1994 INTERNATIONAL SYSTEM DYNAMICS CONFERENCE
Competitive Simulations: Introducing Corporations to Organizational
Learning
Gregory Hennessy
Monitor Company
25 First Street
Cambridge, MA 02141, USA
E-mail: gregory_hennessy @ notes.monitor.com
Abstract
System icists usually view si as a means to help them understand the counter-intuitive
behaviors commonly exhibited by complex systems and to facilitate others’ learning about these
behaviors. However, a second and closely-related application is typically overlooked: the elicitation
of barriers to organizational learning in a i i In our experience, when
clients are placed on teams and p with a simulated future, the izational i
routines which prevent effective decision-making in the real world soon arise in the simulation
setting. A skilled facilitator can then begin to address these behavior patterns in the (relatively) safe
context of the simulation.
Thus, we see tremendous power in the i ion of the applications of si i hnology. First,
the dynamic model used to build a simulation helps both the modeler and the client understand
systemic behavior -- which leads to better strategic choices. Second, the addressing of organizational
defensive routines promotes learning and the conversion of such choices into timely action.
Microworlds, page 31
1994 INTERNATIONAL SYSTEM DYNAMICS CONFERENCE
Competitive Simulati Introducing Corporations to Organisational
Learning
Four years having passed since Senge’s The Fifth Discipline hit the bookstores, it should come as
no surprise that word of system dynamics has spread as far as the executive suites of many
Fortune 500 firms (and their privately held equivalents). Of course, since the field still lacks a
comprehensive and authoritative text describing the state of the science, it should also come as
no surprise that a fair amount of confusion remains regarding what exactly is meant by the term
system dynamics.
Although it is b ing i ingly to find with a strong interest and
ble background in system ics, such are typically, for now anyway, in the
planning or economic analysis groups -- not in the executive suites. Often, those senior
managers and executives who have heard about system dynamics misunderstand it. Systems
dynamics is thought to be some new-agey-kind-of-oneness-thing, more closely associated with
the Gaia hypothesis (for example, see Myers) than with control theory!. They often think the
advantage of simulation has more to do with seeing “the numbers” in a new, more interesting way
than with changing the way they think about the numbers. That is, they see simulation as a way
to achieve single-loop learning, rather than as an opportunity to experience double-loop learning.
But clearly, one of the chief benefits of the simulation experience is the opportunity to learn
about the systemic structure and processes which give rise to the numbers.
Is there a way to make the insights we have to offer real for senior managers and executives?
Traditional hods (education-based, group-model-building, from-the-ground-up efforts) are
labor- and time-intensive, and are an excellent approach, but only for managers with the insight
to see the value of such an approach and with the commitment to see such projects through. But
what about more reticent managers? Can we find a way to help them? After all, those most in
need of help are often the ones most ingrained in their old ways. At Monitor, we believe we have
found a way. Through Competiti imulation, we introduce senior and ‘ives to
system dynamics and organizational learning.
The Task: Not Only Better Decision Making - But Also Creating Action
It is our mission as consultants to convert informed choice into timely action. The first
objective is to help executives to make informed choices based on an understanding of the
dynamic situation they face. Monitor established itself as the firm most capable of putting into
practice Porter’s strategic analysis (Porter 1980, 1985). These frameworks have helped
d d their petitive position and make better business decisions. Today they
are standard practice, and we are looking for ways to provide superior insight; system dynamics is
one of the areas in which we are most interested.
The second objective is to help executives to act in a timely manner and in a way consistent
with their new thinking. As Argyris has pointed out for years (see Argyris 1992) and as will be
briefly described later, d ive routines and izational barriers often dictate a dysfunctional
pattern of managerial behavior, even in light of new, more informed decision-making power.
Before we can help them act in a timely manner, we must help them learn to act appropriately in
the first place.
1Not that either is right or wrong, just that perception is often at the “mystical” end of the
spectrum.
