Change the Dialogue with Management
In looking for approaches to head off embarrassing strategic errors, we found that this is the key.

If the blame game goes as it usually does, criticism about the mistakes that led to the current recession and to poor performance at so many companies won’t stop with the CEO. Aspersions will be tossed at boards, too. To see how this might look, read the withering complaints about the Lehman Brothers board that failed to stop the CEO from running the company out of existence.

While it’s obviously not possible to reset the clock and avoid mistakes that have already been made, it’s possible to learn from errors and protect yourself from future attacks. This is true even for companies and boards that aren’t having problems: You can learn from others’ mistakes.

To identify the lessons from business failures, we assembled a database of the 2,500 biggest disasters of the past quarter century (principally write-offs, bankruptcies and discontinued operations). We had a research team of 20 consultants torture the data for 18 months to make it yield up its secrets. We found that a fatally flawed strategy was at the root of the failure half the time, and identified numerous red flags that can alert companies that they’re heading down a dead end.

Interestingly, directors often saw problems coming but failed to prevent them. In those cases, the board’s interaction with the CEO took one of three forms:

• Because outside directors have less information than management does and generally have less experience in the industry in which the company operates, board members may censor themselves. Why take the chance of looking silly? At Samsung, directors had doubts about moving from electronics into car manufacturing in 1997 but deferred to the chairman, even though there was already too much capacity in the Korean auto market and Samsung had no particular strength in cars. Samsung spent $5 billion on the venture, which went into receivership and was sold three years later for some $700 million.

• Even when board members speak up, the CEO may outmaneuver them. At Ames Department Stores, two directors objected to the purchase of another discount department store chain in 1988, in a misguided attempt to match Wal-Mart’s scale. The CEO cut them out of conversations on the topic and had them go last in any voting, so their opinions wouldn’t influence others. Ames bought the chain — and soon filed for bankruptcy protection.

• Directors sometimes objected so strongly to a bad idea that they went public — and still failed to head off disaster. At Oglebay Norton, directors concurred with a decision to move away from the iron ore business and into limestone, but two objected strenuously when the CEO went on a buying spree and paid inflated prices for limestone operations. The two resigned in protest, but the CEO cowed the rest of the board, arguing that he hadn’t been hired to be a custodian. Oglebay Norton filed for bankruptcy protection in 2004, then was sold.

In looking for approaches to head off embarrassing strategic errors, we found that the key lies in the board’s ability to change the dialogue with management. Boards and management agreed ahead of time that major decisions would be discussed openly, with numerous alternatives on the table, before management committed to a strategy.

Although there are numerous ways to change the dialogue, one we recommend is a “devil’s advocate” review. To avoid bias, the review is led by someone who hasn’t been involved in forming the strategy and has no stake in the outcome. The devil’s advocate mission is to ensure that critical assumptions are made explicit and that key design decisions are examined from a fresh perspective.

One technique for doing this is to get senior executives to debate each other, asking the detailed, tough questions that directors may not be able to summon because they’ve been shielded from information. The devil’s advocate ensures that the debate is robust by, among other things, using role-playing: that way, people can raise issues without fear of hurting their careers. (Unlike what occurred at Samsung.) The advocate then lays out for the board all the key assumptions and stakeholders underlying a strategy, so directors can understand all the ways it could go wrong. Later, if the board decides to proceed with the strategy, the review panel does a deep dive into every aspect, making sure all potential problems are brought to the surface and discussed, rather than being glossed over. (As happened at Ames and Oglebay Norton.)

We’ve used the process to head off any number of bad strategies, and many successful companies do something similar. At Intel, the board discusses possible strategies with management before they get so far along that rejecting an idea becomes a bring-in-the-next-CEO proposition. It’s hard to argue with Intel’s results.

A New York Times piece on BofA focuses, as many articles do, on the clubby nature of many boardrooms–you serve on my board, I serve on yours, and we make nice with each other. But the issue is broader than that. Even if you have a truly independent board with lots of diverse viewpoints, the CEO will still carry the day almost every time. He has all the information and controls how it’s presented to the board. He makes the decisions, while the board pretty much is limited to deciding whether he gets to continue in the job or whether he should be replaced–a call that boards seldom make unless the situation is dire.

Whatever the makeup of the board, directors need to find a way to have a true dialog with the CEO. That means conversing broadly before the CEO has committed so fully to a strategy that a reversal would cause too much loss of face. That means getting information in less digested form, hearing the reasons not to pursue a strategy as well as the factors working in its favor.

In short, that means some sort of a devil’s advocate review, so that boards understand all the assumptions that go into a strategy and can evaluate all the ways things might go wrong. We describe such reviews in great detail elsewhere on this site, so we won’t repeat the information here. But we’ll note that a colleague who is on the Intel board says it uses a devil’s advocate-like process to hold robust conversations with top management, and it’s hard to argue with Intel’s results.

We’ve written an article about the dialog with top management for Directors and Boards magazine, and we’ll post the article as soon as it becomes available. The magazine is going to press this week, so that should be soon.