Harvard Business Review Online | What Do Managers Know, Anyway?

Click here to visit:

>| http://www.hbsp.org

What Do Managers Know, Anyway?

A lot less than they think. Now, the good news.

by John M. Mezias and William H. Starbuck

John M. Mezias is an assistant professor of management at the University of Miami’s School of Business

Administration in Coral Gables, Florida. William H. Starbuck is the ITT Professor of Creative Management at New

York University’s Stern School of Business. For more information on this topic, see their article in the March 2003

issue of the British Journal of Management.

Most corporate decision making assumes that managers know what they’re talking about. But

three decades of research, capped by our own recent studies, support what many people have

long suspected: Managers often have badly distorted pictures of their businesses and their

environments.

Though they receive endless data about their organizations and the markets they operate in,

managers tend to focus on what is happening right now, in their specific jobs, in their specific

business units, operating in very specific competitive situations. Busy among the trees, they lose

sight of the forest. They base their analyses on formal corporate documents (which they often

misunderstand), personal experiences, rumors they hear at the watercooler, conversations

during committee meetings, articles in periodicals, speeches by their CEOs, and other sources of

varied reliability.

As early as the 1970s, studies began to hint at the chasm between managers’ perceptions and

the objective reality of their businesses. Research by University of Florida management

professor Henry Tosi and colleagues in 1973, and by business consultant H. Kirk Downey and

colleagues in 1975, asked middle and top managers to describe how stable their companies’

markets were. Both research groups then compared the managers’ responses with volatility

indices calculated from the companies’ financial reports and industry statistics. In both studies,

the correlations between the managers’ perceptions and the objective measures were near zero

and negative more often than positive.

Later studies of companies in crisis showed that senior managers had great confidence in their

own, often distorted perceptions of the shape their companies were in – despite the accuracy of

their subordinates’ diagnoses. Research also confirmed the common wisdom that many senior

managers surround themselves with yea-sayers who filter out warnings from middle

management.

To Err Is Human

It was hard for us to imagine that managers could be as out of touch with their environments as

those studies suggested. But because none of the research was designed specifically to assess

perceptual accuracy, it seemed possible that the results were misleading. So we conducted two

studies to try to quantify managers’ perceptions, tapping executives from some of the world’s

best-known companies.

http://harvardbusinessonline.hbsp.harvard.edu/b0...5APrint.jhtml;jsessionid=40EK0QCMJBDDSCTEQENSELQ (1 of 3) [01-May-03 11:58:45]

Harvard Business Review Online | What Do Managers Know, Anyway?

In our first study, we surveyed 70 managers in executive MBA courses at Columbia University

and New York University. These managers worked in various industries; they were presidents of

small companies, department heads in large organizations, and technical specialists. We asked

them about their businesses and environments, querying them about sales growth and

fluctuation as well as industry growth, concentration, and homogeneity. We then compared their

answers with published market reports and statistics.

On average, these managers misreported percentage changes in their industry’s previous-year

sales by 300%. Some managers were off by 1,000% or more. And more than half of the

executives made inaccurate statements about sales by their business units. About one-third of

them underestimated sales, with errors ranging from 75% to almost 100%, and about a quarter

of them grossly overestimated sales, with errors of 200% or more. We saw similar inaccuracies

in the managers’ perceptions of other sales measures, and much to our surprise, managers who

had sales experience fared as poorly as managers who didn’t.

To see if our findings applied consistently to top management, we surveyed 47 senior leaders in

an organization whose divisions were large companies themselves. We looked at the managers’

perceptions of their company’s quality-improvement program, a major headquarters-driven

initiative. Each division had its own quality-improvement department. Managers attended

training courses and received quarterly quality-performance reports, and they earned bonuses

for quality improvements made in their divisions. This is a group you’d expect to have a firm

grasp of corporate quality metrics.

But many of the senior managers’ responses to questions about quality performance were way

off. In reporting quality performance as defect-rate percentages, or defects per million, about

one-third of the nonquality specialists and one-quarter of the quality specialists we interviewed

were off by 50% or more. When defect rates were reported using the nonlinear sigma scale – a

common measurement tool in quality-improvement programs – three-quarters of the nonquality

specialists made errors of 50% or more, and half of the quality specialists made similarly large

errors. Some managers could not read the quality-performance reports correctly even after the

program had been operating for two years. In one case, a quality specialist and a finance

manager both gave the same extremely precise but incorrect number – having pulled it from the

wrong section of the quality report.

Our studies confirmed the astonishingly high prevalence of large errors in perception among

executives.

To Forgive Is Judicious

Clearly, effective strategic thinking doesn’t depend on managers having a precise understanding

of their environments. (For more on this idea, see Kathleen M. Sutcliffe and Klaus Weber’s “The

High Cost of Accurate Knowledge” in this issue.) Because managers usually get prompt feedback

on the impact of their decisions, their misperceptions cause only small errors if they respond

appropriately. For example, managers’ ability to improve product quality depends hardly at all

on their knowledge of current quality measures. It does depend on their readiness to adjust

their thinking in response to feedback – which requires a willingness to admit mistakes.

In 1981, Bill Gates observed, “640K ought to be enough for anybody.” Gates famously

misjudged the potential demand for computer memory; but then, almost all managers

underestimate or overestimate many dimensions of their environments. What distinguishes

Gates’s mistake is not that he made it but that he quickly realized his error and went on to build

a software empire predicated on PCs’ ever-expanding storage capabilities.

Owning up to that gaffe posed no threat to Gates. But most managers today fear sanctions for

being wrong. This trepidation poses a great danger for companies. Distorted perceptions are a

fact of management. Simply providing managers with more training and data is unlikely to

change that. Instead, companies must design decision processes that work despite inaccurate

http://harvardbusinessonline.hbsp.harvard.edu/b0...5APrint.jhtml;jsessionid=40EK0QCMJBDDSCTEQENSELQ (2 of 3) [01-May-03 11:58:45]

Harvard Business Review Online | What Do Managers Know, Anyway?

perceptions, just as aircraft engineers try to anticipate pilot error. That means encouraging

managers to admit their errors and to modify their approach accordingly. The greatest asset

that managers can bring to strategic decision making is not their immediate knowledge but their

ability to seek and make wise use of feedback.

Reprint Number F0305A

Copyright © 2003 Harvard Business School Publishing.

This content may not be reproduced or transmitted in any form or by any means, electronic or

mechanical, including photocopy, recording, or any information storage or retrieval system, without

written permission. Requests for permission should be directed to [e-mail], 1-

888-500-1020, or mailed to Permissions, Harvard Business School Publishing, 60 Harvard Way,

Boston, MA 02163.

http://harvardbusinessonline.hbsp.harvard.edu/b0...5APrint.jhtml;jsessionid=40EK0QCMJBDDSCTEQENSELQ (3 of 3) [01-May-03 11:58:45]