Systems Thinking Basics

From Concepts to Causal Loops

Virginia Anderson AND Lauren Johnson


Waltham, Massachusetts

© 1997 by Pegasus Communications, Inc. First printing March 1997

All rights reserved. No part of this publication may be reproduced or transmitted in any form

or by any means, electronic or mechanical, including photocopying and recording, or by any

information storage or retrieval system, without written permission from the publisher. For

additional copies, information about other titles published by Pegasus Communications, Inc.,

or general inquiries, contact:


One Moody Street

Waltham, MA 02453-5339 USA

Phone 800-272-0945 / 781-398-9700

Fax 781-894-7175

[email protected] [email protected]

Library of Congress Cataloging-in-Publication Data

Anderson, Virginia, 1947–

Systems Thinking Basics: From Concepts to Causal Loops / by Virginia Anderson and Lauren Johnson.

p. cm.

ISBN 1-883823-12-9

1. Industrial management—Case studies—Problems, exercises, etc.

2. Systems analysis—Problems, exercises, etc. 3. Decision making—exercises, etc.

I. Johnson, Lauren. II. Title.

HD30. 19. A53 1997

658.4’032—dc21 96-39762


Acquiring editor: Kellie Wardman O’Reilly

Project editor: Lauren Johnson

Production, art, and design: Boynton Hue Studio

First edition.

To the Reader: Why Use This Book? vii


What Is a System? 2

Events, Patterns, Structure 5

L E A R N I N G A C T I V I T I E S 10 1: Identifying Systems 10 2: Reflecting on System Characteristics 13 3: Understanding System Feedback 14


The Principles of Systems Thinking 18

Systems Thinking as a Special Language 20

L E A R N I N G A C T I V I T I E S 22 1: Stretching the Timeline 22 2: The Shape of the Problem 25 3: Is Time on Your Side? 29 4: From Short Term to Long Term 30 5: In the Midst of a Problem 34


Formulating a Problem 38

Identifying Variables 40

Drawing Behavior Over Time Graphs 42

L E A R N I N G A C T I V I T I E S 45 1: The Problem with Price Promotions 45 2: The Case of the Energy Drain 46 3: The Case of the Audio-Electronic Roller Coaster 48




Anatomy of a Causal Loop Diagram 52

Building a Causal Loop Diagram 58

About Multiloop Diagrams 64

L E A R N I N G A C T I V I T I E S 65 1: The Case of the Plateauing Profits 65 2: The Case of the Collapsing Banks 66 3: The “Organic To Go” Story 67 4: The Case of the Restricted Revenues 70 5: The All-for-One Cooperative 73 6: The Problem with Used CDs 75


Characteristics of Complex Systems 78

The Strengths and Weaknesses of Complex Systems 79

ComputeFast: The Story of a Complex System 80


The Do’s and Don’ts of Systems Thinking on the Job 88

Practicing Life-Long Systems Thinking 91

The Learning Journey 94









Why Use This Book?

C ongratulations! By picking up this book, you’ve taken the first step in learning to use the powerful array of systems thinking tools. These tools offer a fresh, highly effective way to grasp the complexities of organi-

zational life and to address the stubborn problems that often confront us in the business world. Systems Thinking Basics: From Concepts to Causal Loops is designed to help you discover the principles of systems and systems think- ing and begin actually using systems thinking tools. There is a broad palette of systems thinking tools available; this book emphasizes two foundational devices: behavior over time graphs and causal loop diagrams.

Why Systems Thinking?

Why take time out of your busy schedule to read this book? Because systems thinking offers a valuable new perspective on our most persistent organiza- tional problems and our role in them. One of the major points that systems thinking makes is that everything—and everyone—is interconnected in an infinitely complex network of systems. When we begin to see the world— and one another—through this “lens,” we start seeing our circumstances in a new light, taking more responsibility for our own role in problems, and identifying more effective ways of addressing recurring difficulties. Systems thinking has a power and a potential that, once you’ve been introduced, are hard to resist. You’ll learn more about the benefits of systems thinking in Section 2.

How to Use This Book

The world of systems and systems thinking marks a dramatic shift from the more linear, analytic way of thinking that many of us are used to. It’s espe- cially challenging to convey the abstract characteristics of systems in print. Because systems themselves are so complex and so interconnected with everything around them, it’s also impossible to capture any one entire sys- tem on paper. For this reason, the tools and activities in this book are meant to offer primarily a glimpse into systems behavior.

