Ghost Stories

Children and Computers
Endorsers

The Manager's Core Work in the New Economy - Peter Henschel

From Mental Models to Monday Morning — Lew Rhodes

Would you like to share your reactions to anything you've read here with others? Email with your comments.

Organizational change efforts are systems themselves. They are rooted in complex interrelationships among a large number of people, inside and outside the organization, widely separated across time and space, but cohesive across a long span of time. To lead a change effort, and maybe simply to live in one, itÕs essential to develop an intensive capability to see (and work with) systems. That capability, in turn, will gain strength and subtlety if you can understand systems with more than one approach.

In learning organizational work, "systems thinking" has always meant the "loops and links" of system dynamics Ñ the notation we use in this book to show reinforcing and balancing processes, along with "Limits to Growth." But some people react to these in frustration. "I know the loops are important," they say, "And IÕve gone to training programs. But I still canÕt do it." They canÕt quite figure out how to take their complex problems and translate them into a relevant, clear set of archetypal structures Ñ let alone a systems model. And they are (rightfully) reluctant to delegate their systems insights to an expert "modeler from MIT."

If youÕve struggled to learn systems thinking, then we are kindred spirits. Experts have told me, "IÕm not sure you understand systems thinking." And yet IÕve known that I am a systems thinker since 1981, when I was trained in family systems therapy. This has furnished me with insights that I had never seen come out of any loop. In the end, there is not one kind of systems thinking. There are probably many, each appropriate to different peoplesÕ attitudes and learning styles.

I personally know five. Most business people have been exposed, to some degree, ot all of them, if only briefly:

• Open systems (encompassing the work of Ludwig von Bertellanfy, Elihu Goldratt, and others) Ñ seeing the world through flows and constraints;

• Social systems (encompassing systems family therapy, the work of Barry Oshry, and more) Ñ seeing the world through human interaction;

• System dynamics Ñ seeing the world through ongoing patterns of influence;

• Process systems (encompassing sociotechnical systems and reengineering) Ñ seeing the world through information flow;

• Living systems Ñ seeing the world as it naturally unfolds.

Since "system dynamics" (the third form) is covered throughout Schools That Learn and the other Fieldbooks, I will concentrate on the other four.

Most people in business seem to have a tacit, unspoken preference for one or another approach. They tend to fall back on that method when asked to think about "systems." Engineers are more comfortable with system dynamics; computer people, with information flow; biologists, with living systems; and O.D. people, with social systems. But each has its own value in practice, they are complementary with each other. Each, like any discipline, offers comparatively easy "ways in" that yield quick results, followed by depth and breadth that could take a lifetime to learn effectively. Eventually, switching between them can become so natural that you might forget to think of them as separate. You can become increasingly multi-dimensional in your intuitive grasp of situations Ñ which, after all, is the goal of any work with seeing systems.

A system, in this context, is anything which takes its integrity and form from the ongoing interaction of its parts. Companies, nations, families, biological niches, bodies, television sets, personalities, and atoms are all systems. Systems are defined by the fact that these elements have a common purpose and behave in some common ways, precisely because they are interrelated toward that purpose. All five approaches to systems thinking are based on the idea that the forces in a system are complex, and their interaction canÕt be predicted in any mechanistic way. These approaches are not designed to provide you with "answers" that you can implement right away; they provide you with conversations that help you think, and act more effectively, no matter what problems and opportunities may face you in the future.

Some general precepts for all systems approaches

• No one is to blame. Systems determine peoplesÕ behavior more than people determine system behavior. Changing the players, or "fixing" their behavior by commanding them to act differently, wonÕt solve your systemic problems. But if you can figure out the right people to involve in conversation, together you can all begin to rethink and redesign your system.

• The system is already doing exactly what itÕs "supposed to do." In some way, the system youÕre trying to understand Ñ your home, work, community, or nation Ñ is operating precisely as it is designed, to satisfy some particular (and often hidden) set of priorities. In one wholesale company, experienced sales representatives kept quitting, leaving customers with na•ve new sales reps Ñ which drove customers away as well. Sales training did not solve the problem, because the system had its own imperative: a built-in policy of cutting territories in half when commissions got too high. This was, in turn, designed to establish the "systemic" goal of not letting sales commissions get "inappropriately" high. If youÕre unhappy with your systemÕs performance, look at the benefits that various parts of the system gain from the performance that makes you unhappy.

• Problems are interdependent. Management systems pioneer Russell Ackoff puts it this way: "Managers are not confronted with problems that are independent of each other." People are confronted, he says, only with dynamic situations that consist of complex systems, of changing problems, interacting with each other. His term for such systems is: "messes." Purely analytical techniques will not serve managers. Their job is managing messes, and they need "the active, synthetic skill of designing a desirable future and inventing ways of bringing it about."