Microworlds, page 32
1994 INTERNATIONAL SYSTEM DYNAMICS CONFERENCE
Our Competitive Simulation product helps us to Figure 1: Taxonomy of Complexity
integrate the two tasks by presenting managers
with a simulation that is rich both in detail and
dynamic complexity (refer to Figure 1). First High System Competitive
of all, because the simulation is very closely Dynamics ‘Simulation
modeled after the executives” business, they CaN pynamic
attach th | to it lly and Complexity
emotionally. In the process of playing the
simulation they learn to make better choices
because they experience the time-compressed Low Speed Databases
feedback processes which are central to learning
about the long run consequences of decisions
made today. lw High
Detail Complexity
A second, and we believe critical, di ion to Ce itt imulations is that interact
with people, not boxes. Thus, as they wrestle with real choices they also experience the
organizational defensive routines which act as barriers to action. A trained facilitator, working
with the team, can point out their ineffective interactions and can help them understand how
they can change their behavior to act with greater effectiveness.
Ineffective Interactions
Ineffective interactions between managers prevent ideas from being discussed openly, lead to false
consensus, and ensure a management environment where new courses of action are impossible.
Exhibit 1 presents a framework which has helped me to understand the dynamics of ineffective
interactions and points to how system dynamics can be integrated with action science to convert
informed choice into timely action?.
How We Think
The Data I Select is drawn from the total Data that is available. My Beliefs and Assumptions act
as a filter to determine which data points I choose to acknowledge. Having selected some data
points and not others, the Data I Select has an important influence on How I Interpret It. My
interpretation, then, influences the Conclusions I Reach. My conclusions then become a part of
My Beliefs and Assumptions, which then affects the Data I Select next time around.
If I assume you are irresponsible, then I focus on evidence which supports my
underlying belief. Because I have a biased set of data, my interpretation and
conclusion confirm that you can’t be trusted with responsibility.
We act based upon our implicit about our envi
If I have luded you are irresponsible, I withhold ibility from you; that
is, my choice of action is dependent upon my conclusions about your ability to
perform.
WW
We each go through our thinking process: beliefs filtering the data we select, influence the
conclusions we reach, which in turn affects our beliefs. And we each act based upon our individual
conclusions about the environment we share. Our interaction, then, adds to the pool of data
available for us to select specific observations.
I, having for some time believed you to be ii ible, withhold
from you. You, having believed for some time that I don't trust you with
important issues, focus your efforts on other projects and give me only a partial
2This framework is based upon Argyris & Schén (1974) and materials developed by Actior
Microw
1994 INTERNATIONAL SYSTEM DYNAMICS CONFERENCE
effort. The result is that I am more convinced than ever that you are
irresponsible, and you are inced more than ever that I am a micromanager.
The Result: Organizational Defense Pattern (ODP)
The way we interact usually results in a pattern of defensive behavior (Argyris 1990). We bypass
discussing sensitive issues because we don’t want to start a fight, and we cover up the fact that we
passed the issue. For le, if we reach very different conclusions about a certain issue --
conclusions which differ because we have selected different data and have interpreted it
differently -- we avoid trying to have a meaningful reconciliation of our conclusions because we
each “know” the other is too closed-minded to listen to reason. Having bypassed the issue, we
then cover-up the bypass by having a perfunctory discussion of the issue, using broad statements
and generalities. Framed with sufficient generality, we reach agreement on completely
unactionable directives.
Most strategic failures result not from insufficient decision-making tools and skills, but from the
inability to discuss important issues and and from the resulting inability to achieve meaningful
change in the izati A tant who enters an ization blind to the implications of
ODP is destined to see the organization stifle any changes he or she recommends.
The Need for Integrated Interventions
The of ineffective i ions requires that, if informed choice is to be converted into
timely action, all three components of the structure be addressed simultaneously. We must help
managers change the way they think, the way the act (based upon their thinking), and (as a result
of changing the first two) the way they interact.
