|
|
Oct. 15, 1996 Issue of CIO Magazine
Joining the Culture Club
If you want to understand employees' behavior, go to the experts
By Tom Davenport
Here's an idea that may not have crossed your mind: The next time you hire a
systems analyst, look for someone with a degree in anthropology. Or the next
time you place an ad for a programmer, after "knowledge of C++ required" put
"some ethnographic experience desirable." OK, stop chuckling. Believe it or
not, many smart companies will hire anthropologists in the future, and corporate
anthropologists are already in demand.
Anthropology, for those of you who spent your college careers in the
underground computer vault, is the study of cultures. Any culture -- not just those
on South Sea islands -- is a respectable topic for anthropological analysis.
Ethnography is one of the primary methods used by anthropologists to understand cultures; it involves living in a culture, observing it carefully and
describing those observations in minute detail. Clifford Geertz, one of the of the world's most respected anthropologists (and
still not exactly a household name), called that method "thick description."
What business do these people have in the hallowed halls of IS? Just ask
yourself (or even better, ask your customers) if your systems developers have a
record of producing systems that fit the culture of the user organization. Think
about whether the designers of new business processes in your organization -- be they external consultants, internal reengineers or systems analysts -- have been sensitive to how work really gets done and what motivates employees to work harder and more
efficiently. Think about how well IS has been able to understand and meet the information requirements of your company's senior
managers. Do you know why some managers like computers and some hate them? Let's be honest. Most IS groups do an abysmal job of producing and
implementing systems that fit an existing culture or move it gracefully in a
desired direction. Why is that? Many methodologies have steps that sound almost ethnographic, recommending that analysts actually observe how work
gets done in an office or factory before building a system to support it. But such
careful observation is usually abandoned in the mad rush to start coding (or to
install a package). Whether we're reengineering or identifying information
requirements, we usually consider our analysis of the existing work environment to be thorough if we spend a half-hour interviewing a "representative sample" of
those doing the work. In reengineering projects, many companies don't even
bother to understand and document the current business process because they
plan to obliterate it. Variations in how tasks are performed between workers
and locations are written off as products of ignorance or poor management. In
most cases, managers haven't a clue how those with jobs slated for automation
get meaning from their work, respond to their work environments and use or
ignore information on a day-to-day basis.
So how can corporate anthropologists help? The basic idea is that current work
practices are worthy of respect. People do things the way they do for specific
reasons, and those reasons should be considered when a company tries to move to a new process or system. The process workers follow and the
information they use can't be discovered through a brief interview; in most
cases, workers can't even explain how they work. It's tacit knowledge that can
be transferred only by watching them, working with them and demonstrating that
you understand their world. This takes a while; after all, how long do you
think it would take an observer to understand your job?
Anthropologists use a variety of pretty straightforward techniques. Some keep
detailed project notebooks, recording what they learn during a period of observation. Others videotape work activity and analyze it in detail later.
Anthropologists get out into the field -- if they want to study the logistics process,
they don't follow around the vice president of logistics; they ride in delivery
trucks.
Anthropologists do not simply observe; they try to ascertain whether a particular
change or system would fit into the work environment. One high-tech manufacturer, for example, used ethnographic analysis to determine whether
artificial intelligence tools would help customer service representatives solve
specific repair problems over the telephone to avoid dispatching a service
technician to the customer site. The answer? A qualified yes; some of the representatives were quite capable of remote problem solving, but many were
unwilling or unable to go so far beyond their job descriptions. Think it's too
time-consuming or too expensive to employ anthropological methods? How much does it cost to build a system that is never used? How expensive is it to
"finish" a system and then have to modify it to do what your customers really
want?
How much does a consultant charge to design a new process that is never implemented? As the Japanese have been saying for years, spending more
money early in the development of new systems and processes means spending less later to fix them. This is not a topic sequestered in the ivory tower.
Hundreds of anthropologists are working on these types of issues. Motorola
Inc., for example, teaches anthropological analysis at its Motorola University; the
anthropologists are credited with pointing out that teenagers were using pagers
as fashion accessories, not as utilitarian devices. Motorola then radically
expanded the range of colors available for its pager line.
At Nynex and Hoffman-LaRoche Inc., anthropologists have helped reengineer business processes in a manner tailored more to actual work behaviors than
most other projects of that ilk are. At Hoffman, anthropologists also studied how
scientists were using an expert system that described the drug development
process.
One of the most influential incubators for corporate anthropologists is the
Institute for Research on Learning in Menlo Park, Calif. A small consulting and
research organization with many Xerox Palo Alto Research Center connections,
IRL has a cadre of ethnographically oriented researchers who study learning
activities in schools and workplaces. IRL experts have worked at Xerox Corp.,
National Semiconductor Corp., Sun Microsystems Inc. and many other companies around the world. They are pioneering a participatory form of
consulting that could spread to mainstream companies. Most large consultancies already have change management practices, but they generally
use traditional approaches to understanding how an organization works, such
as brief interviews or surveys.
Of course, life with anthropologists can have its problems. One I have noted is
that they want to observe too long before coming to a conclusion. Insisting that
you have to observe workers for six months before understanding what they do
is just as untenable as presuming you can figure it all out in 15 minutes. And
anthropologists are often reluctant to generalize from their observations in a
workplace; sure, every work situation is unique, but it's also possible to
concentrate on common facets.
Anthropologists just out of the university may also be naive about other realities of corporate life, such as the need for high-level sponsorship and the fact that lifetime employment is no longer a corporate guarantee. But they can learn quickly in their new surroundings. Just in case you're wondering, I have no ax to grind. I was trained as a sociologist, not an anthropologist. Sociologists often make the same mistakes corporate analysts do: thinking about people only in the aggregate and trying to understand what makes them tick by gazing at a
computer screen. Not just any ol' social scientist will do; you should look for the
real anthro thing. They may be called social anthropologists, cultural anthropologists, even industrial anthropologists. Look out for physical
anthropologists, however; they mostly dig for bones. Whatever type of anthropologist you're searching for, you'd better move quickly. Universities don't
turn out a whole heap of anthropologists each year and post-graduate training in
the field isn't widely available at the local junior college. A fast-moving company
-- maybe yours -- will be able to corner the corporate anthropologist market
pretty quickly, and it will have the most cultured systems of all.
Thomas H. Davenport is a professor and director of the Information Management Program at the University of Texas at Austin. He welcomes reader comments at
davenport@mail.utexas.edu.
Principles of Corporate Anthropology
Cast a wide net. You don't know in advance what aspect of a culture will be most important to your objective, so try to observe everything.
Take some time to observe the people and processes in which you are
interested. True understanding comes with living and working alongside your
research subjects.
It's hard to listen and watch if you're talking. You may feel that you know the
solution to the problems of those you're observing, but it will be better if they
figure it out with little help from you.
Be wary of generalizations. Sure, this office may perform the same function as several others in distant parts of the company. But they may see and do things differently; visit them too if you can.
Pay more attention to what people do than what they say they do. We all want to be viewed as rational, thorough and well-informed decision-makers.
Unfortunately, many of us aren't.
|
