The Successful Adoption of Web-Based Collaborative Software (2005)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

PART 2: THE THEORY YOU NEED

43 CHAPTER 5 BOUNDARY THEORY—UNDERBOUNDED AND OVERBOUNDED GROUPS AND THEIR PROBLEMS Web-based collaborative tools are essentially tools that enable you to move information across boundaries more easily—different companies located in different time zones around the globe in Raytheon’s case, across 30+ different sites in New York and New Jersey in the case of the PANYNJ, and between CTA managers, architects, design- ers, engineers, and contractors in the field in Chicago. Bound- ary theory can help us recognize what is working, why, and what we can do about what is not working. The boundary theory in this chapter is from Clayton Alderfer’s “Consulting to Underbounded Systems.”i Boundaries regulate transactions between a human system (i.e., a group) and its environment; in simple terms, they determine what gets in and what stays out of the system. The concept has its roots in biology, but has been applied by social scientists to define systems and groups. In this context, boundaries are the defining characteristic of a group of people. Without boundaries, talking about “groups” is meaningless. Boundaries can be physical, such as geographic distance or the architectural layout of a building that separates two departments, or they can be psychological. For example, cross- ing into a new group can make an individual feel out of place or unsupported. Or reading highly technical material can be difficult if one is not familiar with the jargon of that field. In both of these examples, there is a psychological boundary that separates the individual from the group with which he or she is interacting. According to Alderfer, “Psychological boundaries tell more about the ‘here-and-now’ of a sys- tem, but their condition is harder to detect than physical boundaries—especially to an outsider.” In cellular biology, the ease or difficulty with which for- eign objects move in and out of a cell is called “permeabil- ity.” An optimum degree of permeability means that the cell can take in what it needs to function and release what it needs to release without difficulty. If its outer membrane is too per- missive and lets in too much, the cell can die. The cell ceases to function because its boundary no longer works to differ- entiate it from the larger environment. But a cell also dies when its outer membrane is too restrictive and causes the cell to lose the ability to take in important parts of its environment. The cell will miss out on essential metabolites and “starve to death.” Success depends on taking in neither too much nor too little, on preserving a fragile balance. When a group’s boundary with its environment is too restrictive, social scientists describe it as “overbounded,” whereas groups whose boundaries are too permissive are called “underbounded.” Another simple way of thinking about this is that overbounded groups have boundaries that are too tight, so it is difficult for the groups to take in people or think- ing or information from the outside. Underbounded groups have boundaries that are too loose, so the groups are not able to keep the turbulence of their environment at bay. When a group’s boundaries are not functional or balanced, the result is a variety of “symptoms” that are much easier to identify than the actual boundaries themselves. These symptoms are deeply familiar to most people because at one time or another most people have belonged to a group with dysfunctional, imbalanced boundaries. And each of these conditions—too rigid and too loose—carries a distinctive threat for the sys- tem. Alderfer writes, The primary threat to underbounded systems (too loose) is that they will become totally caught up in their environmen- tal turbulence and lose a consistent sense of their own iden- tity and coherence. The primary threat to overbounded (too rigid) systems is that they become closed off to their envi- ronments and lose the capacity to respond adaptively to envi- ronmental changes. Thus, being extremely underbounded is a greater threat to a system’s survival, especially in the short run, than being overbounded. Groups that are underbounded or overbounded have dif- ferent problems that have different solutions. In order to determine whether a group in trouble is underbounded or overbounded, one can look at several indicators. Alderfer has identified several variables whose values will be quite differ- ent in underbounded and overbounded systems (see Table 4). HOW TO RECOGNIZE AN UNDERBOUNDED GROUP The following characteristics are typical of underbounded groups—in other words, where boundaries are too loose and i Alderfer, Clayton. “Consulting to Underbounded Systems.” Chapter 11 in Advances in Experiential Social Processes, Vol. 2, eds. C. P. Alderfer and C. L. Cooper. New York: John Wiley & Sons, Ltd. 1980, pp. 269–278.