• Systems thinking is vision-driven. Systems thinking is hard work and can irritate people, if only because it raises annoying knowledge about the "things going on around here." A sense of your common purpose and vision can provide context. A vision you care about, and even love, will give you the focus to pursue your understanding of the systems even through the irritation.

• Problem-solving is expensive. You can use systems thinking to fix everyday problems, and a number of consultants make good livings doing exactly that. But by the time you fix those problems, a new set of problems will have emerged, because each intervention in a system will produce its own set of unintended consequences. ItÕs much more effective to slow down and embed in yourself the capability to move directly (and sustainably) towards the future you want to create.

Open Systems: Seeing the world through flows and contraints

"[One of the most] significant historical events in my lifetime," wrote philosopher/anthropologist Gregory Bateson, "was the growing together of a number of ideas we may... call cybernetics, or communication theory, or information theory, or systems theory.... All these separate developments in different intellectual centers dealt with... the problem of what sort of thing is an organized system." [This quote appears in Steps Toward an Ecology of Mind, quoted in Uncommon Sense by Mark Davidson, p. 191. ]

The Viennese biologist Ludwig von Bertallanfy was the grandfather of general systems theory, which occupied a lifetime of work from the 1930s through the early 1970s. Why, asked Bertallanfy, is the whole more than the sum of its parts? Because of the ways they are arranged. The boundaries of the system, for instance, are defined by the beholder of the system Ñ because all relevant elements are included within. If you work in marketing in Missouri, for instance, what is the boundary of your system? The continental United States? Your desk area? Your company? Your function within the company?

In its emphasis on "optimizing for the whole system," Open Systems foreshadowed the key tenets of the quality movement (whose most prominent figure, W. Edwards Deming, credited Russell Ackoff with helping him learn this aspect of the quality approach). "Suboptimization," a perennial issue in the quality and reengineering movements, is actually a boundary-definition issue: Define your sphere of interest too narrowly and you are apt to produce benefits for, say, your marketing function or this yearÕs revenue goals at the expense of the whole organizationÕs ability to respond to a looming competitive threat.

Open Systems was also a direct philosophical counterattack against the reductionist, mechanistic approach of Frederick TaylorÕs scientific management, and Bertallanfy made a point of saying so. Any human organization, he declared, was not a machine and should not be seen as one. It was a life form, like a biological cell or living entity. To an open systems theorist, that meant a thing that transforms its inputs Ñ everything it eats, breathes, perceives, absorbs and intakes. And they transform the entity as well. To change an open system, learn to understand and influence the things that it takes in, and its relationship with its environment.

Since the environment is also a system, however, open systems work continually looks out beyond the implicit boundaries: beyond the walls of the corporation or the division to see how "outputs" affect the larger world, which in turn affects the inputs. "Open systems" researchers also seek out the unconscious strategies by which the system maintains its integrity and move towards greater self-generated order and complexity.

The whole depends on the parts and the parts depend on the whole. Actions anywhere in the system can bring reactions anywhere and everywhere in the system. Therefore, it is impossible to subdivide responsibility into separate departments.

Right before he died in 1972 at the State University of New York at Buffalo, Bertallanfy was researching the international postal service, a truly cooperative international system that functions even in wartime, crossing political, social, and economic borders. Soldiers, no matter where they are, still get their mail. It cycles through the system. Could there be an application of this type of conflict-free operation that could help undo the friction that leads to war?

Open Systems" work in practice.

Sit down with a pilot group and ask yourselves:

1. What are the boundaries of the system? In other words, how broad is the situation you are trying to understand? Is it just your team? Is it your department? Are you concerned only with one particular process or product line? Or are you concerned with our company as a whole? Or your industry?

2. What are the known behaviors or principles implicit in this system? Often, this will emerge in interviewing people outside the pilot group. In particular, look at the flows into and out of the system, using a diagram to guide you.

What inputs (goods, capital, labor, information) come in from outside? What kinds of transformations, in the general sense of the word, do these go through? And then what outputs (reports, products, services, numbers, waste) does the system generate? How does the world outside respond to those "outputs," and how does that response affect the next round of inputs?

One critical input is information from the marketplace. In a major Fortune 50 company, this exercise revealed that the market research department controlled that information, so protectively that they kept it from R&D until after senior managers had approved it. Since that delayed it to the point of uselessness, R&D developed its own market research department (under a different name, to avoid suspicion.) Neither market research department shows its results to the other; when the marketplace shifts, both groups can get blindsided.

