Karen Ruhleder (contact author)
University of Illinois, Urbana-Champaign
Graduate School of Library and Information Science
Xerox Palo Alto Research Center and
Institute for Research on Learning
Michael B. Elmes
Worcester Polytechnic Institute
Department of Management
Presented at the 1996 Annual Meeting of the Academy of Management
Recipient of the "Best Paper" Award in the OCIS Division
Many management scholars and practitioners have cast the 1990's as a time of profound change in the business climate requiring equally profound shifts in organizational structures and processes. "Familiar strategies, rigid hierarchies, and swollen middle management ranks [have become] the targets of major organizational reform" (Applegate, 1994:15) as successful companies reorganize to become more flexible and responsive (Haeckel, 1993; Kanter, 1989). The resulting organization will be driven by self-directed information specialists (Drucker, 1988:45) and will break away from outdated rules and assumptions to reengineer and bridge traditional functional boundaries (Hammer, 1990). Current technical trends seem to dovetail neatly with these new organizational visions as enterprise networks proliferate, groupware becomes more generally available, and the virtual or boundaryless organization becomes possible (Hotch, 1993; Manzi, 1994; Stedman, 1995; Zuboff, 1991; c.f. also Yates (1989:20) on earlier convergences of management systems and systems of communication).
This paper is based upon an ethnographic study of a small, unusual organization, The Holding Company (THC), that is trying to implement many of these radical approaches in support of its business mission: to reorganize and dispose of a set of holdings while realizing maximum economic gain for a parent company (click here for project description). In order to support that mission, the CEO and a small core of key players created a holding company characterized by a relatively flat hierarchy and a team-based, project-centered work environment. Various forms of groupware, including Lotus Development Corporation's Notes(r) and remote meeting technologies, have been implemented to support synchronous and asynchronous group work across an organization characterized by heavy travel and interaction with geographically distributed holdings.
This study builds on the existing literature on organizational learning and offers insight into the differences between the promises of radical organizational redesign and the realities of implementation. Our central concerns revolve around the relationship between technology, the new organizational structures, and the learning that has to take place in order to implement them effectively. Fieldwork at THC uncovered a variety of strategies through which people were learning about the technology and integrating it into their work in order to support the broader mission of the organization. It also uncovered a set of assumptions about learning and technology use that served as sources of tension and barriers to effective learning. We conclude this paper with some suggestions for turning these tensions into a source of dialogue about technology and organizational learning.
In the sections that follow, we outline some of the literature on which we draw and the methodology used to gather and analyze data about the company. We then describe the company in more detail, paying particular attention to the application of technologies to support specific work processes and a new organizational design. We also look at the barriers to learning that have arisen, and suggest ways of addressing them.
We draw on literatures that approach organizational learning from a narrative or cultural perspective (Cook and Yanow, 1993; Tenkasi and Boland, 1993). Many current models, such as those outlined in Garvin (1993), are based on the assumption that organizational learning can be compared to a brain taking in, analyzing, and interpreting information to generate knowledge about the environment. Learning is an individual conceptual activity, and knowledge resides in the human head (Argyris 1977, 1994; Simon 1991).
In contrast to this cognitive or "computer model of the mind" metaphor for organizational learning (Stubbart, 1989), a cultural perspective considers:
"...the capacity of an organization to learn how to do what it does, where what it learns is possessed not by individual members of the organization but by the aggregate itself. That is, when a group acquires the know-how associated with its ability to carry out its collective activities, that constitutes organizational learning" (Cook and Yanow, 1993: 378).
We believe that learning and problem-solving are situated, collective activities, and that organizational learning is best understood as a phenomenon that is located in the formal and informal social groupings that make up an organization. Of particular importance- since often ignored- are emergent "communities of practice" (COPs) (Brown and Duguid, 1991; Lave and Wenger, 1991):
"...naturally occurring groups that arise more or less spontaneously around a particular task, technology, or enterprise. COPs are self-organizing; they emerge in response to changing conditions and opportunities in the workplace." (Jordan, 1994:6)
In this view, learning and problem-solving involve a process of sensemaking around work practices that relies significantly on the construction of shared narratives and metaphors (Orr, 1986; Elmes and Kasouf, 1995). As such, the mastery of discrete skill-sets is deeply intertwined with and inseparable from the development of shared meanings associated with cultural artifacts (Jordan, 1989). The collapse of sensemaking can lead to the collapse of organization (Weick, 1993). When sensemaking activities take place around a specific set of technologies, the emergence of shared perspectives or "frames" for understanding appropriate use may both foster cohesion within a group and division across groups (Orlikowski and Gash, 1994). Learning thus becomes not a set of facts to be acquired by specified members of the organization, but a web of on-going conversations and developing social relationships that forms part of the infrastructure that supports on-going use (Kling, 1992; Ruhleder, 1995; Star and Ruhleder, 1994).
Our study draws on ethnographic field data collected by the first author over a period of four months in a project directed by the second author. Data collection was carried out through unstructured interviews, participant observation at THC and several business units, and review of paper documents and on-line materials. All 32 THC Associates, most of the seven temporary staff, and some members of the business units associated with THC were interviewed at least once over the course of the four months.
