Advancing the Value of Ethnography

Trajectories of Change in Global Enterprise Transformation


Download PDF

Cite this article:

Ethnographic Praxis in Industry Conference Proceedings 2011, pp. 134–151.

This paper reports on the efforts of a global IT services company to transform the way it delivers IT outsourcing services. The change initiative was designed to bring about a radical transformation in the how work gets done across the enterprise with the expected benefit of delivering greater service quality and reliability at a lower cost. In addition, the standardization of processes and tools would allow work to move more freely from one location to another thus creating flexibility to meet changing demands. Based on a study of the impact of this initiative on four global delivery centers we explore how change occurs within organizations both as an ongoing achievement and as the result of explicit corporate initiatives. Taking account of the particular historic, geographic, demographic, socioeconomic, and cultural characteristics of individual delivery centers we trace trajectories of change with the aim of providing both a broad synoptic view given these differences in delivery centers characteristics and a detailed performative account of the transformation from the perspective of the day-to-day actions of employees.

Everything flows, nothing stands still


The contingency of globalization may originate in its production but it can be discovered in its variation.

Burawoy, 2001


Enterprise transformation is on the mind of corporate leaders these days as globalizing markets, the availability of skilled workers worldwide, the decision to outsource services, and growth through mergers and acquisitions are on the rise. Companies are looking for ways to integrate across the enterprise as customers, workers, and other resources are widely distributed and where local histories and differing social and economic contexts create challenges for unifying around shared objectives and similar ways of doing things. Different strategies have emerged to bring about enterprise-wide transformation, but most involve top-down definitions of common goals, policies, processes, and technologies.

The enterprise transformation that is our focus involved radical changes in how IT outsourcing services1 were to be delivered. The new work standards and divisions of labor were designed by a corporate “center of competency” in collaboration with an outside management consulting firm. The new model of work, here referred to as the Service Delivery Standard (SDS)2, was being rolled-out across the enterprise to delivery centers (DC) located around the world. The changes were intended to bring about a transformation in how work gets done with the expected benefit of delivering greater service quality and reliability at a lower cost. In addition, the standardization of processes and tools would allow work to move more freely from one location to another thus creating flexibility to meet changing demands. The new model of work was based on Lean principles (Womack and Jones, 1996) and involved segmenting the work, pooling human resources, and using metrics to identify inefficiencies and prevent defects (Anderson et al., 2011). A fundamental goal of the transformation was to make the way services are delivered “the same” no matter the location from which they are delivered. To insure that this goal was achieved part of the implementation strategy involved the regular assessment of employees to ascertain how their work conformed to the model’s specifications.

Our interest was to understand what changes were occurring within the enterprise in response to this initiative to transform work viewed both as an ongoing achievement and as the result of explicit corporate dictums. In particular, our concern was in how specific historic, geographic, demographic, socioeconomic, and cultural characteristics of individual delivery centers were shaping trajectories of change with the aim of providing both a broad synoptic view of the changes and a more detailed performative account of the transformation from the perspective of the day-to-day actions of employees. We wanted to explore the role of enterprise initiatives in directing organizational change while acknowledging that change in the end involves “an ongoing improvisation enacted by organizational actors trying to make sense of and act coherently in the world” (Orlikowski, 1996:65).


Enterprise transformation has been defined as organizational change that is “… not just routine change but fundamental change that substantially alters an organization’s relationships with one or more key constituencies, e.g., customers, employees, suppliers, and investors (Rouse, 2005, 275).” Rouse goes on to suggest three ways to transform organizations which entail changes in the work of the enterprise. The first involves improvements in how the current work is performed (e.g., establishing process standards to reduce variability); the second introduces changes to the work itself (e.g. proposing new divisions of labor regarding tasks and job roles); and the third shifts the work to another entity all together (e.g. outsourcing work to another organization). SDS involved deploying all three levers of transformation.

