How Smart Does Your CEO Have To Be?

by davidcreelman

How smart does your CEO have to be?  I’ll give you the answer up front; they need to be smart enough to know they are not that smart.

Sydney Finkelstein, a professor at the Tuck School of Business, did a study of 51 cases of massive business mistakes in his book Why Smart Executives Fail.  Each story has its own flavour.  Rubbermaid focused on outstanding innovativeness while retailers were simply demanding lower costs.  Long Term Capital Management took enormous (and fatal) risks based on their faith in their own economic models.  When Roger Smith ran General Motors he was sure that robotics was the magic answer to Japanese competition.  In each case these proved to be poor strategic choices.  However, bad strategy was never sufficient in itself to cause disaster.  Disaster only happened when the management was so convinced they were right, so confident they had the answers, and so sure they were doing a great job that they ignored the ample evidence that they were on the wrong track.

The confidence of the CEO was usually supported by great team spirit.  The whole management team saw things the same way, or when they had doubts managers were too team-spirited to sow discord but voicing their concerns.

It’s not that the CEOs involved were not smart. They were among the smartest managers in the world.  In the case of Long Term Capital Management two of the principals were Nobel Prize winners.  The problem is that they were not smart enough.

How can it be that the smartest people in the world are still not smart enough?   It’s because the world is an exceedingly complex and subtle place.  What worked one place will not work someplace else.  What worked last year may not work this year.  Measures that once provided helpful guidance can prove dangerously misleading. It is dangerously arrogant when a CEO thinks they’ve got it all figured out.

Dr. Karl Weick of the University of Michigan studies high reliability organizations like nuclear power plants and aircraft carriers.  They have to be very reliable because the risk and consequence of an accident is so high.  What is interesting about high reliability organizations is that they operate in a mode of perpetual nervousness.  Any hint of something odd sends them pondering and investigating—it’s the polar opposite of the sense of “knowing it all” that Finkelstein found in corporate failures.  They achieve high reliability by constantly questioning themselves and being constantly alert for the slightest sign that something may be going wrong.

So what do CEOs need to do to avoid the massive disasters that have humbled otherwise good firms?  First they need to accept that no one person ever has all the answers.  They need to continually seek input from outside advisors.  These advisors should represent a diverse set of viewpoints.  A twenty-year old skateboarder may know things that the management team would never see. Dr. Saj-Nicole Joni, a consultant who studies advisory networks, has one particularly memorable piece of advice—staff these networks with the same seriousness you would staff positions in your firm.

A strong advisory network is one step.  It will help make CEOs smarter.  However, even a group of smart individuals is not so smart that they tower over the complexities of the world.  CEOs ought to live each day with a certain nervousness.  They must be suspicious of winning formulas, sure things, and track records of success.

Henry Mintzberg has commented that management programs should leave participants with a sense of humility.  The work of Finkelstein and Weick support his emphasis on that old-fashioned virtue.  Sadly, everyone loves the confident CEO who has it all figured out.  Financial analysts worship a track record of success and the media celebrates the “brilliant” leaders who have devised an infallible strategy.  We all want to believe that there is a winning formula which once found will deliver us from the uncertainties of business life.  The truly smart CEOs are the one who resist these forces, who say, “I smart. Smart enough to know I’m not smart enough.”