Stanford University News Service



CONTACT: Stanford University News Service (650) 723-2558

Balancing tradition with innovation main task for managers

STANFORD -- One after another, giant corporations long held as indestructible fall victim to changing economic times. Most recently and dramatically, Sears and IBM have been forced into drastic reorganizations after sustaining huge losses.

Other companies, such as Intel, seem to thrive on the economic chaos of the early 1990's. Why do some thrive while others perish?

The difference may be whether the organization relies solely on "induced process," traditional top-down management, or also makes use of "autonomous process," with ideas and experiments coming from the bottom up, said Robert A. Burgelman, a professor in the Stanford University Graduate School of Business.

To survive in changing times, a company must successfully balance its corporate focus and principal markets with exploring new avenues of marketing and research, Burgelman said at a session of the American Association for the Advancement of Science meeting Feb. 15 in Boston.

Not all management can do that well.

"To be successful, a declared strategic transformation has to be preceded by internal experimentation and selection," he said.

Organizational science has identified two primary types of internal structure within large organizations, such as corporations:

  • One Burgelman calls the induced process. Actions are primarily the result of strategic thinking - plans formulated by top management, consistent with the stated corporate direction and carried out down the ranks.
  • The opposite is organizational ecology - bottom-up management. Burgelman calls this the autonomous process. New ideas spring up out of lower-level management or the work force itself, such as proposals for new product lines and marketing directions. Like genetic mutations, these new ideas must face a series of trials before proving they are adapted to the changing environment.

Splitting with the traditional corporate goals, these autonomous processes play the key role of providing needed room for experimentation, Burgelman said.

Dabbling in a new market while still focusing primarily in the company's main direction allows for the testing of new waters and development of new capabilities, he said. Successful experiments can be pursued and grow, while unsuccessful ones can be quietly folded away and written off.

When large companies are forced to change their corporate strategy radically in response to market demands, those that did not pursue autonomous processes often fail, Burgelman said.

Other companies, those which allowed the working of autonomous processes, have experience in new fields to guide them through the transition. In fact, a highly successful autonomous process may eventually become part of the induced process.

Anatomy of a success

Burgelman cites the Intel Corporation as an example of a company that successfully used autonomous processes to develop new markets and survive a market change. Intel had historically been a large and profitable manufacturer of DRAMs, dynamic random access memory, for computers.

However, within the company some researchers were pursuing projects seemingly outside the induced process: the development of microprocessors, which were invented at the company. But during the early 1980s, this was a secondary objective at Intel; the company still considered memory to be its primary focus.

By the mid-1980's however, American memory manufacturers were facing the prospect of a flood of low-cost Japanese DRAMs. Correctly predicting its inability to compete with the new threat, Intel in 1985 announced that it was abandoning the production of DRAMs and focusing now on microprocessor technology.

This allowed Intel not only to survive a market change that killed many other American firms, but to soar. In the 23 years between 1968 and 1989, the firm's sales rose from $1 million dollars to $4.2 billion (up 420,000 percent). From 1969 to in 1991, the balance sheet rose from a loss of $2 million to a profit of $800 million.

Intel included in its induced process what had initially been an autonomous one.

Intriguingly, there now is a new autonomous process at work within the company, one proposing a return to memory development. This advocates the development of FLASH memory, a new and possibly revolutionary portable data storage system.

While Intel successfully balanced autonomous and induced processes, may other companies have not done so, Burgelman said. IBM's recent troubles are a classic example of a company unable to promote autonomous processes within the basic strategic framework.

IBM continued to try to push mainframes to its business clients, instead of the newer distributed desktop systems introduced by its competitors. Not emphasizing the marketing of these systems, IBM religiously pursued the path stated by its induced process.

"They lined up the troops and marched them off a cliff," Burgelman said. "What happened to the ideas that could have saved them?"

Lacking in any experience outside the induced process, IBM is scrambling to catch up in a changed market.

Just as dangerous are autonomous processes run amok, cautions Burgelman. Though it is wise to explore new avenues, there must be a sufficiently strong induced process to insure some degree of focus. Following every fad of the marketplace results in spreading resources too thin for survival.

Testing the waters

Autonomous processes should be allowed to operate in a limited way, testing the waters. Those that fail to be profitable or are clearly fads must be trimmed away, and successful ones encouraged, eventually becoming part of the induced process, as in Intel's case, said Burgelman.

The difficulty lies in upper-level management simultaneously accepting two seemingly opposed ideas.

"One sign of intelligent life is the ability to entertain opposing ideas," said Burgelman, implying that the mark of a successful high-level manager is one who can balance the induced process of the corporate strategy with the autonomous processes suggested by his subordinates.

"The internal selection must reflect external selection pressures," said Burgelman. -jns-

This story was written by R. Nicholas Strauss, a science writing intern at the Stanford News Service.


This is an archived release.

This release is not available in any other form. Images mentioned in this release are not available online.
Stanford News Service has an extensive library of images, some of which may be available to you online. Direct your request by EMail to

© Stanford University. All Rights Reserved. Stanford, CA 94305. (650) 723-2300.