In addition, the ethnographer observed a variety of meetings, including project reviews and workgroup sessions, and conducted informal observations of situated learning and use of local technologies. She also traveled to five business units for 1-2 days each to attend a variety of meetings, including small group discussions, technology demonstrations, and a formal operations review. During the course of the fieldwork, she had access to the majority of the technologies available to THC employees including email and most on-line databases. Initial interviews identified technical learning and effective use of technologies as key issues within THC. Further interviews and field observations centered around better understanding of how people learned to use the technologies, and how they learned to integrate them into their work. Photographs, video tapes and documents supplement and illustrate these data.
Although THC is unusual in terms of its mission, its organization, and its use of technology, we believe that this single and unique case will provide generally useful insights for corporations in similar situations (c.f. March et al., 1991). Many of the difficulties faced by THC either support or cast a different light on the findings of other organizational studies. For instance, in an organization that stresses technical learning and experimentation across the board, some forms of power over users traditionally vested in the information systems function become diffused (Davies and Mitchell, 1994; Markus and Bjorn-Andersen, 1987); however, new loci of symbolic power emerge. Similarly, Eisenhardt has discussed the importance of case study research for generating "novel theory." By juxtaposing contradictory or paradoxical evidence between existing theory and the data, case study research tends to 'unfreeze' thinking about, in this case, the promises of new information technology-supported organizational forms and the realities of implementation in human settings (Eisenhardt, 1989).
In 1992, a major corporation hired a new CEO to manage one of their companies, a centrally-managed insurance company with several franchise entities in various commercial insurance businesses, with a mission to realize maximum economic return within a five-year time frame. These entities were streamlined and reorganized around specific business missions. After analysis, the decision was made to fully decentralize the company and pursue sequential disposition of the resulting business units. In order to do this, major pieces of the former "home office" were moved to the companies: legal, actuarial, HR, taxes, accounting, financial portfolio management, etc.
Out of this strategy emerged the vision of creating The Holding Company (THC)  as a management oversight organization physically and operationally removed from the insurance companies themselves. THC would coach the newly-created business units in learning to carry out their new responsibilities. One of the Managing Directors (MDs) explains, "While much of the experienced home office staff had moved to the entities, there would be a big learning curve ahead for these companies to stand on their own successfully in order to maximize value." Finally, the CEO wanted to carry out this complex mission with a holding company structure as lean as possible.
The process of forming the holding company took well over a year, and involved a physical move from the East Coast to the West Coast. Not everyone was prepared to adapt to the proposed changes in structure and work process (described more fully below), and "...many opted out [along the way] when it became apparent that management styles would change from hierarchically driven, closed structures to a flat, open, no perks," with 24 hour, 7 day availability (MD). In the process, the central management oversight team went from 250 to 32 in what this same MD describes as "an aggressive drive to a very different operating environment as well as culture." The final team included the CEO, five MDs, and 26 other highly skilled and motivated THC "Associates," characterized by another MD as, "the A-team, no question."
Given their collective responsibilities and constraints, and given the urgency of the mission, THC Associates had to be able to communicate and participate in both the insurance companies and the holding company 24 hours a day, 7 days a week from any location. "We had to find a way for everyone to participate in particular meetings from remote sites as well as to keep up with the work of other colleagues remotely ... [and to] aggressively mandate the rules to play/participate in the environment in order to execute this strategy" (MD). The work pace and style also required a new set of technologies that would support a "24 by 7" flexible work environment, that would provide information access regardless of time or place, and that would allow for meaningful organization of mission-critical information.
These needs were met by a series of organizational arrangements designed to eliminate traditional functional and hierarchical barriers while promoting teamwork and open access to people and information. These organizational arrangements were further supported by state-of-the-art computing and communication technologies, including various forms of groupware and remote meeting technologies. In contrast to other organizations, the CEO and a small core of key players actively championed these technologies and their ability to support the business mission. The CEO in particular was generally considered to be someone who is not only exceptionally knowledgeable about the industry as a whole, but "...picks up on the strengths of a technology and understands what the business application is" (manager is business unit). Below, we outline the convergence of organizational and technological innovations implemented to support the business mission.
The move from one coast to the other facilitated the creation of both a physical and virtual landscape that exemplifies the corporate philosophy of openness and equal access. Traditional status cues such as square footage, corner offices and various grades of furniture are eliminated. Instead, everyone (including the CEO) share a large, open room, and each person has an identical desk and workstation. There is no hierarchy of "perks"; everyone uses the same lunchroom, everyone gets a parking sticker. And everyone has a brilliant view of a bay through large windows surrounding the space on three sides.
Various meeting rooms and living room clusters provide formal and informal meeting spaces of varying degrees of privacy. In the open environment, "you get to see people and talk to people more" (MD), and "it makes it easier to work and to interact [with others]" (Associate). The benefits people mention include serendipitous interactions, overhearing conversations, and the ability to observe how people with other skills or greater expertise carry out their work. These physical arrangements suit an organization characterized by frequent travel, making people an immediate part of the organization when they are "in residence."
