What Do Managers Know, Anyway?
Harvard Business Review, 81 (5):
16--18 (2003)Abstract
A lot less than they think. Now, the good news.
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 water cooler, 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.
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 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.
John M. Mezias is an assistant professor of management at the University
of Miami's School of Business Administration in Coral Cables, 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.
~~~~~~~~
By John M. Mezias and William H. Starbuck
Harvard Business Review and Harvard Business School Publishing content
on EBSCOhost is licensed for the individual use of authorized EBSCOhost
patrons at this institution and is not intended for use as assigned
course material. Harvard Business School Publishing is pleased to
grant permission to make this work available through "electronic
reserves" or other means of digital access or transmission to students
enrolled in a course. For rates and authorization regarding such
course usage, contact permissions@hbsp.harvard.edu
Description
MBA_Decision_Making
Links and resources
Tags
community
@shangnan's tags highlighted