44 the group or system is in danger of being swamped by the chaos in the surrounding environment: • Goals. In order to be clear, goals need two qualities: they need to be clearly stated, and they need to be sup- ported by a consensus. Underbounded systems have neither clarity nor consensus about their goals. People in these groups may experience their group as flounder- ing without a sense of direction. • Authority. Larry Hirschhorn, at CFAR, has conceptu- alized all leadership as the management of the group’s boundaries, which in turn promotes the work of the whole system. In underbounded systems, the authority may be fragmented and unclear. People in these systems may ask, “Who is really in charge?” “Who is responsi- ble for doing this work?” “To whom can I turn to clar- ify my job?” • Economic conditions. Economic crisis can have a pro- found effect on psychological boundaries. Systems that have healthy boundaries when they are prospering will often transform into underbounded systems when their territory or economics are threatened. Or underbounded- ness may itself cause financial problems. A group that is unable to organize itself may miss opportunities or waste resources. People in underbounded groups may be wor- rying about their job security or about their compensa- tion, making it even more difficult to focus on the work. • Role definitions. For individuals in an underbounded group, the clarity and expectations of their roles tend to feel fragmented, conflicted, and isolated. There may be conflicting demands on individuals coming from differ- ent sources, or there may be a lack of explicit perfor- mance expectations. People in these groups often won- der how their contributions will be evaluated, if they are working alone, or what they are supposed to focus on. • Communication patterns. Communication has to do with the way in which valid information is given and received in order to do the work. In an underbounded system, it is even difficult to determine who should talk with whom or to organize a meeting that gets everyone together. Once an exchange of information actually takes place, it may be characterized by withdrawal or outbreaks of simultaneous talking. There may be a great deal of conflict simmering just below the surface. • Human energy. In underbounded systems, people’s energy may feel fragmented and diffuse. It may be diffi- cult to harness energy and effectively channel it toward the group’s goals. Variable Overbounded Systems Underbounded Systems Goals Goals clear, priority unequivocal Goals unclear, priorities equivocal Authority Monolithic Multiple and competing Economic conditions Minimal, short-term stress Impending economic crisis Role definitions Precise, detailed, restrictive Imprecise, incomplete, overlapping Communication patterns Difficulties with openness when people meet Difficulties in determining who can and should meet Human energy Constrained, blocked Diffuse, exhausting Affect/feelings Positive inside, negative outside (“We’re great, you’re terrible”) Negative inside, negative outside (“We’re terrible and so are you”) Intergroup dynamics Task groups dominate Identity groups dominate Time span Long Short Cognitive work Single theory ideology Multiple theories or no theory or ideologies Source: Alderfer, Clayton. “Consulting to Underbounded Systems.” Chapter 11 in Advances in Experiential Social Processes Vol. 2, eds. C. P. Alderfer and C. L. Cooper. New York: John Wiley & Sons, Ltd. 1980, p. 278. TABLE 4 Properties of overbounded and underbounded systems

45 • Affect/feelings. Overall, the balance of the feelings in an underbounded system is less positive than it is in an overbounded system. The system may be facing (or feel that it is facing) an imminent threat. People often feel a lack of confidence in themselves and in their leader, and there may be feelings of futility. Members of under- bounded groups are often very stressed because they spend a great deal of their time organizing the chaos that surrounds them. In times of systemwide change, mem- bers of underbounded groups are likely to fear the dis- solution of their group. Promoting the discovery of pos- itive feelings in an underbounded system may allow people to see that they do share common concerns and that they are able to control the chaos that is threatening to overwhelm their system. • Intergroup dynamics. Underbounded groups tend to experience less conflict between task groups (e.g., main- tenance, purchasing, engineering, and marketing) than between identity groups (e.g., age, gender, race, and eth- nicity). This may make it difficult to achieve a sustained sense of direction. • Time span. Underbounded groups tend to have a short time perspective. Some organizations are capable of planning and thinking ahead, and others are more short term oriented. This may be the result of external threats in the environment making underbounded groups unsure about whether they will be able to survive. • Cognitive work. People in groups and organizations need a rudimentary theory to explain what they experi- ence in organizations. Without it, they would be over- whelmed by confusion and not know how they should behave. Underbounded systems may have no unifying theory at all, or they may have multiple theories that compete with each another. HOW TO RECOGNIZE AN OVERBOUNDED GROUP Being a member of an overbounded group is often a lot more pleasant than being a member of an underbounded group. But, by definition, overbounded groups have problems, too. They are too rigid, too impermeable, and, although these characteristics may initially feel comfortable for the group’s leaders and members, the characteristics make the group too vulnerable to changes in the environment and too difficult to collaborate with. This type of group is not optimal, either. Overbounded groups have the following characteristics: • Goals. Overbounded groups and systems tend to boast an unequivocal clarity about their goals and priorities. In order to increase the rigidity of a group’s boundary, increase the clarity of its goals and priorities. • Authority. Authority in overbounded groups and sys- tems tends to be centralized and hierarchical. Most resources are controlled by a single source, usually at the top of the organization. These organizations are char- acterized by a unity of purpose, direction, and control.ii Therefore, to increase the strength of a group’s bound- ary, leaders need to exert a directive style of authority. • Economic conditions. Overbounded groups are likely to be facing smoother economic conditions. When eco- nomic conditions are favorable, the group’s territory and technology may improve and it may able to keep mem- bers and attract new members. • Role definitions. Expectations in overbounded systems tend to be precise, detailed, and restrictive. People in these groups may feel confined and constrained and expe- rience a lack of creativity and stimulation, especially at lower levels of the organization.iii • Communication patterns. When information is not being given and received as needed to do the organiza- tion’s work, then the communication in a system is con- sidered suboptimal. The clear roles established in over- bounded groups establish clear communication links between people and groups. Getting people together on a regular schedule is usually not difficult. However, in overbounded systems there is a strong tendency to with- hold information or to distort it in order to present one’s own position in a favorable light. Bad news is withheld, especially from authority figures. • Human energy. In overbounded systems, people often feel that their energy is confined and that it is difficult to release it for work. If boundaries are tightened further, the effect is to further restrict the available energy. The feeling in an overbounded group is that this is not an environment that welcomes individual creativity. • Affect/feelings. This characteristic is particularly com- plex for an overbounded group. On the one hand, the feeling in an overbounded group is generally quite pos- itive. The future looks rosy, the group’s goals and pri- orities are well understood, the leadership is in control, and people’s roles are clear. However, this positive feel- ing can often be the result of repressive forces, which discourage people from voicing any criticism or nega- tive feelings. Members of overbounded groups are likely to feel that things are the way they are and cannot be changed. Internal criticism is muted in an overbounded system, and, as a result, a negative affect is directed out- ward at other groups. This may make it difficult for an overbounded group to collaborate with others, to share information or take in information across its boundary, ii Gulick, L., and Urwick, L. Papers on the Science of Administration, 1937. As cited in Alderfer, Clayton. “Consulting to Underbounded Systems.” Chapter 11 in Advances in Experiential Social Processes, Vol. 2, eds. C. P. Alderfer and C. L. Cooper. New York: John Wiley & Sons, Ltd. 1980, p. 271. iii Argyris, C. Personality and Organization, Harper and Row, 1957. As cited in Alderfer, Clayton. “Consulting to Underbounded Systems.” Chapter 11 in Advances in Experiential Social Processes, Vol. 2, eds. C. P. Alderfer and C. L. Cooper. New York: John Wiley & Sons, Ltd. 1980, pp. 269–278.