The "virtual space" of the company extends beyond the boundaries of THC headquarters. Each person, including the support staff, has an identical home set-up, and travelers carry laptops kept up-to-date by the information technology staff. People not only use email remotely but can also replicate the Notes environment at home and on the road:
"Everyone now has a home set-up, and a laptop for travel. They could be anywhere and do their work. At this point, it wouldn't matter where people lived physically. In essence, THC is ready to be a virtual corporation now." (CEO)
These kind of arrangements are designed to support the "24 by 7" operating environment with its needs for heavy travel, asynchronous communication, and near-instantaneous turn-around time. In particular, they support off-site work.
THC is constructed as a team-based, project-centered organization. In the course of flattening the organizational hierarchy, functional titles such as CIO or CFO have been eliminated. Each of the five "Managing Directors" (MD's) is responsible for one or more business units and for projects involving one or more functional areas. For instance, one MD oversees both the IS function and the Human Resources function. MD's receive a fixed annual salary without bonuses, but will share equally in the final profits when the mission is completed. For other Associates,
"THC works on a pay for performance basis. The salary is fixed, and depends on where you are in terms of the ability to influence things. There are end-of-year bonuses based on which salary band people are in, and then a formula based on which of their annually planned projects they completed, and what percentage they were responsible for in that completion." (MD)
The salary bands referred to above place people into one of five "Bands" according to the scope of their responsibilities. Annual bonuses at all Band levels are directly tied to Return on Equity (ROE), a figure available to all Associates and integrated into the bonus calculation formula.
There is no limit to mobility between Bands, and one's Band does not necessarily correspond to any kind of reporting structure. Band 1s, the "highest" Band, play a special role in that they tend to coordinate particular functional areas and serve as project leaders, reporting directly to the MD's. Associates at all Band levels combine functional tasks (i.e., tax reporting) with work as members of cross-functional project teams reporting to a project leader and/or the project MD. Depending on the nature of their work, some may spend 100% on functional tasks (i.e., secretarial support), others 100% on project work (i.e., strategic planning). Most Associates' work combines the two; the distribution is not necessarily tied to Band level. While THC is ideologically committed to a flat organization, these elements of structure represent the need for clear responsibilities for functional and project work. The structure is still far more fluid than traditional hierarchies, as people may be shifted around based on project needs, and report to multiple individuals. THC also employs seven temps and consultants to provide additional secretarial and technical support.
THC's daily work is generated by over 200 projects, created by the CEO and the MDs, that are formally evaluated on a yearly basis; additional projects are created throughout the year on an as-needed basis. Projects vary in nature and scope, and are the vehicle for carying out the work of the organization, especially non-routine forms of work. A project might reexamine leasing versus purchasing options across several business units, another might implement organizational changes within a single business unit. Each project is assigned a responsible MD, a project leader, and a project team which may include members of the business units. In order to support both functional responsibilities and cross-functional teamwork, THC makes a wide range of information available to all Associates:
"There are databases for planning, monitoring .... On the road, I use the phone, the fax, email, and databases: the quarterly reports, the balance sheet, the income statement. These are the databases that [key people who work with me] use, too." (MD)
Technology thus becomes a vehicle for inscribing and implementing ideas about organizational structure in support of the broader organizational mission.
A variety of computing technologies have been implemented to facilitate various modes of work. These include technologies familiar to most Associates, such as standard word processing and spreadsheet applications, presentation tools, and graphics packages. They also include groupware technologies, which were unfamiliar to most Associates prior to coming to THC, and which require not only new technical skills but the development of new ways of working. Technologies such as shared Notes databases and a calendaring/scheduling tool support asynchronous group work; the LiveBoard(r) (a networked electronic whiteboard), PictureTel(r) (remote video linkages across multiple sites), and desktop conferencing (joint use of a single application running on two desktops) support synchronous group work.
People regularly use Notes mail for interpersonal communication, to share memos and spreadsheets, to keep people "in the loop," and to document activities or decisions. Notes has also been pushed down into the business units, and serves as a conduit between the business unit, the MD liaison, and other THC Associates. Business units submit reports electronically, templates for financial statements are distributed via email, and meetings are coordinated on-line. Again, the ability to access mail anywhere, any time, supports a distributed, travel-oriented organization.
The various databases encompass routine and non-routine work, and form a kind of glue that holds different groups together. One key database is Work Manager, a Notes database developed to track the activities of THC's cross-functional teams, and to share project information across the organization. All THC projects are listed in Work Manager along with a project description, economic value, team composition, current status, relevant documents, and related email. Once a Work Manager project entry is created, all team members can add status updates.
Work Manager is designed to keep people up-to-date on project activities and give them an on-going overview of the work of the organization, facilitating horizontal and lateral communication. It is accessible to all THC Associates. An MD noted,
"In terms of technology, you use Work Manager. That's where you see where projects are going, communicate their status."
And someone else stated,
"It's a key source of information. The fact that the projects are in there is important, plus the status, so that I know where we are on something."
Work Manager is also intended to serve as a mechanism for clarifying corporate priorities. For instance, a new version of Work Manager contains the "EV" or economic value for each project as well as a better layout for project information.