For an economic organization to succeed in today’s marketplace, it is not sufficient for them to
have a great strategic plan -- they might be unable to act in a manner consistent with the plan.
Nor is it sufficient for them to be able to act on just any old plan -- such a plan might move them
in the wrong strategic direction. Fuller states (1993, p. 44),
“The key objective in competition ... is to improve your organization’s
per along these di i
* to generate better information than your rivals do
* to analyze that information and make sound choices
* to make those choices quickly
* to convert strategic choices into decisive action”
In order for an organization to be successful, it must do all of these things well. Piecemeal
pp h iated with traditional strategic ting, system dy ics, process Itati
and action science just won’t do the job.
Traditional Approaches
Traditional approaches fail to provide the integration of informed choice and timely action
Strategy consulting, system dynamics, and action science & process consultation approaches arc
considered in turn.
Traditional Strategy Consulting Intervention
Historically, strategy consultants have worked to affect the How We Think section of the system
Specifically, consultants have sold their ability to help managers: select better data, interpret i
better, and reach better conclusions. The projects under analysis might involve which markets t
serve, how to organize to face those markets, or any number of other common themes. Bu
fundamentally, they focus on the “informed choice” part of the problem.
Microworlds, page 34
1994 INTERNATIONAL SYSTEM DYNAMICS CONFERENCE
More recently, “implementation” has been the for Iting firms. All
want to see their recommendations actually come to life within the client organization. But
standard approaches to implementation end up doing one of the following:
* consultants work side-by-side with existing middle- and line-management for
an extended period of time to make sure the letter of the new strategic law is
followed, or
* senior management replaces the existing middle- and line-management with a
new set of managers “more suited” to working under the new strategy
While it is to de that such i are addressing How We Act and How We
Interact, they are byp g the fund 1 ional issues which caused inaction in the
first place (Argyris 1990). ” Once the bisgeeter have left, the old, ineffective behaviors will
resurface (though perhaps in a new form), and the or once again unprepared for the
next strategic challenge, searches for strategic advisors.
Thus, strategy consulting as commonly practiced, cannot help managers learn to convert
informed choice into timely action.
System Dynamics
System d ddi some of the deficiencies of traditional strategy ting and
complements the value provided by traditional strategic advice. Like traditional strategy
system d: helps select better data, interpret it better, and reach
better Lusi But the ad ge of system d is more than just better data,
interpretations, and conclusions; it is that managers learn to question the implicit beliefs and
assumptions which filter the data they select in the first place. The systems view completely
changes the way they think about their problems. Thus, the strength of system dynamics lies in
its ability to improve How We Think.
Many system dynamicists have migrated into Action Science in the pursuit of changing How We
Act.3 Those that have are likely to present an integrated approach similar to the one discussed
here. However. system dynamics purists are still hesitant to take that step and are, as a result,
being less effective at achieving change than they might otherwise be.
Interventions based upon process consultation and action science (see Schein and Argyris) focus
on How We Think (about each other), How We Act and How We Interact. They don’t address How
We Think with respect to the strategic challenges organizations face. They are sound methods for
achieving change in the client. but rely upon the client’s internal ability to know what to change
into or where to move strategically. In some cases this reliance may be justified, but often,
external strategic advisors can provide valuable persp and
Thus, no single traditional approach offers a full solution. What i is needed i is a combination of
strategy consulting. system dynamics and action science. C lation is a technol
which supports an intervention based upon the three schools of thought.
Converting Informed Choice Into Timely Action
How does an organization begin to do all of these things? Fuller (1993, p. 45) suggests that
companies need to improve performance in three related areas:
* gather better information -- that is. information that is dynamic, that cuts
across organizational boundaries. and that exists “in real time”
* establishing a framework for making decisions -- that is, creating a business
version of military doctrine
* practicing the integration of the pieces -- that is, learning to use competitive
simulation or business “war games”
3For example, see Rufat-Latre (1993), Rufat-Latre, et. al. (1993), and Senge (1990)
Microworlds, page 35
1994 INTERNATIONAL SYSTEM DYNAMICS CONFERENCE
In other words, companies wishing to succeed need to understand: data (detail complexity),
doctrine, and d lexity (system d: ics, systems thinking).
