A Learning-Based Approach to Leading Change · The PricewaterhouseCoopers Endowment for The Business of Government A Learning-Based Approach to Leading Change Barry Sugarman Lesley

The PricewaterhouseCoopers Endowment for

The Business of Government

A Learning-Based Approach

to Leading Change

Barry SugarmanLesley UniversityandSociety for Organizational Learning

G r a n t R e p o r t

December 2000

About The EndowmentThrough grants for Research and Thought Leadership Forums,The PricewaterhouseCoopers Endowment for The Business ofGovernment stimulates research and facilitates discussion onnew approaches to improving the effectiveness of governmentat the federal, state, local, and international levels.

Founded in 1998 by PricewaterhouseCoopers, The Endowmentis one of the ways that PricewaterhouseCoopers seeks toadvance knowledge on how to improve public sector effec-tiveness. The PricewaterhouseCoopers Endowment focuses on the future of the operation and management of the publicsector.



A Learning-Based Approach to Leading Change 1

A Learning-Based Approach

to Leading Change

Barry SugarmanLesley University

andSociety for Organizational Learning

December 2000

2 A Learning-Based Approach to Leading Change

TABLE OF CONTENTS

Foreword ......................................................................................3

Executive Summary ......................................................................4

Introduction ................................................................................5

What Is a “Learning Organization”? ............................................8

A Learning-Based Change Model................................................14

Case Studies ..............................................................................27The SoL-Federal Government Partnership ............................27Case: Office of Special Education/

Department of Education ..............................................28Case: Office of Surveillance and Biometrics/

Food and Drug Administration ......................................32Case: Environmental Protection Agency

Regional Office ..............................................................35

References ..................................................................................42

About the Author........................................................................43

Key Contact Information ............................................................44


ForewordDecember 2000

On behalf of The PricewaterhouseCoopers Endowment for The Business of Government, we are pleased topresent this report by Barry Sugarman, “A Learning-Based Approach to Leading Change.”

The new economy has challenged organizations in both the private and public sector to shift from older,more traditional organizational models of bureaucracy to new, more flexible models. The government sec-tor has experienced major change in terms of significantly increased demands from its “customers.” The traditional model of bureaucratic organization may no longer be sufficient to meet the demands forimproved service and greater efficiencies in large-scale programs. These changes demand a new model —that model, posits Professor Sugarman, is the learning organization.

The learning organization is one that is “continually improving its ability to be more effective in meetinggoals that are important to its members,” and capable of reinventing itself when necessary. To move anorganization towards becoming a learning organization, Professor Sugarman presents the “Learning-BasedChange Model.” This model takes the approach of introducing new ideas, then nurturing and protectingthem while they grow stronger. It relies to a great extent on the power of the grass roots rather than onchange from the top down.

Professor Sugarman presents three case studies of ongoing learning-based change initiatives in three differ-ent federal government agencies. These organizations were working in partnership with the Society forOrganizational Learning to pilot a new model of learning-based change.

It is our hope that this report will prove thought-provoking and helpful to senior executives, middle man-agers, supervisors, work group leaders, human resource development staff, and others who are involved in organizational change efforts.

Paul Lawrence Ian LittmanPartner, PricewaterhouseCoopers Partner, PricewaterhouseCoopersCo-Chair, Endowment Advisory Board Co-Chair, Endowment Advisory [email protected] [email protected]




This report presents a brief version of the “learning-based approach” to organizational change leader-ship, based on the ideas of the learning organization.For the practical executive and change leader atany level, it presents 14 key elements that are necessary for planting the seeds of a learning orga-nization and helping an organization to shift in that direction.

The new model and new thinking that are involvedin this approach are illustrated concretely throughthree case studies. They are ongoing learning-basedchange initiatives in three different federal govern-ment agencies. Leaders within five federal agenciesbegan a partnership with the Society for Organiza-tional Learning (SoL) to try out this approach. Inter-views and observations were conducted in order to understand better the process of change in these agencies.

These cases represent some early results from organizational change initiatives that have not yetachieved full fruition. Unlike many other reports of change projects, these cases were not selectedas the pick of the crop of proven successes. This is an interim report and think piece based on anexperiment still in progress. These cases offer thechance to observe, reflect upon, and analyze some of the complexities of guided organizationalchange in government settings.

Executive Summary


The concepts and management tools of “the learning organization” and “the learning-basedapproach to change leadership” have been found tobe very useful for better understanding and improv-ing the ways that organizations change themselves.Today all organizations face unprecedented levelsof demand for better results — in the government,as much as anywhere else — for new levels of ser-vice and response, for greater efficiency to producemore with less resources, and to take advantage ofnew opportunities, such as the Internet. While mostof our experience so far with the learning-basedapproach to change management comes mainlyfrom the business sector, it is important that we alsoevaluate this approach in the public sector. That iswhat this report begins to do through several casestudies in the U.S. federal government.

Leaders within several federal agencies began apartnership with the Society for OrganizationalLearning (SoL) to try this approach, starting justover a year ago. That came about as the result of anearlier four-year partnership between SoL and theNew England Regional Office of the EnvironmentalProtection Agency (EPA). Interviews and observa-tions were conducted in order to understand theprocess of change in these agencies. These data areanalyzed in the light of our experience with earlierchange initiatives in the business world, sponsoredby SOL and its predecessor, the MIT Center forOrganizational Learning.

The information presented here represents earlyresults from organizational change initiatives thatare expected to take much longer to reach full

fruition. Unlike many other reports of change proj-ects, these cases were not selected as the pick ofthe crop of proven successes. This is an interimreport and think piece based on an experiment still in process.

Demands for Change in the GovernmentThere have been significant changes (improve-ments) in the management of government agenciesover the past decade, and there continue to beenormous pressures for these agencies to changemore for a variety of reasons:

• The gap between ever-increasing demands andrestricted resources constantly expands as theconsumer revolution creates demands forhigher standards of service to citizens.

• Changes in technology and other external fac-tors create new threats and opportunities.

• Industry groups and others who are subject to government regulation have become moredemanding, and elected officials have becomemore receptive to their demands.

• Meanwhile, there is new thinking about therespective roles of federal, state, and local gov-ernments; corporations and industry groups;and private, nonprofit service and advocacygroups. For example, there are notions of part-nership between government and privategroups, as well as between levels of govern-ment, in many areas (such as environmentalprotection; regulating the safety of food, drugs,

Introduction


and medical devices; and ensuring that ser-vices are provided to disabled children andtheir families) which require great change inhow government managers define and conducttheir work.

• Lastly, the vast changes in our world (manybased on new technology) present never-end-ing new challenges to government — geneticresearch and new questions of ethics and regu-lation, the AIDS epidemic, the economic stresson all health-care systems, protection of theinfrastructure of the Internet now that both eco-nomic and national security systems dependon it. In short, the demand for more change ingovernment will only get greater.

Basic Ideas About ChangeLeadershipIn spite of the huge and relatively recent demandfor more change in government, the role of changeleader is unfamiliar to many public managers andmany have serious misconceptions about it.People’s ideas and assumptions about change andleadership are very important because they greatlyaffect how people act (or refuse to act) as leadersand as followers who shape what leaders canaccomplish. One deeply rooted assumption is thebelief that change depends on some rare, extraordi-nary kind of individual who calls to us from a highposition in the organization (“the hero on the whitehorse”) and that we must wait for the arrival of thischarismatic senior change leader before we canmake any contribution to change. This idea ishighly questionable and may be very harmful. Adifferent view is the idea of the learning organiza-tion and the learning-based approach to changemanagement, which suggest a different view of thechange process in organizations. This report isoffered as a brief introduction to that view.

This introduction is intended to be relevant anduseful to several groups of readers, not just tosenior executives but also to middle managers,supervisors, work group leaders (including officialleaders), human resource development staff, indi-vidual contributors who are frustrated by the limitsthe organization places on their talents, and to“live wires” of all kinds. Hidden (or perhaps nothidden) within government agencies are plenty ofpeople with good ideas and the ability to develop

significant improvements, but there are serious bar-riers to these potential change agents being heardand working together. Therefore, the role of thechange leader is basically to encourage them andto remove those barriers.

Lucky senior executives may find members of theiragencies coming forward with proposals for changethat they want to implement, just asking for a littlesupport; but more often the senior executive whowants to promote meaningful change will need tosignal to staff that such efforts will be appropriatelysupported. What is “appropriate”? What does ittake to create a change initiative? That is the focusof this report. At a very introductory level we shallpresent some of the necessary elements needed ina learning-based organizational change initiative.

Important ideas are often truly simple at their core,but they get overlaid by related but non-essentialelements. In this case the central idea is the follow-ing: There is sufficient talent and energy at thegrassroots level of most organizations to make sig-nificant improvements, but long experience withorganizational barriers has led those people to be frustrated and discouraged, which then leadsthem to develop beliefs that change is impossible in this agency.

It is impossible, they believe, because “the others”(their colleagues and bosses) would not support orallow it. “They” appear to be major barriers. Butoriginally, they were just like the would-be innova-tors, full of ideas and energy to make things better.Over time, many have become trapped in a viciouscycle of disappointment, in which we refuse to gethurt anymore, so we protect ourselves behinddefensive beliefs that cloak the fact that we careabout making things better. This creates a cycle inwhich we each confirm the fears of our colleagues(“they would not stick their necks out to support anew idea”), while we all retreat into defensive cyni-cism. In the business world the competitive market-place tends to blast employees out of that situation,with big rewards for those who succeed and jobloss for those who fail.

The Public Sector’s AdvantageThe public sector faces a big challenge to meetvastly increased expectations with tighter resourcesand with some significant structural barriers not


faced in the private sector. Yet the public sectoralso has one special asset — its public, common,patriotic purpose. It exists to create public value —for example, safe food and drugs, security againstcrime and external enemies, clean water, safe airtravel, a banking system and financial institutionswith integrity. This work is essential to the founda-tions of our civic society and to the survival of thenation. It is work to be proud of, and many govern-ment workers choose it for that reason. This com-mitment to the ideals of the public good, held bymany government workers, should be a big asset to those who wish to light more fires of change in government agencies. For the learning-basedapproach believes that one crucial element in lead-ing change is to rediscover how much we reallycare about making things better and to help othersto rediscover that they do, too.

We shall now introduce the learning-basedapproach to leading change in two parts. First, we present the idea of the “learning organization” as the ideal state towards which we are aiming toove. Then we present a very simplified version of the change process involved, the process that leadersat all levels can understand and use, which we call“the learning-based approach to leading change.”


Organizations that understand how fast the world is changing and what that means for their surviv-ability aspire to be “learning organizations.” ForPeter Senge [Senge, 1990, 1999], a “learning orga-nization” is an organization that is continuallyimproving its ability to be more effective in meeting goals that are really important to itsmembers. A one-time improvement, howeverimpressive, does not make a learning organization;it requires a state of continuous improvement andimproving the way that improvements are made. A learning organization is capable of reinventingitself when necessary.

The learning organization can also be called a“new model” organization. Either way, the differ-ence between “old model” and “new model” organizations is profound. The “old model” organi-zation operates by a basic formula that defines how to run the organization: how to obtain thenecessary inputs (including resources, licenses, and charters), how to transform them into theappropriate products and services, and how to get them into the hands of those who need them,earning appreciation or at least support for theorganization’s continued existence. That broad definition should cover both government agenciesand businesses in the free market.

The “old model” formula-driven organizationspends little time or effort on innovation orimprovement. The “new model” or “learning

organization” is the opposite. It devotes a majorpart of its efforts to discovering new ways, notpresently known to it. It seeks new formulas, know-ing that they can last only a limited time beforeneeding replacement. The way that the old modelorganization depends on its embedded formula,without giving it any thought, does not exist in thenew model, learning organization. Here there ismuch thought given to how and why things aredone a certain way, because they must beimproved on continuously, and because they mayneed to be replaced in the event that suddenchanges in the environment should make the oldway unviable. The new model, or learning organi-zation, lives in a state of permanent change.Naturally, it requires a very different kind of organi-zation and very different behavior and thinkingfrom all its members.

This contrast between old model and new model is deliberately over-simplified in order to make animportant point. We are not describing actual casesright now. In fact, these two models or types do notexist in the real world; they are mental constructswhich we may use to help us understand the realworld — or we may reject them. Most real organi-zations have some features of both types; there arevery few, if any real examples that are as simple oras extreme as the models. Yet they can be useful inhelping us to understand how organizations differin the way they function. The old and new modelsmay be considered as end points on a continuum

What Is a “LearningOrganization”?


where few examples lie at the ends, but many areclose enough to be classified as nearer to onemodel than the other.

The “New Model” or “Learning”OrganizationOrganization researchers, looking at both govern-ment and business sectors, have contrasted the “oldmodel” of modern administration, known proudlyas “bureaucracy,” with newer models of organiza-tion that have been emerging in the latter part ofthe 20th century. By contrast with the old bureau-cratic model that laid the foundations of the mod-ern state and the industrial revolution in the late19th and early 20th century, these newer models oforganization aimed to add new kinds of flexibilityonto the base of dependability which was a largepart of the legacy of the old model.

The business sector discovered its need for a “newmodel” of organization and management as itentered an era of intense global competition, theinfo-telecom revolution, and the knowledge-basedeconomy. (Burns and Stalker, Lawrence and Lorsch,Mintzberg, Ouchi, Pascale and Athos, Quinn)Whereas the old model of production and distribu-tion had enabled the Allies to win the SecondWorld War and to rebuild during the post-war era of shortage, the new economy of abundant supply, fierce global competition, empowered consumers, and more talented workers demandeda different model. That new model was, in effect,the learning organization.

The government sector experienced its own shake-up in terms of significantly changed demands fromits “customers.” The traditional model would nolonger suffice here either, in the face of new levelsof demand for improved service and efficiency inlarge-scale programs, and for new effectiveness inspecially challenging areas, such as preparing forY2K, mapping the human genome, defeating globalterrorism, exploring Mars, and reducing welfaredependency. Max Weber’s classic model of stable,honest, rule-governed bureaucracy is no longer a suitable standard for an excellent governmentagency in the new millennium. In the words of a recent commentator, “Reinventing governmentmeans more than creating a government that worksbetter and costs less. It means going beyond the

industrial-age model of governance.” It requiresnew thinking and a new model.

We may contrast the new and old models on sevenkey points. The same differences in organizationaldesign between new and old models will be foundto be important, I maintain, in both governmentand business sectors.

The basic strategy of the old model organization isreplication or mass production, always followingthe basic formula, pattern, or rule book, while thenew model is innovation as needed to meet (oranticipate) the requirements of the customer, cur-rent or anticipated.