The enterprise transformation literature builds on organizational change theories that focus on managing the change required when internal or external forces threaten the stability and viability of the firm. These organizational theories have been directed at understanding how stability is maintained, viewing change as something that disrupts the status quo and presents challenges for management even when initiated from within the organization. In this sense change is viewed as a necessary evil that allows organizations to survive in turbulent times. More recently however, organizational scholars have been rethinking the ontological primacy of organizational stability and instead are coming to view change as an ordinary and ongoing process of organizations, not as an exceptional response to outside forces (Tsoukas & Chia, 2002).

Similarly, the literature on the management of organizational change has been dominated by synoptic accounts that attempt to identify the antecedents of particular trajectories of change, asking what conditions enable change to take place and determine the success of the changes (e.g. leadership, skilled workforce, adequate investments, etc.). However there is a growing cadre of organizational scholars who argue for a more performative account of change that focuses on how change is enacted in practice (Barley, 1990; Feldman, 2000; Kling and Lamb, 1999; Olikowski, 1996, 2000; Weick and Quinn, 1999; etc.). This line of inquiry looks to understand how change emerges from the situated actions of people over time and views change as a normal and necessary part of organizational life. As such the processes of enacting change are also the processes that enable stability in the sense that they allow for organizational continuance.

Synoptic accounts allow us to notice patterns at different points in time, while performative accounts allow us to see the emergence and accomplishment of change in and through the actions of organization members. As Tsoukas & Chia (2002) suggest both synoptic and performative accounts contribute to our understanding of organizational change and are useful in guiding actions that direct particular change trajectories.

The current focus on enterprise transformation is related to the rise of the multinational corporation with its ability to tap into markets and resources around the globe. Some argue that multinational corporations are giving way to globally integrated enterprises where differences between subsidiaries, suppliers and partners are minimized through “… standardized technologies and business operations all over the world, interlinking and facilitating work both within and among companies (Palmisano, 2006:129).” Integrating across the enterprise drives efforts to standardize business processes and technology infrastructures, leading to initiatives like the rollout of SDS, the new model of work. Because multinational organizations, qua globally integrated enterprises, cross institutional and national divides, they create what Morgan (2001) has identified as “transnational social spaces” where flows of people, ideas, resources and practices intersect.

It is precisely in this context of transnational social spaces that ethnographic research can help reveal differences in the meanings, practices, and actions of actors who participate in these flows (Sharpe, 2007). Furthermore, as Churchill and Whalen (2005:180) point out focusing on the “detailed, everyday work practices and the day-to-day sociality of employees” facilitates not only understanding the accomplishment of organizational change, but also its management and enablement. Noting the limited involvement of ethnographic researchers in change management Holme (2010) argues for an anthropological agenda focused on understanding the enablers of change. We aim to contribute to the exploration of the role of the ethnographic perspective on understanding and affecting enterprise transformation.


Efforts at standard making are problematic if conceived of as simply the implementation of standard tools and processes without acknowledging the requirement of translation and localization (Cefkin et al., 2007). By ignoring the multiplicities and inconsistencies in the social order the goals of standardization are undermined. A central question raised by our study of the rollout of SDS concerns how to achieve integration across the enterprise while at the same time acknowledging and enabling local interpretation and enactment of standards.

While recognizing the limits of standardization, it is also evident that the globalizing of economic activity is leading to a nominal embrace of shared business and technology standards that purport to give companies greater options as to who will perform the work of the enterprise and from what locations. Competition for workers and resources likewise creates pressure to adopt uniform policies and work standards (Tempel & Walgenbach, 2007) and these become the expected norms for workers and employers alike. However, the push to standardize organizational forms and management policies, driven by the desire to align tools and process across the enterprise, risks disempowering employees who operate in differing social, cultural, linguistic, and economic contexts. The imposition of hegemonic ways of doing things3, originating from corporate headquarters often without consideration for local contexts, can lead to ill-fitted practices, and unhappy and unmotivated workers.

Given the interest in developing and implementing standards, questions remain about how standards are achieved, what mechanisms are available to bring about adoption, including normative pressures, imitation, and even coercion. As Tempel & Walgenbach (ibid:14) note “… multinational companies can exert strong coercive pressures on local subsidiary management and employment representatives to introduce standardized human resource management and work organization policies despite national institutional differences.” Conversely we must also ask how standards are resisted, hybridized, transformed, and indigenized through the local agency (Sharpe 2007). In this highly contested terrain of globalization and top down directives, enterprises nonetheless evolve and standards are translated and become integrated into local institutional contexts. How this occurs is the focus of the current study.