We also hope you will see the book as the starting point to a long, learn- ing-filled journey. Systems research and theories are constantly evolving, and the concepts and tools in this book offer only one of many methods for looking at the whole. When you finish the book, you can continue your adventure into exploring systems thinking tools by using some of the resources listed in Appendix F.

In writing Systems Thinking Basics, we not only needed to decide what tools to include; we had to choose a plan for sequencing the sections. We


viii TO THE READER ➤ Why Use This Book?

opted for a format that allows each section to build on the one that came before, because that arrangement seemed best suited for new systems thinkers. For this reason, we recommend that you read the sections in numerical order, although you are certainly free to choose a different order depending on your interests and your familiarity with the material. We also hope that you’ll work as often as possible with others on the readings and activities—it’s when people use systems thinking tools together that these tools are their most potent.

Your journey through Systems Thinking Basics will begin with an explo- ration of the unique characteristics of systems (Section 1), and then move on to a definition of systems thinking and an explanation of its importance, especially to organizational life (Section 2). In sections 3 and 4, you will have the opportunity to create your own diagrams of systemic structures, step by step. Section 5 then gives you a taste of complex systems. Finally, Section 6 answers the question, “What next?,” by offering advice on prac- ticing systems thinking in your day-to-day life.

About the Learning Activities

Mastering systems thinking tools requires practice with lots of real-life examples. Accordingly, each section in this book contains a wealth of illus- trations from the business world. Most sections then present a series of learning activities that challenge you to apply your new knowledge. The learning activities can be done as self-study or in groups, but again, we encourage you to work in groups as much as possible. We invite you to share with each other your responses to the activities and the insights you gain as you work with the activities.

The learning activities range from reflection and discussion to graphing and diagramming, and are filled with case studies adapted from recent analyses published in leading business journals, including The Systems Thinker™ newsletter. There is also a section (Appendix A) that offers extra learning activities should you desire additional practice. For the majority of the learning activities, you can check your responses against the “Learning Activity Key Points and Suggested Responses” in Appendix B. Remember, however, that there is no one right way to describe and diagram a system. These activities, and the suggested responses, are meant to spark your imag- ination and serve as a starting point for you to think about systems.


The work contained in this book is built on the prior work of many others. First, we acknowledge Peter Senge for introducing the concept of systems thinking to a broad management audience in The Fifth Discipline. In many ways, the tremendous success of his book is what made the writing of this workbook possible in the first place. We also acknowledge the many people who were involved in the development of systems thinking principles, con- cepts, and tools before they ever appeared in The Fifth Discipline, including John Sterman, Dennis Meadows, Dana Meadows, Michael Goodman, David Kreutzer, Jenny Kemeny, Jack Pugh, George Richardson, James Lyneis,

TO THE READER ➤ Why Use This Book? ix

Nathaniel Mass, Barry Richmond, and others. In addition, we thank all the systems thinking workshop participants and contributors to The Systems Thinker newsletter, who have given us valuable insights into understanding and applying systems thinking to organizational life. Finally, we honor Jay Forrester’s brilliance and insight in founding the field of system dynamics, which we believe has much to contribute to the practice of management.

We’d like to thank the following colleagues who kindly reviewed the manuscript for this book. Each of them offered valuable insights and thoughtful suggestions that greatly strengthened the book:

Turina Bakken, MAQIN

Polly Bashore, General Motors

Dave Buffenbarger, New Dimensions in Learning

Lou Ann Daly, Innovation Associates

Richard Karash, Speaker, Facilitator, Trainer, “Towards Learning Organizations”

Daniel H. Kim, The MIT Center for Organizational Learning, and Pegasus Communications

Keith Perry, McClellan Air Force Base

Kellie Wardman O’Reilly, Pegasus Communications

Carol Ann Zulauf, Suffolk University

and, finally, all the folks at Coca-Cola:

Michael Canning, Connie Carroad, Dianne Culhane, Teresa Hogan, Scott Moyer, Cheryl Oates, and Rodolfo Salgado

We hope you’ll find your adventure into systems thinking stimulatingand rewarding on both a personal and a professional level. We also invite you to contact us with any comments, questions, or suggestions about using this book or about systems thinking in general.

Bon voyage!

Virginia (“Prinny”) Anderson ([email protected]) Lauren Johnson ([email protected])

S E C T I O N 1

What Are Systems?