The basic structure of the old model organizationis hierarchy, with emphasis on respect for the chainof command. At the top of the hierarchy are thefew who alone have the whole picture of the orga-nization’s strategy, so staff must depend on them forapproval of any initiative. The new model approachof structure is heterogeneous, with many projectteams. Many people (possibly all of them) have thebig picture.

The basic system of the old model organization isformalized, with explicit rules, policies, and proce-dures being used. It runs by its standard operatingprocedures, which are documented and availableto all, unlike the new model where systems aremore informal, because they are frequently evolv-ing and changing. Staff mutually adjust and coordi-nate informally (without waiting for supervisors totell them how) in the new model.

The basic style of the old model organization isone of conformity and “please the boss,” whereasthe new organization emphasizes learning and cre-ativity to please the customer above all. This issometimes described as a shift from a vertical (hier-archical) emphasis to a horizontal (customer-focused) emphasis, with the value-chain flowinghorizontally across all the components of the orga-nization as they contribute to the final result, whichis delivered to the all-important customer.

The staffing of the old model organization is basedon clarity about roles and duties (i.e., job descrip-tions are taken very seriously), whereas the neworganization is flexible about job boundaries,


allowing staff to adapt their jobs as they innovate.The new organization encourages staff to be passionate about their work, unlike the old modelbureaucracy which expected staff to be impersonaland dispassionate, leaving their personal “baggage”at home.

The skills model of the old type organization wasto have narrowly specialized roles, but the newmodel organization prefers skill clusters and cross-training of staff so they can be more versatile andsupport a flexible system, with workers switchingtheir duties as the need arises. Therefore, the num-ber of different job categories tends to be reducedin the new model.

Finally we compare the old and new models interms of their purpose or superordinate goals. Inthe old model these ultimate goals are different forthe public and private sectors. In the private (busi-ness) sector the old model goals are to create prof-its and rising share values for investors, while ingovernment the old model goals are to administerprograms and policies authorized by the legislatureand created by senior officials of the executivebranch. Because the thinking of the new modelabout organizational purpose is quite different, it ispossible to combine both public and private underone set of superordinate goals: to fulfill a sharedvision of the organization’s mission through cre-ative work in a community of employees, partners,and stakeholders (see Figure 1).

This view of the purpose (superordinate goal) of apublic (government) agency is idealistic. It presentsthe ideal of an organization staffed by workers,managers, and leaders who strive to develop ashared understanding of its mission, together withthe various customers and other stakeholders. Theircreativity is required both for negotiating thisshared vision and for the continuous improvementof performance, with ever tighter resources. Adegree of success in this context offers consider-able intrinsic rewards, more so than did the oldmodel. Instead of being just a program administra-tor, as in the old model, the new model publicmanager tries to act as a significant player in aprocess of negotiating among diverse stakeholders.“Instead of simply devising the means for achievingmandated purposes, they become explorers whoseek to discover, define, and produce publicvalue.” [Moore, 1995, p. 20.]


Figure 1

Strategy

Structure

Systems

Style

Staffing

Skills

Purpose(SuperordinateGoals)

Old Model Bureaucracy

Replicate.Follow the formula.Mass produce.

Hierarchy. Vertical.Chain of command.

Big picture held at the top.

Formalized.Coordination by rule book.Standard operating proceduresvery important.

Conformity.Please the boss. Politics.“Everything in its place.”

Role clarity.Dispassionate.

Narrowly specialized.

Create profits and rising sharevalues for investors. (BUSINESS)Administer programs and poli-cies authorized by the legisla-ture. (GOVERNMENT)

New ModelLearning Organization

Innovate to please the customer.

Networks. Horizontal.Many project teams. Maybe matrix.Everyone has the big picture.

Informal. Coordination by mutual adjustment.

Creativity. Learning. Participation. Dialogue.Politics (rivalry) over priorities andstrategies.

Flexible job boundaries.Passion about their work.

Versatile.Cross-trained.

Fulfill a shared vision of organizationmission through creative work in acommunity of employees, partners,and stake-holders. (BOTH GOVERN-MENT AND BUSINESS.)


“AutoCo” is one of the big three Detroit car makers,and “Epsilon” was its program to create a newdesign and production tools for one of its luxurypassenger vehicles (Roth and Kleiner). The programmanager and his team were responsible for creat-ing the design and being ready to launch produc-tion by a date some three years away, meetingmany quality standards, and for staying within acost budget. This would require the coordinatedefforts of never less than 300 and as many as 1,000 employees, mainly engineers.

The program director came in with a personal goal,not only to produce a great car design, but toimprove drastically the process of program devel-opment and the production launch. Having workedon several prior new car programs, he was familiarwith the customary period of panic, out-of-controlstress, and pandemonium in the last phase, justbefore launch, where everyone would be workingextreme amounts of overtime, trying (always unsuc-cessfully) to launch on schedule. He was con-vinced that this was unnecessary and could beavoided by better management. Exactly how, hedid not know, but he was fortunate that his deputyhad become interested in learning organizations,especially the version being developed at MIT.

A consulting research team from MIT agreed towork with the Epsilon program leaders. While theMIT team had many ideas and tools to offer, theydid not have a fully developed “package,” and itwas agreed that they would work as partners withthe Epsilon team in a joint learning experience.

A “core learning team” was formed with 10 Epsilonmanagers and several staff from MIT, which metevery month or two for a period of eight months.They conducted joint assessments on their team

working issues and began learning and practicingthe basic concepts and tools of organizationallearning. Before asking the rest of the program staffto become involved in changes, members of thisleadership group first engaged themselves in somevery serious learning and change. When it wastime for the learning labs, some senior Epsilonmanagers from the core learning team acted asteachers and coaches for the other staff, with thehelp of internal AutoCo consultants.

The training agenda and content were developedby the core learning team. Several members inter-viewed other program staff about their “greatestchallenges and strengths.” The core team used thatdata to study the central question: “Why are ourparts always late?” Working together they created asystems map (causal loops) of many factors, whichled to discovering the root cause and point ofleverage. It was the fact that engineers who werehaving a problem with their component would notreport this until very late, which would cause otherdependent elements to be even further delayed,compounding the problem. Had the problem beenrevealed earlier, others could have helped to speedup the solution and prevent the escalation effect.

Concealing problems in one’s own area, theylearned, was a consistent pattern. The core teamdiscussed why this happened, concluding that itwas a combination of “engineering culture” (don’treport any problem until you know the solution)and a company culture in which reporting prob-lems would be held against someone, downgradinghis or her performance appraisals and reputation.The Epsilon core team wanted to change that andrealized that it would require establishing greatertrust among the program staff, trust that “bearingbad tidings” would be safe. This required change

Creating a Learning Organization in (part of) an AutoCompany (Case Study)


in some other supporting norms and beliefs — thebelief that no one has all the answers (even man-agers) and that cooperation for the good of thewhole program was more important than theembarrassment of some individuals because theirpart of the work was not going well at the time. Inother words, the Epsilon team was aiming to createa culture very different from the one that they hadexperienced over many years at AutoCo. In thenew learning organization culture, managersexpected engineers to make their own decisions(instead of demanding to control them) and cross-functional collaboration was expected andrewarded. These insights were developed firstamong the core learning team over eight months of regular meetings.

Several learning labs were held, eventually involv-ing about one quarter of the entire staff, and many briefings and discussions were held with all of them. Much of the focus was on changingthe norms of communication. The learning labsincluded various experiential methods. What madethe biggest impact on employees, though, was see-ing senior Epsilon managers actually change theirown behavior. They became less authoritarian andmore open to other viewpoints, and when engi-neers reported a serious problem delaying theirwork, senior managers made sure that they gothelp and did not suffer for their honesty.

The changed work process and culture was suc-cessful on the bottom line. Launch of the newmodel went smoothly and on time (as intended)instead of the usual last-minute panic and pande-monium — an unheard of event in the recent his-tory of this very large company. The project wasalso well under budget, saving some $60 million in re-tooling costs. Customer reaction and various

quality measures on the new vehicle were alsowell above previous levels. In three years — a shorttime for serious change in organizational culture —this program achieved impressive strides towardsbecoming a learning organization.

Note: The two change leaders, who were also thetwo senior executives of the Epsilon car program,Fred Simon and Nick Zeniuk, left the companyshortly after the Epsilon launch to become consul-tants in this field. Subsequently, they were asked toassist the EPA New England Office with their initia-tive, which they did over some five years, throughlearning labs and continuing consultation with themain leaders. When the partnership between SoLand the Committee for High Performing FederalAgencies began, they became the main consultantsto each of those projects.


In the previous section, we summarized the maindifferences between the old model with whichtoday’s generation of managers grew up, and whichthey took for granted, and the new model, whichincreasingly seems to have answers to the shortcom-ings we find in our organizations. Making changesin the direction of the learning organization is ahuge challenge, but it has begun, including some ofthe Reinventing Government change initiatives.

On a much smaller scale, the partnership betweenthe Society for Organizational Learning and five fed-eral agencies has directly taken up this challenge ofmoving to the new model. Senior managers who arealso natural leaders for such change initiatives weresought out and supported in preparing themselvesfor their new role. Outside experts were brought inwho had prior experience in leading similar changeinitiatives and in training and coaching change lead-ers. For the most part, though, the leaders of thesegovernment change initiatives would have to learntheir new roles in the same way that their predeces-sors had learned — not in the classroom but in thelearning lab and on the job, experimenting in howto apply the new principles and tools, and makingmistakes along with their colleagues. Leadership andlearning are intimately related in the new model oforganizational management and change — “thelearning organization.”

Having defined the learning organization, the newmodel or ideal, we must now define the change

process that is required to move in that direction.This will be called the “Learning-Based ChangeModel,” which differs in some important ways fromthe traditional or mainstream model of organiza-tional change. We shall highlight some of these differences before we present a checklist of themajor steps involved in applying the Learning-Based Change Model, based on the principles ofthe learning organization.

In its simplest terms, the Learning-Based ChangeModel relies on the approach of introducing thegerms of new ideas, then nurturing and protectingthem while they grow stronger, propelled largely by their intrinsic appeal. It is a model of changethat relies as much as possible on the power of the grass roots. The traditional model, by contrast,relies on the “leader on the white horse” settingdirection and impetus from the top. A more system-atic breakdown into two columns can be seen inFigure 2.

Key Success Factors for theLearning-Based Change ModelWhat is actually involved in cultivating a learningorganization, following the learning-based changemodel? The following checklist highlights some ofthe key building blocks needed in order to be suc-cessful in that challenging undertaking. If the firstsection aroused or rekindled the reader’s interest tothe point of asking, “What specifically is involved

A Learning-Based ChangeModel


in this?” here is a brief answer. It is based on theexperience of a dozen change projects guided by the ideas of the learning organization, plus the experience of other change experts, twoapproaches which agree in some areas and not in others.

Since this report is designed for public managersand others with a practical interest in the subject,not for scholars, we shall not attend to these differ-ences, but go straight to a combined checklist ofkey factors related to success. We do not promisesuccess if they are followed but we confidently pre-dict problems if careful attention is not paid to allof them.

The Change Model ChecklistEach of the items on the checklist will be describedbriefly. This explanation offers a very “applied” ver-sion of the learning-based theory of change, butyou will note at one point where I borrow carefullyfrom the traditional (mainstream) theory in order to

fill in a weakness in the learning-based theory aswe have known it. The goal of both change theoriesmight seem similar — to move closer to the idealof the learning organization — but only the learn-ing-based theory makes the change in a way thatreally leads to a strong learning organization.

PHASE ONE: Getting Ready

First requirement: Leaders who want to do thisNo factor is more important than the sincere com-mitment of the leaders to this change effort. Theymust be volunteers; it is not enough to be a “goodsoldier” leading this because you were assigned by the boss. The ideal leader for a learning-basedchange project does not wait to be assigned but cre-ates the assignment for him/herself and then seeks asenior sponsor or champion. In terms of position inthe organization, the “ideal leader” is not one per-son but a coalition of three individuals: (1) one is in a line (rather than staff) management position, (2) one is an internal consultant or networker (e.g.,

Figure 2

Traditional Change Model

Change is pushed from the top level.

Change program presents a full set of answers(formula).

“Here’s my (top level) vision ... This is how weshall do it.”

Change behavior through changing policies andreward systems.

Exhort, encourage, push, threaten.

Urgency to show clear, tangible results.

Setting a tough mandate and deadline (from the top).

Learning-Based Change Model

Change bubbles up from mid-levels and grass roots.

Change approach offers some key answersplus opportunity to learn together the restof what is needed.

“Here’s my (top level) vision ... How does thisconnect to what is important to you? Howwill you contribute to this vision?”

Develop new thinking, which will shape new behavior.

Put down roots, feed them, let them spread.

Emphasis on building a foundation for sustained change and improvement.

Work on several local level areas of improvement and what arises.


someone in human resource development (HRD),and (3) one is a senior level sponsor. If you, thereader, fit one of these standards, you may nominateyourself for the leadership role. Just be sure that youalso find the other two partners you need.

Example: In the partnership between SoL and theCommittee for High-Performing Federal Agencies,the Committee played the role of seeking out poten-tial leadership candidates in several agencies, lineleaders (mostly from an office of around 100 staff)who would welcome this opportunity and challenge.

Help leaders build skills, understanding, and com-mitment: find advisors, coaches, or consultantsThis is the special responsibility of the internal con-sultant (HRD, staff position), the second member ofthe leadership group — in the sense that this per-son has some knowledge of change managementand good access to other resources for training,coaching, and consulting. Leading change and creating change together with other people isextremely demanding in terms of skills and in termsof character. Some people start out with more nat-ural aptitude than others and some pick up related“management development” gains in this area dur-ing their careers. But few (even the most fortunate)would choose to jump unprepared into this role ifthere were a chance to prepare. SoL’s experiencehas been that several days spent preparing them-selves off-site with other leaders who are approach-ing such a change venture can be very valuable.This was in fact provided to the volunteer leadersthrough the Society for Organization Learning’score competencies course.

Establish senior level supportThe senior sponsor (or “champion”) is the thirdessential member of the leadership team. This per-son’s main function is to give legitimacy to the ini-tiative, in case it should be challenged by others,even at senior levels. So it is important that thesponsor understands and accepts the rationale forthe initiative and the theory of change. The spon-sor’s contact with the change leaders can be quiteinfrequent, unless this is the leader’s direct boss.Sometimes the sponsor is the first to get the idea of the change initiative and encourages the others;sometimes the line manager is the first one; some-times it is the HRD internal consultant.