46 or to experience the growth that depends on under- standing the power of dissent. In times of systemwide change, members of overbounded groups will tighten their boundaries as much as possible to the point of becoming wary of outsiders. • Intergroup dynamics. Overbounded groups are more likely to experience conflict on the task boundary, such as when people from one department clash with people from another department, than on their identity boundaries (e.g., race, gender, and ethnicity). One way to loosen the boundary around a group that is too rigid is to encour- age people to come together in their identity groups so that they can experience their similarities outside of their department or task groups. • Time span. Feeling relatively secure, it is easier for overbounded groups to take a longer time perspective than underbounded groups. • Cognitive work. Overbounded groups are much more likely than underbounded groups to have a single unify- ing body of theory or ideology that members are expected to master. This may be invoked as “the company line,” which influences everything and is introduced to new members. If the group is too rigidly bounded, it may make it difficult for the organization to experiment or adopt innovation. CONSTRUCTION: AN EXAMPLE OF MULTIPLE BOUNDARIES AT WORK Because boundaries describe a relationship between a group and its environment, it is sometimes necessary to look at more than one group to figure out what is going on. You may need to look at two groups interacting where one has more rigid boundaries than the other. You may need to con- sider why a group may be underbounded when it exists inside a company or agency that is generally overbounded. To understand how groups in a system affect one another and how boundary theory applies, let’s look at the example of a construction project. According to Scott Unger of Con- structware, “historical inefficiencies, cost over runs and inter- party disputes often characterize the building process.”iv Typ- ically, these problems occur on the boundaries between groups—between architects and designers, between contrac- tors and their subcontractors, or between contractors and the owner or agency. The new trend to create partnering events at the beginning of a large construction project is an attempt to create a new system whose members are all the participants on a particu- lar build. Three or four days may be devoted to a meeting at which everyone agrees on a common vision and purpose, vows to work collaboratively, and gets to know one another. If this work is done well, participants may emerge feeling ter- rific about their new partners, but it is all too easy for the effects of a good partnering session to wear off long before the project is complete. Why do these effects wear off? Imagine that each participant has a primary membership in his or her own firm—whether architect, transit agency engi- neer, or lighting designer. The build creates a temporary sys- tem where all participants come together for work—much like a task force in a corporation that cuts across different divisions or departments. The boundary around this new group is quite a weak one compared with the boundaries of the pri- mary groups. The original, primary groups claim most of the individuals’ allegiance, and the primary group cultures dic- tate how the individuals behave on the new team. Figure 4 shows these boundaries at work. It is easy to imagine that the new “partner” group on the build will face many of the challenges associated with under- bounded systems: conflicting priorities, multiple and some- times competing authority, overlapping role definitions, dif- fused human energy, short time perspectives, and multiple ideologies with no unifying theory to hold them together. The social system of a transit construction project will be made up of several subgroups with varying levels of boundary permeability. A classic problem in construction projects is the lack of communication between one group and other groups in the project, causing someone to act on a false assumption and leading to waste and delays. Another way to think about this example is to imagine information as an object that must transfer from within one group across the boundary to another. In order for this information to be transferred, an exchange between the group members must be arranged. If information is not passed on or is misunderstood, there is a flaw in the exchange process. In an ideal construction project, groups will have all the necessary information and understand what information all other groups need and when they need it. Consequently, the groups will freely give important information to everyone who needs it when they need it. When a web-based knowl- edge management technology is applied to that ideal project, iv Unger, S. “The Trend Towards an Internet-Based Communication Standard in the A/E/C Industry.” Constructware, January 2002, p. 1. http://www.constructware.com/ common/downloads/standardization_white_paper.pdf Figure 4. Multiple boundaries at work.