W elcome to the world of systems and systems thinking! You may be asking yourself, Why is it important to explore systems? One reason is that we live in and are influenced by systems all around us, from

the natural environment to healthcare, education, government, and fam- ily and organizational life. Understanding how these systems work lets us function more effectively and proactively within them. The more we build our understanding of system behavior, the more we can anticipate that behavior and work with the system to shape the quality of our lives.

This section introduces you to the idea of systems and what makes them unique. In the learning activities at the end of the section, you will have the opportunity to identify some major systems in your own work life and to think about typical system behavior.


2 SECTION 1 ➤ What Are Systems?


A system is a group of interacting, interrelated, or interdependent compo- nents that form a complex and unified whole. A system’s components can be physical objects that you can touch, such as the various parts that make up a car. The components can also be intangible, such as processes; rela- tionships; company policies; information flows; interpersonal interactions; and internal states of mind such as feelings, values, and beliefs.

In an organizational setting, for example, the R&D group is a system made up of people, equipment, and processes that create new products to be manufactured by the production system and sold by the sales system. The components of the R&D group have to interact with one another to perform their function and thus are interdependent. In turn, the R&D group interacts and is interdependent with other systems within the com- pany. A system such as the R&D group always has a specific purpose in rela- tion to an even larger system—in this case, the entire organization (Figure 1.1, “Interdependent Systems Within Interdependent Systems”).

Your body is another example. Within it, your circulatory system deliv- ers oxygen, nutrients, hormones, and antibodies produced by other systems and carries waste to the excretory system. The circulatory system is made up of the heart, veins and arteries, blood, and a host of supporting elements. All of these components interact to carry out their purpose within the larger system—your entire body.

Both of these examples raise an intriguing point about systems: We can think of all systems as nodes embedded in a giant network in which every- thing is connected. For example, the company described above, with its interdependent R&D, production, and sales systems, is itself a large system that is interdependent with an even larger system—industry as a whole.



Pr oc

es se


R & D




F I G U R E 1 . 1

Interdependent Systems Within Interdependent Systems

What Is a System? 3

And industry is interdependent with an even larger system—the econ- omy—and so on. The more we widen our view in this way, the more we see that everything—from the tiniest subatomic particle to the universe (and maybe beyond!)—is intertwined.

We can also distinguish between natural systems and human-made, nonliving systems. Natural systems—a living being’s body, human societies, an ecosystem such as a prairie—have an enormous number and complexity of components and interactions among those components. They also have virtually an infinite number of connections to all the systems around them. Human-made systems—cars, for example—can also be quite complex, but these nonliving systems are not as intricately linked to systems around them. If a car breaks down, the impact of this event is not nearly as far- reaching as if a species were removed from a prairie ecosystem (although you may disagree if it’s your car that breaks down!). Put another way, human-made nonliving systems are more self-contained than natural sys- tems, which we can think of as more open in their connections to sur- rounding systems.

Defining Characteristics of Systems

Systems have several essential characteristics:

1. A system’s parts must all be present for the system to carry out its purpose optimally.

If you can take components away from something without affecting its functioning and its relationships, then you have just a collection, not a sys- tem. For example, if you remove a cashew from a bowl of mixed nuts, you have fewer nuts, but you have not changed the nature of the collection of components. Therefore, a bowl of mixed nuts is not a system.

Similarly, if you can add components to a collection without affecting its functioning and relationships, it’s still just a collection. So, if you add pistachios to your bowl of mixed nuts, you have more nuts and you have a different mix, but you still have just a collection of nuts.

However, if you assign new tasks to an R&D group or redefine the job descriptions of its staff, you will likely change the group’s functioning and relationships—whether for the better or worse. The R&D group is not just an assortment of people, equipment, and processes; it is a system.

2. A system’s parts must be arranged in a specific way for the system to carry out its purpose.

If the components of a collection can be combined in any random order, then they do not make up a system. For example, in a bowl of fruit, the oranges can go at the bottom, in the middle, or on the top without chang- ing the essential nature of the collection of fruit.

However, in a system such as a company, imagine what would happen if the parts shifted around randomly—if, for instance, the accounting spe- cialists suddenly decided to work on the production line, and the produc- tion specialists decided to write marketing copy. Of course, people do change jobs within their companies, but only after training and much tran- sition time. Most companies function best when people are working in jobs

4 SECTION 1 ➤ What Are Systems?

that match their skills and experience, and when the staff is organized according to a specific plan.