The study of the rollout of the Service Delivery Standard was initiated at the request of the vice president of human resources who wanted to assess how employees understood the new model of work, how the model was affecting their day-to-day experiences at work, and what changes might be made to the model and its implementation to improve the situation for the workers. The project also was supported by the executive who managed the “center of competence” responsible for the design of the SDS model of work and for the success of this transformational initiative. Their interests were in finding ways to both evolve the model and make sure its implementation was delivering on the promise of greater service quality and reliability at a lower cost without creating negative impacts on employees’ work environment.

Data collection and analysis

The study focused on four global delivery centers, two in the United States, one in the Central Europe and one in India, chosen because they represented delivery center variation in terms of size (from 300 to 8,000 employees) and length of time employees had been working under the new model (from less than a year to over 2 years). Our analysis was based on in situ interviews and facilitated discussions with more than 150 employees across the four delivery centers, interviews with six members of the center of competency responsible for the design and implementation of the new model, observations of employees at work, and demographic and performance data for the four delivery centers.


TABLE 1. Delivery Center Participants

We chose an ethnographic approach to address our interest in how enterprise-wide transformation initiatives were adopted, resisted, and altered by workers as they responded to the everyday realities of their work situations. We were able to gain delivery center assistance in arranging field visits, including in situ interviews and facilitated discussions with data center employees in part because of the project’s high level of sponsorship. That said we were also careful to make sure that the employees who participated represented a broad spectrum of workers with varying experiences with the company and with the new model of work (see table 1 for a breakdown of study participants).

Based on conversations with the vice president of human resources and others, we knew that the new model of work had encountered some resistance, with concern among employees that the model might have an unfavorable impact on their work situation, including the possibility of being reassigned or even losing their jobs. Under these circumstances it was essential that the anonymity of participants be strictly maintained and that we establish clear boundaries between ourselves and the sponsors of the research. We were careful not to be too closely aligned with members of the center of competency who regularly visited the delivery centers to assess compliance with the model. We took great care in explaining who we were and how the results of our study would be used. We verbally committed to making sure the issues raised by participants would be communicated with the explicit aim of improving their situation. While there was some skepticism that the input provided would actually result in change, employees nonetheless welcomed the opportunity to share their experiences with the hope of contributing to enhancements to the model and its implementation.

Our concern with maintaining participant anonymity meant that we chose not to audio record the interviews and facilitated discussions, but instead the two researchers took as close to verbatim notes as possible and these records were combined to provide the primary data that informed our analysis. In addition, notes were taken during and after observations on the “shop floor” and these too contributed to our analysis. Finally demographic information (e.g., length of time employed by the company, time in the workforce and in current position, etc.) and performance information (e.g. customer satisfaction measures, service level agreement attainment, etc.) for each of the delivery centers were collected.

The interview notes were reviewed and initial themes identified. Individual ‘quotes’ were then associated with these initial themes and in the process the themes were combined and revised. During a week-long working session the two researchers grouped the themes into six overarching issue areas and for each area a composite “voice of the employee” narrative was created drawing on the language used by employees in interviews and facilitated discussions. Differences in delivery center characteristics were defined and assessed for their impact on employee experiences. In our final report to our sponsors we focused on 6 issue areas4 as they played out in the 4 delivery centers and offered initial recommendations for ways of addressing problems identified. Since our sponsors were committed to implementing a standard model of work, in our report we focused on ways to improve the model and its implementation, without recommending a radical rethinking of the goals or the practicality of achieving the transformation.