An Implementation Checklist

PHASE ONE: Getting Ready

• Find leaders who want to do this (firstrequirement)

• Help leaders build skills, understanding,and commitment: find advisors, coaches,or consultants

• Establish senior level support

• Engage others who want to do this(“partners”)

• Have partners get to know each other

• Form a core learning/leadership group

PHASE TWO: The Pilot Project

• Select pilot project, ensuring alignmentwith the strategic direction of the parentorganization

• Implement pilot project and initial goals

– Check out assumptions and “mentalmodels”

– Identify systemic issues

• Assess your progress frequently, bothresults and process

• Help all members build their skills,understanding, and commitment

PHASE THREE: Building on Initial Learning

• Make the necessary changes to struc-tures, policies

• Build more capacity

• Keep the neighbors (and boss) informed

• Renew the vision and feed the passionbehind the effort for change

• Make and keep contact with othergroups making similar change efforts,create a mutual learning support system


Example: In our case at an EPA Regional Office theinitiative suffered greatly because the senior spon-sor (who was the regional administrator) and thechange leader did not have a good shared under-standing of what would be required to make thechange work. They agreed on the grand vision ofthe change but not on how it would be managedand how the active resistance of some other man-agers would be handled. As a result, the initiativewas more limited in its success until the regionaladministrator left; and the burden of personal stresswas much greater than necessary.

Engage others who want to do this (“partners”)All change management models emphasize the needfor the leader to present a clear and compellingvision for the change needed — why it is necessary,what the future state will be like, and how they canget there. Traditional change models expect theleader to push hard for all employees to follow, to adopt the same vision, and to act accordingly,offering rewards for those who do. However, thelearning-based change model seeks to get as muchmileage as possible out of the voluntary support ofthose who find this new vision genuinely attractive(even though it may be a small proportion).

These small groups of enthusiasts will encourageeach other; their commitment will lead to someearly creative efforts and successes. These start areinforcing cycle or positive growth loop, whichmay attract other volunteers. The leaders allow others to participate in developing and shaping theinitiative. It is not a top-down, pre-ordained pro-gram. Although this way may take longer to get upsteam, it seems to have more lasting power to keepon trucking over the long haul, whereas top-downinitiatives are more vulnerable to collapse as soonas the leader’s full attention waivers.

Early adopters of the learning organization conceptare low-key evangelists, always willing to talk abouttheir ideas and ready to invite colleagues who areinterested to attend learning labs or other meetingswhere they can see more of what it is about.

Have partners get to know each otherNot only do the initial group of change leaders andsupporters work together in a collaborative, partici-pative way, but they get to know each other as

individuals. Our experience indicates that peoplelearn more and work more creatively when theygive up trying to leave “personal stuff” at home andare able to “be fully present” in the work group.Whereas the old model of organization wouldstrive for a complete separation between the offi-cial role and the private life of the office-holder, the learning organization prefers that workers andmanagers acknowledge their private lives, interests,and values, while still being held to the usual stan-dards of performance and fairness. Research showsthat organizations in fact get better results whenthey give consideration to the personal lives ofemployees [Bailyn, 1997].

Form a core learning/leadership groupThis is a very important part of the entire learning-based change leadership process. It is where thefounders get together and develop their sharedvision, which guides the change initiative. Sincethey are working with a new model involving newconcepts, they need to experiment with them. Theyare learning about each other as individuals andhow they each think, building trust and sharedcommitment. All these activities depend on unusu-ally strong skills at conversation, which participantsand change leaders must continue to develop.

At the beginning (before the pilot project), the coregroup of change leaders and fellow enthusiasts willprobably be small enough (say 4 to 10) to functionwith everyone participating in all discussions. Later,during and after the pilot, the numbers of peopleand groups involved grow larger and a steeringgroup may become useful.

• Leaders need support and feedback fromtrusted colleagues and reflective partners.

• They need to model how to be good learners,acknowledging what they do not know, beingwilling to risk making mistakes, and learningfrom those mistakes.

• The learning process within this group sets apattern that can spread through their followersto “infect” wider areas of the organization.

Example: In the case of the Office of SpecialEducation Programs (OSEP) in the Department ofEducation we see a clear example of such a coregroup. Formed around Patty Guard (the senior


career civil servant) is a group of four (includingherself and two of the three next-level managers(her direct reports), plus their in-house HRD con-sultant). This group is part of the OSEP manage-ment team. Core group members went through theSoL orientation and training. They meet frequentlyto plan the work of the office and the organiza-tional learning activities, and they meet followingmany of these activities to de-brief.

PHASE TWO: The Pilot Project

Select a pilot project, ensuring alignment with thestrategic direction of the parent organizationA major outcome or culmination of the core learn-ing group’s initial gestation period is to select apilot project in which it will apply the newapproach to change and improvement. The kind ofgoal statement needed for the pilot project involvestwo things: (1) some improvement in an area that isimportant to the parent organization, and (2) to beachieved through improvements to the basic busi-ness processes involved. This alignment of theprocess improvement with strategic goals of theparent organization is crucial, though sometimesnot receiving sufficient attention. (This is an areawhere the learning-based approach can learnsomething important from the mainstream.) Havingestablished that we are pointed in the right direc-tion, the next item looks at the implementation ofthe project itself.

Implement pilot project and initial goalsCheck out assumptions and “mental models”

Pilot projects typically have goals requiring majorimprovements, more like 40 percent than 10 per-cent gain, the kind that cannot be achieved just byworking harder or by other incremental effort, butby requiring a different approach, a different way ofthinking about how the work is done. The assump-tions (usually unspoken) that workers (includingmanagers) make in the workplace are brought tothe surface and examined by them in reflectiveconversations. In this way, the mental models thatpeople tacitly use in thinking about the program, itspurpose and methods, its customers and how theyare approached and treated, its suppliers, etc., areall reviewed and tested for their validity. Alternativeassumptions are considered and may become thebasis for a new approach, promising major improve-

ments. An example of this can be seen in the Officeof Special Education case. (See “The Power ofReassessing One’s Mental Models,” p. 19.)

Identify systemic issues

The causes of problems are not always obvious,since events can have multiple causes and multipleconsequences, and some of the consequences canbe delayed and unnoticed. Some of these unin-tended and unnoticed consequences can looparound and bite us in unexpected places. The usualmethod of making a list of causal factors, thentackling them one at a time, sometimes fails to fixthe problem. A series of “solutions” that eachworked for only a short time may be an indicationthat this problem needs stronger (more systematic)medicine. As participants in a problem situation,we must listen carefully to each others’ “stories”from different roles, as we sort out the specific con-sequences of things that happened and map them.Then we construct a systemic view from whichnew insights and solutions can be derived. In alearning organization there is careful communica-tion and analysis using systems thinking.

Often, dedicated work is entirely focused on “our”piece of the organization, but not on the purposeof the whole — sometimes to the detriment of thewhole — perhaps without our even realizing whatwe do. It takes a little systems thinking to gain per-spective. The EPA spent many years pursuing pol-luters to impose penalties. After 20 years, many ofthe worst offenders were dealt with, and the gen-eral public learned from the example of these pros-ecutions. Then some EPA strategists, taking asystems view, began to ask whether more could beachieved to stop pollution and protect the environ-ment by negotiating broader agreements with cer-tain industries instead of enforcing just what thelaw provided for. This went against the inclinationof the enforcement division, since it had alwaysbeen judged on how many court actions it won.This shift in perspective is at the heart of the EPAchange case.

Assess your progress frequently, both results and process

One of the defining marks of the learning organiza-tion is that members frequently check on “how are we doing?” This usually falls into two areas:


At the time they attended the first learning lab,the staff of the State Monitoring Division in theDepartment of Education’s Office of SpecialEducation (OSEP) was struggling unsuccessfullywith a very sticky issue concerning the state reports,which are at the heart of that division’s work. Whatthe division does, in simple terms, is to go out tothe various states and monitor their delivery of spe-cial education services. They come back from eachmonitoring visit with “tons of information.” Theproblem has been how to organize that informationeffectively (and efficiently) into a state report, whichcan help state officials to move into correctiveaction or improvement planning. The issue of effi-ciency was very much weighing on the minds of the staff at the time of the two-day learning lab,because the division was seriously behind schedulein producing these reports and the backlog was get-ting worse all the time.

With the new energy, skills in organizational learn-ing, and insights that they brought back from thatlearning lab, staff spent “a relatively short amount of time” looking for (and finding) a new and betterapproach to the situation. They considered theirvision for how the state monitoring process andreports to the states should ideally be functioning.They also spent some time examining their mentalmodels about the work of the division: the assump-tions they made about the state agencies, their moti-vation, and how they believed the state staff viewedthe monitoring process. This produced many pagesof chart paper, full of notes on mental modelsaround this issue. From April to August of 1999small groups of division staff met and came up with a strategy for solving the backlog of reports.

They went “outside the box” for a solution. Whatled them to the solution was to question the realpurpose of the reports. The breakthrough camewhen they realized that the reports were not anend in themselves, but a means to improve thequality and effectiveness of services to students

with disabilities. By focussing on the results thatmattered, on the true purpose of their program,they were able to view the problem in a totally different light. Once the problem was reframed, asolution that was previously inconceivable nowbecame obvious. Stated with a little irony, the solu-tion to having too many reports to produce was toproduce fewer reports — because they saw that thereports per se did not really matter. What matteredwas the corrective action that the reports were supposed to lead to — and there were more directways to get there. (This is explained more fully inthe OSEP case in the Case Studies.)

It takes courage to go forward with such a solution.“It was pretty radical thinking for us,” said RuthRyder, director of the division. “For 20 years we’vebeen going out, gathering data, doing reports, devel-oping corrective action plans, and going back tostates four or five years later and doing the samething — and a lot of times coming up with the same problems in the states.” New thinking alwaysrequires hard adjustments — in this case from thosewho might think that it means that staff have beenwasting much of their energy for all those years, car-rying an unnecessary load. Another thought amongsome staff was concern that the new approach mightappear as if the agency were “backing off” from itsresponsibility to enforce the law. For some groups,that discomfort would be enough to squash the pro-posed innovation, if it should ever be suggested. This highlights the fact that innovation requires morethan just smart creative thinking. It also requires thestrong personal commitment of members to the mis-sion and values of the organization (enhanced by thevision discussion in this case) and to doing the rightthing, what is best for the “customers” (the disabledchildren) — even if it causes some discomfort to the staff. These staffers did care more about a bettersolution to the problem, regardless of their own discomfort, and they are on their way to creating a learning organization in the Office of SpecialEducation Programs.

The Power of Reassessing One’s Mental Models

(1) how do we feel about the way we are workingtogether? (work process), as well as (2) how wellare we progressing towards our major goal toimprove the standard of what we do for our cus-tomers? (performance results).

Progress on performance goals must be measured.The host organization usually has its own mea-sures, but the work group may supplement thesewith its own. These internal measures, which donot have to be shared outside the work group, maybe “fuzzy,” broad, and subjective — as applied toboth the output goals and work process — pro-vided that the group members are ruthlessly honestwith each other. This is possible if a culture of trusthas really been developed, along with a fiercecommitment to the performance goals. Where this

exists, the fuzzy, gut feelings of members abouthow they are progressing on a tough, complex taskmay be more valid than the precise but narrow offi-cial measurements.

Help all members build up their skills, understand-ing, and commitment

This is the special responsibility of the secondmember of the leadership group, the internalhuman resource development consultant. This person should have some knowledge of changemanagement and ready access to other sources for training, coaching, and consulting. This personneeds to know a variety of qualified people whocan be trusted and who know and accept yourapproach as important.


A meeting involving seven “learning leaders” andsenior executives of the Office of Surveillanceand Biometrics (OSB) in the Food and DrugAdministration’s (FDA) Center for Devices andRadiologic Health (CDRH) took place in March2000 to review the progress of the organizationallearning initiative. The meeting was called byLarry Kessler, director of OSB. Among the exam-ples narrated of areas where the new learninghad been applied were the following:

I have noticed a lot more meetings thatI’ve attended where we tend to behavedifferently — using the ladder of infer-ence* and trying not to climb up too far,keeping in mind our behavior and how itreflects on others and how people canreact to that.

I see the use of the ladder* [of inference]on a day-to-day basis, often in a light-hearted manner.

We seem to be more careful than wewere before about climbing the ladder of inference. An example is talking aboutexploring other people’s perspectives —I think that’s something we didn’t donearly as well three years ago.

More recently [my division has] begun tothink about how we can use SoL ideas toaddress how our partnership, our rela-tionship is with this other office [which isour major customer…]

According to this discussion, the main changes atOSB since their involvement in the SoL partner-ship are not so much different products, proce-dures, or policies (at least not so far) but “how wedeal with our colleagues.”

One of our senior staff said to me, “Ithink SoL is the way we do businessevery day, and I try to have my divisionwork like this anyway, and we try to keepthese kinds of principles living and grow-ing in the way we do business.”

As Kessler summarized it: “This [organizationallearning] has become part of our culture here.”

* The “ladder of inference” is a learning tool or metaphordesigned to increase our awareness of the degree to whichwe go beyond the hard facts in everyday perceptions andjudgments. At the bottom of the “ladder” we stick close toverifiable facts with minimal interpretations. At the top ofthe ladder we use major assumptions or biases in order todraw interesting but speculative conclusions. Both can beuseful — at times — but it can be valuable to be aware ofwhen we are betting on a major assumption in case it is notcorrect for this instance.

Reviewing Progress in Process Improvement


Learning from peers and learning in the actual context of work (“adult learning” or “experientiallearning”) are to be encouraged. “Learning labs”that present concepts and tools of change manage-ment in the context of work issues — and that sup-port learners while they apply the new learning —have been found to be an effective approach.Coaching takes place in the labs and during normalwork time. It is done initially by internal consul-tants and then, increasingly, by their line managers,colleagues, and supervisors. As the pilot projectexpands, the organization must also expand itsinvestment in learning labs and on-the-job coach-ing for the new roles that the change initiative is introducing.

Example: In the OSB case, the change leadersdecided to create a full-time detail position, with90-day rotations, for a staff member to support their organizational learning initiative and to help it take root. (So far this approach is unique amongthe partnership projects). The first appointee was a branch chief within the office, who left thoseresponsibilities for three months while she servedon this detail. She estimated that during the detailshe spoke to about 80 percent of all staff members.Some approached her, asking for help, but she alsoasked to be invited into some regular meetings tospeak to work groups. (See “Apply OrganizationalLearning.”)

PHASE THREE: Building on Initial Learning

Make the necessary changes to structures, policiesOrganizations have many structures that guide thebehavior of their members along certain paths andaway from others simply by making it easier to getone kind of resource rather than another, undercertain conditions rather than others. In addition toall these indirect structures, there are also directreward systems, especially those based on perfor-mance appraisal.