3. Systems have specific purposes within larger systems.

All systems have a specific purpose in relationship to the larger system in which they’re embedded, as we saw in the examples of the R&D department and human circulatory system above. Because each system has its own pur- pose, each is a discrete entity and has a kind of integrity that holds it together. In other words, you can’t force two or more systems together and get a new, single, larger system. Nor can you subdivide a system and auto- matically end up with two smaller identical, functioning systems.

As the saying goes, if you divide an elephant in half, you don’t end up with two smaller elephants. And if you put two small elephants together, you don’t have a new, single, larger elephant (although some day you may end up with a new system—known as a herd!).

4. Systems maintain their stability through fluctuations and adjustments.

Left to themselves, systems seek to maintain their stability. Your organiza- tion does its best to maintain a designated profit margin just as most human bodies work to maintain a temperature of about 98.6 degrees Fahrenheit. If you examined your organization’s revenues against expendi- tures every week or graphed your body temperature every five minutes, you would probably draw a wobbly, fluctuating line that nevertheless holds steady overall. Margins appear and disappear as a company pays its suppli- ers and collects checks from customers. Your body temperature rises and falls depending on your mood and your level of physical exertion. On aver- age, however, your body temperature remains stable. And, with reasonable management and no cataclysmic change, your organization’s margin also remains stable overall.

Systems achieve this stability through the interactions, feedback, and adjustments that continually circulate among the system parts, and between the system and its environment. Let’s say a corporation receives an unusually large stack of suppliers’ invoices (external stimuli) in the mail. The accounts payable department responds by paying the bills. As the checks go out the door, the accounting department, alarmed, compares rev- enue versus expenditures and gives feedback to management: Expenditures are up and revenues aren’t covering them. Management then adjusts the system by reminding key customers to pay overdue invoices. Similarly, if you go for a run, your exertion warms your body. The sensation of heat is fed back to your sweat glands, which begin to work. Over time, sweating readjusts your temperature back to the norm.

5. Systems have feedback.

Feedback is the transmission and return of information. For example, imag- ine that you are steering your car into a curve. If you turn too sharply, you receive visual cues and internal sensations that inform you that you are turning too much. You then make adjustments to correct the degree of your turn. The most important feature of feedback is that it provides the catalyst for a change in behavior.

What Is a System? 5

A system has feedback within itself. But because all systems are part of larger systems, a system also has feedback between itself and external sys- tems. In some systems, the feedback and adjustment processes happen so quickly that it is relatively easy for an observer to follow. In other systems, it may take a long time before the feedback is returned, so an observer would have trouble identifying the action that prompted the feedback. For example, if you sunbathed a lot in your teens, you may develop skin prob- lems after age 40—but because so much time passed between the two events, you may not recognize the connection between them.

Finally, feedback is not necessarily transmitted and returned through the same system component—or even through the same system. It may travel through several intervening components within the system first, or return from an external system, before finally arriving again at the component where it started.

For instance, imagine that the company you work for is suffering finan- cially and decides to lay off 20 percent of the work force. That quarter, the layoff does indeed improve the looks of the financial bottom line. On this basis, the upper management might decide that layoffs are a reliable way to improve the financial picture.

However, let’s say you survived the layoff; how would you describe your state of mind and that of your other remaining colleagues? Besides cutting costs, layoffs are also famous for damaging morale and driving people to “jump ship” in search of more secure waters. Eventually, as low morale per- sists, you and your colleagues might start coming to work late and leaving early, and caring less and less about the quality of your work. Productivity could drop. In addition, everyone who leaves—whether voluntarily or by being laid off—takes valuable skills and experience with them, so the over- all capability of the work force goes down, further hurting productivity. Lowered productivity leads to expensive mistakes and lost sales from dis- gruntled customers. All this eats away even more at the company’s revenue, tempting management to think about having even more layoffs to cut costs.

In this example, the feedback that made layoffs look like good policy was returned quickly—probably within one quarter. The feedback about the long-term costs of layoffs went through more steps and took a lot longer to return. Yet this information was essential for the management team to see the full impact of their decisions.


In reading all this information, you may be wondering what actually gives rise to systems. Systems are built on structures that leave evidence of their presence, like fingerprints or tire marks, even if you can’t see them. But what is structure, exactly? The concept is difficult to describe. In simplest terms, structure is the overall way in which the system components are interrelated—the organization of a system. Because structure is defined by the interrelationships of a system’s parts, and not the parts themselves, struc- ture is invisible. (As we’ll see later, however, there are ways to draw our understanding of a system’s structure.)