47 A HYPOTHETICAL TRANSIT CASE: SUPPLY CHAIN CONFLICTS To illustrate the point, let’s examine a hypothetical case set in a large, metropolitan transit agency. The director of purchasing is frustrated. Despite many attempts, he or she cannot convince the maintenance depart- ment to produce reliable forecasts far enough in advance. Maintenance in turn sees the director as unresponsive. When maintenance workers need a piece of equipment, they know that it can take more than 24 months to produce it. If pur- chasing can’t manage its supplier relationships better than that, they figure, why should we knock ourselves out creat- ing more elaborate forecasts for them? The director of pur- chasing has several times suggested that he or she place one of the purchasing employees in the maintenance division to learn more about the way maintenance works and to bring back early intelligence, but each time the director of pur- chasing suggests this, he or she is rebuffed by maintenance. The director of purchasing would also like to have more influ- ence over the suppliers’ projections and wonders if it would help to have someone “on the ground” working directly at the supplier site. Perhaps he or she needs to pull together a task force, with representation from maintenance, purchasing, and some of the suppliers? But that’s the last thing anybody wants—another committee and set of meetings to attend. In boundary terms, it makes sense that purchasing very likely has looser boundaries as an organization than main- tenance does—after all, purchasing must interact with all of the departments at the transit agency, as well as all of the suppliers. It couldn’t do its job if it had rigid, impermeable boundaries that did not permit information to flow easily. Maintenance in this case has a much more rigid boundary— it doesn’t like the idea of letting an outsider into the depart- ment, and it is very regimented about the way it wants infor- mation to flow out. Maintenance organizations often enjoy strong, hierarchical leadership, clearly defined roles, and strict protocols—this develops naturally as a way of working when the cost of error is perceived to be very high. (Think of the difference between the tight norms that a cardiac surgeon establishes for what is “right” behavior in the operating room versus a general practitioner seeing patients in the office.) An overbounded maintenance organization would try to suppress or expose weaknesses in any information that wasn’t favor- able to maintenance. What may happen if the director of purchasing puts together a task force that includes representation from purchasing, maintenance, and suppliers? To be successful, the task force will need to counteract many forces that work to make the group underbounded, chaotic, and disorganized because these forces can make it hard for the group to accomplish its task. The task force may need to establish clear authority for its work from someone at the agency who is higher up than either purchasing or maintenance. This sponsor will need to be clear about the group’s purpose and goals and hold the group groups will give and take from the web constantly, thereby creating an up-to-date and complete information archive of the construction project. In a less-than-ideal world, however, problems can still exist when groups do not give information in a timely manner or do not give it at all. If the subgroups within the social system of the transit construction project are all overbounded, they will have trouble knowing what information other groups need at what time. This trouble is not due to incompetence or indifference toward the other groups on the project, but is a simple lack of familiarity with the members of other groups and their expectations. When the common membership in the social system of the construction project is emphasized over membership in the different subgroups, the boundaries of each of the subgroups will loosen and information will begin to get where it needs to go. However, because of the inherent temporary nature of the social system created for construction projects and the more permanent nature of the subgroups involved in the projects, emphasizing common membership in the social system is easier said than done. Web-based knowledge management technologies are some- times thought of in terms of easing the process of commu- nication between groups that have common goals or tasks. When you use the theoretical framework about boundaries and groups, however, you attempt to create a new group that includes all of the people who need information about the project and who are trained to use the new web-based tech- nology. Technology is the communication enabler of this new group of all decision makers on a project. This new group spans across all the subgroups of the system by includ- ing a few key members of each. If any of the subgroups are overbounded, then members of those groups will be less likely to contribute to the project’s knowledge. They will be the “hold outs.” Old familiar communication breakdowns will take on a different form through the use of technology. At the same time, using web-based collaborative software can help a temporary system like a construction project to strengthen its boundaries. Implementing new technology to help carry information across group boundaries is not enough by itself. To be successful, the social system needs to be eval- uated and possibly changed to make sure that the boundaries of the groups are neither too loose nor too tight to exchange information. For example, if the group can’t work together productively and you believe that it’s because the construc- tion team needs to have stronger boundaries, you can recom- mend a more directive style of leadership for that group, or you can physically move the group so that it works more closely together, or you can arrange for the group to spend time together face-to-face to help the group establish per- sonal trust. You might examine the reward system to make sure that the group was rewarded for collaboration instead of for protecting the interests of individual companies. Changing the social system at the same time you change the technological system is the primary idea behind socio- technical design and is discussed further in the next section.

