When Gary Loveman arrived at the headquarters of Harrah's Entertainment, in 1998, as chief operating officer, most of its employees weren't prepared to wager that the Harvard Business School professor had what it takes to succeed in the gaming business. "Most people thought I'd leave in two years and go back to Harvard," Loveman recalls. "They thought this would be like a kidney stone: it would hurt for a while and then it would pass."

The skeptics stopped taking bets long ago. In five years, Loveman has reinvigorated the company—largely by developing a successful gaming-business strategy rooted in smart retailing. In the process, he led a transformation of his company's casino culture, created a national brand unique to its sector, and doubled revenues and earnings.

Loveman acknowledges that such results were hardly a sure thing when he arrived. A decade ago, as a handful of states began to liberalize their gambling regulations, Harrah's seemed to be on a roll: it quickly expanded outside Nevada, opening casinos and riverboat-gambling venues in several Midwestern and Eastern locales. But as the opening of new jurisdictions ground to a halt, competitors noticed the profits Harrah's was making outside its traditional Nevada and Atlantic City markets. They soon entered the arena with more attractive casinos, hotels, and amenities.

With most of Harrah's markets under increasing competition, the company began to struggle. Loveman—then a consultant to Harrah's—and its then-CEO, Philip Satre, began developing a bold new corporate strategy based on customer loyalty. For Harrah's, the strategy was born in part of necessity, since the company had neither the time nor the money to "upgrade its merchandise," recalls Loveman. Harrah's therefore had to persuade gamblers to spend less at competitors and more at Harrah's. Satre asked Loveman to join the company and bring this customer loyalty strategy to life.

Loveman took action on four major fronts. First, he overhauled Harrah's marketing, replacing the industry veterans with customer-relationship-management "rocket scientists." Now, Harrah's uses a loyalty card program and sophisticated technology to collect and analyze data from customer interactions. The insights thus obtained guide every marketing plan and investment decision. Second, Loveman raised the bar in one particular way: despite Harrah's widespread recognition as the best casino operator for customer service, he persuaded his property general managers that it could—and had to be—improved. Third, leveraging Harrah's unique national distribution network, Loveman invested heavily in building the Harrah's brand. Finally, he shook up Harrah's culture with a new human-resources approach that valued brainpower and leadership over industry experience.

Loveman became Harrah's CEO on January 1, 2003. Chatting with visitors in his office, near the Las Vegas strip, he freely mixes Harvard erudition with the blunt talk of a gambling-floor boss. In this interview, conducted soon after he moved into the CEO's suite, he shares with McKinsey's David Becker his thoughts on implementing Harrah's turnaround strategy, the pivotal role of marketing, and the company's future.

The Quarterly: What was the strategy you helped to develop and were hired to implement?

Gary Loveman: The gamble was to make the existing businesses grow without access to new casino development. This was a retail strategy—a loyalty-oriented, same-store-sales-growth strategy—which had no precedent in the casino business. We said that you don't primarily need more gaming; you need more loyalty among players who already know you. That meant orienting the company toward influencing consumer choice. So this whole strategy was about modifying consumer behavior. We set about building a bunch of self-reinforcing mechanisms that enveloped players, as we call customers, with reasons to be more loyal. Part of this effort was to create a brand they could be attached to, and that required a significant improvement in service quality. And part of it was using relationship-marketing tools that constantly try to develop closer and more valuable interactions with players.

The Quarterly: What does a gambling brand look like?

Gary Loveman: I've always been fascinated by the power of brands to influence consumer decisions. The gaming business is a service that provides deep enjoyment. People are very caught up in gaming. I mean, it's every bit as personally rewarding as fragrances, fashion, automobiles, resort destinations—you name it. It's an experience that cries out for a brand, yet nobody was stepping up to that. Caesars had done so in the past with its Vegas operation, but nobody else had. We could step up by creating a national gaming brand, and no one else could. We operate 26 casinos in 13 states, and with a brand we could influence players to visit a casino close to them or to come visit us anywhere, coast to coast. We wanted a brand that really inspired customer loyalty.

We also had the opportunity to create a powerful brand because we focus on just one thing: a great gaming experience. We are not primarily for families or for destination getaways. We're a gambling joint. We're there for people who want to gamble, and that's where we wanted to center the brand. The economics are very transparent. So we're just for gamblers. Our property-positioning statements are centered specifically on gaming and the thrilling, exciting experiences our customers describe while visiting our casinos. We call this positioning "exuberantly alive" because it focuses on the customers' raw energy and enthusiasm for this form of entertainment. All of our brand advertising clearly illustrates this feeling and resonates extremely well with gamblers. It's all about gaming.

The Quarterly: Harrah's had a rewards program in place when you joined. Did it help build the brand and customer loyalty?