In terms of representing work, however, it is not equally effective from everyone's perspective. Contributions through routine work (such as financial reporting or payroll functions) and support work (such as answering technical questions or making sure a meeting goes smoothly) are not easily reflected in Work Manager. A number of alternative databases (described in more detail below) have emerged to capture the aspects of project work not well-suited to Work Manager representation (c.f. Grudin, 1994; Grudin, 1989). Finally, some people find it hard to understand the often cryptic descriptions and updates:
"It's good to go through the Work Manager projects in a meeting. If you just read through them, you don't have an appreciation. In a meeting people explain them. You don't have to speak [the language of another functional area] to understand." (Associate)
Work Manager (projected via LiveBoard) literally serves as the focal point for periodic project review meetings. These kinds of public discussions around the projects may alleviate a problem the CEO identified towards the end of the first year of project-based work, when people got into a rush and "...were doing what the words said, not what we wanted to accomplish with a particular project."
The primary remote meeting technologies are LiveBoard, an electronic whiteboard which allows for shared viewing and manipulation of files between multiple sites, and PictureTel, which projects video images between sites. At THC, the LiveBoard and PictureTel set-up is located in the large conference room, with an additional LiveBoard in a medium-sized conference room. Most of the business units have these technologies in place or are in the process of implementing them. While many face-to-face meetings still occur, and people's work lives are characterized by heavy travel, these technologies were implemented to enable meetings in situations where it would not be cost- or time-effective to bring all the participants together in one location.
These technologies often support specific group work activities. For instance, members of the accounting group were developing an application together with members of ISBU , located on the East Coast. The initial meeting between the ISBU groups leader and two of the accountants took place face-to-face on the West Coast; subsequent meetings were conducted via PictureTel, using the LiveBoard to bring up a shared file with specifications. Desktop conferencing was later used to demonstrate a prototype remotely and get feedback from the accounting group. Another group met across four sites via a multi-point PictureTel linkage; this meeting involved THC, two business units, and the parent company.
A major benefit of these technologies is that they extend participation in important business events. They enabling Associates and members of the business units to attend meetings whose content relates to their own work, but in which they may not have been included if travel had been required. The ability to attend these reviews is especially valued by people who, in a conventional company, would not be at a high enough level to participate in these kinds of discussions.
The quarterly operations reviews serve as the best example of extending participation. Each quarter, the CEO and key Associates travel to the business units for the formal presentation of the quarterly reports. Reviews are broadcast live back to THC via PictureTel and LiveBoard and can be attended selectively. Quarterly reports are available beforehand via Notes, and both the reports and the meetings extend Associates' understanding of corporate operations:
"I try to go to two, three reviews a quarter...and to read all of the quarterly reports at a high level...You can select which reviews you will benefit from attending personally, or will add value to the work you do." (Associate)
"You learn a lot from the questions [the CEO] and others ask, and from the responses. Each quarter there is a different focus." (Associate)
Up to 20 additional THC Associates "attend" each of these meetings through the PictureTel and LiveBoard.
THC's Associates are well-educated, highly motivated, and hard-working. They continue to build their formal educational backgrounds through conferences and user group meetings, on-going professional programs, and journals. They seek out career opportunities that will offer them new challenges:
"[At my previous company] I was offered a position to head up an entrepreneurial branch of the company. The focus was on new products, new marketing methods. It was an incubator. It was a great learning experience, a chance to do something new. I came [to THC] to do special projects, with a SWAT team mentality." (Associate)
Learning more about the tangibles and non-tangibles of business including corporate operations, groups dynamics, and financial reporting structures was often cited as a reason for joining THC:
"I came here to learn. I looked at my career of over 20 years in finance and came away with the conclusion that there wouldn't be many chances this good in the future to hook on with something that was leading edge." (Associate)
"[This is] my first learning experience of how the financial function works with the systems it has and the data it creates, and how it supports management." (Associate)
However, few people outside of the IS staff came to THC with a lot of experience in using computers. While many were familiar with basic PC applications and a few had used mainframe-based applications in the past, some had never used a computer before joining THC. Shortly after the move into the new space, the IS staff offered a series of brown bag lunches on key technologies in order to get people started, and now answers specific questions as they arise. The CEO and others recommend books or articles that might be useful for understanding broader technology trends.
Still, many Associates feel that they do not fully understand all of the technologies available to them at THC. They feel overwhelmed by the number of new applications, the changing technical environment, and the need to balance technical learning with learning other aspects of the job. The need to use new forms of information technology was made clear to people when they were asked to join THC, and willingness to learn about new technologies was considered a condition of employment. However, people unfamiliar with computers, or those whose experience was limited to single-user applications, often found it difficult to articulate what kinds of skills they needed or what they needed to understand about the applications available to them in order to use them more effectively. Their sense of what they need to know thus does not always match well with the level at which others need or would like them to function. (This is akin to the dilemma faced by early user communities trying to work with systems developers to establish requirements: users didn't know enough about computing to know what kinds of thing they could ask for; similarly, Orlikowski and Gash (1993) found that users familiar with single-user applications had trouble reframing their view of the technology to use groupware effectively).
At the same time, a group of "techies" has emerged, including some people with no previous computer background. They also do not know all of the applications, but usually learn one particular application thoroughly enough to become technology mediators (Okamura et al., 1994) and tailors (Trigg and B¿dker, 1994). Their problem is a different one, in that the development of their technical skills is to a greater or lesser extent peripheral to the work they were originally hired to do. As such, their interest in the technology may conflict with others' sense of boundaries and may require negotiation to define authority and responsibility more clearly (Hirschhorn and Gilmore, 1992).