Example: In the case of the EPA Regional Office,where the change initiative was centered on inte-grating the work of separate environmental pro-grams around common results (cleaner air andwater) in place of the old program activity goals,the change leaders realized that they should incor-porate these new expectations directly into the per-formance goals of senior managers and so on down

Apply Organizational Learning

In an interview, Mary Brady, the first persondetailed to this position, stated: “I felt peopleneeded to bite off a little at a time. People inthe government don’t handle change well.I’ve been in government 16 years, and thereis a lot of this type of training — one morething the government is going to talk aboutand not act on. But I also felt they would be overwhelmed.

“The role I played was to remind people of what they learned in July. I started inNovember and I asked: What can we suc-ceed in right away that will show somethingtangible? What can we work on that will beindividual and not have to add to their dailyactivities? So I thought of the mental models.Everyone seemed to enjoy that part the best. And what I did was focus on how webehave, how we react to others, how we’rethinking — just being more cognizant of thaton a daily basis. It might take six to eightweeks just for this part. I put them through a dog-and-pony show at all meetings and Iworked with individuals who requested it. Itwas all voluntary. You don’t need to comply,but you need to be aware of it when talkingto others. We looked at the ladder of infer-ence, mental models, [left-hand] columns*.[My aim was to get them] to practice inmeetings, in and outside the organization,and see how it feels to be using this. For themost part, everyone thought it was a goodidea learning how to deal with difficult peo-ple, keeping your left-hand column intact,how can you deal with others? … I ended up talking to virtually the whole office — 80 out of 100…”

* The “left-hand column” exercise asks people to writedown a conversation they were dissatisfied with. Inthe right-hand column they write what each personsaid; in the left- hand column they write what theythemselves were thinking or feeling but did not say.This exercise was developed by Chris Argyris. [1985;1996, first published 1978].


the management hierarchy. However, it is importantto realize (as they did) that this kind of structuralchange only works when combined with fosteringand teaching the new thinking that goes with it.This requires engaging managers in dialogue abouthow they can rethink their work plan, focussing on the new priorities. This cannot simply beimposed from above. Then each manager who hasredesigned his/her own performance goals musthold that dialogue with his/her own reports. Only if there is meaningful dialogue and participation at every level can the change work. As a last resort,those managers who refuse or fail to comply mayhave to be removed or transferred; the success ofthe change, however, depends on how many canbe engaged in the new way of thinking and expressit through their performance goals.

Where dysfunctional policies and structures of theorganization itself form barriers to improved perfor-mance based on a systematic analysis of the rootcauses of problems, these should be changed.Often this is much harder than it sounds, becauseorganization-wide policies are involved. Now wediscover how much investment the highest levels ofmanagement have in the goal of becoming a learn-ing organization. In large organizations, top man-agers have little awareness of the change initiative,off in a corner of one of their divisions. The influ-ence of the senior sponsor (and how close this per-son is to the top levels) can be significant here.Many change leaders will learn that it is much eas-ier if they confine their ambitions to changes thatcan be managed within their own area, withoutseeking collaboration from other parts of the largerbody, i.e., without causing that entity to changeitself. Initiatives that were sponsored from the topwith the intention of instigating wide change (quiterare creatures) will be treated differently.

Build more capacityThis is a continuation and expansion of the item inphase two named “help all members build theirskills, understanding, and commitment.” It is thespecial contribution of the internal human resourcedevelopment consultant, who should have readyaccess to sources for training, coaching, and con-sulting. As the pilot project expands, the organiza-tion must also expand its investment in learninglabs and on-the-job coaching for the new roles that

the change initiative is introducing. This is neededboth for the early-adopter, the enthusiastic group asit expands, and also for the larger group of employ-ees who are complying like “good soldiers” due topeer pressure, the urging of top managers, and theirsense that “this is the way things are going aroundhere” — although they don’t (so far) feel the depthof conviction of the enthusiasts’ group.

Keep the neighbors (and boss) informedThis is not just common sense and good mannersbut can be a matter of life and death for a new ini-tiative. The change initiative introduces some verydifferent ideas and role behavior. Outsiders whosee or hear about things that are happening due tothe initiative, without understanding the contextand reason, could get a very wrong impression.

Example: In the Epsilon new car project there wasa big effort to get engineers to report right awaywhen they anticipated problems, so that all couldget involved in resolving them and so that otherswould not count on a component stayingunchanged when there was a known problem withit. This openness had never happened before, dueto lack of trust. Building up trust was a central goalfor the Epsilon program managers, and higher num-bers of reported problems were an indicator (tothem) of increased trust. Seeing more problems inthe open than previously (before the start of the ini-tiative) was interpreted within the initiative as agood sign. But to their boss, who had paid littleattention to the nature of the change initiative andresisted their efforts to keep him informed, the largenumber of problems was very distressing, because(to him) it meant that the program was “out of con-trol.” That is what such a number traditionallymeant in the old culture.

The new behavior in the change project (e.g.,workers sharing duties in new ways that theydevised and acting more autonomously by notchecking every move with a supervisor) can alsoseem threatening to those still accustomed to theold ways. Just the enthusiasm and esprit de corps ofthe pioneers can seem like a “cult” to those whodo not know their plan and purpose. This “publiceducation” for neighbors and bosses also helps thenew knowledge of other ways of managementbecome available to other parts of the organization.


In the first year of the EPA New England initia-tive, the strategy was designed to be vision driven. After engaging with the vision, staffmembers were expected to design the modifiedstructures and policies that would be needed tocreate an integrated, results-oriented environ-mental system. However, powerful distractorsgot in the way and not enough managersbecame sufficiently engaged in this process tothe point where they felt joint ownership of thevision. In one major area, however, the initiativewent ahead very much according to plan. Thiswas in the newly created “state offices,” whichmanage relations between the EPA and each ofthe states (to provide the state environmentalagencies with “one-window shopping”). These“state offices” embodied fairly well what thenew strategy and vision called for. As for the restof the regional office, however, progress in get-ting collaboration and integration of effortsacross units was quite limited.

In the third year of this change initiative a group of change leaders/senior executives whowere looking to accelerate the change processdecided to use the staff performance manage-ment system (PMS) as a tool for this purpose.PMS had existed on paper for several years butwas ineffective. It was an empty, required ritual.In the first year of the change initiative, man-agers were, in effect, invited to participate in

conversations about the new vision and mentalmodels, but they were not pushed to set perfor-mance goals for their units and themselves inline with the new thinking. Now, in the thirdyear, managers and their reports were expectedto talk together about the goals for their unitsand themselves, and then to set goals appropri-ate to the new thinking. Training was providedfor all levels of staff to help them with this.

In the new (third year) strategy it was consideredessential that every manager have good conver-sations about which unit and personal perfor-mance goals would fit with the agency’s newvision of integration around environmentalresults. Each top manager would have this conversation with his/her direct reports, andtherefore every manager and employee wouldeventually have such a conversation with his/herboss. In the new strategy staff members weretold that they and their supervisors would beevaluated both on the way these goals werewritten (now) and on how well they were met(later). Thus the PMS was intended to provide astructure for shaping and supporting the behav-ior and thinking of staff (at all levels) who werenot opposed to the new direction (vision) butwere not going to enact it without more help.For staff who were opposed to moving in thisdirection, PMS might be experienced as a coercive structure.

The EPA New England Initiative: Changing thinkingand organizational structures

To get the best results from an organizational change it is usually necessary for participants tochange their thinking (purpose and mental models) at the same time as organizational structuresare changed.

Renew the vision, and feed the passion behind theeffort for changeThis item is a reminder that basic to the learning-based approach is the notion that a major energysource for the change process is the personal visionand deeply felt aspirations of individuals, the goalsand values they really care about. In addition to the

importance of skill development to meet these new challenges, there is also a need for periodicreminders to participants about why they areinvolved in this change effort, invoking symbolsthat rekindle the passion. Examples include cele-brating recent successes, commemorating old ones,retelling the stories of early struggles, and hearing


The following factors impact the ability to imple-ment change. Use this list either to select the morepromising locales for change initiatives and/or use itto prepare the ground, developing a more favorableenvironment for learning-based change initiatives.

When an organization has the following factorsoperating at a high level, it will be harder toimprove organizational learning and to build moreeffective learning organizations — whether it is inthe public or private sectors. It is often assumedthat these negative factors generally characterizethe public sector in contrast to the business sector,but we do not need to make that assumption. (Infact, those who are familiar with the inner work-ings of Fortune 500 corporations will instantly rec-ognize most of the pathologies on this list.) Thepoint is, it will be harder to promote learning-based change and to move towards the learningorganization in situations where the following conditions exist:

• The work environment makes use of blamingwhen mistakes happen — especially when theblaming is personal, punitive, and public. Allthat is bad enough inside the organization. It’s10 times worse when your mistakes are consid-ered fair game for the mass media and may be featured in provocative headlines or on the6 p.m. news.

• Credits for successes are not balanced against afailure — this leads to a risk-averse culture.

• Stakeholders have the power to intervene andmicromanage (“Congressman X’s office calledabout … “).

• Change-resistant workers feel that they canwait out (and outwit) the reformist leaders. This is a specific reason for not depending too much on top-level (politically appointed)agency executives as change leaders, becausetheir tenure is short. While they are indispens-able for certain kinds of change — e.g., wherelegislation must be changed — the career service officials can play a broader role aschange leaders.

• Dysfunctional structures are locked in by legis-lation or other hard-to-change means. This iswhere the political leverage of the politicalappointee can be valuable in getting the legisla-tive changes. Most effective is an alliancebetween senior career officials, who may do thegroundwork, both in research and in the prepa-ration of staff for the changes in operationalmodels (changes in thinking) over several yearsprior to the arrival of the new appointee. (Thishappened, for example, in the OSEP case.)

• Senior managers micromanage or permit stake-holders to interfere in matters of administrationthrough them.

• Poor performers (at any level of the organiza-tion) remain long-term without effective evalu-ation, help, or sanction. This creates a doubledrag effect on the morale of colleagues and onthe performance of the unit.

• Accountability is weak and staff members feelthey are invulnerable. Change comes as asevere threat to this situation.

• The workforce feels neglected and/or abused.Curiously, this may overlap with the previoustwo situations.

“It’s different here. You don’t understand.”

Environments That Are Unfavorable to the New Model


• Workers consider themselves more expertand/or more dedicated than their bosses.

• The vision of workers is limited to their ownjob description (“it’s not my job”), with noregard for the purpose or mission of the wholeprogram, and no concern for those whodepend on its outputs (their “customers”).

• Inadequate feedback loops and informationsystems leave staff “flying blind.”

• A significant part of the workforce has a differ-ent vision for the agency from the official one.(This is a serious situation, even more than theothers listed here.)

• The intangibility of the work leaves workers in disagreement about the nature of their cus-tomers, about what they do, and about what isor is not effective, with no ready evidence onwhich they could agree to settle the matter.

• The customers, those who depend on the out-puts of the agency, have no way to reward orpenalize the agency based on how well itmeets or does not meet their needs. This is amajor difference between traditional govern-ment services and the world of the market-place. And this is why there has been so muchinterest in privatizing public services, makingthem compete and risking the loss of their fran-chise if they fail. This approach may work forservice programs but would not work the sameway for regulatory or security functions.

• The payer (third party) is different from the cus-tomer or beneficiary. In effect, there are twocustomers/stakeholders with possibly differentinterests, feedback loops, etc.

• Key players remain in position only a shorttime, often replaced by someone holding quitedifferent assumptions, principles, and policies(discontinuity of key players and policy). Thisoccurs in the electoral system of politicalappointees at the top levels of governmentagencies, with a democratic rationale, as thelarger political system adjusts to a shift in voterpreferences. With increasing volatility in thebusiness world, there may be an element ofthis discontinuity of key players with or withouta shift in policy.

These 17 factors unfavorable to change are hard toavoid — especially (but not exclusively) in the pub-lic sector. Perhaps the main value of the list is toenable the prospective change leader to anticipatethe nature of the barriers to be faced.


from people who represent personal and poignantexamples of the need to which the work of theorganization is dedicated. From a shared vision likethis, the wells of positive motivation run very deepand strong.

Make and keep contact with other groups makingsimilar change efforts, create a mutual learningsupport systemChange leadership in this model is all about learn-ing together to do things one has never donebefore. It involves making mistakes and learningfrom them. It involves recognizing the assumptionsthat invisibly guide our behavior and choices, andsometimes revising them. That reflection occursboth individually and jointly, sometimes in groupsand sometimes between two individuals. It is a nat-ural extension of learning within each group andchange initiative to include learning across severalinitiatives. It is useful to share learnings and bestpractices, and to get advice or help with difficultproblems. The perspective of the outsider can bevaluable in spotting problems that those on theinside couldn’t see.

Example: Among the five federal government agen-cies participating in the partnership with SoL thereis a conference call every three months involvingthe two main leaders, several members of theHigh-Performing Federal Agencies Committee, SoLstaff, and consultants. There have also been twolarger, longer, face-to-face meetings in less thantwo years. These meetings share encouragementand best practices.


The Partnership Between theSociety for Organizational Learningand the High-Performing FederalAgencies Committee

The idea of trying a learning-based approach in the federal government arose out of conversationsbetween the Society for Organizational Learningand the High-Performing Federal AgenciesCommittee of the Inter-agency Human ResourceDevelopment Council (HRDC). This idea built onthe pioneering efforts begun in 1995 at theEnvironmental Protection Agency’s New EnglandRegional Office in its partnership with SoL and sev-eral of its consultant members. Georgie Bishop,one of the leaders of the New England EPA initia-tive, was a member of the High-Performing Federal Agencies Committee and shared with hercolleagues some of their early experiences. Thisaroused their interest and led in 1998 to the start of a partnership between the committee and SoL.

The committee sought out and selected other government executives interested in becominginvolved with the partnership and developing theirown initiatives. These volunteers, senior executivesfrom four different agencies, with the consent oftheir superiors, joined the partnership. Theyattended a five-day introduction course in the “corecompetencies” of the learning organization, alongwith one or two staff from their respective units, allof them carefully selected. At the “core course”they met colleagues from many other organizations

— mainly non-governmental — on a similar careerjourney. Upon their return, this group planned withSoL consultants an in-house workshop (“learninglab”) to introduce some of their own staff to thenew approach. Each unit then planned, with thehelp of its consultants, how to take the initiativeforward. Four times during the first year severalleaders from each of the four projects talkedtogether, three times via conference call and onceat an all-day in-person meeting. Other times thisnetworking group might meet when invited to aninitiative-related event at one of the participatingagencies. For example, the Department ofEducation invited them when they hosted a presen-tation by the two EPA leaders to their own staff inJune 1999. Members of the networking group alsowould attend meetings at SoL. In fact, the SoLannual meetings in 1999 and 2000 included pre-sentations by leaders of government projects aboutthe work being done in the federal government.