Gary Loveman: The Total Rewards program was the first thing I worked on when I got here. It was a customer-recognition rewards program then, not a loyalty program. It did not have loyalty incentives. If you were a $500 customer, we would give you $100 in goodies—discounts on rooms, free meals, and so on—every time you came in. But if you didn't come for a year and meantime visited competing casinos, still, the next time you came in again we treated you like a $500 guy and gave you $100 in goodies. We changed the program by building in loyalty incentives so that as a customer, each time you think about visiting a casino, you end up visiting Harrah's because it's better for you. Or to put it in the negative, if you don't visit Harrah's you lose something. So there's this clear pecuniary and nonpecuniary signal that influences decision making.

Total Rewards is now a three-tiered loyalty program, with gold, platinum, and diamond levels. You can consolidate all your gaming with us in any of our casinos. The more of your gaming you give us the bigger the rewards, and you go up the tiers. The results of that program have been stunning. People who are close to a gate—from gold to platinum, from platinum to diamond—aspire to get over those gates by consolidating their business with us. They know that if they make ten visits to Atlantic City—and they used to give us, say, three of those visits—they know that to get a Diamond Card they have got to give us six visits, and they do. If you look at our same-store-sales growth and our overall revenue growth, it is disproportionately among those who have advanced through the tiers and consolidated their business with us. It's exactly what we set out to do.

We also collect a tremendous amount of information on what players do with us. We know when you arrive at a casino, what you do there, and when you leave. We have information on 26 million customers. And we measure everything. Having studied a lot of consumer businesses, the only one I know that comes close to measuring as much as we do is Wal-Mart. Testing and measuring is important to us. When our employees use the words "I think," the hair stands up on the back of my neck. We have the capacity to know rather than guess at something because we collect so much information about our customers. And if we say, "I know," let's really make sure we know.

The Quarterly: What critical steps did you take to implement the overall strategy?

Gary Loveman: Well, the number-one, most important thing in this whole process was this: Phil Satre brought me in as the chief operating officer, not the chief marketing officer. Chief marketing officers have ideas, and they try to sell them to the guys with line authority. They spend all their lives selling. Most of their work is not done against competitors; it's done internally, trying to get the P&L people to let them do things. That's a losing battle. And if you're trying to generate change in an operating business, it's almost a doomed exercise. As chief operating officer, I had the authority to say, "Here's how we're going to engage customers." We also proved to the property general managers that our approach would work, starting with the early experiments we ran in December 1998. As operators, property general managers are greedy buggers. If something works and they can make more money, they'll get on it at some point. But having the authority to mandate this made all the difference.

The other really important thing we did was to replace the existing marketing people. We took out practically the entire corporate marketing department. They were never going to get our program where it needed to go. They were never going to build the decision tools or be able to plot out the mathematics of this program the way we needed. So we brought in the kind of people we have now, who have the horsepower to do this kind of work.

'I made it clear when I came that we're going to be something other than a holding entity' for operators

Architecturally, we did a couple of other critical things. The corporate office was not adequately serving the operating businesses. It was isolated, had a lot of ineffective people in it, and the property general managers hated it. So I made it clear when I came that we're going to be something other than a holding entity for operating businesses. I tried to make that point tangible by holding all of my meetings in the field to make it clear that we knew where the customers were and where the money was. We weren't confused about that.

We really tried to change the whole gestalt around the corporate office, and we implemented some things that helped. For example, to raise money to support the brand-advertising campaign, I instituted a 1 percent brand tax on the properties. Now, if you were a McDonald's franchise you'd have to give 9 percent, so lucky you, you're only giving 1 percent. We're going to put it in the budget, and Richard Mirman, our chief marketing officer, will figure out what to do with it.

The Quarterly: How else did you increase the role of corporate marketing?

Gary Loveman: We created a marketing council. I chaired it as the senior operator and the senior marketer, and it brought together three of the four senior corporate marketing people and the four senior field marketing people as well as our outside agencies, our PR agencies, and our senior technology person, because so much of our marketing runs through technology systems. In the early stages, the marketing council met about every other month. It now meets quarterly and has continued to meet every quarter in the five years that we've been doing this. It's an effort to make sure that all of our marketing work is a collaboration between corporate and the field and that everybody owns these decisions.

Also, I created a position—divisional vice president of marketing—for each of our three operating divisions. There had never been such a position before. And to those three people I said, "You're the head marketer for your division. If I see anything going on in this division that's badly marketed or that doesn't use our marketing tools adequately, you're responsible." The division presidents understand what the senior marketing people are responsible for.

For example, we don't really market our business much through food. So if I'm in a property in the Midwest and I see that our advertising position is around food, I'm going to want to know what the hell is going on, because that's not our gig. That has worked extremely well. The division presidents have come to count on these divisional vice presidents of marketing as probably their single most important subordinates—more so, in some instances, than their property general managers.