Their emergence has also set up a dynamic in which people ascribe to the "techies" of the organization (including the CEO) not just a particular set of expertise but an innate, hence not learnable or duplicable, ability to work with technology. A techie is "a real genius" who takes to technology "like a duck to water" while others "lag behind" significantly and are "idiots compared to these guys on the fringe." The time invested in learning the technologies and working with other techies or the IS staff is often overlooked by or invisible to those who attribute superior performance to innate ability.
Everyone at THC uses email, spreadsheets, a word processing package, and a set of Notes databases. Some of these applications are extremely malleable and lend themselves to tailoring for specific needs. THC management has developed pan-THC applications, such as Work Manager, that they expect to see used by everyone. At the same time, they also encourage people to learn about the available technologies and apply them innovatively to their own work. Not surprisingly, we found that work groups were tailoring these technologies to local needs, and developing alternate procedures for using them effectively, in some ways generating tensions between the desirability of pan-THC applications and the inherent appropriateness of locally adapted technologies. We have seen this phenomenon in a variety of other companies where front-line workers find top-down developed technologies not sufficiently responsive to their needs and begin to develop their own systems or workarounds (Gasser, 1986; Aronson et al, 1995; Bishop et al., 1994).
One issue for learning organizations, of course, is how to take advantage of local expertise and to examine locally developed applications for their usability in wider arenas. Local development has the advantage of responsiveness to user needs and intensive user input during the development and modification cycles. It avoids the most vexing problem of technology transfer, the lack of user buy-in that is so often deadly for technologies developed externally to the user community. However, neither criteria for appropriate adaptation nor a sense of how to strike an effective balance between using local and pan-THC technologies has emerged so far. Local experimentation contributes to a dialogue around the development of meaningful criteria.
We illustrate the kinds of experimentation taking place through three examples, each of which illustrates how groups within the organization are trying to adapt pan-THC technologies or concepts to specific local needs and situations. Accountants are exploring the THC-wide interest in relational databases through the development of an ad hoc reporting application. Work Manager "workarounds" led to a new set of Notes databases that extend THC-wide project organization and tracking concepts to local group work. And the community of practice that has emerged around the use of remote meeting technologies has created new forms of coordination work. These activities allow group members to learn both about the technologies and about their applicability to specific organizational settings.
Accountants and relational databases. The accountants combined an interest in relational databases with a need for better ad hoc financial reporting. They worked with ISBU over the course of several months to develop a suitable application. The process proceeded in almost textbook form, with the development of a needs analysis, a data model, and a prototype. They also explored remote meeting options for facilitating small group work. The initial meeting to discuss the users' needs took place in person, but subsequent meetings were conducted using PictureTel, LiveBoard, and desktop conferencing. It was the first time members on the accounting side had initiated meetings using these technologies, and the first time that ISBU members had used desktop conferencing to demonstrate a prototype. The participants considered this development effort as an application of technology that fits the broader vision:
"I'm spearheading a coordinated effort with ISBU to develop a relational database [application] for doing the balance sheet. That will free [us] up to do more analysis. We want to free up the process, move towards a new model." (Accounting Associate)
"You have to understand what it means to automate vs. to really reengineer. That's what they are looking at, what is the data source, where is it located, who are all the people involved? What is duplicate information, do they only need a subset of what they get?" (IS Staff)
Their experiences are being evaluated by others, who may build on both the application and the process to guide their own thinking about possible relational database applications:
I'd like to upgrade [a certain form of] reporting. I'm going to see what kind of expertise [they] are developing. (Associate)
These kinds of local interpretations of a technology take place alongside larger, "official" relational database development projects within the business units.
Development of Work Manager offshoots. Another set of activities focuses on applying the Work Manager concept to tracking and coordinating local activities. Many see Work Manager as appropriate for larger projects only, and not for short-term projects or sub-projects that "don't rise to the rank of Work Manager" (MD). Instead, they carry out several forms of adaptation work, modifying the official technology to track specific project or functional activities, while adjusting their work habits to incorporate updates of the "formal" technology based either on a personal schedule or external events such as project review meetings. Associates from two different groups explain their use of local Work Manager adaptations:
"We actually have our own little system. The Work Manager is for bigger projects, and a lot of our projects can be carried out in two weeks or a month. Every time I give [people] a project, I say put it on the system. When I'm on the road, I can pull this database up. I do this every week, every Monday morning." (MD)
"I want to keep [the group's project database] a Monday through Friday view for me, that's how I want to use it. A project that can't be done this week, I want to push it down [enter it into a database for longer-term projects]." (Associate)
A locally-acknowledged Notes expert and Work Manager "tailor" has contributed to the construction of many of these databases, some of which extend into the business units. They are used regularly, and their use corresponds with membership in a particular project team or functional area. Use involves negotiation around the granularity of entries into the workgroup databases, and around processes for monitoring and updating them. These systems occasionally generate tension between their use and use of the "official" Work Manager database, as they capture many of the activities and interactions intended for Work Manager.