Detail and dynamic complexity are familiar terms to most system dynamicists, but what about
“doctrine”? Why is it essential to learning how to integrate informed choice with timely action?
Relati
Doctrine is an established framework for making decisions; it is not an operations manual or a
detailed list of IF ... THEN ... ELSE statements. In the military, it is what allows a platoon
leader to act without having to wait for orders in each new circumstance. In business, doctrine
allows a manager to make decisions and then act in a timely fashion whenever new information
arrives. It is not minutely prescriptive, but rather a mixture of philosophy and practice, and the
connections between the two. Thus, an organization possessing an effective doctrine sees every
employee contributing to the strategic advancement of the organization each time a decision is
made and an action is taken.
Doctrine has an important role i in the learning organization. Senge & Sterman (1992) warn that,
“local decision making and i lead to anarchy unless managers
account for the interconnections and long-term side-effects of their local decisions. Laudable
goals such as ‘empowering’ and ‘enabling’ individuals often prove counterproductive unless
managers can act locally and think globally.” Substitute “platoon leader” for “manager” and you
can see that this is fundamentally the same problem that the military addresses through its
doctrine.
Specifically, doctrine i is related to, but aistinet from “Shared Vision” (Senge, 1990). Shared Vision
is described as a “ eld by of the it is “a Vision that
many people are truly committed to, because it reflects ‘their own personal vision.” ” (p. 206).
Once the organization possesses a Shared Vision, it needs a framework for turning the vision into
action -- such a framework is its doctrine.
It is easy to worry that doctrine only ion in a set d ing
framework which later serves to hamper change. Indeed, Martin (1993), while identifying il
stages that troubled companies ve typically experienced, discusses the role of “Steering
isms” in such pany’s Steering Mech: -- it’s doctrine -- keep the
company on track. While they are anal d loped “to keep the ization aligned with the
founder’s vision and to keep the vision aligned with the economic environment,” before long the
economic times have changed and the mechanisms are interfering with the organization’s ability
to adapt to customer needs. The resulting Disrupted Feedback leads managers down the wrong
path, asking the wrong questions. “Digital Equipment Corporation, for example, gathered
on what ‘s wanted from its proprietary word-processing software
without realizing that the age of proprietary minicomputers was over.”
However, identifying a business's existing: | ne dated and counter-productive doctrine and
g it can be an imp ing a troubled izati Toffler
(1993) describes convincingly how the U s. iis establishment, following the Vietnam war,
completely reinvented itself (to use the language of business) in a concerted effort to create a
“Third- Wave” military doctrine. That the prototype of modern, hierarchical, American
ions could lish such a drastic haul says a lot about the power of recalibrating
ks of deci: king.
Of course, it would be inadv isable tol institute a framework for local decision-making without first
testing its rob Cc aids in the development of doctrine by providing
managers with a practice field to find. out what works and what doesn’t.
‘Together with the previously ioned decisi king envi that i is tich in detail and
ity, Competitive ion helps ization: d its data,
Microworlds, page 36
1994 INTERNATIONAL SYSTEM DYNAMICS CONFERENCE
and doctrine. As a result, the organization becomes better at converting informed choice into
timely action.
Competitive Simulation
“Strategy ... is an interactive dynamic .. . [and] competitive simulation provides
an opportunity to practice that i It trains to
anticipate the unexpected. It helps to build a management team that can roll with
the punches, deal with unanticipated situations, and work together when things
don’t go according to plan.” (Fuller, 1993)
Competitive Simulations are an intervention which allows Monitor to integrate informed choice
and timely action and to introduce the concept of doctrine. The intervention relies on two
things. First, a dynamic strategy simulation to provide managers with a basis to improve their
decision-making. Second, facilitators trained in action science who can help managers address
organizational barriers as they arise during the simulation. The intervention is best described by
example.