We've used this approach in other ways. We've built a number of capabilities, many of them highly mathematical, like our hotel yield-management systems and our slot-pricing systems. We get field and corporate people together with outside experts, build the stuff, and then make it mandatory throughout the company. The notion is that we've got to be better in every one of our casinos because we operate so many of them. So with the way we price the slot machines, the way we yield the hotel rooms, the way we procure things, we ought to be better than the local casinos we compete against in many of our markets, because we ought to be a lot smarter.

'If they weren’t using the right tools to figure it out or weren’t using the tools effectively, they were dead'

I routinely enforced using the tools. I would come into an operating meeting at a property and say, "OK, what's wrong with segment 452 here in this city?" And if they weren't using the right tools to figure it out or weren't using the tools effectively, they were dead.

The Quarterly: You said that improved customer service was the third element in the new strategy, after branding and marketing. How did you change customer service?

Gary Loveman: Casino service generally is disappointing all around. Service is hard to deliver in a casino. Employees are under strict rules to ensure there is no corruption. For example, dealers might want to give you a hug but can't, because you might slip something into their pockets. It's not like a hotel. So the business had always grown up around control. Service came way down the list, after control. Customers are losing. They're tired. It's a complex service-delivery process. So there are a lot of things that can get in the way of delivering good service.

But we had to convince our employees that we really cared about customer service. We were convinced that customers would be more loyal if we provided better service and that, if we got just a little bit more share of wallet from loyal players, we would be massively more profitable.

One thing we did was change the compensation system for our general managers. A quarter of their bonuses depends on customer satisfaction results. But we also had to train employees, so we developed a service curriculum, which came out of research with our best customers on the issues that really motivated their loyalty. For the first time in the company's history, every single employee attended this training. We paid them their tipped wages while they were in training; dealers make most of their money on tips. If you put them in training and only give them their hourly time, they get the message: this isn't really important. If you pay them their tipped wages—for the first time in most of these people's lives they were in training at tipped wages—that is a huge deal. And we ran all these programs 24 hours a day because we had people working all shifts. At the Rio, in Las Vegas, for example, we ran 200 sessions with 20 people in each to get through 4,000 employees in just five months. At the end of the program, you had to pass a test—otherwise you could not keep your job. You can imagine the anxiety that percolated through the system.

Coupled with this, beginning in 1999 we started paying out a bonus to every nonmanagement employee of the casino if his or her property improved its customer service scores by 3 percent over the same period a year earlier. And as long as the property is at 80 percent of its operating-income plan or higher, everyone gets a bonus. We've paid out $40 million in bonuses to employees across the system—anywhere from $75 to $300 each, each quarter. Again, if I hadn't been chief operating officer, this would have been a dead program. But it's a big deal to employees. In the employee areas there are graphs to let them know their service numbers, which are based on customer satisfaction surveys. The data come in each week, and employees check to see how they're doing.

The Quarterly: How do you link it back to customers?

Gary Loveman: We can track the customers who fill out surveys. We can track their gaming behavior, so we can assess whether a player who rates us better this year than last year also plays more. And the answer is remarkably positive. The people who get happier with our service play much more with us, and the people who become unhappy play much less with us. Market by market, where our profitability and revenues greatly exceed our relative market position, there's no question but that the results are largely service driven.

We also worked on reducing employee turnover. We're very careful about who we hire and are doing a better job of nurturing people through their first 90 days with us because that's where we've been losing everybody. And a lot of it comes from making sure employees know what they're being hired into. We take people through what we call realistic job previews and get them acquainted with the work before they start. And we check in with employees the first week they're on the job, the second week, the fourth week, the eighth week. We work very hard on supervision reviews and so on. We've managed to reduce turnover quite a lot, which in turn helps our customer service scores.

The Quarterly: New strategy, new people, new operations. A new culture at Harrah's?

Gary Loveman: I think the single biggest cultural change has come by instilling a meritocracy. I'm very pleased with where we've come on this. Once, people were considered adequate if they were meeting the minimum standards of the job. I do talent reviews annually with all of our operating and corporate people, and, frankly, I say to general managers, "Do we really want you to spend your time trying to make a barely OK person successful? Why wouldn't we rather have a case where employees are so good that they are putting some heat on you and teaching you some things and pushing you around a little bit and making you better?" If you're not working with people who make you better, something is wrong. I wanted to instill the notion that jobs didn't belong to people; jobs belong to a company. It was the company's responsibility to get the most capable person it could find into the job.