Office staff and remote meetings. Finally, a community of practice has emerged around the activities involved in setting up remote meetings among office and IS support staff located across distributed sites. In theory, Associates could run their own remote meetings. In practice, successful meetings require a good deal of prep work. This includes setting up and testing connections, checking formats on LiveBoard files, and providing various forms of logistical support. Testing is a time-consuming, multi-step, multi-person process distributed over several sites, and rarely goes smoothly. It involves talking through the process and assigning responsibilities for the various functions. Teaching newcomers the mechanics of establishing a connection between two sites involves talking technically non-literate individuals through cryptic error messages, figuring out why things are "not staying clicked," or identifying the correct drive while trying to locate a file:
"The r-drive was correct. Some file on the [business unit] side had the address listed [incorrectly]. [Their staff] had changed all the files except this one. This is why we do these tests, because they never go smoothly." (Associate)
Problems are resolved by looking for peers who have gone through the process before. If they cannot be resolved jointly by the office support staff at the two sites, THC technical support staff provides assistance. The office support staff have devised new roles for themselves as meeting coordinators and facilitators. They are guided by a growing body of informal procedures and protocols, which may become codified over time. For instance, THC has recently started experimenting with meetings involving multiple sites, which has generated a written set of testing and meeting protocols developed by the office support staff in conjunction with the IS staff.
These activities reflect an on-going dialogue around the appropriate use and the usefulness of pan-THC's technologies for specific forms of work. They are a response to the tension between technologies implemented from the top which may support some organizational goals, but do not always support local work practices. They also form an important basis for organizational learning and sensemaking around technology as groups across the organization try to collectively establish a balance between local and global expectations for technology use. As such, they provide an important source of information for the organization about what it takes to truly implement and use these kinds of technologies on a regular basis.
People cite tremendous gains in their own technical skills, and continue to look for ways to enhance their ability to carry out collective activities organized around aspects of the business mission. Their experimentation is centered on the intersection of particular tasks and technologies. Yet, despite this on-going experimentation and collaboration, concerns persist across the organization that people aren't developing the desired level of technical skills, and that they aren't using the available technologies appropriately to facilitate the mission. We believe that these concerns and the tensions they engender are rooted in the way in which the organization views technical learning, and in the scope of dialogue around what kinds of learning should be going on.
THC embeds socially-based learning opportunities within the work of the organization through open quarterly operations reviews, informal mentoring, and access to a wide variety of information. Yet, individualistic explanations for resistance to technology also persist, and lack of use of particular applications or technical arrangements is often cast as a problem residing within the individual, rather than as lack of fit with local arrangements or work practices. For example, Work Manager's underutilization is attributed to inappropriate use of the technology- "it's all a matter of how people use it" (CEO), and "not everyone uses it effectively" (MD)- rather than serving as a jumping-off point for reexamining the fit of the tool with the requirements of work.
Many of the examples above, in contrast, strengthen the idea of learning as a social and cultural phenomenon. In this view, learning and adaptation are produced through interactions between people within a particular environment, and appropriate forms of learning are constructed collectively with respect to specific work settings and tasks. Formal work groups, such as the accounting group or the legal staff, form the basis for informal interactions around how a technology can be used, who will use it, who will create it and who will maintain it. Formal events, such as remote meetings, are sustained by informal agreements, unwritten procedures, and peer-based learning. The examples illustrate the use of local technologies to meet the needs of a formal or informal work group: getting better reports, keeping track of short-term projects or finely-grained project activities, making sure that remote meetings run smoothly.
All of these activities also represent important forms of sensemaking around technology, as the product and experience of one group becomes a springboard for others ideas about applying technologies to similar work settings. The adaptation of Work Manager, for instance, propagates not only a specific application of the Notes technology, but further stimulates ideas about organizing work and coordinating activities. And potential modifications to the accountants new application for use in the business units will generate a new set of relationships and expectations between the accountants at THC and their contacts in the business units. Sensemaking involves a broad range of organizational members who share ideas, trade stories, and demonstrate new applications for each other.
Finally, a social model of learning also suggests the need for dialogue around appropriate skill levels. Despite broad general interest in technical skill mastery, none of the above learning situations centered around deep technical learning on the part of the participants. Instead, people chose to link up with others whose expertise supplements or complements their own technical skills and functional knowledge. Several of the above examples involve "techies" or IS staff shaping and executing the ideas of people with fewer technical skills. The people involved consider this a legitimate form of collaboration that leads to mutual learning, though not necessarily the across-the-board enhancement of deep technical skills that others would find desirable.
The kinds of informal activities and relationships around tasks and technologies we outline above cannot be created. Their emergence within the organization can, however, be nurtured and sustained. A first step may be to recognize the multiple modalities of learning present within the organization; a second may involve broadening the range of voices taking part in a dialogue around legitimate forms of technology use and technical learning. Skill acquisition might occur more easily if positioned within specific work contexts, and informal activities may constitute the most appropriate base for constructing formal learning opportunities. The most important questions may be questions directed to these informal groups themselves, asking them what forms of learning they need to pursue, and how THC can best support them in doing so.
