fairbank says.

The intellectual ferment starts at the top. Although Fairbank is clearly the senior partner, he and Morris share power on almost every front. They use what they call the “major/minor principle.” Morris explains: “I major in the international operations because that’s a passion of mine, and Rich minors in those. I have the main responsibilities there, but we make the decisions together. And we’re always in touch with each other’s businesses.” From the beginning, Morris says, the duo wanted to build “a place where people share a spirit of openness, intellectual honesty, and a fact-based approach to problems.”

Capital One’s reliance on data over gut feelings has served it—but it also has cost the company some opportunities. Capital One was the first U.S. card issuer to enter the British market—but “while they tested this and tested that, [rival] MBNA came in and put its foot to the floor,” says analyst Berry. “MBNA now has a larger share of that market.”

But Fairbank and Morris say thorough testing has saved the company from expensive mistakes. Take their approach to the Internet. When other card companies were paying top dollar to lock up prime advertising slots on Internet portals in 1999, Capital One was almost nowhere to be seen. The Web’s model—customers find you—flies in the face of Capital One’s targeted marketing strategy. Worse, early research showed that people who sought cards via the Web were worse credit risks than mail applicants—and 10 times as likely to commit fraud. So Capital One launched intensive tests to find key factors, such as the clickstream that brings a customer to its Web site, that help identify fraudsters, computer-generated applications, and poor risks. It also refined its pitches and tested ad strategies: “When everyone else was just buying up space, we had to convince portals to let us test ads on Tuesday against ads on Sunday,” Morris says.

The payoff: By mid-2000, with the dot-com world imploding and online ad costs plummeting, Capital One was ready to ramp up. By the end of the year, it was the largest Internet advertiser, and in November and December, capitalone.com was the most-visited financial Web site. Online account servicing—for which 2.5 million customers have signed up—offers even greater rewards. Those customers log on 2.7 times a month—nine times as frequently as offline customers contact the company, creating nine times as many opportunities to pitch targeted offers for new cards or other products.

Now Capital One sees two great advantages on the Internet. First, credit cards are the currency of the Web, promising to eclipse all other forms of payment—cash, checks, and fund transfers—as more and more commerce flows onto the Net. Just as important, the Internet fuels the data machine—and the direct marketing revolution that Fairbank hopes to ride far into the future. “We’ve hitched ourselves to an enormous macro trend, and I don’t see any end to it,” he says. Knowing Fairbank, he’s got the data to prove it.