6 SECTION 1 ➤ What Are Systems?

F I G U R E 1 . 2

The Events / Patterns / Structure Pyram




Why is it important to understand a system’s structure? Because it’s sys- tem structure that gives rise to—that explains—all the events and trends that we see happening in the world around us.

Perhaps the best way to grasp the role of structure is to explore the Events / Patterns / Structure pyramid, shown in Figure 1.2.


We live in an event-focused society (Figure 1.3, “The Tip of the Pyramid”). A fire breaks out in the neighborhood; a project misses a deadline; a machine breaks down. We tend to focus on events rather than think about their causes or how they fit into a larger pattern. This isn’t surprising; in our evolutionary development as a species, this ability to respond to immediate events ensured our very survival.

But focusing on events is like wearing blinders: You can only react to each new event rather than anticipate and shape them. What’s more, solu- tions designed at the event level tend to be short lived. Most important, they do nothing to alter the fundamental structure that caused that event. For example, if a building is burning, you would want local firefighters to react by putting out the fire. This is a necessary and essential action. How- ever, if it is the only action ever taken, it is inadequate from a systems think- ing perspective. Why? Because it has solved the immediate problem but hasn’t changed the underlying structure that caused the fire, such as inad- equate building codes, lack of sprinkler systems, and so on.

By uncovering the elusive systemic structure that drives events, you can begin identifying higher-leverage actions. The next step to comprehending systemic structure is to move from thinking at the event level to thinking at the pattern level.


Whereas events are like a snapshot, a picture of a single moment in time, patterns let us understand reality at a deeper level (Figure 1.4, “Moving from Events to Patterns”). Patterns are trends, or changes in events over


F I G U R E 1 . 3

The Tip of the Pyramid


F I G U R E 1 . 4

Moving from Events to Patterns



Events, Patterns, Structure 7

F I G U R E 1 . 5

Graphs of Patterns

Time Time Time

N um

be r o

f E ng

in ee

rs L

ea vi



% P

ro je

ct s

O ve

r B ud

ge t


To ta

l S al



F I G U R E 1 . 6

The Complete Pyramid




time. Whenever you see a pattern of events—for example, sales have been declining over the past few years, or two-thirds of the department’s projects have gone over budget in the last year, or several senior engineers have left the company recently, most of them in the last six months—you’re getting one step closer to grasping the systemic structure driving that pattern.

In each of the above examples, you could draw a simple graph to repre- sent the trend (Figure 1.5, “Graphs of Patterns”).

What is the advantage of thinking at the pattern level, as opposed to the event level? Detecting a pattern helps you put the most recent event in the context of other, similar events. The spotlight is then taken off the specific event, and you can focus on exploring how the series of events are related and begin thinking about what caused them. In the end, to anticipate events and ultimately change a pattern, you need to shift your thinking one more time: to the level of structure (Figure 1.6, “The Complete Pyramid”).


To move to this deeper level of understanding, let’s reconsider the above example of the senior engineers’ exodus. You might begin digging for the structure behind this pattern by asking, “What’s causing more and more senior engineers to leave?” In this case, suppose a change in corporate pol- icy has cut both the budget and the number of administrative assistants for the engineering group. The engineers’ workloads have ballooned, and they’ve begun grumbling more and more about their job pressure. Worse yet, as some of them leave, those left behind get even more upset as their workloads expand further. It’s a vicious cycle that you might sketch as shown in Figure 1.7, “The Engineering Exodus,” p. 8.

As we saw earlier, all systems are systems, so it’s impossible to capt entirety on paper. Nevertheless, t depicting parts of systems in a dia glimpse how a system works and its behavior. One such way is to c diagram, or CLD. (Figure 1.7 is an kind of drawing.) These diagrams place for discussing and thinking events or patterns, and for openin addressing problems differently. I help you gain insight into system they identify ways you might cha behavior. After all, it is changes m level, rather than at the pattern o often prove to be the most long-l sustaining.

It’s important to remember, h representations of systems are jus

8 SECTION 1 ➤ What Are Systems?

Whenever we ask questions like, “Why is this pattern happening?” or “What’s causing these events?” we are probing at structure. Thinking at the structural level means thinking in terms of causal connections. It is the struc- tural level that holds the key to lasting, high-leverage change. Let’s return to our example about a house catching fire, to see how this works. To fight fires at the event level, you would simply react to quell the fire as soon as possible after it broke out. You would probably then repair any smoke and water damage, and put the incident out of your mind.

Engineers’ Workloads


Engineers Leaving