Achieving Joint Optimization
Work is done, or at least overseen, by people. This may seem like a blindingly obvious statement, but failure to recognize the consequences of this simple fact is the primary factor distinguishing successful projects from unsuccessful ones. Whenever a work system is designed or modified, both technical and social components are affected. When a project falls short of its objectives, the root cause is invariably a failure to address satisfactorily one or the other of these components. The recognition that social and technical systems have to be jointly, concurrently optimized lies at the heart of the most powerful of all approaches to the design of work systems—Sociotechnical Systems Design, or STS.
In a recent special report on management, Business Week referred to STS as “an immensely important trend, one that is producing a new model of job design and work relations that will shape the workplace well into the 21st century.” In 2008 the National Science Foundation, recognizing that cyberinfrastructure strategies built on the basis of hardware and software alone could not ensure the effectiveness of virtual work, established the VOSS Grants Program to explore Virtual Organizations as Sociotechnical Systems, which is now in its second year of funding. In short, STS is often hailed, these days, as “the next big thing” because it achieves unprecedented results in the entire range of work environments from traditional manufacturing to white-collar information work to today’s virtual work systems.
Those coming to STS for the first time often experience it as a kind of revelation and are surprised to learn that STS has a long and highly successful project history. The approach was first articulated by Eric Trist and Fred Emery over fifty years ago in research on shortwall mining techniques in the UK. Since that time, STS has been applied with enormous success by a few forward-thinking managers in every major industry sector, and in the nonprofit, government, and military sectors as well. Ironically, the basic principles of what became STS were well understood by such pioneers of scientific management as Frederick Taylor and Edward Deming, though their followers often failed (sometimes catastrophically, as was the case with much of the Business Process Reengineering of the 1990s) to grasp and implement these key but less widely understood aspects of the pioneers’ thought.
Projects go wrong when people fail to recognize what the great thinkers in the history of scientific management understood implicitly—that technical and social systems must be addressed concurrently. The central goal of STS, like that of other improvement methodologies, is, of course, to reduce error and eliminate variances, and experts in process control have long understood that errors and variances typically result not from worker inadequacies but from problems in work system design. The difference between STS and other approaches to work systems design lies in the degree of attention STS pays to understanding the whole system before taking action. STS practitioners use the standard process quality control tools and techniques to understand and optimize the technical system being implemented, but they also attend with equal care to the “as is” and goal states of the social system. Social system factors studied include those traditionally attended to—staffing, training, and compensation—as well as those typically neglected—issues of status, personal control and autonomy, work structure, decision-making processes, coordination mechanisms, physical space, work environment, communication flow, and job and organization design.