48 necessarily short term. Rather than continuing to allow the situation to continue to unravel, Swanson and his team took steps to impose a unifying framework on Raytheon. In bound- ary terms, they simultaneously • Clarified goals and priorities, • Asserted clear top-down authority, • Imposed clear role definitions on managers (Raytheon Six Sigma training and language), and • Improved the negative feelings of self-doubt in the system. All of these actions served to tighten the boundaries around the new Raytheon and weaken the boundaries around the four original companies. Swanson and his team succeeded—at the time of this writing, Raytheon’s stock price was $37.23. WHY TO STUDY BOUNDARY THEORY If you are looking at a struggling group, it helps to under- stand the theory underlying some of the things you are see- ing, because the theory will help you decide what course of action to take to correct the problem. For example, there is a great deal of rhetoric in business texts that urges managers to “empower their employees.” A common strategy is to bring many levels together to brainstorm vision and strategy for a unit or organization. But many of these ideas are based on research in the organizational behavior literature, where the study of overbounded systems is more prevalent. For over- bounded systems, breaking down barriers to communication and pushing decision-making power down in the organiza- tion are generally helpful ideas. However, if you are dealing with a group or a system that is underbounded, applying these same concepts can be disastrous. Dealing with overbounded groups always requires loosening the boundaries, and deal- ing with underbounded groups always requires tightening the boundaries. For underbounded groups, it may be more use- ful to create a strong leadership structure that consults with the team but retains the authority over strategy, mission, and vision. The classic tools of project management—setting inter- nal deadlines, monitoring them, and seeing deliverables—may be much more appropriate. accountable. People serving on the task force will need to have clearly defined roles for their contribution to the work. A clear meeting schedule and group membership will have to be established, and members will have to undergo training or adopt a new unifying theory for their work, or “common language.” Can this group come together as a virtual, web-based col- laborative group and satisfy some of these criteria? The answer is a qualified “yes” if the group follows the best prac- tices outlined later in the report. AN ACTUAL BOUNDARIES CASE: RAYTHEON SIX SIGMA In this study, Raytheon provided the clearest example of what happens when a company lacks a unifying framework. After the merger of the four companies, there were compet- ing ways of working and talking and much unproductive conflict. Alderfer writes, “Usually the multiple theories are associated with the various warring groups whose conflict keeps the system in constant turmoil. Finding a means to deal with the theoretical differences also reduces boundary per- meability and aids the system to establish a greater sense of wholeness.” The four original companies were the primary groups in the system, and the “new Raytheon” that encompassed them had far weaker boundaries. Membership in one’s original com- pany defined one’s reaction to others. For example, a best practices database was originally developed at Texas Instru- ments. Texas Instruments thought everyone would be thrilled to have the new database, but nobody looked at it but Texas Instruments. The new Raytheon was an underbounded group at the beginning, with the economic crisis to prove it. Think about the way that Bill Swanson worked to impose Six Sigma on the organization in a very authoritarian, top- down way. He flew around the globe to meet with almost every Raytheon employee to emphasize the importance of the ini- tiative. Many middle managers were trained; in fact, becom- ing a Six Sigma expert was encouraged for promotion. The impetus behind this Six Sigma reform was economic crisis; the stock price had plummeted from $70 a share to $17 a share, another characteristic of underbounded systems. Thinking was

49 CHAPTER 6 PRINCIPLES OF SOCIO-TECHNICAL DESIGN The principles of socio-technical design are based on two foundational ideas: • Organizations should design their technical and social systems concurrently. • The people who work in these systems should partici- pate in the system redesign. Fred Emery, one of the founders of socio-technical design, said, “The simple truth is that organizations harnessing the mental power of all their employees outperform organiza- tions that have brilliant leadership yet fail to harness the men- tal power of all their workers.”i Research and practice highlight at least four principles of socio-technical design that are relevant to thinking about how to introduce new technology. First, identify the diver- gent interests that different users have. Don’t work under the assumption that there is only one paradigmatic user. Second, let people who will work with the new technical system par- ticipate in building the social system that will work with it. Third, focus on the system that will deliver a new technol- ogy, as well as the technology itself. Fourth, do not over- design the implementation; use minimum critical specifica- tions. Let us briefly examine each of these principles. IDENTIFY THE DIVERGENT INTERESTS THAT DIFFERENT USERS HAVE In designing new technologies and their organizational supports, we need to take account of the variety of users who interact with the technology. Designers are frequently coun- seled to find out what the user wants, but most often there are different user subgroups whose needs and interests may con- flict. A good designing process takes an up-front account of these competing needs so that the resulting design optimally resolves the tensions and differences between groups of users. The design embodies a consensus. Thus, for example, main- tenance technicians and operators in many settings typically have different information needs from a series of displays. The operators (when they are at a console in a factory or in the con- ductor’s cab of a train car) want systems-level data that high- light the links between all the relevant components; mainte- nance technicians want more in-depth, below-the-surface data on particular subcomponents. In designing the circuitry, mem- ory banks, and displays for components, designers may need to take account of these different needs so that the result- ing informational capacity of the component can reason- ably satisfy both groups. Moreover, when the components are designed with participation from both groups, each group and the designer develop a deeper appreciation of the others’ needs and interests. The design process itself has integrating effects. LET PEOPLE WHO WILL WORK WITH THE NEW TECHNICAL SYSTEM PARTICIPATE IN BUILDING THE SOCIAL SYSTEM THAT WILL WORK WITH IT Participative methods are essential in redesigning a social system that complements the new technical system. In an interview with CFAR Principal Larry Hirschhorn, Harold Salzman described a study of the design of machining cells in an aerospace factory. The findings from the study showed how effective participative methods are. In the study, a design team containing machinists and other professionals made a storyboard with the proposed layout of cells, hung it on a wall of the shop, and invited machinists to suggest improvements to the basic placement of machine tools and the flow of parts. Machinists proposed changes on such issues as the bunching of tools at workstations, the type of tools to be used (e.g., hydraulic versus electrical), and the nature of certain operations. Rotating off and on the design team, machinists reviewed proposed vendors and helped select new equipment. Salzman reported that when the machining system came up online, workers were in much closer proximity to each other than before. They informally cross-trained one another, several decided to take math courses at a junior college to “trig” out problems, and they participated actively in a post- implementation suggestion system, sometimes doubling their income by participating in the cost savings resulting from their suggested improvements. Thus, a participative design process can not only create an effective technical system but i Participative Design, by Fred and Marrelyn Emery, adapted by Robert Rehm from Participative Design for Participative Democracy, a Tavistock anthology, edited by Eric Trist and Hugh Murray, Vol. 2, The Socio-Technical Perspective.