The role of the Committee for High-PerformingFederal Agencies (HPFA) was crucial in three areas:1) in sponsoring and legitimating the idea of thisexperimental partnership from the outset, whichincluded the recruitment of participants; 2) in facil-itating the contracting process between SoL and thefederal Office of Personnel Management; and 3) infinding suitable pilot agencies and appropriateleaders within the agencies to participate.

In their individual HRD roles, several committeemembers have also taken part in facilitating learninglaboratories and providing ongoing on-site consulta-

Case Studies


tion with the application of the concepts and toolsin the specific agencies. The quarterly conversationsamong project leaders were planned and sponsoredby the HPFA Committee. The leadership role of theCommittee on High Performing Federal Agencieswas recognized by their peers who selected them to receive the Distinguished Service Award “ForService to the Federal HRDC Community” from the Training Officers Conference in June 1999.

Four pilot projects for the first year of the contractwere started within the following agencies:

• National Aeronautics and Space Agency (NASA)

• Department of Veterans Affairs: VocationalRehabilitation and Employment (VRE)

• Department of Education: Office of SpecialEducation Programs (OSEP)

• Food and Drug Administration: Center forDevices and Radiological Health, Office forSurveillance and Biometrics (OSB)

Historically, the first project was undertaken at theEnvironmental Protection Agency, New EnglandRegional Office, which began its own partnershipwith SoL in 1995. The leaders of that project,Georgie Bishop and David Fierra, are consideredvalued mentors within the group and are regularparticipants in meetings of the partnership. The lastcase study in this report focuses on the work doneat the EPA. The first two case studies highlight ini-tiatives at OSEP and OSB.

The recruitment of government executives to be“learning leaders” for these initiatives (through theHPFA Committee) represents the first of three keyelements required for introducing organizationallearning initiatives into the federal government. Thesecond key element is the availability of internalhuman resource development consultants whocould coach and support the projects; and the thirdis the outside expertise and support brought in bySoL supporting staff and SoL-affiliated consultant-trainers (primarily Fred Simon and Nick Zeniuk).

Each of the four agencies taking part in this partner-ship followed its own timeline, but the sequence ofmajor steps was the same: after attending the intro-ductory course in the Boston area, the learning

leaders and internal HRD consultants next sched-uled a “learning lab” for a group of their staff mem-bers, to be presented by outside consultant-trainersrecommended by SoL. This learning lab, presentedlocally for some 25 to 40 staff members of eachpilot agency, introduced them to some basic organi-zational learning methods in the context of a “prac-tice field” where they could begin working out howto apply these methods to a selected challenge thatthey faced. Not only the learning leaders (execu-tives) but also the staff and supervisors involved inthese pilot initiatives were all volunteers — a basicprinciple of the SoL approach.

As we shall see in the case studies, participantslearned to improve their effectiveness in communi-cating and working together. At NASA, for exam-ple, the human resources development staff (whichwas the host for one of the initiatives) noted that itswidely dispersed membership was able to increaseits productivity as a group, as a result of developingtogether a shared vision and strategy in the learninglab. At the Department of Education, the divisionresponsible for monitoring state programs of specialeducation and early intervention for children withdisabilities was able to improve significantly itswhole system for working with state agencies,based on rethinking some of its basic assumptionsand mental models.

The informal assessment of the first two years’achievements, both by federal participants and SoLpartners, is that much important progress has beenmade and that the beginnings of healthy roots havebeen put down. The SoL approach is to invest heav-ily upfront and to continue patiently, building up ahealthy orchard to support extensive future harvestsand growth. SoL’s program manager for this part-nership is Jeff Clanon.

Case Study: The Office of SpecialEducation Programs (OSEP) in theU.S. Department of EducationOverviewThis change initiative went through the steps out-lined earlier. During phase one (getting ready) a corelearning team was formed to lead the effort, headedby Patty Guard, deputy director of OSEP and seniorcareer servant in OSEP, and included two of her


direct reports and an internal consultant. Their pilotproject (phase two) turned out to be redesigning thework of the State Monitoring and ImprovementDivision. This project was led by Ruth Ryder, direc-tor of the division, with the rest of the core learningteam staying in the background to act as her supportgroup and reflective partners. The core learning (andleadership) team in OSEP then moved the initiativeinto phase three, expanding its scope from the pilotproject in one division to a more complex projectthat cut across several divisions of OSEP. The impres-sive progress made by the pilot project encouragedthe three-person leadership team to get more ambi-tious and take on a larger and more complex projectthat involved several segments of OSEP that neededto collaborate in new ways. The success of bothphases gave credibility to the goal of making OSEPinto more of a learning organization.

Phase One: Getting Ready About one-third of OSEP staff (about 30 people)took part in a two-day learning lab run by two con-sultant-trainers from the Society for OrganizationalLearning, presented on-site at the agency, to intro-duce them to the basics of some tools of organiza-tional learning. For three OSEP executives and theirinternal HRD consultant, their involvement beganat least two months earlier when they attended thefive-day “core competencies course” in Boston,along with other change leaders from a number oforganizations. These three took the lead in plan-ning the in-house learning lab and in OSEP’s whole venture into organizational learning with itsoutside partners (consultants) from the Society forOrganizational Learning. The four who attendedthe core competencies course became the corelearning team for OSEP’s learning-based changeinitiative: Patty Guard, two of her direct reports —Ruth Ryder and Lou Danielson — and their internalconsultant, Bette Novak. From the start they hadthe support of the OSEP office director (a politicalappointee), and when he left the position, Guardtook care to acquaint the new director with whatthey were doing, so that he too became a firm sup-porter of this change initiative.

Phase Two: Pilot Project in the State Monitoringand Improvement Planning DivisionImmediately following the two-day learning lab inMarch, staff who had participated began to experi-

ment with applying the tools and concepts to theirwork. One group in particular found itself in apainful bind: They really wanted to put their newlearning to use, but they faced a huge backlog ofoverdue reports on the state reviews for which theywere responsible. They felt that they dare not takethe time necessary to plan an organizational learn-ing initiative. Yet they really felt the need to improvethe way they managed the situation instead of con-tinuing to push the boulder up the mountain.

They felt this dilemma acutely. But it was not longbefore they broke through “the wall” with a questionabout their mental models around the state monitor-ing process: How might they have caused themselvesto be in this bind? And what would be another men-tal model for what they were trying to do that wouldsomehow reduce this dreadful workload?

That proved to be the essence of the solution theyfound. It still took a lot of hard work, talking andthinking about the purpose of their program and theassumptions they made about their counterparts inthe state agencies. The rhetoric of recent Americanpolitical philosophy called on federal government to think of their state counterparts (and even the pri-vate sector) as “partners.” But the long history ofmore adversarial roles between state and federalagencies (not to mention the government and pri-vate sector) carried a legacy of less trusting mentalmodels, which supported a particular approach tomonitoring, accountability, and reporting back tothe states. What if they could truly adopt a partner-ship model? How could that change the monitoringand reporting process? And could that be less oner-ous to both sides, and perhaps more useful to thestates and more beneficial to the real customers —the disabled children and their families? The“magic” of mental models work —the surfacing oftacit assumptions that invisibly guide people andreflecting on their implications — makes possiblesuch significant breakthroughs in problem solving.This work is not unilateral, of course, for the part-ners also need to be involved in the discussions and to agree to the changes that emerge.

Under the leadership of Ruth Ryder, the Division ofMonitoring and State Improvement Planning staffmet weekly for several months, applying these toolsto their “wicked problem” of how to manage theever-increasing backlog of state reports. An innova-


tive solution was created, and the next steps forimplementation are being developed. In this workthey have used a variety of tools: mental models,shared vision and team learning, and advocacy,inquiry, and reflection. The staff was very enthusias-tic about the process. The comment heard manytimes was that they felt they had “been given per-mission to think outside the box.” This work hasincluded the entire staff of the division, a signifi-cant number of whom did not participate in thelearning labs.

Phase Two: Assessment“We found that the Vision Deployment Matrix(VDM)1 was a very structured and labor-intensivetool — not one that we would use for our day-to-day work, but one we could effectively use forlarger projects such as the ECTA (Early ChildhoodTechnical Assistance) work group,” said BetteNovak, internal consultant. “The work group andlearning leaders found that the VDM process andother organizational learning tools surfaced manycomplex and broad-based issues related to OSEP’srole in providing technical assistance that may nothave been identified without using this process. Themajor outstanding issues that needed additionalattention included work with our sister federalagencies which also have programs for young chil-dren, including providing technical assistance, andwork with our external customers and partners todevelop a ‘shared vision’.”

Meeting with the supervisors, Guard remarked jokingly that in their current task they would againbe using the VDM — “a tool that has become nearand dear to our hearts” (laughter). “We seem to useit for everything around here. We are getting betterat it all the time and we have learned a lot fromusing this tool.”

Some further assessment of the OSEP experiencewith the VDM tool was offered at a cross-agencymeeting by Lou Danielson, another member of thethree-person leadership team for OSEP and directorof the Research to Practice Division, who washeavily involved in the ECTA task force, as well asall of the management team meetings. In his assess-ment, they were “extraordinarily compulsive” thefirst time they used the VDM; in every cell theypushed the discussion until they ran out of steam.On reflection, that was not necessary, since pointsthat might be missed the first time can easily beadded later without loss of effectiveness. Sincethen, they have learned to work much faster withthe VDM: in one case achieving in four hours whatthey would earlier have taken four days to accom-plish. (Researcher’s comment: This does not neces-sarily represent a ten-fold increase in skill becausethe increased speed probably comes from theirincreased confidence and the realization that someof their earlier meticulousness was unnecessary.)Ruth Ryder noted (at the same meeting) that theyhad learned several things about using the VDM: to avoid spending time on discussions of “whichbox to do next?”; to avoid “wordsmithing” theirphrases; to allocate one question per cell; and touse a time-clock to help them keep to a schedule.Although the VDM is a tool designed to fosterthinking, not hasty decision-making and just “fillingin the boxes,” it is possible to be efficient and time-conscious about collective thinking — especiallyafter some practice and with growing skill in thearts of thoughtful listening and dialogue.

Another feature of the VDM, noted by Danielson, isthat it can be used effectively without much training— at least in some instances. He cited the exampleof one staff member who did not attend the learninglab but found herself in a project leadership rolewhere the VDM had been suggested. With only anhour of instruction from a colleague and the benefitof reading the Daniel Kim article on the VDM used in the lab, Bonnie Jones proceeded to experi-ment with its use in her project group. The groupresponded positively to the way it helped them tothink about their work and the resulting product.

Staff response to the introduction of organizationallearning tools has been very positive. LouDanielson observes that, for once, staff do notcomplain about the extra time that the application

1 The Vision Deployment Matrix (VDM) is a trademarked toolfor organizational learning developed by Daniel H. Kim andDiane Cory. It was described by Kim in an article in TheSystems Thinker, 1995, vol. 6, no. 1. The VDM provideschange leaders and work groups with a powerful method oforganizing the efforts of a change team, using their under-standing of current reality and their aspirations for a differentfuture, focusing on the gaps (and questions) between them,and on the actions they will take to narrow the gap. All that isrepeated on five different levels of perspective, from events topatterns to systemic structures to mental models to vision. Intotal this yields a 5 x 6 matrix, though it is possible to workeffectively in a subset of the whole.


of organizational learning tools requires. Anothersource of data comes from staff feedback collectedfrom OSEP supervisors on how they felt aboutusing the organizational learning tools to do thework of OSEP, including the ECTA task force led byGuard, Ryder, and Danielson. According to staff,the use of new tools changed their experience ofthe work process because it:

• Increased the feeling of inclusiveness

• Showed evidence of the collaboration andcommitment of the three learning leaders tousing the tools and concepts

• Engaged everyone in the conversations anddecision-making process; it promoted creativityand out-of-the box thinking

• Helped large groups to get a common under-standing of the complex area of technical assistance.

Phase Three: Building on the Initial LearningThe success of the State Monitoring andImprovement Planning Division pilot project wasan inspiration to their colleagues in the other halfof OSEP and encouraged two other formal initia-tives, using the organizational learning tools. Athird day of the learning lab (held in June 1999)provided the opportunity for staff to review theconcepts of mental models and systems thinkingand to apply them to work that cuts across all com-ponents of OSEP. The work in the learning labfocused on how OSEP currently does and couldprovide technical assistance and information dis-semination in order to get useful research intopractice. The Vision Deployment Matrix was intro-duced and applied more deeply by the manage-ment team. Subsequently, OSEP asked forvolunteers to serve on a work group to develop apriority to fund a $4 million grant for technicalassistance in the early childhood area. This workgroup used the VDM to consider the future of earlychildhood technical assistance. In addition, theyused other organizational learning tools includingdialogue, advocacy, inquiry, reflective thinking,mental models, and systems thinking in their work.

The Vision Deployment Matrix is one tool of orga-nizational learning that has been very useful toOSEP. It looks like a planning tool, but the way it

works is different from conventional planningdevices whose purpose is to get decisions made.Here the main purpose of the tool is to facilitatecollective thinking (the formulating of vision, thesurfacing of mental models, the analysis of sys-temic structures, and the alignment across these)that will lead to shared vision across the group,with shared strategies and goals and structures thatare in alignment.

Conclusion: Overall Assessment Results so far would include the following:

1) Reorientation and improvement of the statemonitoring program (which represents at leasthalf of the total OSEP workload and mission)

2) Design of a new Early Childhood TechnicalAssistance System by a task force representingmany areas of OSEP

(Note: Both of these items are considered“results” of the work in applying organizationallearning methods, since they go significantlybeyond “business as usual” both in theirapproach and in the extent of the challengetaken on.)

3) A significant increase in collaboration amongthe top three career executives in OSEP aroundthe core work of the office.

Overall, these three changes represent an increase inthe ability of OSEP staff to make improvements inthe way they work. Beyond these specific changes,an enhanced capacity is being created for continu-ous improvement. Obviously, the evidence to sup-port this hypothesis will have to be collected in thenext few years. But it is important to state thehypothesis now, so that even the data at this earlystage may be examined with this possibility in mind.