A Case Study4
Consider a key business unit in one large high-technology company that recently completed a
Competitive Simulation. Fifteen people on three continents worked for four months simply to
“map the battlefield.” Their preliminary work produced a rough-cut description of the
competitive shape of the industry, the assets of the competitors, and their company’s own assets.
It took another two months to turn that rough map into a dynamic computer model in which the
industry evolved in plausible ways and strategic choices produced credible results.
The parameters of the game and the structure of the exercise were straightforward. The
simulation covered a seven-year time horizon during which two teams -- a blue team and a green
team, both rep ig the - peted against seven rivals across 15 product lines.
The game progressed in four | strategic moves: the first move represented one year of
competition, each of the next three moves represented two years. To make a move, each of the
two teams made choices in four categories: business unit choices such as pricing and advertising;
infrastructure choices such as deploying new technologies; strategic choices such as acquisitions
and alliances; and corporate-level choices such as issuing debt or stock. In all, each team
controlled roughly 300 mathematical inputs with as many as 3,000 independent variables that
represented their choices in allocating company resources.
That was the part of the game the teams saw; behind the scenes, game referees and “control”
converted their choices into financial results and market share consequences using software
programmed with 40 pages of algorithms and more than 1 million data points -- and seven NeXT
workstations crunching the data.
To add even more credibility to the simulation, during each move the teams received extensive
factual information on their performance: accurate financial and market data broken out by
product category, corporate income statements, balance sheets, cash flows, and key performance
ratios.
The teams also Teceived information on what their rivals were doing. But here, in keeping with
the “fog of war,” the information was limited, lly sketchy, and ilable only after a
time lag. The simulation included updates on new products, cost-reduction opportunities,
acquisition candidates, and an industry newsletter that carried news, gossip, and rumors concerning
major trends. Just like in real newsletters, the articles were a mixture of accurate reports and
erroneous items. Just like in real business, the teams had to, discern fact from fiction, information
from misinformation.
This section draws heavily upon Fuller (1993).
Microworlds, page 37
1994 INTERNATIONAL SYSTEM DYNAMICS CONFERENCE
As the game neared its end, all the substantive and strategic issues were on the table. At this
point, the teams were testing themselves -- and being tested -- on whether they had developed a
consistent and coherent view of the world. Had they, in fact, fashioned a strategy that could fit
rapidly changing circumstances? Had they consistently executed it? Could they convert
informed choice into timely action?
Two different styles of teamwork emerged. The blue team worked as one group to develop its
arena-by-arena decisions. The green team broke into smaller groups, each of which focused on
different sets of choices. One team was reflective and contemplative in its work style; the other
was boisterous and aggressive. One thing was common; both teams had driven themselves into
the dard or routines which afflicted them in their “real life”
interactions. A trained facilitator pointed out their ineffective behavior and worked with them to
improve their team interaction and effectiveness.
After the game, the two teams came together to Jearn the results and reflect on the experience.
In fact, the of the d the larger point of the exercise: the value of
competitive simulation is not in the computer’s output but in the players’ input. There was no
clear winner. The teams had settled on vastly different strategic directions. The green team had
developed a shrewd strategy for working with competitors who were also valued customers. The
blue team had focused on heavy investments in new technologies and future growth opportunities.
The results: the green team won based on stock price, the blue team won based on other financial
measurements. Neither strategy was ultimately “more right” than the other.
The investment in the simulation produced a fluid, dynamic, credible process backed by
d, powerful technology. But the real lessons of the war game were only partly -- and
least signifi icantly -- quantitative. The real value was interp 1 and organi: I
exposed the pany politics and undi: ble human issues that exist in every large-
scale organization. The players learned much about strategy; the learned even more about
teamwork.