also stimulate a process of continued learning and improve- ment. This is critical to workers’ ongoing success as they transform from fixers to planners. FOCUS ON THE SYSTEM THAT WILL DELIVER A NEW TECHNOLOGY, AS WELL AS THE TECHNOLOGY ITSELF In her study of the dissemination of an expert system designed to help sales representatives configure computer systems, Dorothy Leonard-Barton found that many sales rep- resentative rated the program poorly because other compet- ing programs overloaded the computer networks that trans- mitted program data to local terminals.ii Because the sales representatives didn’t understand that the configuration program competed with others for space on the network, the representatives attributed the slow response time that they got from the expert system to the expert sys- tem itself. In other words, they confused the technology with the system for delivering it. In Chicago, the CTA found that one of the earliest prob- lems of implementation was the way that subcontractors in the field were using their scanning equipment. Using their equipment to scan drawings into the system was especially difficult for subcontractors. The CTA staff and consultants went into the field to check the equipment, sometimes deter- mining that the scanner wasn’t heavy duty enough to do the work, sometimes setting up the equipment for people in their trailers, and sometimes training people on the proper use of the equipment. The CTA staff and consultants then issued detailed guidelines for what types of scanning equipment should be purchased. Without this elaborate technical sup- port, the success of the whole web-based system was in jeop- ardy. People would likely have attributed the failure of the system to the new collaborative technology rather than to the scanning equipment that was necessary to deliver the collab- orative technology. As the above two examples suggest, new tools can be used and new technologies can be managed only if the context sur- rounding the tool, its delivery system broadly described, is redesigned or reconfigured. Component designers may take great pride in a particular module, technology, or tool, but users rarely distinguish between the component and the sys- tem that makes the component useable. This will be no less true of subway car operators and riders. When prototyping the social system to match the technical system, we design the tool or technology and its delivery system concurrently. 50 In this way, both the design and the resulting rise of new tools and technology are optimized. DO NOT OVER-DESIGN THE SYSTEM; USE A MINIMUM OF CRITICAL SPECIFICATIONS Eric Trist, one of the founders of socio-technical theory, said, When you come to designing something new, you don’t know what to do. When you’ve got change in a technologi- cal process and in the social process accompanying it, you can’t lay it all down in the beginning. You can’t program it and detail it. So we came up with the idea that you should only decide those things that, at any particular time and phase, have to be decided. You leave everything else open. . . . You specify the least that you have to, and then you get on with that, and that produces the next set of problems, and you fill in the next. . . . This is a critical truth because things are over designed and they have to be undone.iii Whenever an organization chooses to pilot a new approach, it is acknowledging its debt to this idea that some things can be learned only through experience. With the introduction of new technology, it is rarely possible to know all of the chal- lenges you will face and all of the effects it will have, even with meticulous planning on the part of knowledgeable people. We believe it is a best practice to create pilot projects and early forays to learn from experience before trying to broadly imple- ment new technology. If early participants can be recruited into the process as learners and improvers, you will have the benefit of their experience and encounter less resistance. The PANYNJ case exemplifies this idea. The leadership there consciously set out not to over-design the implemen- tation of new technology. While the leaders solicited input from many groups, they did not put together a formal process redesign team, believing that such a team would slow down progress. Instead, the leaders instituted a pilot project. Once there was experience with the technology, the leaders began to shift roles and create new roles to reflect what they had learned. This in turn changed their work processes. Pradip Mehta explained it this way: Put a system out there and it will illuminate the facts and it will help you. You have to make a very careful, intelligent decision to what degree you’re going to solve your business problems—I knew that if I had to wait to solve our work order tracking problems, [the new technology] would have taken me 3 more years to implement. You have 10 different entities involved! It is what it is. Just put it out there and get everyone to see the problems—otherwise we’d still be sitting here with our process flow charts. ii Leonard-Barton, Dorothy. Wellsprings of Knowledge: Building and Sustaining the Sources of Innovation. Harvard University Business School Press, 1995. iii Eric Trist in “An Interview with Eric Trist, Father of the Socio-technical Systems Approach,” by William M. Fox in The Journal of Applied Behavioral Science, Vol. 26, No. 2, 1990, pp. 259–279.

