Going Out of Business Tales

Learning from inexcusable business failures.

Paul B. Carroll and Chunka Mui have impeccable timing. When the pair began work on “Billion Dollar Lessons” (Portfolio) three years ago, the American economy was relatively healthy and catastrophic business failures were few and far between. But in the three months since the release of their new book the economy has been in meltdown mode and “bailout” has become, quite literally, “Word of the Year.” Never has a detailed examination of business failures seemed more apropos. 

In “Billion Dollar Lessons,” Carroll (author of “Big Blues: The Unmaking of IBM”) and Mui (co-author of “Unleashing the Killer App”) review what they have identified as the most “inexcusable” business failures of the past 25 years, identifying patterns and providing readers with the intellectual tools to avoid repeating the same mistakes. It’s the kind of after-action analysis that is often given lip-service, but rarely receives serious consideration. With this in mind, Failure interviewed Carroll about mistakes of business past, and how we can learn from them.

Why did you decide to focus on failure when most business books focus on success?
That’s pretty much the reason. Lots of people draw good lessons from success, but there have to be lessons from failure as well. I’m also a bit of a contrarian, and when we started work on this book it just seemed like everything in the economy was too good to be true. And you know what? It was. 

The subhead of the book refers to “inexcusable mistakes.” How do you define an inexcusable mistake?
We tried to be pretty careful about that because anybody can be a Monday morning quarterback. We started with 2,500 of the biggest failures, which we defined as bankruptcies, big writeoffs and discontinued operations. To make sure we had a representative list by industry and timeframe, we cut it to 750 using various screens. Finally, we had a research team of almost 20 people who spent a year-and-a-half going through the data. 

We also tried to be careful to make sure it was possible to see at the time that an idea was a non-starter. For instance, when Blue Circle [Cement] decided it would make lawnmowers on the theory that cement is used in homes and homes have lawns … to me that’s a stupid idea. On the other hand, we didn’t ding the airlines for not understanding that 9/11 was going to happen. 

We started out thinking that maybe one out of five failures could have been foreseeable. In fact, almost half should have been foreseeable.

Why don’t businesses do a better job of learning from mistakes?
It’s partly psychological and partly institutional. The psychological part is that humans by nature are optimists. There are all kinds of studies that show how people tend to discount failures as bad luck or a product of unforeseeable circumstances. Whereas if something goes well people tend to say, “I’m a really smart guy.” 

There are also institutional issues. A CEO is going to make more money if he or she runs a bigger business. So CEO’s tend to want to buy companies even if it might be more appropriate to sell. Also, if you consider the way options are set up, they can’t be worth less than zero, but they can be worth a lot of money. Options encourage CEO’s to swing for the fences, when it might make more sense to be more cautious.

Are some types of businesses better or worse than others at learning from mistakes?
If you look at some of the great companies in history, like IBM in the Tom Watson Jr. era or GE in the Jack Welch [Jr.] era, you find cultures that fostered disagreement. At one point, I had the pleasure of meeting Watson. He was a terrific guy, but he loved a good fight. He talked about liking harsh, scratchy people, and he was a harsh, scratchy guy. In his era, there wasn’t any need to sugarcoat things for fear he would get mad. 

We found that companies that encourage disagreement are much more likely to avoid mistakes than companies where everybody is trying to take whatever idea the CEO has and make it work.

Can you explain your concept of “danger zones”? 
The idea is that there are seven types of strategies [synergy, financial engineering, rollups, staying the course, adjacencies, riding technology and consolidation] that are most commonly associated with failure. If you understand that these seven areas are potentially dangerous you can watch for the red flags we identified that let you know you may be pursuing a bad strategy.

With all the potential pitfalls it’s a wonder that anyone succeeds. Can business failures really be avoided?
I think so. And there is a developing literature on this. Robert F. Bruner has a book called “Deals From Hell” (Wiley) which shows that if you weed out the really bad mergers and acquisitions [M&A] you can take the statistic about the percentage that fail and turn it on its head and show that M&A can actually be pretty successful. 

Is there a company in the news right now that is making the same mistake as one of the companies you cover in the book?
The one that springs to mind is Bank of America (BofA). For the past couple of years we have been doing a blog that makes predictions about companies that seem to be making mistakes. A couple of months ago when BofA offered to buy Merrill Lynch we said that seemed to be an adjacent-to market mistake [see Chapter 5, “Misjudged Adjacencies”]. BofA, which knows retail banking very well, decided that qualified them to operate a retail brokerage, despite knowing almost nothing about retail brokerages. BofA also decided that it could somehow get into investment banking, even though that’s a totally different market. BofA bought too quickly; they agreed after about an hour of conversation to spend $50 billion. And, at the time, BofA was glossing over problems with the securities that Merrill Lynch owned that were backed by mortgages. Since then stories have started to appear about culture clashes between the BofA and Merrill Lynch folks, and it’s been said that BofA clearly overpaid.

