Failed Strategies

When Tom Watson Jr. was chief executive of IBM in the early 1960s, he summoned an executive to his office after the man lost $10 million in a venture. Watson asked the man, "Do you know why I called you here?" Knowing of Watson's legendary temper, the man replied: "I assume you're going to fire me."

"Fire you?" Watson asked. "I spent $10 million educating you. I just want to be sure you learned the right lessons."

There are plenty of educational business failures out there, but executives seldom learn the lessons to be had. That's because the tendency in business writing is to look at success stories and say, "Here's how to be like those guys." Almost never do people take a systematic look at failures and say, "Here's how not to be like those guys." Only looking at success stories is like interviewing winners at the roulette wheel and assuming that their strategies will help you win, too. To get a true picture of your odds of success, you also have to interview losers.

To fill in the gap in business literature, we spent nearly two years with a team of researchers investigating the 750 biggest business failures of the past 25 years. We found several patterns that can help chief executives avoid repeating others' mistakes.

The broadest insight is that almost half (46 percent) of the catastrophes stemmed from misguided strategies not from bad luck or from poor execution, even though conventional wisdom is that good execution is the key to business success. That percentage was more than double what we hypothesized going into the study.

More specifically, the research found that seven types of strategies are most commonly associated with failure. So, anyone pursuing one of these seven strategies should take a hard second look. It's not that these strategies never succeed. They can and do. It's just that they lead to failure often enough that executives should look for the warning signs.

The research also found numerous methods for preventing problems. The biggest was to introduce disagreement into the strategy-setting process in formal ways, so that all potential problems are brought to the surface and evaluated rather than glossed over as executives try to get others to buy into a plan. 

Seven suspect strategies:  

• Synergy. It's tempting to think that one plus one equals three, but sometimes one plus one equals one. Sears got tripped up in the 1980s when it bought the Coldwell Banker real-estate services business and the Dean Witter stock brokerage. The theory was that customers who came to Sears for power tools would also buy financial tools. Some called the strategy "socks and stocks". But customers select realtors and stock brokers using criteria other than whether they're sitting in the middle of a department store. Sears sold both businesses after frittering away the management attention that should have been focused on the looming onslaught named Wal-Mart. 
• Financial engineering. Sometimes, companies get too clever and come up with a strategy that works in the short term but can't be sustained. Green Tree Financial led a surge in lending for mobile homes in the 1990s by offering 30-year mortgages versus the standard 15-year loan. But mobile homes can have a useful life of just 10 to 15 years. After three years, a $50,000 home could be worth $25,000 yet have $49,000 still owed on it. A wave of defaults not only took down Green Tree but also bankrupted Conseco, which bought it for $7.6 billion in 1998. 
• Rollups. People sometimes decide they can buy hundreds of small businesses and operate them efficiently as one national business, but find the complexity overwhelming. It was hard to find examples of successful rollups but easy to find failures. Loewen Funeral Homes bought more than 1,100 funeral homes and 400 cemeteries in North America over 12 years but found few efficiencies and amassed so much debt that the company went into bankruptcy proceedings in 1999. Many rollups devolved into fraud as executives tried to hide the problems. 
• Staying the course. Companies sometimes decide they're focusing on their core business when they're actually ignoring a threat. Kodak was well aware of digital photography as early as 1981, when the company conducted a detailed study. But the company was addicted to the 60 percent-plus profit margins on its traditional film, chemicals and paper businesses, so it viewed digital technology as something that could enhance the existing business, not as an all-out assault. (The New York Times reported that, when a Kodak researcher invented a key piece of digital technology, management's reaction was, "That's cute but don't tell anyone about it.") While competitors Agfa and Fuji acted decisively and well, Kodak lost 75 percent of its market value in the past decade and shed more than two-thirds of its workforce. 
• Moving into an adjacent market. While companies can benefit greatly from stretching themselves, as GE did under Jack Welch, they sometimes overestimate the value of what they bring to the new market and underestimate its complexity. Laidlaw thought its expertise in operating school buses would transfer to a different form of transportation: ambulances. What Laidlaw found was that ambulances were actually a medical services business, subject to regulation and contractual issues that Laidlaw hadn't faced. Laidlaw took a $1.8 billion write-off in 1998. 
• Using technology. Although information technology can lead to highly profitable strategies, executives often fumble the technology, perhaps by underestimating how much competing technologies will improve. Federal Express made this mistake when it introduced Zapmail in 1984. Zapmail picked up documents at a customer's office, faxed them to a FedEx office near the delivery point, then had a courier drop them off less than two hours after being picked up. FedEx forecast that Zapmail would generate $1.3 billion in annual revenue, or a third of its total, in 1988. Instead, fax machines improved in quality and dropped in price so fast that customers didn't need FedEx. Zapmail lost almost $700 million in its two years of operation and was shut down in 1986. 
• Consolidation. It's an axiom that as industries mature, they consolidate. But executives sometimes focus on the assets they will get by purchasing big competitors and underestimate the problems they're buying. Ames Department Stores actually went into bankruptcy proceedings twice because of problems with acquisitions. Its founders had the idea for discount department stores four years before Sam Walton, and Ames became one of the four biggest chains in the country before liquidating in 2002.

To test whether one of these seven types of strategies or any other is going astray, it's necessary to introduce disagreement into the strategy process. That's because the natural tendency of both individuals and institutions is to settle on a probable plan quickly and then try to build support for it, glossing over problems. There are numerous, relatively simple, ways of seeking disagreement. People can be encouraged to be informal devil's advocates. Anonymous surveys can be taken to see what people really think. Channels can be created that let objections be sent to top executives. Prediction markets can be set up.

All the informal approaches have limitations, however. A self-appointed devil's advocate, for example, may hold back for fear of offending the chief executive. Lyndon Johnson encouraged devil's advocacy in meetings on the Vietnam War, but did so just to assure himself he was considering all options; he made clear that no real objections would be tolerated.

So, any major strategy should be subjected to a formal devil's advocate review, which uses debate to get all important assumptions and objections on the table, where they can be considered objectively.

The design of a review is key. Otherwise, institutional defense mechanisms are too powerful as one executive reminded us, "When someone wins an argument, that means someone else lost." To avoid internal antibodies, the review needs to have a clear and limited charter that is to raise questions about a proposed strategy but not to come up with a better one. Letting the review propose an alternative would set it up as being smarter than those who came up with the strategy, which would be much too contentious. In addition, the review needs to be led by someone with credibility who doesn't have a stake in the strategy, whether that's a senior executive, a director or an outsider.

Because the CEO is typically the one who commissions a review, he controls the process and can use it as an adviser. That can be a crucial asset, given that quality advisers are so hard to come by seeking too much advice from senior executives or directors can signal weakness, while investment bankers and many consultants have financial incentives to recommend action, even if it's inappropriate. In our experience, CEOs also use reviews to build support among the executive team and the board allowing serious questions to be raised and then addressed in an open manner provides a mechanism for building true consensus.

History, reaching all the way back to the ancient Persians, shows how well reviews can work. The Persians made decisions twice, once while drunk and once sober. If the decisions matched, the Persians proceeded and they ruled the world for 300 years.

While executives aren't drunk as they set strategy, the emotion and social interaction that are involved can put businesses in a delicate state. The sober evaluation provided by a devil's advocate review can let executives spot any problems before it's too late.