In addition to the self-assessment of key partici-pants, which we have depended on mostly here, itcan be useful to observe them in action, as I didduring a three-hour meeting of the OSEP supervi-sors’ team in May 2000. Thirteen OSEP supervisorswere present at the meeting, which was led byDeputy Director Guard. The group worked in avery focused way. It started on time, with no late-comers. They used the Vision Deployment Matrix toaddress a new issue concerning the changing roles


of their support staff. Members were attentivethroughout and highly focused on the work planpresented to them that was used to guide the meeting. There was humor and a lot of informalparticipation after the purpose and procedures hadbeen explained by the leader, but there was anintense focus on the task throughout. Members did a lot of listening. Several times when differentmembers were having trouble getting into the dis-cussion, other members intervened tactfully to helpthem get a chance to speak. No one went off ontangents. At the cost of negotiating the break timedown to barely 10 minutes, the agenda was cov-ered. At the end of the meeting, which went 10minutes overtime, the group did not protest whenthe leader asked for extra time for them to assesswhat they thought of the way they had workedtogether. That was done quickly and (to thisobserver) hurriedly, focusing on how the work wasorganized but not on the interpersonal dynamics.

The bottom line is that this group covered in threehours an amount of work that I estimate would typ-ically take at least a full day. This led me to infer(incorrectly) that this group of 13 supervisorsworked together on a regular basis. In fact, I wastold it does not normally work together as a singlegroup — though its two divisions and other subsetsof the group each do so. In addition, it is very sig-nificant that its three senior leaders, the membersof the core learning team, have learned a lot in thepast two years about working closely together.

After the meeting I asked Patty Guard: “How muchdifference would I have seen in this group one, twoyears ago?” She answered that their involvement inthe SoL partnership has “made a significant differ-ence” in the way the three senior (career) leaders ofOSEP work as a team on issues that cut across theorganization. They work much more smoothlytogether, which has been remarked on by staff.They meet frequently to plan the work of the OSEPand the organizational learning activities, and theymeet following these activities to debrief.

In a reflection on her personal experience of thiswork, Guard concluded: “My involvement in theSoL partnership has been one of the most challeng-ing and rewarding professional growth experiencesof my career. It is an outstanding opportunity forsenior executive professional development.”

Case Study: The Office forSurveillance and Biometrics (OSB) in the Center for Devices andRadiological Health (CDRH) of theFood and Drug Administration (FDA)OverviewThis case followed a fairly typical phase one (get-ting ready), but its phase two was not focusedaround one pilot project — not in the way itoccurred at OSEP. While there was a pilot project,there was more emphasis on many, small, dailyapplications of the new organizational learningmethods. Another contrast with the prior case isthat there is not the same type of cohesive corelearning team among the leaders and senior man-agers, although there are just as many strong sup-porters of the organizational learning initiativehere. In assessing their progress, these leaders areclear that organizational learning has become partof the culture in OSB.

Phase One. Getting Ready 1. All staff (about 100) were invited to learning

labs (some 300 person-days of staff develop-ment). Plus a review workshop was given, pri-marily for managers, by Rebecca Pille (aninternal consultant from another federalagency).

2. Emphasis is placed on building internal capac-ity among line staff, as opposed to using exter-nal consultants or even internal ones.

3. Creating the detail position was a major part ofthe capacity-building strategy.

4. Encouragement of the use of organizationallearning approaches in varied aspects of thework of the Office — especially during the firstyear at the micro-, interpersonal level.

5. “Dialogues” open to the entire OSB have beenscheduled regularly, about every three weeks.(Note: This is an unusual locus or format fordialogue — a “town meeting” open to anyOSB staff as opposed to a small group withmore focus and more consistent attendance.There were some difficulties with guidelinesand attendance.)


Phase Two: Pilot ProjectAfter the first year of the initiative at OSB, there arenot any specific projects, achievements, or quanti-fied results that would impress the outside world.There was one top-priority project that was selectedto be an organizational learning effort. Its toughand important mission was to re-engineer the entireprocess of postmarket surveillance of regulatedmedical devices, i.e., most of the work of OSB,which was closely interlocked with other Offices inthe Center (the mother organization within FDA).Compared to the state monitoring project at OSEP,this task was entangled with more major externalplayers and faced much tougher constraints. Thecentral reengineering office, which exerted verytight control, was a powerful stakeholder.Integrating the organizational learning approachwith the rigid demands of the re-engineering officeproved more than this group could manage, espe-cially given the large number of different groupsthat had to be coordinated. One lesson learnedfrom this experience concerns the wisdom of thischoice for a first-year project. Given that the leaderwanted to take a risk since the project was soimportant to the organization, this raises questionsabout what kind of leadership is needed in such acase — both from the project leader and from theoffice director.

Assessment and Impressions of Cultural ChangeA meeting involving seven “learning leaders” andsenior executives of OSB (several occupied bothcategories) took place in March 2000 to review theprogress of the organizational learning initiative.The meeting was called by Larry Kessler, director ofOSB, at the suggestion of the researcher, who waspresent at the meeting. Among the examples nar-rated of areas where the new learning had beenapplied were the following:

I have noticed a lot more meetings that I’veattended where we tend to behave differ-ently — using the ladder of inference* andtrying not to climb up too far, keeping inmind our behavior and how it reflects onothers and how people can react to that.

I see the use of the ladder* [of inference]on a day-to-day basis, often in a light-hearted manner.

We seem to be more careful than we werebefore about climbing the ladder of infer-ence. An example is talking about explor-ing other people’s perspectives — I thinkthat’s something we didn’t do nearly aswell three years ago.

More recently [my division has] begun tothink about how we can use SoL ideas toaddress how our partnership, our relation-ship is with this other office [which is ourmajor customer ...]

What about results? According to this discussion,the main changes at OSB since their involvementin the SoL partnership are not so much differentproducts, procedures, or policies (at least not sofar) but “how we deal with our colleagues.”

One of our senior staff said to me, “I thinkSoL is the way we do business every day,”and I try to have my division work like thisanyway, and we try to keep these kinds ofprinciples living and growing in the waywe do business.

* The “ladder” is a learning tool designed to increase awarenessof the degree to which we go beyond the hard facts in every-day perceptions and judgments.

Mission Statement of OSB“Our mission is to constantly improve patientand provider outcomes related to the use ofmedical device and radiation-emitting elec-tronic products.

We accomplish this by:• collecting, analyzing, and generating sur-

veillance information to identify ... prod-uct issues;

• ... facilitating development and imple-mentations of postmarket problem solv-ing strategies;

• providing statistical and epidemiologicalexpertise and support in pre- and post-market product evaluation ... “ (from OSBAnnual Report, Fiscal Year 1997.)


As Larry Kessler, director of OSB, said: “This [organizational learning] has become part of our culture here.”

Phase Three: Building CapacityOne important practice used at OSB to help theirinitiative take root (so far unique among the partner-ship projects) has been to create a full-time detailposition, with 90-day rotations, for a staff memberto devote full time to the initiative. So far twoappointments have occurred. An interview with the first appointee will be used here. The secondappointee has just completed her time, and no deci-sion has been made about continuing the practice.

Mary Brady, the first appointee, is a branch chiefwithin OSB and left those responsibilities for threemonths to fulfill the detail with much energy.During her detail, she estimates that she spoke to80 staff members. Some people approached her,asking for help, and she took advantage of someregular meetings to insert something onto theagenda where she could. She also asked to beinvited to speak to work groups. We could nick-name this the “Jenny Appleseed” approach to pro-moting organizational learning, as opposed to onethat focuses efforts around specific projects ordepends on an executive-led, top-down approach.

In an interview Mary Brady stated her approach:

Larry’s [the director’s] expectations were dif-ferent than mine. I felt people needed tobite it off a little at a time. People in thegovernment don’t handle change well. I’vebeen in government 16 years, and there is alot of this type of training — one more thingthe government is going to talk about andnot act on. But I also felt they would beoverwhelmed. Larry wanted me to integrateSoL into our strategic plan. I told him rightaway to put it on hold now. We need timeto figure how to integrate this and get staffbuy-in. See what we can do, little by little.People aren’t going to understand massivechange and where it is in their priorities. He was OK with that.

The role I played was to remind people of what they learned in July. I started inNovember and I asked: What can we suc-ceed in right away that will show something

tangible? What can we work on that will beindividual and not have to add to their dailyactivities? So I thought of the mental mod-els. Everyone seemed to enjoy that part thebest. And what I did was focus on how webehave, how we react to others, how we’rethinking — just being more cognizant ofthat on a daily basis. It might take six toeight weeks just for this part. I put themthrough a dog-and-pony show at all meet-ings and I worked with individuals whorequested it. It was all voluntary. You don’tneed to comply, but you need to be awareof it when talking to others. We looked atthe ladder of inference, mental models,[left-hand] columns. [My aim was to getthem] to practice in meetings, in and out-side the organization, and see how it feelsto be using this. For the most part, everyonethought it was a good idea learning how todeal with difficult people, keeping your left-hand column intact, how can you deal withothers? ... I ended up talking to virtually thewhole office — 80 out of 100...

Mary Brady’s approach to change leadership goesbeyond being the open-handed teacher and coach(Jenny Appleseed). She also sees the need torecruit, support, and coach other leaders who canextend and multiply the effect. “I looked in eachdivision for someone who could be my goodwillambassador. I keep reinforcing with them, being acheerleader and keeping it in the front of theirmind, asking them to keep pushing it,” Brady says.

Some Lessons Learned and Further Questions• Using the tools of organizational learning,

managers and staff at OSB believe they haveimproved the quality of workplace relationsand the quality of work in several areas, asthey learned to question their assumptionsmore often and to become more aware of howunverified assumptions and feelings may enterinto perceptions and judgments, causing mis-understandings.

• An effective way of providing support for OSBstaff in the process of incorporating organiza-tional learning (OL) into their daily lives hasbeen to detail an interested volunteer, a middlemanager, or a recent learner to play an at-largesupport role.


• The role of the full-time detail support person,plus the leadership of all those at OSB whobelieve in the value of organizational learning,in addition to learning labs made available toall staff, have all combined to make some ele-ments of OL “part of our culture.”

• Where is the leadership team for the OL initia-tive? There are a number of dedicated leaders,but they do not function as a team with fre-quent or regular communication (comparedwith other initiatives). It could be combinedwith the OSB director’s senior managementteam, if they want to serve that function, or itcould be a separate group.

• They chose reengineering as their pilot project.This was a high visibility, high stakes projectthat required a high degree of cooperationacross boundaries, with powerful outside stake-holders. Was this too tough an assignment forinexperienced change leaders, given the diffi-cult constraints?

Case Study: The EnvironmentalProtection Agency (EPA): NewEngland Regional OfficeOverviewThis case extended over four years, twice the timeof the previous two cases, and it operated on amuch larger scale, involving more than five timesthe number of staff (800). Still, it can be summa-rized in terms of the same three phases. In phaseone, a senior executive conceived a vision for trans-forming the overall structure of this EPA regionaloffice, aiming to integrate its components around aclear focus on results. He found a major ally in asenior HR colleague. Together they attended athree-day course in learning-based leadership andenlisted outside help in training and consultation,for themselves and later for the pilot project.

In phase two, a major pilot project was launched toinaugurate a newly formed department with majorresponsibilities for integrating services in entirelynew ways. The new department was mandated bythe new regional administrator as part of his sweep-ing reorganization of the entire office, rather thanemerging out of the deliberations of a core learningteam, as in the classic version of the model. There

was much learning from this pilot project (much ofit painful), which contributed to the design of phasethree, in which the change leaders shifted theirfocus from the new office (within the regionaloffice) to the whole regional office.

Phase three also featured a strategic partnershipwith the human resources function to implement aredesigned performance management system thatgave more impetus and direction to the drive forchange. Finally, there is a phase four, going beyondthe basic model, in which a surprise change of topleadership in the organization creates new possibil-ities for the change initiative.

Phase One: Getting Ready In 1995, David Fierra, a senior executive in the EPANew England Regional Office, found himself chair-ing a taskforce to design a new Office of EcosystemProtection (OEP), intended to integrate the work ofthe various programs within the regional office. Hehad sought this assignment, which he felt stronglywas essential to this agency regaining a clear focuson its original purpose, but he knew it was boundto be an uphill struggle. Historically this agency(established in 1970) had grown out of a collectionof separate federal laws and programs (Clean AirAct, Clean Water Act, etc.) each with its own legis-lation, funding, structure, sanctions, staffing, andaccountability. The EPA had 25 years of accom-plishment, but by 1995 it was apparent to a grow-ing number of thoughtful EPA employees and otherenvironmentalists that the environment could notbe protected piecemeal and that a more integratedapproach would be necessary. This was a formida-ble challenge since the staffs of each program hadmostly spent their whole careers in that programand indeed many of them had been the founders oftheir program. While they might see some theoreti-cal merit in the integration idea, they were stilldeeply committed to defending the integrity of theirseparate programs, as originally conceived, andthey felt little resonance with the new vision atfirst. Making the challenge even more difficult wasthe fact that EPA headquarters was still organizedin program “silos” for accountability purposes.

Fortunately, Fierra found a major ally in this toughleadership challenge in his colleague, senior humanresources development specialist Georgie Bishop,


who led him to the second major asset: theapproach and methods of organizational learning.”The two of them attended a three-day course,which their thought would be invaluable as anapproach to managing the significant organizationalchanges that they believed to be necessary at theEPA New England Regional Office. And so it provedto be. Georgie Bishop was already acquainted withorganizational learning and systems thinking, havingheard Peter Senge five times at conferences (ASTD,Linkage, etc.). From the HR training budget she hadbought copies of The Fifth Discipline and audiotapes. She had organized a study group on The FifthDiscipline within the HR department. “It made methink,” she said, but exactly how it could be appliedwas not yet clear to her — until 1995. Fierra soughtout her help with his leadership challenge. Had henot done so, she would have offered, since she wasimpressed by his unusual openness with staff andtheir trust in him. [Interview 7/8/99]

Phase TwoThe launch of the new office was assisted by athree-day “learning lab” conducted by outside con-sultants in organizational learning for all 20 of thetop staff of the new Office of Ecosystem Protection.As director, Fierra introduced weekly three-hourexecutive staff meetings (called “EcoLabs”) thatdevoted significant blocks of time to thoughtfulconversation aimed at identifying the core issues ofthe new approach and at examining the “mentalmodels” they were each using to think about prob-lems and solutions, so they could begin new waysof thinking towards better, integrated solutions.Dialogues in the EcoLab also focused on the ques-tion of how staffers in different areas of responsibil-ity would interrelate at work. The purpose of theseEcolabs was “not to ‘practice the tools [of organiza-tional learning]’ but trying to keep in mind whatwe had learned as we did our work and trying tobecome a ‘learning team.’” These meetings weredifferent from the kind of administrative meetingstraditional to this agency, meetings occupiedmainly with operational details and tactics. Instead,they kept a focus on defining and realizing thevision. They took themes from the vision statementto work on and they kept a focus on strategy andoutcomes, as well as tactics. The meetings werestructured but flexible. While they always had aprepared agenda, Fierra was flexible and wouldchange the schedule if a topic with much emotionor energy behind it appeared. [Interview 14/7/99]

There was some initial resistance to Fierra’s insis-tence that the executive team of the Office ofEcosystem Protection should together generatesolutions to major issues. He tried to make themchallenge their own thinking, instead of looking to him, as director, to make all the important deci-sions. He challenged them to understand theirinterdependencies and how they all contributed (orcould contribute) together to achieving better envi-ronmental results. He was working from a mentalmodel unfamiliar to them, one that assumed thatthe leader’s proper role is not to make the majordecisions but to coach and coordinate the efforts ofthose who are closer to the situation and who col-lectively have far more of the relevant knowledge.[Interview 14/7/99]

Results of Phases One and TwoAbout a year after the establishment of the Officeof Ecosystem Protection in the EPA Regional Officein New England, there were indications that vari-ous results had been accomplished.