This study builds on the existing literature on organizational learning and offers insights into the differences between the promises of the reengineered organization and the day-to-day realities of implementing that vision within an organization characterized by time-critical, intense forms of work. THC is a unique organization that has reinvented itself around teams, cross-functional project work, and advanced technologies in support of its business mission. It has understood, from the very beginning, that organizational redesign is not a one-shot affair but a continuing process of change that requires an equally continuous learning process around the technologies that support the change. While it has not necessarily achieved the level of technical learning originally expected, multiple local experiments with the technology have emerged. What, then, are the implications of our investigations for THC and other companies launched on similar trajectories and working under similar conditions?
One of the major resources for learning that they have not yet explored fully comes from understanding learning as a social, rather than an individual cognitive process. Many of the restrictions and impasses that the organization constructs inadvertently have to do with seeing the individual as the locus of learning. This leads to an emphasis on individual effort and responsibility for acquiring particular kinds of knowledge and expertise, a view that is at best partial. To complement it, a social view of learning asks: what does the social group (be that an informal community of practice or a formally constituted team) need to learn in order to do its work, what resources does it have, how does it divide up its work, and what levels and divisions of expertise make sense for it? This view turns the process around, focusing not on knowledge transferred, but rather on organizational support for people's learning needs. In this model, learning is driven by the community of people who actually do the daily work. It is generated by the requirements of the work, rather than determined by abstract standards of what people ought to know.
Taking this view seriously requires an environment in which managers and resource providers listen hard to understand the ways in which new and pre-existing technologies fit into peoples working lives and increasingly into their non-working lives as well. It is recognized that what may be a good fit and appropriate for one team or one person may not be for another. This view honors diversity of work requirements, of people's talents and prior knowledge, and of the realities of their lives on and off the job (Garvin, 1993; Sachs, 1995; Kelley and Caplan, 1993; Orr, 1986).
Within this framework, technology is always embedded in practice, and the success of a particular technology, as well as the ease and readiness with which it is acquired, depend on its fit with people's ongoing endeavors. Rarely are these solitary endeavors. Massive amounts of informal, tacit, unacknowledged learning happens in the daily interactions of people- face-to-face, electronically, or through remote communication channels. The relevant issues for organizations must be, first, to recognize and legitimate multiple modalities of learning and, secondly, to support these informal communities and connections as productive venues for learning and work.
The case of THC suggests that managing the cultural and learning aspects of new organizational forms requires a deep understanding of the fundamental inseparability of technology, practice and community. It also suggests that reconceptualizing learning as a social process will provide organizations with a major resource for fostering productive use and application of technology to the broader organizational mission. This reconceptualization may be difficult in settings characterized by high pressure and little time for experimentation, yet we feel it is essential to the kind of broad dialogue required to develop and sustain the 21st century organization.
The authors wish to thank Randi Markussen, Susan Stuckey and members of THC for their careful, thoughtful commentary. This research was sponsored by Xerox Palo Alto Research Center, the Institute for Research on Learning, and THC.
Applegate, Lynda M. "Managing in an Information Age: Transforming the Organization for the 1990's." In Transforming Organizations with Information Technology, eds. R. Baskerville, S. Smithson, O. Ngwenyama, and J. I. DeGross. 1594. Elsevier Science B. V. (North-Holland), 1994.
Argyris, Chris. "Double Loop Learning in Organizations." Harvard Business Review (September-October): 115-125, 1977.
Argyris, Chris. "Good Communication that Blocks Learning." Harvard Business Review (July-August): 77-85, 1994.
Aronson, Meredith, Libby Bishop, Melissa Cefkin, Brigitte Jordan, Nancy Lawrence, Lindy Sullivan, Connie Preston and Julia Oesterle. Reflections on a Journey of Transformation: Learning, Growth, and Change at Xerox Business Services. Systemic Assessment Project, Final Report. Palo Alto, CA: Institute for Research on Learning. June, 1995.
Bishop, Libby, Melissa Cefkin, William Clancey, Brigitte Jordan, Julia Oesterle and Beth Tudor. A Learning Organization in the Making: A Report on the Work Practice and Design Project at the Xerox CAC in Dallas. Project Report. Palo Alto, CA: Institute for Research on Learning. Also available as Technical Report SPL 94-030 P94-00014, Xerox Palo Alto Research Center, Palo Alto, CA. 1994.
Brown, J. S. and P. Duguid. "Organizational Learning and Communities-of-Practice: Towards a Unified View of Working, Learning, and Innovation." Organizational Science 2 (1): 40-57, 1991.
Cook, Scott N. and Dvora Yanow. "Culture and Organizational Learning." Journal of Management Inquiry 2 (4) 373-390, 1993.
Davies, Lynda and Geof Mitchell. "The Dual Nature of the Impact of IT on Organizational Transformation." In Transforming Organizations with Information Technology, eds. R. Baskerville, S. Smithson, O. Ngwenyama, and J. I. DeGross. 243--261. Elsevier Science B. V. (North-Holland), 1994.
Drucker, Peter F. "The Coming of the New Organization." Harvard Business Review 66 (1): 4553, 1988.
Eisenhardt, Kathleen M. "Building Theories from Case Study Research." Academy of Management Review 14 (4): 532-550, 1989.
Elmes, Michael B. and Chickery J. Kasouf. "Knowledge Workers and Organizational Learning: Narratives from Biotechnology." Management Learning 26 (4): 403-422, 1995.
Garvin, David A. "Building a Learning Organization." Harvard Business Review (July-August): 78--91. 1993.