3. What key interactions have organized the system?

With this question, you begin to convert your "open system" diagram into a model of the business that everyone can understand. Why are the various inputs, outputs, and throughputs set up the way they are? What are their costs? What do they contribute to the businessÕ financial viability, or to its social viability (the extent to which they make it a better place to work)?

4. Take the thinking of your team further out by expanding your systemÕs boundaries one level. What interactions or insights now come into focus?

Bertallanfy had suggested this practice in his own writings about social change. By deciding to consider the space beyond your conventional team or organizational boundaries, you'll usually find insight. Typically, an "external" personÕs concerns come into the picture: A critical stakeholder, customer, or policy-maker, perhaps.

5. In the system as it currently stands, who is aware of the picture you have created? If your organization is typical, youÕll find that the answers to these questions have traditionally been privileged information, kept hidden by one part of the system. Purchasing understands the flow of raw materials and tools. Finance understands capital costs. HR understands the "input" of knowledge and capabilities. R&D and operations are responsible for the throughput of manufacturing. Sales, marketing, distribution and product services handle different aspects of "output." Nobody focuses on the "renewed input," the ways in which outputs affect inputs later. ItÕs left to senior management to develop a feeling for the system as a whole. But since their view is so lofty, a great deal of waste slips through the cracks, as different members of the system duplicate or even undermine each otherÕs work.

6. Now consider the current status of people who have knowledge of the whole system. How are they treated by the system as a whole? How can the system make better use of their knowledge? (As you ask these questions, you begin to move into systems thinking form #2, social systems.)

7. Figure out if one element of the system gets more attention than others. For instance, is the organizationsÕ attention focused disproportionately on one type of input (finance? materials?) when another type of input (human knowledge?) could yield more leverage? How do different types of inputs and outputs influence each other over time? (As you ask these questions, you begin to move into systems thinking form #3, system dynamics.)

8. Each of these parts of the system has information that might help other parts. What kinds of information need to flow more completely and how? (As you ask these questions, you begin to move into systems thinking form #4, sociotechnical systems.)

Resources for "Open Systems" work

BertallanfyÕs own writings are dense and difficult to read. Two introductory books date back to the 1970s. The first is a biography, an in-depth guide to his life and work: Uncommon Sense, by Mark Davidson. Published in 1983 by J.P. Tarcher, this book is sadly out of print. The second, The Social Psychology of Organizations, by Daniel Katz and Robert L. Kahn (1963, 1978, Wiley), is the pre-eminent textbook, with in-depth discussion of BertellanfyÕs concepts Ñ inputs, outputs, throughputs, boundary definitions, and cycles of feedback and homeostasis. Many managers will recognize Open Systems from the 1970s and 1980s, because the Katz-and-Kahn influence filtered out into organizational training throughout that period. Unlike most management fads of that period, it remains useful today.

Elihu Goldratt: another pair of open system "glasses"

The books of Elihu Goldratt provide a series of methods for applying "open systems" work without feeling that you are stuck in an academic treadmill. Goldratt delivers his lessons through a series of engaging novels, explictly designed to emulate the Socratic method. (He points out that if he just told people the answer, theyÕd be much less likely to put into practice. The novels make them struggle with it until they grasp the implications.)

Under GoldrattÕs guidance, you begin to recognize the bottlenecks and constraints that tend to be hidden, precisely because they are systemic. The most basic of these interrelationships concerns the relationship between inventory, throughput (the speed with which your work moves through the system), and operating expense. To improve the system as a whole, improving throughput offers the greatest leverage.

He tells a great story about a group of kids hiking through the woods. They struggle to keep up with each other, and itÕs obvious that the slowest kid, Herbie, is holding the rest of them back. No matter how fast they hike, no matter, where they place Herbie in line, they canÕt travel any faster than Herbie travels, or else theyÕll spread too far apart on the trail (with, as Goldratt points out, exactly the same wasted effort and expense as a glut of inventory). The solution? Everybody sits Herbie down to figure out why heÕs so slow. It turns out that his pack is loaded down with canned goods and tools. Dividing them up, they lighten HerbieÕs pack, and then they can all move much faster through the woods....

Each of GoldrattÕs books is full of stories like that, mostly in real industrial or financial settings (one of his most memorable books, Critical Chain, is set in a business school). He spells out the theory in a slim hardback called Theory of Constraints. He is generous in pointing out how the various types of "systems thinking" all reinforce each other, including his own. And he looks at the emotional impact of problems in throughput in a genuinely helpful way. In person, Goldratt comes across as arrogant and self-absorbed (I have seen him stand onstage, facing the audience, while a younger assistant read aloud from his novels). But the books themselves, all of them, are not only valuable, but engaging, respectful, and downright humble in their generous intent to educate. ÑArt Kleiner

Use Links at beginning of article to read about other types of sstems thinking.