51 CHAPTER 7 SOCIAL RESISTANCE THEORY When the introduction of new technology disappoints or fails, it is often attributed to resistance on the part of workers who are still attached to the old way of doing things. That resistance may take active forms, as when someone with some authority in the organization questions the value of the tech- nology or flatly refuses to use it, but many times it takes the form of passive resistance. The psychological term “resis- tance” is in wide use in the culture, but in order to effectively respond to it, we need to understand it. Unless otherwise noted, the social resistance theory in this chapter comes from Robert Graham’s classic article, “Understanding the Benefits of Poor Communications.”i RESISTANCE MAY BE RATIONAL One perspective is to take a broader and more systems- oriented view of resistance and why it may be occurring. Graham argues the following: From the literature on innovation theory (and from common sense) we know that the criterion of acceptability of an inno- vation is the conviction on the part of potential recipients that the innovation will, in sum, contribute more importantly to the satisfaction of a network of wants and needs than to their frustration. Graham further points out that different people and groups within an organization have different interests, and, there- fore, “With many innovations it is often true that the needs and wants of one group of people will be satisfied while those of another group will be frustrated. That is, one person’s solu- tion can often become another person’s problem.” As management consultants, we often hear individuals in organizations attribute the organizational problems to “poor communication.” The underlying belief seems to be that if only communication could be improved or even made per- fect, there would be no conflict. In his article, Graham draws on the earlier work of psychoanalyst Charlotte Kursch to sug- gest that nothing could be farther from the truth. When some- one is calling for better communication, it is possible that one or more of four erroneous assumptions are at work: 1. Poor communication is a result of faulty technique. In reality, the problem may be the message rather than the technique. 2. Better communication will reduce strife and con- flict. It is sometimes the case that more clarity only serves to underscore conflict and bring it out into the open. Where people or groups are conflict averse, they may rely on poor communication to obscure their dif- ferences because they lack the skill or the courage to deal openly with them. 3. When conflict continues for a long time, lack of communication must be one of the basic problems. Graham points out that “During any prolonged conflict there is usually plenty of communication but just not much agreement.” Graham further says that “clarity has the disadvantage of tying the executives’ hands and limiting flexibility in dealing with contingencies as they arise.” One seemingly rational remedy would be to identify the particular motive or anxiety that was con- tributing to the stalemate, so that the benefits of clear communication are seen to outweigh the benefits of poor communication. However, any experienced con- sultant or manager will be able to think of examples when that approach has not sufficed. 4. It is in everyone’s interest to attain clarity. In real- ity, it is often in the interest of one or both parties to leave the situation fuzzy. Graham claims that fuzzy communication provides the following benefits: • Minimizes impact of poor planning (do not let others know you do not know what you are doing), • Permits faster decision making and minimizes objections, • Preserves the freedom to change your mind, • Allows you to say two things at the same time, • Allows you to say no nicely, • Helps you avoid confrontation and anxiety, and • Helps minimize opposition and criticism. In the case of the first and second erroneous assumptions, the answer is to first diagnose accurately that the resistance is quite rational, although it may not be publicly expressed. Improving the communication technique and getting the mes- sage clearer in cases like these will not only not improve the i Graham, Robert. “Understanding the Benefits of Poor Communications.” Interfaces, Vol. 11, No. 3 (June 1981), pp. 80–81.

situation, but it may exacerbate long-existing or suppressed conflict in the system. The remedy is to acknowledge that some resistance is ratio- nal and use a negotiation framework to address the different motivations and interests of the groups or departments. Dis- missing the resistance as bad or irrational will ignore the underlying conflicts it reflects. It is helpful to consider an alternative stance on conflict: Conflict is an opportunity to make progress on business outcomes and relationships and to strengthen the organization’s capacity. Conflict is often less about personalities than about the system. In this frame, con- flict can actually be seen as an entry point for productive and creative work on business issues. Can conflict be acknowl- edged in a nondestructive way? Yes, if you adopt a negotia- tion framework. It is helpful to look at underlying forces that may be at work when we encounter resistance. FORCES THAT UNDERLIE RESISTANCE Edgar Scheinii,iii believes that there is an inherent paradox of learning: Anxiety inhibits learning, but anxiety is also neces- sary if learning is going to happen. He distinguishes between two types of anxiety: “learning anxiety” and “survival anx- iety.” Learning anxiety comes from being afraid to try some- thing new for fear that it will be too difficult, that we will have to change old habits, or that we will look stupid attempting something new. Schein believes that you can’t talk people out of their learning anxieties and that these anxieties are the basis for resistance to change. Survival anxiety, however, is the realization that in order to survive, we will have to change. Schein believes that it takes survival anxiety to push people to the point where they are willing to take the risk to change. He argues, “The basic principle is that learning only happens when survival anxiety is greater than learning anxiety.” You can increase survival anxiety by threatening people with loss of jobs or valued rewards. This is the basis for moti- vation behind top-down (i.e., mandated) change programs where everyone from the CEO on down