Case Study

www.computerworid.com MANA6EMENT January 9 , 2 0 0 6 COMPUTERWORLD 37

Bounded awareness' can cause you to ignore critical information when making decisions.

Success in IT, as in any field, is all about focus. But in this month's Har- vard Business Review, MAX H. BAZERMAN a n d Dolly Chughposit that focusing too tightly can cause you to miss critical information that's right under your nose. From the Challenger disaster

to the Vioxx debacle, bad decisions often can be tracked back to a failure to con- sider information that was readily avail- able. Bazerman, a professor of business administration at Harvard Business School, talked with Kathleen Melymuka about how to take the blinders off.

What is "bounded awareness"? It's the tendency to fail to see critical informa- tion in our environment because we're overly focused on some subsegment of what's out there. We're so focused on a specific task that we miss other infor- mation that's extremely relevant.

In your article, you give an example from the Ulric Neisser study. He has two videos transposed on top of each other. One has three players in white T-shirts passing a ball. The other has three players in dark T-shirts passing a ball. Sinee they're superimposed, they never pass between the colors. People watch- ing are given the task of counting the number of passes among the white T-shirts. As the film goes on, a clearly visible woman with an open umbrella walks through the frame. She is so visi- ble that normally everyone in the room would see her, but when they're busy counting, the vast majority of people don't. In Neisser's study, only 21% saw her. My experience with executives is closer to 3%. Neisser was looking

at what people fail to see literally, but we're looking at what people fail to see figuratively.

You write about several causes of bounded awareness. The first is the failure of decision- makers to seek information. On the face of it, that sounds silly. It does, but there are situations where people use the infor- mation in the room where they should be identifying information they need to make this decision optimally. One clas- sic example is the Challenger disaster, where decision-makers at NASA didn't ask for relevant information to analyze whether low temperature related to O- ring failure. They used the information that was available.

To Increase Your Awareness

KNOW what you're looking for.

DEVELOP (or pay for) an out- sider's perspective.

CHAUENGE the absence of contradictory data.

OVERSEARCH in contexts where an error would be extremely difficult to recover from.

THINK about the full context of the situation; if you overemphasize one area, you may discount important information in another.

ASSUME that the information you need exists.

ASK for each person's unique information.

MAKE information-sharing the norm.

- MAX H. BAZERMAN AND DOLLY CHUGH

How can a CIO avoid that kind of error? See Neisser's video, or go to [http://viscog. beckman.uiuc.edu/djs_lab/demos.htmU, where my colleague Dan Simons has 12 of these visual illusions. I find it's very valuable for executives to see a visual illusion. It tells us that there are things going on in our minds we really don't understand. If I were to tell a CIO,

"There's important information out there, and you're missing it," he would probably say, "And your evidence would be what?" It's useful to unfreeze people with the visual illusion. Then they can more readily see that there are situations where smart people miss opportunities to bring the right Infor- mation to a decision.

You write that another cause of bounded awareness is failure to use information. Can you give me an example? In many companies, the information is there, but somehow it doesn't get used. In the Vioxx story, it's clear that information about the medical risks existed at Merck long before the public became aware. It seems they had it but didn't use it.

You say that success in one technical area impairs companies' use of new technologies outside that area. Mike Tushman at Har- vard Business School has written about the Swiss watch industry. Switzerland owned tbe watch market for decades, and they had quartz technology long before anyone else had it. But their suc- cess with mechanical watches got in the way of their developing a replace- ment market. They essentially gave the quartz technology away, and the Japanese cleaned up as a result. Their success helped put on the blinders to using the information they had about an upcoming technological change that was going to occur with them or with- out them.

You note that members of a team often dis- cuss only the information they're all aware of and don't share their unique information. Why? It seems crazy, because the whole reason we put together teams is to get information that only one or two of us may have. Our best guess is that when you say what everybody knows, you get positive reinforcement. When you bring up something that is unique information, people sit there or it's un- clear what happens next. So people are reinforced more for saying things other people already know.

How can an IT manager make sure all these mistakes don't happen? The role of the devil's inquisitor can be of use here. Too often, we end up using the infor-

TRY IT erman an

s use this exercit adapted from psychologist P.O. Wason, to ilfustrate ths failure to seek the right information:

The instructor writes the sequence "2-4-6" on the board and challenges students to guess the rule by calling out other sequences of three numbers. The instructor wiij tell them whether each sequence they offer follows the rule. Students can query as many sequences as they iike, but each gets only one chance to guess the rule.

The volunteers usually offer only a few sequences before making their final guesses, and the guesses are invariably incorrect because the students have sought only evidence that would con- firm their guesses- For example, one may say, "80-86-90" to confirm that the rule is "even numbers," or "1-3-5" to confirm that the rule is "numbers that go up by two." When they get that con- firmation, they guess - incorrectly.

The rule is actually "any three as- cending numbers." So. Bazerman says,

"it is rare that we will have answered no to a sequence." Solving this problem requires people to accumulate contra- dictory rather than confirming evidence

- something most of us are not good at.

-KATHLEEN MELYMUKA

mation available and agree to the same points and bring people into the deci- sion path who have a lot in common. The devil's inquisitor is assigned to ask, "What information that we need isn't here? What information might we have that we're not using? What might we know thai we're not sharing?"

I can imagine readers hecoming paranoid about never having enough of the right infor- mation. How do you balance all this with the need to decide? We don't mean to create analysis paralysis, but too often we use the information we have rather than ask, "What information do we need?" People are always doing cost-benefit analyses about whether to look for moje information or stop with what they have. But we're not even arguing for collecting more, but for collecting the right information. At NASA, they had an enormous amount of informa- tion, but not the right information. •

This is the latest in a series of monthly discus- sions with Harvard Business Review authors on topics of interest to IT managers.