The Challenge - Entry Strategies for Reaching Executives
Our goal in using C i imulati is a very aggressive one: help senior managers learn to
convert informed choice into timely action. To do so requires an understanding of traditional
strategic analysis, of system dynamics, and of action science. Each of these areas is important
and each presents organizations with an opportunity to leverage the organizations’ existing
strengths.
However, companies are so far from integrating these pieces that an even greater opportunity
exists in helping them to do all three The ion is that while the goal is
, the obj d with each particular skill area can be modest. We are noi
looking to turn clients into master strategists; rather, we help them to increase incrementally the
strategic thinking skills which they have already acquired. We are not looking to turn clients intc
system rather, we the importance of a few simple, first-order feedback
processes. We are not looking to turn clients into Chris Argyris; rather, we start them thinking
about the ineffective structures for teamwork which they themselves create.
Many senior and ives aren’t ii d in talking about or learning about systen
dynamics, action science, or for that matter “integration.” They are interested in talking anc
learning about things that will help them run their business, today.
Competitive Simulation is geared toward managers with this “bottom-line” focus. If they won’
listen to abstract lessons, customize the simulation to look like their world. If they have nc
interest in great leaps of insight, focus on providing small, manageable lessons. In this way
Competitive Simulation performs the important function of introducing system dynamics an
Microworlds, page 38
1994 INTERNATIONAL SYSTEM DYNAMICS CONFERENCE
organizational learning to managers who might not otherwise be receptive to these concepts.
Maybe they will even learn more than they bargained for.
Note:
Mark Fuller, CEO of Monitor Company, inspired the selection of this paper’s topic and outlined
many of the themes presented here. John McClellan, also of Monitor Company, provided
valuable suggestions and feedback during the writing of this paper.
While at Monitor Company, Greg Hennessy has researched the application of system dynamics
and gy in strategy and has par d in projects within the
I services and tele ions industries. Prior to joining Monitor, he received the
Master's degree from MIT’s Sloan School of Management and the M.S. degree in Social Science
from the California Institute of Technology.
References
Argyris, C. 1990. Overcoming Organizational Defenses: Boston, MA: Allyn & Bacon.
Argyris, C. 1992. On Organizational Learning. Cambridge, MA: Blackwell
Argyris, C. & D. Schin. 1974. Theory in Practice. San Francisco: Jossey-Bass.
Fuller, M. 1993. Business as War. Fast Company November, 1993.
Martin, R. 1993. Changing the Mind of the Corporati Harvard Busii Review Ne t
December 1993.
Myers, N. 1984. Gaia: An Atlas of Planet Management New York: Doubleday.
Porter, M. 1980. Competitive Strategy: Techniques for Analyzing Industries and Competitors.
New York: Free Press.
Porter, M. 1985. Competitive Advantage: Creating and Sustaining Superior Performance New
York: Free Press.
Rufat-Latre, J. 1993. An Intructional Approach to Learn from Interpersonal Interactions.
Master’s Thesis: MIT Sloan School of Management.
Rufat-Latre, J., M. Jamieson, M. Mora. 1993. Transferring Systems Thinking and Circumscribing
Problems: A Case Study. System Dynamics 1993.
Senge, P. 1990. The Fifth Discipline: The Art & Practice of The Learning Organization. New
York: Doubleday.
Senge, P. & J. Sterman. 1992. Systems Thinking and Organizational Learning: Acting Locally
and Thinking Globally in the Organization of the Future. European Journal of Operational
Research 59 pp. 137-150.
Toffler, A. & H. Toffler. 1993. War and Anit-War. Boston, MA: Little, Brown and Company.
Microworlds, page 39
1994 INTERNATIONAL SYSTEM DYNAMICS CONFERENCE
[Exhibit 1: a Fi for ing System Di ics and O izati Leaming
= =
ee
How | Interpret It How You Interpret It
Microworlds. page 40