Before describing the SDS model of work it is useful to give a sense of what is involved in doing IT service delivery. System administrators who typically perform this work are responsible for maintaining and operating a company’s computer systems and networks; doing such things as installing and maintaining servers; managing the applications that run on servers, desktop computers, and mainframes; responding to service outages; diagnosing problems and identifying solutions; and managing systems-related projects (Barrett et. al. 2004). Some of the work of system administration is organized by problem and change tickets that define problems to be fixed or changes to be undertaken to enable the ongoing operation of the IT system. It is the responsibility of the system administrators to respond to tickets, assembling the people and resources required, and then reporting back (closing the ticket) once the work is completed. System administrators usually have post high school technical training (e.g. university or technical school) although some learn on the job.

As mentioned earlier, the new model of work was based on Lean principles. Specifically it involved four main elements: (1) work segmentation, (2) pooling human resources, (3) using metrics to identify inefficiencies and prevent defects, and (4) worker co-location.

Work segmentation

Work was segmented into three main levels of difficulty from simple to complex. Workers were similarly grouped into three distinct job categories depending on the skill level required to perform the work. As tickets came in a dispatcher (often a system administrator) would assign the tickets to available system administrators with the required level of skill as determined by their job role. So the simplest tickets or service requests would go to those in the least skilled job category and the most complex to those in the highest job skill level (see Figure 1).

Pooling human resources

System administrators were placed in pools made up of employees with a mix of job skill levels and responsible for a particular service line (e.g. server management, network security, or mainframe). Pools ranged in size, but typical was around 25 workers per pool. Pools were responsible for providing service to multiple clients, the number in part determined by the size of the service contract and the type of service provided. In some cases for very large contracts or those with special security needs a pool might be dedicated to a single client, but this was an exception to the model which sought to distribute skill, knowledge, and experience across client accounts by having system administrators provide service to more than one client. It is important to note that in the old model or work system administrators were dedicated to a single client.


FIGURE 1 New model of work

Metrics to identify inefficiencies and prevent defects

As mentioned earlier SDS was built on Lean principles that emphasize finding ways to eliminate waste and prevent defects (continual improvement). To achieve these goals it was important to have uniform metrics that could be used to identify opportunities to cut costs and systematically respond to defects or recurring problems. The new model had a defined set of metrics that pools were asked to produce with the help of specialists called quality analysts.


As part of the SDS model workers were required to be co-located, most often in delivery centers, where those assigned to a given pool would sit next to each other facilitating the daily pool meet-ups5 that were also specified in the model. Co-location was stipulated in the model to better enable learning and sharing of experience and expertise. For some workers this was not a significant change as they had already been working from delivery center locations, but for others co-location meant a considerable adjustment as previously they had worked from their homes, from client sites, or from one of the company’s mobility centers.


As mentioned earlier the new way of working was being rolled out globally with the expectation that delivery centers would adopt the SDS model of work. For our study we selected four delivery centers that differed from one another with respect to a broad set of characteristics including the history of the delivery center, employee demographics, job opportunities inside and outside the company, types of customers served, and mix of services delivered. We now explore how these differences shaped trajectories of change, offering both synoptic and performative accounts. The synoptic accounts focus on differences in delivery center responses to the new model of work and the performative accounts show the emergence and accomplishment of change in and through the day-to-day actions of workers.

Synoptic accounts of model implementation

Synoptic accounts characterize change as a series of accomplishments whose variations are tied to particular causal antecedents (in our case things like the prior experiences of employees, the history and location of delivery centers) and view change as occurring in stages often depicted as a series of episodic events during which the organization transforms, with each stage representing a point in time. Synoptic accounts link characteristics of organizations to the changes that have occurred and offer explanations for particular change trajectories (Tsoukas & Chia, 2002). Below we provide a few examples of how particular characteristics of the four delivery centers shaped responses to SDS.

USA-1 – This delivery center was located at a large company facility that housed many employees with the delivery center’s relatively small number of employees (approximately 300 workers) making up only a fraction of the overall workforce at this location. The workers at this DC had lengthy histories with the company, most being seasoned and experienced workers very familiar with the old model of work. Many of the employees had previously worked from home or a customer site before being asked to relocate to the delivery center site as specified by SDS.