One of the most interesting concepts you address in the book is confirmation bias. Why are people prone to ignoring uncomfortable facts?
Again, it’s both psychological and institutional. People have what psychologists call a confirmation bias. If you think you know the answer to something you tend to look for ways to confirm that you’re right. Even though, logically, the only thing you can do is test lots of ways that you might be wrong. 

And if you think about the way organizations work, even though strategy departments are supposed to vet ideas carefully, we all know what happens. Somebody in senior management has an idea and puts it out there. Then everybody below them tries to find a way to make it work. CEO’s tend to see themselves as powerful leaders, and as a result they don’t broker dissent. So the people below them are afraid to offer dissent. 

In the book you describe “The Devil’s Advocate” process. How can that help bring problems to the surface?
The Devil’s Advocate process is our way of making sure that people agree up front on a process that brings problems to light. The idea is to convene a panel led by somebody with no stake in the game. Through role playing and debate you get people at senior levels pitted against each other so that they argue in ways that make it clear what the assumptions are for strategy and where the potential pitfalls are. You make somebody an advocate for a potential strategy and then other people say, “Yeah, but that’s assuming the economy is going to do this.” Or “that’s assuming our competitors are going to do that.” Once you get all the assumptions on the table you see if anything looks like a bad bet and investigate further. So the basic idea is to have an outsider foster a discussion in a way that not only makes sure all the assumptions are brought to the surface where they can be tested, but does it in a way that doesn’t threaten the power of the CEO.

Are there any ways to incentivize people to learn from failure? 
That’s an interesting question. And the answer has to be yes. The question is how you do that. We think the Board [of directors] has a lot of responsibility to make sure that people get things right. And the CEO, under the direction of the Board, has to create an environment in which employees understand that they have to learn from failure. And if they make a new mistake rather than an old mistake they are certainly not going to be punished. 

There’s a great story from my days covering IBM. As I said, Tom Watson Jr. was this ferocious character. There was an IBM employee in the early 1960s who had run a business unit that lost 10 million dollars. Watson called him in to headquarters. The guy walked into Watson’s office kind of weak-kneed. Watson said, “Do you know why I called you here?” He said, “I assume you called me here to fire me.” Watson said, “Fire you? Hell, I just spent 10 million dollars educating you. I just wanted to be sure you learned the right lessons.”

Is it okay to make new mistakes? 
Absolutely. We’re all in favor of new mistakes. There is a deal between American Airlines, British Airways and Iberia that would combine their Atlantic operations without actually merging. It may be right or it may be wrong, but at least it’s different. And part of what is smart about it is that they are doing it in a way that limits the downside. 

In the course of your research did you find one particular cause of failure that was more common than any other?
When companies are making an acquisition or considering a move into a new market, they tend to focus too much on benefits and not enough on the potential failures. This often shows up in companies overestimating the benefits of scale and underestimating complexity. 

A year-and-a-half ago Oshkosh Corporation, which makes heavy-duty trucks, decided it was going to buy a company that made lifters [for lifting construction materials]. Their rationale was they were going to get increased economies of scale because they were going to buy a lot more steel. On the face of it, it was kind of silly because at the time they bought a quarter of one percent of the world’s steel production. Going to one-half of one percent wasn’t going to impress anybody. And in the meantime, we said [on our blog] this is very different equipment; the sales forces won’t mesh well and the manufacturing processes won’t mesh either. 

As it turns out, Oshkosh bought this company for $3 billion in cash and even before the market fell apart this summer the combined companies were worth a billion dollars. So the idea is that companies focus so much on benefits that they ignore the problems.

In light of recent economic developments, do you think you might want to change the title to “Trillion Dollar Lessons”?
[Laughs]. It’s funny you say that because we have said that. Time magazine, in a review, said we should be working on a sequel. 

In the course of your research, did anything surprise you? 
Probably the biggest surprise was how many failures could have been prevented. When I was at The Wall Street Journal we used to say, “There are no new stories, just new reporters.” In some ways, it seems like there are no new failures, just new companies and executives that make the same old mistakes.