'I had a lot of bloody battles where I'd have to tell a general manager to fire this person tomorrow'

People had a very, very hard time with this. I had a lot of bloody battles where I'd have to tell a general manager to fire this person tomorrow or I would do it myself. It was ugly. I mean, for a long time there was a lot of antipathy among a lot of people, and it continues to some degree to this day. It was a big change in the history of Harrah's. People say this used to be a safe, family company, and now that damned professor has turned this into a place where nobody can feel safe. And there's an element of truth to that, because it is results that make any of us safe. If we don't produce, the shareholders and the other employees deserve better. The result of this meritocratic approach has been that we have assembled the best management team in the industry. And management talent has been key to our success. We look for people who are very, very smart.

The Quarterly: Where have you stumbled?

Gary Loveman: My own biggest mistakes have been the failure to make management changes in time. When we bought the Rio, we got lulled into two problems: staying in some businesses we shouldn't have been in and keeping the existing management team. Those were tightly connected blunders. But for a while after we bought the Rio, everything looked great. The tide was high and everyone was swimming. Then in early 2000, baby, the tide went out and we saw that we had a lot of naked people swimming in there. We got creamed at the Rio in 2000. The people in there weren't doing what we asked them to do. They were dragging their feet on making changes because they thought what they were doing was just fine. Since the moment we replaced them with people who did what needed to be done, it's been a steady climb up for the Rio. We were trying to finesse the problem—to make these businesses part of a larger entity and, at the same time, not a part. It didn't work.

The Quarterly: What about the route your competitors pursue—the big hotel-casinos with lots of show business spectacles?

Gary Loveman: We make most of our money in markets outside Vegas, and there isn't much spectacle in those markets. Vegas is a whole other beast. To give you a sense of the parameters, we'll cash-flow just under $100 million per year at the Rio here in Vegas. We'll cash-flow $180 million at Harrah's in Atlantic City on an investment of half as much. People just gamble in Atlantic City. People come to Vegas for lots of reasons. My competitors use spectacle to drive visitation in what is increasingly becoming a hotel business.

We estimate that Harrah's Las Vegas has one of the highest wins per slot machine on the strip even though we have half the number of rooms that MGM has sitting on top of its machines. And the reason is that we fill our rooms with casino gamblers. That's the only reason to be there; there's not much to look at: no fountains, no battleship show going off every hour, no circus. On the other hand, MGM mainly operates in Las Vegas, Detroit, southern Mississippi, and that's it. They don't have national distribution. So, for example, their marketing approach isn't focused on building relationships and loyalty with gaming customers in Chicago as a way to grow their Vegas business—that's not a strategy for them. Their business is much more oriented toward travel, a destination business. It's not the $100 slot players in St. Louis. That's our gig. I like it because there's a sense of strategic complementarity. We're not all beating on the same issue, as you'll find in some industries.

The Quarterly: What's next for Harrah's?

Gary Loveman: I think this business will become a consumer distribution business, and I want us to be the leaders of that process. We're the only casino company currently that will offer you gaming in a land-based traditional casino, a riverboat, a Native American reservation, and a "racino"—a race track with a casino. Many of the Northeastern jurisdictions, like Pennsylvania and New York, seem to be going the way of racinos. Some of my large-cap competitors are not in that business, but I love that business.

Our research shows that casino customers really like to pursue "safe risks." They like uncertainty and the excitement of being on the edge. They aren't for over-the-top risks, however: for example, they like reading mysteries and watching Survivor, not bungee jumping. The recent success of reality-TV shows demonstrates that there is broad pent-up demand for the type of entertainment we offer. But unlike other forms of entertainment, gaming is hard to buy. It's legal only in a limited number of states, and even there supply is constrained. I want to create a world where Harrah's competes with other forms of entertainment to satisfy these desires.

For example, I want to have gambling on cell phones and interactive TV. I want to distribute the ability for consumers to take safe risks. It's nothing but a distribution argument, and it's not specific to casinos. Financial-services companies are putting teller machines in grocery stores and giving you the ability to debit your account at the gas pump. Dunkin' Donuts is selling doughnuts at the gas station, transforming it into a coffee shop. There's the Starbucks revolution. You name it. This industry—my industry—has just not tapped into that. I mean, it's ripe for that. I know the future lies somewhere in that direction.

We're going to have to press the regulatory frontiers a bit and try to be innovative. Right now, the biggest challenges we face to expansion are mostly political. There is opposition to gaming's expansion. Gradually, as the country's attitude toward gaming matures, that opposition is falling down. I think big growth from the liberalization of traditional gaming restrictions will occur in states like Massachusetts and Rhode Island, Pennsylvania and Maryland. Internationally, it will take longer.

The Quarterly: How has your style changed now that you're the CEO?

Gary Loveman: My job is largely externally oriented now. It involves articulating the strategy, developing growth alternatives for the business, engaging with the political wing of our world, with potential business partners, with trade unions. I'll continue to be the voice for the critical internal strategies with employees. But last year, if we weren't getting enough local 55- to 70-year-old women to St. Louis on Fridays, I would have been in St. Louis trying to figure out what the hell's wrong. That's primarily someone else's job now.