• OEP managers understood that the results theorganization should be measuring and empha-sizing included cleaner air, cleaner water, anda safe environment (as opposed to numbers ofpermits issued, or enforcement actions taken,for example).

• The heads of various programs learned aboutother programs in the regional office.

• OEP managers and staff developed some skillsin “the five disciplines” of organizational learn-ing [Senge, 1990].

• A new sense of openness and trust grewamong OEP managers.

• Performance Partnership Agreements weredeveloped with each state in the region.

• The states (as “customers”) were more pleasedwith their dealings with the regional office. Infact, there were glowing reports from the statesabout the new emphasis in the regional office.

• A new strategic alliance formed between exec-utive change leaders and human resourcedevelopment staff.

Phase Two: Further DevelopmentsIn phase one, we told the story of a successfulbeginning to what was intended to be a process of


transformation in a large, complex federal govern-ment agency. We told the truth, but not the wholetruth. We did not indicate the scale of resistancethat Dave Fierra met from certain members of hismanagement team, nor the failure of his boss, theregional administrator, to back him up, nor theconsequent level of personal pain and stress suf-fered by Fierra as he struggled with his leadershipchallenge in the face of seriously inadequate sup-port. In organizational learning initiatives that areassisted by consultants within the Society forOrganizational Learning, this is an uncommonoccurrence. One of the basic guidelines is that ini-tiatives should not be undertaken unless they havethe commitment of a three-part leadership coali-tion, including a line manager (Dave Fierra in thiscase), an internal consultant (Georgie Bishop), anda senior sponsor.

The new regional administrator (head of theregional office), appointed by the first Clintonadministration, enjoyed an outstanding public reputation as a dedicated environmentalist and atough and brilliant public sector leader. He sawclearly the need for integration of EPA’s programsand for more partnering outside the agency, andapproved the introduction of a new Office ofEcosystem Protection within the regional office,which represented a radical change for that office.However, when Fierra found that two of his seniormanagers in the new office continued to opposehis efforts, by lobbying senior executives to com-plain about the changes and by actively disruptingEcoLab meetings on a regular basis, the regionaladministrator gave him no support in counseling,transferring or disciplining them. In fact, he paidvery little attention to this major initiative to realignand integrate the regional office through the newOEP. That is, he did not concern himself with over-seeing or supporting the management effort neces-sary to implement that vision. That was consistentwith his management style, which was to providevery strong, directive leadership to his own high-priority projects, with a heavy emphasis on activityoutside the agency. How the regional office wasmanaged to support his strategies was not a subjectthat ever got his consistent attention.

The position of deputy administrator might have pro-vided the support that the struggling head of the newOEP needed, but that did not happen. An unusual

situation existed where the new regional administra-tor had divided the functions of the Deputy’s officebetween two different managers, with an “acting”deputy. This significantly diminished the authority ofthe office and, as far as Dave Fierra was concerned,its ability to back him up effectively.

Earlier the office of deputy had become vacant dueto the sudden death of the long-serving, much-loved previous deputy administrator. This loss was a great blow to the people of this agency. It coin-cided with budget cuts and the threat of major lay-offs, as well as the arrival of the brash new regionaladministrator, who demanded sweeping changesright away, with little time for consultation. It was adeeply disturbing time for all of the employees ofthe EPA New England Regional Office.

For the main protagonist of this story, Dave Fierra,the result was that he faced debilitating oppositionto his efforts to change the organization without thesupport of a supervisor, either for regular review orfor backing up his authority to take drastic action inthe last resort (such as transferring managers whorefused to cooperate). The fact that he was in thissituation was not due to a failure in planning.Rather it was bad luck that the deputy administratorposition was not functioning normally at just thistime and that the regional administrator directedhis leadership to issues outside the agency. That leftFierra in a most difficult situation. Fortunately hehad the steady support of his internal consultant,Georgie Bishop. She acted as advisor, planningpartner, coach, link to occasional help from outsideconsultants, and confidante.

Phase Three: Strategic Alignment and PerformanceManagementIn 1997, after two years as founder-director of theOffice of Ecosystems Protection, Dave Fierra wasreassigned by the regional administrator to a newposition as director of strategic planning, chargedwith building a team of all the office directorsaligned around environmental results instead ofprograms. From his two years of work in establish-ing the Office of Ecosystem Protection he hadcome to see a need to focus integration efforts onthe other offices within the regional office. Fromhis OEP position he had begun to impact the rest ofthe regional office, fostering increased collabora-tion across offices and programs — but mainly at


levels below that of office director. In his new posi-tion he would work mainly with them, but withouthaving line authority over them. For a year hetalked extensively with these managers, both indi-vidually and collectively, and began to see somepositive changes — at least cognitively.

The office directors agreed to undertake a piloteffort to integrate the work of the regional officeacross four environmental results areas. Theseincluded safe drinking water, restoring water qualityin 14 watersheds, eliminating environmental risks inthree urban areas, and protecting wetlands in NewEngland. To provide direction and accountability forthe pilot stage, workplans were developed for eachresult area in the spring of 1998. Each work planwas to contain a specific quantifiable environmentalobjective and a listing of regional work needed toreach it. It was expected that the middle managersfrom the various offices would provide leadership in integrating the programs to accomplish the objec-tive. Unfortunately, many middle managers (particu-larly in OEP) resisted collaborating with otherprograms to integrate their work around environ-mental results. As a result, the first year of the pilotwas largely unsuccessful. The new OEP office director did not take action against the recalcitrantmiddle managers, stating that these new roles werenot clearly defined.

It was now clear to Dave Fierra that many of themanagers in the region were not motivated tobecome leaders and advocates of the newly re-defined mission of the organization as he hadhoped. “I now believed that it was time to focusthe leadership and behavioral changes needed tointegrate the work through the performance man-agement and planning system.” As it happened, justat this time a mandate from EPA headquarters intro-duced a change in the format of the performancemanagement system (PMS) to take effect in 1999.This helped to get senior managers in the regionaloffice to pay attention.

Much work went into designing and negotiating anew PMS. Here the HR function enters this storyas a strategic partner. A key person was GeorgieBishop, who has been a principal advisor to DaveFierra since 1995. She now assumed a new role infacilitating a new and more committed approachto performance management in the Regional

Office. A key role was also played by the RegionalHuman Resources Council, set up in the mid-1980’s in response to serious morale problems inthe Agency. The Council’s function was to advisesenior managers in the regional office on any mat-ters affecting employees.

Membership of the Council included all levels ofemployees, including union representatives, andwas chaired by the deputy regional administrator.The council could be used, in effect, as a perma-nent focus group to design and test ideas for newHR policies and programs. So the members wereasked how to make the performance managementsystem (PMS) truly relevant to the mission of theagency at the regional New England level.

The old performance management system waswidely considered pointless, unproductive, andpart of a paperwork ritual that was not taken seri-ously by staff at any level. So the HR council metin March 1998 to consider two questions: 1) howto get everyone to focus their work plans and goalson environmental results and 2) how to change theculture of the office.

The council created a large training program tolaunch the new performance management system,recognizing (in the words of Georgie Bishop) thatthis was an attempt at revolutionary change andculture shift. A total of 750 staff received training in 15 sessions in the fall of 1998. Eighty managersreceived separate training and then were alsoinvolved in the training of their own staffs.Consequently, in January 1999 the entire staff ofthe Boston Regional EPA Office was engaged inrewriting their individual performance standards,something that had not been done for many years,if ever. Everyone was directed to write their plan soas to answer two key questions: 1) what results do Iseek to achieve, and 2) with whom do I need tointeract so that this can happen?

Despite the major training effort, leaders of this ini-tiative were not naive enough to think that thatwould suffice, given the magnitude of the culturalshift involved. Even so, they were surprised at thedifficulty experienced in the first year. A QualityAssurance (QA) program was implemented inFebruary and March in which a sample of the 190 plans of individual employees were reviewed,


using a general standards guide. This included areview of all plans for management and supervi-sory positions by the Strategic Planning Office. Asample of 70 performance plans were reviewedacross the regional office from all levels and areas.QA reviewers were provided by HR, the manage-ment team, and the Office of Strategic Planning(Dave Fierra). Written reports were provided toeach office with feedback based on the sampleplans reviewed. QA reviewers also met with themanagement team of each office. Fierra offered tomeet with each of the office directors individuallyto discuss changes needed in the individual perfor-mance standards submitted by their managers. Oneoffice director declined to meet, which resulted inunnecessary confrontation when the QA reviewersmet with managers of that Office. At the meetings(about 90 minutes) the reviewers discussed thereview of plans sampled from that office and alsotalked about the need to support this performanceinitiative with other HR efforts, especially profes-sional development planning and resources.

Later Fierra met with the deputy administrator toreview the initial implementation of the perfor-mance management system, including the rolesplayed by senior executives in that process. Somehad not met their duty to implement the new sys-tem to the best of their ability. He was concernedabout how this would be handled in the perfor-mance reviews of those executives who werereviewed earlier, as Senior Executives, on a differ-ent schedule. Would there be accountability forthis area, which had been determined by the topmanagement of the regional office to be an essen-tial requirement for their performance plans?

The challenge before the EPA regional office was toshift the vision of its employees to one where theyintegrated the efforts of all their programs with afocus on environmental results. An important partof their strategy to bring about that change was toimplement a performance management systembased on the new goals and standards, whichwould make each manager accountable not onlyfor their own efforts but also for holding their ownsubordinates accountable for the same goals andstandards. From an outsider’s viewpoint, this wasnot so much a matter of redirecting the perfor-mance management system but of creating aneffective one for the first time and giving it the nec-

essary direction. In Dave Fierra’s own words, “Thekey was to link the regional and national strategiesto the individual performance of each individual.”In this struggle one could feel the weight of alegacy culture in which these ideas are still quitealien and unwelcome.

Phase Four: Unexpected ChangeIn November 1999, without any warning, theregional administrator (a political appointee)announced his resignation to take a position at alocal university and to enter the private sector. Thisnews came as a complete surprise to the staff. Hehad recently received some criticism in the pressregarding alleged favoritism on a decision (BostonGlobe, Nov. 16, 1999, p. A37) and on the drop inthe number of enforcement actions and permitsissued (consistent with the new vision). But he hadalso won major acclaim for negotiating a hard-fought, major clean-up agreement with GeneralElectric (the toughest of opponents) and for forcingthe Army to stop using its firing range on Cape Codbecause of ground water contamination. The beliefin-house was that he would be around for a while,and there was even speculation that he might bere-appointed by the new administration. He wassuch a strong factor in terms of policy and leader-ship style that his departure could hardly fail to besignificant. If his replacement were someone whobelieved in the new vision and also gave it somereal support as a top leader, what a difference thatcould make!

The deputy administrator, who would be promotedto regional administrator, had only been in thatoffice some two years. In that time she had cometo understand the need for a more integrated andstrategic approach to achieving environmentalresults, as Dave Fierra had been advocating. MindyLubber, the new regional administrator, saw how itmade sense in terms of EPA’s mission and purpose.And unlike her former boss, she understood how itneeded to be implemented — that it involved get-ting staff to work together in new ways and to thinkabout the work of the agency in new ways. Sheunderstood well the need to create managementinfrastructures for staff development and perfor-mance assessment to enable the new direction inhow this EPA Regional Office would approach itsmission. On January 11, 2000 she held a meetingwith the top 80 to 100 managers. Listed first among


the half-dozen priority topics she featured was afocus on environmental results and the need tointegrate programs. Also on that list was the needfor an improved performance management system.She and five other senior managers recentlyattended a three-day course in organizationallearning, the same “leadership and mastery” course attended by Bishop and Fierra four yearsago, when this story was just beginning.

ResultsThere is progress in this long, hard struggle. “We arenow completing our second round of performanceplan reviews with each office,” says Dave Fierra.“The quality of the dialogue and overall understand-ing is improving.” These are the crucial metrics —not numbers of performance standards completedwithout errors (for example) but the quality ofthought and understanding demonstrated by staff.The new behavior sought from staff needs to beguided by new thinking on their part. So there isprogress being made — but considerable resistancecontinues — and Fierra notes that, too. A changeleader needs much patience and forbearance andneeds to be capable of living with encouragementand disappointment mixed together.

In more than half a dozen visits to this agency forinterviews, I rarely saw the same mood twice insuccession: Sometimes there was frustration, butthe next time there would be joy; it went back andforth, like a pendulum. We spent many hours talk-ing about these events and sometimes speculatingabout the dynamics associated with them.Sometimes I heard fatigue and pain, but I alsoheard passion for working together for a clean,healthy, beautiful environment. I never heard cyni-cism and I never heard them consider giving up.

Dave Fierra’s initial approach to transformation atEPA depended on the power of vision. However,his clear personal vision failed to connect with thepersonal visions of some of his reports. In the firstfew years, there was no performance managementsystem to help carry the burden of alignment thatvision alone could not carry. Assessment of organi-zational performance must focus on the quality ofthe dialogue about goals and standards, and thequality of understanding — not just alignment ofstandards and goals on paper — and then test forchanged results.

Note by comparison that OSEP (working on a smallerscale) had success with the Vision DeploymentMatrix as a very structured way to formulate andbuild upon vision, and also had the advantage of asmall, unified top management group. This may havestarted as a slight advantage but through the organi-zational learning experience at OSEP, the three-per-son leadership team became much more cohesiveand they then helped the 12-person supervisors’group to become more cohesive also.

ConclusionReflecting together on “how are things going?” and“why do we think so?” is an essential part of theprocess of managing change — both because weneed to learn from our mistakes and successes, andbecause we need to learn from each other. That is,we need to learn how to understand each otherbetter — the people we are supposed to work with.We need to understand better how they think (andhow we think), how they see things (and how wedo), and how they feel about things (compared tohow we do). These are the keys to changing theway we behave with each other, especially how wecollaborate more effectively across the old barriersthat traditionally separate us — barriers of culture,ethnicity, gender, class, occupation, and all the dif-ferent identities formed within an organization.Improved results for the organization depend onpeople working together more effectively acrossthese boundaries. We can learn to do that if we aredetermined. It’s not like learning computer science,codified knowledge that already exists and justneeds to be internalized; it’s more like creatingnew knowledge together — knowledge about usand how we can collaborate better.