needs to “get with the program” or risk the loss of his or her job. However, the use of survival anxiety as a motivator also has some draw- backs. If corporate change efforts come and go, this type of message can make workers cynical. In response to the next change program, they may sink into even deeper passivity. As another approach, Schein argues that you can decrease learning anxiety by creating a safer environment for unlearn- ing the old way and learning the new way. The creation of 52 psychological safety, however, is very difficult when a com- pany is downsizing or reorganizing. During these times, peo- ple are fearful about their future and their jobs. Schein believes that real change does not begin until the organization’s survival is threatened in some way. This threat creates high levels of survival anxiety and learning anxiety. The remedy, Schein suggests, is for the CEO and top execu- tives to be willing to acknowledge their own vulnerabilities— first to share the risks they feel from the environment and the dangers they face. By acknowledging their own vulnerabili- ties and placing the imperative and the threat outside of the organization, the leadership can simultaneously create sur- vival anxiety and a safer learning environment. This kind of survival anxiety is linked to forces outside the organization, to the “burning platform” rather than to a simple top-down mandate, which is the organizational equivalent to “because I say so.” Survival anxiety can lead to what Schein calls “transformational learning,” but only if the CEO and top executives are willing to acknowledge their own vulnerabil- ities first. Schein advises, “If leaders really want workers to learn new things, they have to educate them about economic realities in a way that makes their messages credible.” One of the barriers to creating effective communities of practice or other forms of web-based group interaction is that openly discussing problems, sharing half-baked ideas, or thinking aloud in public doesn’t come naturally to most of us. As one community member said, “It’s hard to talk about your problems in front of a lot of people you don’t know.”iv In frustration, a senior executive at Raytheon complained, “Some people don’t like to show an unfinished product. . . . It’s like a personality defect. What would a psychologist call this? Some of these guys won’t even contribute in a meet- ing because it will look like they don’t have a well-formed thought—they want to wait until the end and then critique it.” It is particularly difficult for professionals (physicians, engineers, lawyers, etc.) to jump in and collaborate at the beginning of a thinking process, and it is easier to wait until someone else has formed a “straw man” product that they can critique. Larry Hirschhorn, CFAR Principal and author of several works on organizational behavior,v believes that when peo- ple are afraid to look incompetent in front of others, the ii Coutu, D. L. “The Anxiety of Learning” (Interview with Edgar Schein), Harvard Business Review. March 2002. iii Schein, Edgar. “Three Cultures of Management: The Key to Organizational Learn- ing,” Sloan Management Review. Fall 1996. iv McDermott, Richard. “Knowing in Community: 10 Critical Success Factors in Building Communities of Practice.” Community Intelligence Labs (CoIL), 2000. http://www.co-i-l.com/coil/knowledge-garden/cop/knowing.shtml v Hirschhorn, Larry. Beyond Mechanization: Work and Technology in a Postindus- trial Age. MIT Press, 1984, 1986. Hirschhorn, Larry. Managing in the New Team Environment: Skills, Tools, and Meth- ods. Addison, Wesley, Reading, 1990. Hirschhorn, Larry, and Carole K. Barnnett. The Psychodynamics of Organizations. Temple University Press, 1992. Hirschhorn, Larry. Reworking Authority: Leading and Following in the Post-Modern Organization. MIT Press, 1997. Hirschhorn, Larry, P. Noble, and T. Rankin. “Socio-Technical Systems in an Age of Mass Customization,” Journal of Technology and Engineers Management. Vol. 18, 2001.

most basic emotion underlying their reluctance to con- tribute is shame. “Shame,” he said, “is about deficits—the opposite of pride—when people fear their response is a reflection of their competence and self-worth, they feel inad- equate, they will hold back.” The remedy, he suggests, is to make some collective acknowledgment of the difficulty of the task (“Hey, no won- der we’re having difficulty doing this—it’s hard”) and fos- tering an environment of trust and feedback. Richard McDermott goes a step further and orchestrates meetings so that senior members of the community who are well respected ask for help from the group. After several events like this over the first few meetings, it becomes easier for others to start asking for help openly. The climate of evaluation can contribute to the problem. In a good authority system, you have both a system of eval- uating performance and a system for containing the anxiety about evaluation. The boss communicates when you will be evaluated, and it is not every minute, every second. In a good system, there is a sense that evaluation will be fair. In a loose and disorganized authority system, people will be more anx- 53 ious about evaluation. It may seem that it is coming from many different groups and in many different settings. This fearfulness about evaluation will in turn create barriers to collaborations, which involve taking risks. Professionals often work in organizations that are, by their nature, more diffuse and loosely structured (e.g., physicians in a medical center or attorneys in a private practice law firm). These environments can create the kind of evaluation climate that makes people less willing to take risks for fear of appearing foolish in front of their peers. Finally, there may be an identity piece that contributes to the anxiety about collaboration. Engineers, for example, are trained to be very precise in their knowledge and judgment. As one executive at Raytheon put it, “As an engineer I was trained to think four or five steps past anything I was willing to say out loud and nine or ten steps beyond anything I would commit to paper.” The training to be so conscientious makes sense when you think about the potential consequences of failure. In this sense, engineers are trained like physicians, to rely on the power of their own judgment to make life or death decisions and to stand by their decisions once ventured.