Gasser, Les. "The Integration of Computing and Routine Work." ACM Transactions on Office Information Systems 4 (3): 205-225, 1986.
Grudin, Jonathan. "Why groupware applications fail: Problems in design and evaluation." Office: Technology and People 4 (3): 245-264, 1989.
Grudin, Jonathan. "Groupware and Social Dynamics: Eight Challenges for Developers." Communications of the ACM 37 (1): 92-105, 1994.
Haeckel, Stephan H. and Richard L. Nolan. "Managing by Wire." Harvard Business Review (September-October): 122--132, 1993.
Hammer, Michael. "Reengineering Work: Don't Automate, Obliterate." Harvard Business Review (July-August):104-111, 1990.
Hirschhorn, Larry and Thomas Gilmore. "The New Boundaries of the 'Boundaryless' Company." Harvard Business Review (May-June): 104--115, 1992.
Hotch, Ripley. "Managing from a Distance." Nation's Business (February): 24--26, 1993.
Jordan, Brigitte. "Ethnographic Workplace Studies and Computer Supported Cooperative Work. Institute for Research on Learning. IRL94-0026. 1994.
Jordan, Brigitte. Cosmopolitical Obstetrics: Some Insights from the Training of Traditional Midwives. Social Science and Medicine 28:9:925-944, 1989.
Kanter, Rosabeth Moss. When Giants Learn to Dance. New York: Simon and Schuster, 1989.
Kelley, Robert and Janet Caplan. "How Bell Labs Creates Star Performers." Harvard Business Review 71 (4, July-August): 128-139, 1993.
Kling, Rob. "Behind the Terminal: The Critical Role of Computing Infrastructure in Effective Information Systems' Development and Use." In Challenges and Strategies for Research in Systems Development, eds. William Cotterman and James Senn. New York: John Wiley, 1992.
Lave, Jean and Etienne Wenger. Situated Learning: Legitimate Peripheral Participation. Cambridge: Cambridge University Press, 1991.
Manzi, Jim. "Groupware: Future Prospects and Scenarios." Telecommunications (January): 17--18, 1994.
March, James G., Lee S. Sproull, and Michael Tamuz. "Learning from Samples of One or Fewer." Organization Science 2 (1): 1-13, 1991.
Markus, M. Lynne and Niels Bjorn-Andersen. "Power Over Users: Its Exercise by System Professionals." Communications of the ACM 30 (6), 1987
Okamura, Kazuo, Masayo Fujimoto, Wanda J. Orlikowski, and JoAnn Yates. "Helping CSCW Applications Succeed: The Role of Mediators in the Context of Use." In Transcending Boundaries: Computer Supported Collaborative Work (CSCW'94) in Chapel Hill, NC, edited by John B. Smith and F. Don Smith, ACM, 5566, 1994.
Orlikowski, Wanda J. and Debra C. Gash. "Technological Frames: Making Sense of Information Technology in Organizations." ACM Transactions on Information Systems 12 (2): 174-207, 1994.
Orr, Julian. "Narratives at Work: Story Telling as Cooperative Diagnostic Activity." In Conference on Computer Supported Cooperative Work in Austin, TX, ACM 1986.
Ruhleder, Karen. "Computerization and Changes to Infrastructures for Knowledge Work." The Information Society 11 (2): 131-144, 1995.
Sachs, Patricia. "Transforming Work: Collaboration, Learning, and Design." Communications of the ACM 38 (9): 36-44, 1995.
Simon, Herbert A. "Bounded Rationality and Organizational Learning." Organization Science, 2(1):125-134, 1991.
Star, Susan Leigh and Karen Ruhleder. "Steps Towards and Ecology of Infrastructure: Complex Problems in Design and Access for Large-Scale Collaborative Systems." In Computer Supported Cooperative Work in Chapel Hill, NC, ACM 1994.
Stedman, Marina. "A Practical Guide to Lotus Notes Implementation." Information Management & Technology 28 (3): 112--116, 1995.
Stubbart, Charles I. "Managerial Cognition: A Missing Link in Strategic Management Research." Journal of Management Studies, 24 (4): 325-347, 1989.
Tenkasi, Ramkrishnan V. and Richard J. Boland, Jr. "Locating Meaning Making in Organizational Learning: The Narrative Basis of Cognition." in Research in Organizational Change and Development Vol.7, eds. William Passmore and Richard Woodman. Greenwich, CT: JAI Press, pp. 77-103, 1993.
Trigg, Randall H. and Susanne B¿dker. "From Implementation to Design: Tailoring and the Emergence of Systematization in CSCW." In Conference on Computer Supported Cooperative Work in Chapel Hill, NC, ACM Press, 45--54, 1994.
Weick, Karl E. "The Collapse of Sensemaking in Organizations: The Mann Gulch Disaster." Administrative Science Quarterly 38 (December): 628-652, 1993.
Yates, JoAnne. Control through Communication. Baltimore: The Johns Hopkins University Press, 1989.
Zuboff, Shoshana. "Informate the Enterprise: An Agenda for the Twenty-First Century." National Forum 71(3), 1991.
Copyright © 1996 by Karen Ruhleder, Brigitte Jordan, and Michael E. Elmes.