The “lessons” listed here may or may not be cor-rect; they may not be the most important lessons tobe drawn from these change efforts. They are not somuch conclusions as questions for further consider-ation. The point is that every change initiative mustcreate its own reflections and lessons. That is howlearning-based change happens — through peopleusing their thinking and reality-testing abilities togrope forward through the murky, unmapped terri-tory. A learning organization creates new futures,because the old ones do not work well enoughanymore. And the only way for this entity we callan “organization” to create anything new is for itsmembers to create it, using their ability to learn.


Change leaders, core learning teams, and everyoneinvolved in learning-based change —all have learn-ing to do. This is the point of the learning-basedapproach. It is not about following some guidelines

to “get it right.” That’s the old model. The newmodel is about using our capacity to learn in orderto move from bad places to better ones, improvingour organizations for the benefit of all.

To change EPA New England from a focus onenforcement activities to a focus on environmentalresults requires coordinated changes in severalareas:

• changes in the vision that members have ofthe agency’s purpose

• changes in many of the members’ mentalmodels

• changes in the agency’s formal organizationalstructure, including setting up the Office ofEcosystem Protection and (within it) StateOffices, especially changes in the measure-ment system or performance appraisal system(changing the individual performance mea-surement system should have begun earlier)

• changes in resource allocation priorities(away from traditional enforcement work)

• changes in the agency’s relations with (at leastsome of ) its major stakeholders, especiallystate agencies, congressional representatives,other federal agencies, businesses, nationalenvironmental groups, local citizen groups

• new learning opportunities for the new rolesand duties, including the nature of communi-cation between staff and managers at all lev-els — change leaders also need opportunitiesfor dialogue and feedback

• support for the learning of new roles in several ways

– “training” (e.g., how to write the new per-formance standards and goals)

– “practice fields” and “learning labs” wherecoaching and support are provided (byinternal or external consultants) while peo-ple work on real problems

– “reflective partners” for the most activechange leaders — internal human resourcedevelopment consultants would be part-nered with individual change leaders (as

Georgie Bishop was with Dave Fierra) andwith a change leadership team (as BetteNovak was with the three OSEP leaders

• reassignment or termination of staff who donot make the necessary changes after a fairinterval and proper resources

To make these major changes requires effectivecommunication via participation:

• explaining why change is necessary, thedirection of the change, and the new think-ing behind it

• listening while staff react to this new mater-ial, express their concerns, and reflect onwhat it means to them

• jointly developing ideas for implementingthe new direction (even reinventing thewheel) so that commitment develops to thechange goals

• explaining specific implementation plansonce they are adopted

• listening to feedback on problems found withthe new changes and making use of it

This communication (dialogue) requires theactive participation of staff and leaders on allsides of the conversation, in an effort to under-stand the intent and assumptions of others.

Dilemmas Basic to OrganizationalTransformation

1. How do you save the change leaders/pio-neers from getting “murdered” by the oldmeasurement system? (The more successfulthey are under the new model, the worsethey look to the old system.)

2. How do you keep essential services operat-ing while changing to a new system?

(Common solution: keep the old reward system.Hence we create problem #1.)

Some Lessons Learned and Questions For Further Reflection


This short list of useful references combines sourcesthat were cited in the text and other related worksthat are recommended as useful to readers begin-ning to explore this subject.

Argyris, C., D. A. Schon (1996). OrganizationalLearning II. Reading, Mass., Addison-Wesley.

Bailyn, L., J. K. Fletcher, D. Kolb (Summer 1997).“Unexpected Connections: Considering Employees’Personal Lives Can Revitalize Your Business,” SloanManagement Review.

Miles, R. H. (1997). Corporate Comeback. SanFrancisco, Jossey-Bass.

Mintzberg, H. (1989). Mintzberg on Management:Inside Our Strange World of Organizations. NewYork, Free Press.

Moore, M. H. (1995). Creating Public Value:Strategic Management in Government. Cambridge,MA, Harvard University Press.

Nadler, D. and M. B. Nadler (1998). Champions ofChange: how CEOs and their companies are mas-tering the skills of radical change. San Francisco,Jossey-Bass.

Nevis, E. C., J. E. Lancourt, et al. (1996). IntentionalRevolutions: A Seven-Point Strategy for Transform-ing Organizations. San Francisco, Jossey-Bass.

Ouchi, W. (1981). Theory Z: How AmericanBusiness Can Meet the Japanese Challenge.Reading, MA, Addison Wesley.

Pascale, R. T. and A. G. Athos (1981). The Art ofJapanese Management: Applications for Americanexecutives. New York, Simon and Schuster.

Quinn, J. B., J. J. Baruch, et al. (1997). InnovationExplosion: Using Intellect and Software toRevolutionize Growth Strategies. New York, Free Press.

Senge, P. M. (1990). The Fifth Discipline: The Artand Practice of the Learning Organization. NewYork, Doubleday/Currency.

Senge, P. M. Charlotte Roberts, Rick Ross, BryanSmith, Art Kleiner (1994). The Fifth Discipline Field-book: Strategies and Tools for Building a LearningOrganization. New York, Currency/Doubleday.

Senge, P. M. Charlotte Roberts, Rick Ross, GeorgeRoth, Bryan Smith, and Art Kleiner (1999). TheDance of Change: The Challenges of SustainingMomentum in Learning Organizations. New York,Currency/Doubleday.

Young, G. J. (2000). Transforming Government: The Revitalization of the Veterans HealthAdministration. Arlington, VA, ThePricewaterhouseCoopers Endowment for The Business of Government.

References


Barry Sugarman is currently involved in a partnership between the Society for Organizational Learning andseveral federal government agencies. He is the author of four books and numerous articles. He earned a Ph.D. from Princeton University in sociology and anthropology, an M.A. from Southern Illinois University, and a B.A. from the University of Exeter, England.

Dr. Sugarman has been a professor of management at Lesley University in Cambridge, Massachusetts, since1979. He is now a research associate at the Verdict Center, supported by the Veterans Health Administrationat the University of Texas, San Antonio, to study and improve the dissemination of new medical knowledge.

Formerly a research associate at the Center for Organizational Learning at the MIT Sloan School ofManagement, Dr. Sugarman is conducting a 20-case, comparative study of change initiatives in businessand government organizations that emphasize participation and a learning-based approach to change lead-ership. Among the topics being investigated are the roles of formal and informal leaders in introducing newways of learning in the workplace; the shifting of shared understandings about the boundary between“work” and “learning”; and ways of more effectively creating, refining, and sharing knowledge in workgroups, communities of practice, and other social networks.

The first five years of Dr. Sugarman’s career were spent at Oxford University, England, and focused on thesociology of learning. Towards the end of that period he became deeply interested in what was then theradically new model of the therapeutic community for addiction recovery. He was one of the first to studyand assist those therapeutic communities in the early 1970s and published several pieces in this area. Thatinterest led to several years of working outside the academy, gaining experience in both the nonprofit sec-tor and county government. In Dayton, Ohio, he developed an accountability system for all the nonprofitagencies that contracted with the county board of mental health and mental retardation.

At Lesley University Graduate School, he taught mid-career professionals in health services and general management areas. He also designed and directed the unique master’s degree program in management of substance abuse services.

Barry Sugarman grew up in London, England and began to explore the United States as a graduate student.He is now a U.S. citizen. He has been a consultant to a variety of organizations on strategic management,program assessment, and the improvement of learning systems. He is now teaching a course on organiza-tions in the Internet economy at the Harvard University Extension School.

About the Author


To contact the author:Barry Sugarman77 Oakdale RoadNewton, MA 02459(617) 332-0642

e-mail: [email protected]

To contact the Co-chairs of the High-Performing Federal AgenciesCommittee of the Human Resource Development Council:

Blanco T. HighInternal ConsultantU.S. Department of Veterans Affairs(202) 273-9760


Ruth SalingerInternal Performance ConsultantU.S. Department of Health and Human Services(202) 690-5549


To contact the coordinator of SoL-Federal Government Partnership:

Jeff Clanon

Society for Organizational Learning222 Third Street, Suite 2323Cambridge, MA 02142(617) 492-9205


Key Contact Information

To download or order a copy of these reports, visit the Endowment website at: endowment.pwcglobal.com

Managing Workfare: The Case of the Work Experience Program in theNew York City Parks Department (June 1999)

Steven CohenColumbia University

Results of the Government LeadershipSurvey: A 1999 Survey of FederalExecutives (June 1999)

Mark A. AbramsonSteven A. ClyburnElizabeth MercierPricewaterhouseCoopers

Credit Scoring and Loan Scoring:Tools for Improved Management ofFederal Credit Programs (July 1999)

Thomas H. StantonJohns Hopkins University

The Importance of Leadership: The Role of School Principals(September 1999)

Paul TeskeMark Schneider State University of New York at Stony Brook

Leadership for Change: Case Studies in American Local Government (September 1999)

Robert B. DenhardtJanet Vinzant DenhardtArizona State University

Managing DecentralizedDepartments: The Case of the U.S.Department of Health and HumanServices (October 1999)

Beryl A. RadinState University of New York at Albany

New Tools for Improving GovernmentRegulation: An Assessment ofEmissions Trading and Other Market-Based Regulatory Tools (October 1999)

Gary C. BrynerUniversity of Colorado School of Law

Determining a Level Playing Field forPublic-Private Competition (November1999)

Lawrence L. MartinColumbia University

An Assessment of BrownfieldRedevelopment Policies: The MichiganExperience (November 1999)

Richard C. HulaMichigan State University

Religious Organizations, Anti-PovertyRelief, and Charitable Choice: AFeasibility Study of Faith-BasedWelfare Reform in Mississippi(November 1999)

John P. BartkowskiMississippi State UniversityHelen A. RegisLouisiana State University

Business Improvement Districts andInnovative Service Delivery (November 1999)

Jerry MitchellBaruch College, The City University of New York

Profiles in Excellence: Conversationswith the Best of America’s CareerExecutive Service (November 1999)

Mark W. HuddlestonUniversity of Delaware

San Diego County’s InnovationProgram: Using Competition and aWhole Lot More to Improve PublicServices (January 2000)

William B. EimickeColumbia University

Using Activity-Based Costing toManage More Effectively (January2000)

Michael H. GranofDavid E. PlattIgor VaysmanUniversity of Texas at Austin

Innovation in the Administration ofPublic Airports (March 2000)

Scott E. TarrySouthern Illinois University

Transforming Government: TheRenewal and Revitalization of theFederal Emergency ManagementAgency (April 2000)

R. Steven DanielsUniversity of Alabama at BirminghamCarolyn L. Clark-DanielsBirmingham, Alabama

Transforming Government: Creatingthe New Defense Procurement System(April 2000)

Kimberly A. HarokopusBoston College

Entrepreneurial Government:Bureaucrats as Businesspeople (May2000)

Anne LaurentGovernment Executive

Leaders Growing Leaders: Preparingthe Next Generation of Public ServiceExecutives (May 2000)

Ray BluntCouncil for Excellence in Government

Trans-Atlantic Experiences in HealthReform: The United Kingdom’sNational Health Service and theUnited States Veterans HealthAdministration (May 2000)

Marilyn A. DeLucaNew York University

Reflections on Mobility: Case Studiesof Six Federal Executives (May 2000)

Michael D. SerlinInternational Institute of BusinessTechnologies

Implementing State Contracts forSocial Services: An Assessment of theKansas Experience (May 2000)

Jocelyn M. JohnstonBarbara S. RomzekUniversity of Kansas

Transforming Government: TheRevitalization of the Veterans HealthAdministration (June 2000)

Gary J. YoungBoston University School of PublicHealth and Department of VeteransAffairs

Managing Telecommuting in theFederal Government: An InterimReport (June 2000)

Gina VegaMerrimack CollegeLouis BrennanTrinity College, University of Dublin

Using Virtual Teams to ManageComplex Projects: A Case Study of the Radioactive Waste ManagementProject (August 2000)

Samuel M. DeMarieIowa State University

The Auction Model: How the PublicSector Can Leverage the Power of E-Commerce Through Dynamic Pricing(October 2000)

David C. Wyld Southeastern Louisiana University

Rethinking U.S. EnvironmentalProtection Policy: ManagementChallenges for a New Administration(November 2000)

Dennis A. RondinelliUniversity of North Carolina at Chapel Hill

The Challenge of Managing AcrossBoundaries: The Case of the Office ofthe Secretary in the U.S. Departmentof Health and Human Services(November 2000)

Beryl A. RadinState University of New York at Albany

Corporate Strategic Planning inGovernment: Lessons from the UnitedStates Air Force (November 2000)

Colin CampbellGeorgetown University

A Learning-Based Approach toLeading Change (December 2000)

Barry SugarmanLesley College and Society forOrganizational Learning

Supercharging the EmploymentAgency: An Investigation of the Use of Information and CommunicationTechnology to Improve the Service of State Employment Agencies(December 2000)

Anthony M. TownsendUniversity of Delaware

ENDOWMENT REPORTS AVAILABLE

*

*

2000 Presidential Transition Series*

*

For additional information, contact:Mark A. AbramsonExecutive DirectorThe PricewaterhouseCoopers Endowment for

The Business of Government1616 North Fort Myer DriveArlington, VA 22209

(703) 741-1077fax: (703) 741-1076e-mail: [email protected]: endowment.pwcglobal.com

1616 North Fort Myer DriveArlington, VA 22209-3195


The Business of GovernmentBulk Rate

US Postage

P A I DPermit 1112

Merrifield, VA

About PricewaterhouseCoopersThe Management Consulting Services practice ofPricewaterhouseCoopers helps clients maximize theirbusiness performance by integrating strategic change,performance improvement and technology solutions.Through a worldwide network of skills and resources,consultants manage complex projects with globalcapabilities and local knowledge, from strategythrough implementation. PricewaterhouseCoopers(www.pwcglobal.com) is the world’s largest profes-sional services organization. Drawing on the knowl-edge and skills of more than 150,000 people in 150countries, we help our clients solve complex businessproblems and measurably enhance their ability tobuild value, manage risk and improve performance inan Internet-enabled world. PricewaterhouseCoopersrefers to the member firms of the worldwidePricewaterhouseCoopers organization.

Documents

A Learning-Based Approach to Leading Change · The PricewaterhouseCoopers Endowment for The Business of Government A Learning-Based Approach to Leading Change Barry Sugarman Lesley