In "What Stays in Vegas", Adam Tanner uses Caesars as a case study of how a business can make use of what has become known as Big Data—the analysis of vast amounts of quantitative information in search of useful patterns. The title is unfortunate, because "What Stays in Vegas" has little to do with gambling and even less to do with Vegas: The book is about how corporate America amasses and uses information about its customers. Mr. Tanner's findings, based on interviews and, in some cases, on Internet detective work, are unpleasant, but don't bother being alarmed. It's too late for that. Las Vegas, he writes, is less a sin city than "a vast data collection machine."
At the center of Mr. Tanner's narrative is Gary Loveman, a former Harvard Business School professor. In the late 1990s, Mr. Loveman took on a part-time consulting gig training employees of what was then Harrah's Corp. in customer satisfaction. Shocked by the company's lack of sophistication, he suggested to Phil Satre, then the company's chief executive, that Harrah's use data it was already collecting to build customer loyalty. Mr. Satre responded by making Mr. Loveman his chief operating officer, a heady position for a young academic who had never run much of anything.
Mr. Loveman set to work, not necessarily to his loyal customers' benefit. In an elevator at Harrah's in Las Vegas, he met gamblers complaining that the slot machines were too "tight," paying off less than those at Harrah's in Atlantic City. Mr. Loveman knew that the opposite was true, that the company kept seven cents of every dollar pumped into the slots in Atlantic City but only a nickel in Vegas. From this chance conversation came the sort of brainstorm by which fortunes are made: If customers don't know the odds, they probably won't know when the odds worsen. Today, Caesars Entertainment keeps 8% of its slot machine take in Las Vegas instead of 5%. Those three extra cents on the dollar are pure profit. The gamblers don't seem to have noticed.
At the center of Caesars's data-collection effort is Total Rewards. Loyalty programs with rewards for repeat customers go back at least to the 1880s, when the Great Atlantic & Pacific Tea Co. gave buyers coupons that could be exchanged for clocks or tableware displayed in its stores. Total Rewards, which began in a rudimentary form in 1997, is a program of a different order. The member offers up his number each time he sits down at a poker table or eats in a restaurant. The details—you spent three hours playing blackjack, never bet more than $50 on a hand and lost $750 in an evening—end up in Caesars's computers, which crunch them to identify useful patterns. Your reward, at least in theory, is that Caesars will market to you in ways it expects will please you, whether that means having the manager come offer a personal hello when you're at the roulette wheel or sending you a coupon for a free dinner at the sushi bar, where you dine every time you visit. Behind the scenes, computers are evaluating which rewards are likely to make you want to spend more money. As Mr. Loveman explains: "We should be able to give you things that you care about—not have you littered with things you don't care about—and have it work out profitably for us."
Customer relations by algorithm represented a revolution in the casino business. The savvy manager whose instincts led him to offer a free cocktail to a big bettor has been replaced by a computer that reckons that the small bettor who comes every Thursday night is actually more profitable to the casino.
Why does it work? The story of Dan Kostel, a salesman at a Los Angeles asset-management firm, sheds light on that question. Mr. Kostel loves playing blackjack in Las Vegas. He also thinks that Caesars Palace is a bit stodgy. But a few months after he spent an evening there, he received a letter offering a free room and $1,000 in chips on his next visit. The freebies brought him back. Once the computers identified him as a regular, the offers diminished. So Mr. Kostel learned the game. He played elsewhere for a few months, and Caesars Palace upped the offers. He checked into his free room at Caesars even when he was staying in a free room elsewhere, because he would receive more credit toward future rewards if Caesars thought he was staying there while gambling in the hotel's casino. As Mr. Tanner observed, "for Kostel, winning comps was part of the overall game." Of course, Caesars knows that if it has evaluated Mr. Kostel's behavior correctly, it will win in the end.
Not all data collection is so benign. Casino operators collect information about their customers from many other sources beyond loyalty programs; how deeply they probe Facebook FB -0.56% profiles and divorce-court records depends on the operator. Mr. Tanner explores an obscure company called Global Cash Access, which specializes in operating automatic teller machines and cash desks at casinos. If you use its services, it may (for a fee) tell the casino how much cash you withdrew there last month and how much you withdrew at other casinos. This is golden information for a marketer, but gamblers who use the teller machines may not understand that their transactions are far from private.
Mr. Tanner's engaging book is realistic; he knows that this particular genie cannot be stuffed back in the magic lamp. At the same time, he shows how harmful it is when private companies compile electronic dossiers on their clients. Data collectors, he writes, "should be clear about what they are doing, and customers should have a choice about the extent to which they participate." It's a sensible response. But, as "What Stays in Vegas" shows, the collection of personal data is now so widespread that the choice has already been made for us.