Responses to the model at this delivery center were related to the deep familiarity workers had with the old model which they felt provided them more autonomy, better use of their skills, and a more rewarding relationship with the clients they served. In addition, co-location had created a hardship for some workers who now had to commute long distances to work. Over time however workers began to appreciate the camaraderie and knowledge sharing opportunities co-location provided and most had come to terms with this new work requirement. Furthermore, this was the first delivery center to undergo the transition to the SDS model of work and that combined with the high level of worker experience and skill meant that employees at this DC often found themselves in the position of having to help newer employees at other DCs. While this work was sometimes rewarding, it added to dissatisfaction with the model because these efforts were not explicitly or adequately rewarded, falling outside the standard SDS performance metrics.

USA-2 In contrast to USA-1 this delivery center was located at a new company facility in the midsection of the United States. The site was acquired specifically to house the new delivery center. Most of the workers were hired to fill job roles defined by the new model and as such they had little or no experience working under the old model. For a number of these new employees this was their first job after college. Others were making career or job location changes necessary to take advantage of the new job openings at this DC6.

Employees’ lack of experience working in a large global company and their newness to the specific requirements of the job created a different set of challenges for the adoption of the SDS model. Without a history of providing service to particular clients, these workers needed to quickly come up to speed on the specifics of clients’ technical and organizational infrastructure. This requirement was exacerbated by the fact that the pool structure meant that system administrators had to provide service to numerous clients, multiplying the learning necessary. The lack of work experience of some employees coupled with limited client specific knowledge sometimes gave clients the impression that the system administrators working on their problem were not completely on top of things.

Central Europe This delivery center had been in operation for a number of years being one of the first delivery centers in Central Europe. As such there was a group of highly skilled technical workers who had experience with the old model where system administrators were dedicated to a single client and where there was no strict segmentation of the work into skill levels and competency areas. This delivery center mainly served clients headquartered in Europe which included a number of smaller firms in addition to the large multinationals. In addition, because of their distance from corporate headquarters in the US and because the client base was primarily European, the delivery center previously had been left on its own to develop operational strategies and the tools and processes suited to their particular context.

For these workers the new model of work was a double edged sword in that they welcomed the opportunity to become more integrated with the larger company, now able to more easily connect with co-workers in other locations since they were sharing the same technical language, were implementing the same processes, and were adopting the same technologies as specified by the SDS model. Their expectation was that they would be able to help define and evolve the company’s new service delivery model drawing on their experiences and the innovations they had made in processes and enabling technologies. While welcoming the opportunity to become more integrated into the enterprise, some of these highly skilled technical workers felt frustration that their ideas, including tools they had developed to enable SDS, were not being taken up. They were disappointed that they were not having greater influence on changes in how services were being delivered globally.

India This was the largest of the four delivery centers with over 8,000 employees and growing. Because of rapid growth and high turnover at the delivery center many of the employees were recent graduates from Indian technical universities with little practical job experience. These workers were extremely focused on their careers and opportunities for advancement seeing their current position as a stepping stone to a better job, requiring greater technical expertise. Given the fact that there were many high tech companies with operations nearby, there was a great deal of competition for top technical talent.

The Indian workers were realistic about why the company had introduced the new model of work, understanding that the SDS model was in part designed to reduce costs and one way to do this was segment the work so that lower skilled and therefore lower paid workers could perform some tasks, reserving the more complex tasks for a fewer more highly skilled and paid workers. However, the Indian employees were not content to remain in entry level positions for long, demanding opportunities for advancement with some already believing they had the skills required for the more complex work. The workers at this DC were on the look out for other jobs with other divisions within the company or other outside firms. This not only contributed to turnover as workers left for better opportunities, but it also led some to be impatient with restrictions in the model that specified which job roles could perform which tasks.

Performative accounts of worker enactment of change

Synoptic accounts of change are not focused on the enactment of change in the everyday actions of people instead look to identify factors that define particular change trajectories. Performative accounts on the other hand depict what actually occurs as people and organizations respond to new constraints and opportunities and adapt their actions to “fit” the situation at hand. They show the agility and creativity with which people get on with things in face of new contingencies. To begin to get access to these more performative aspects of change, it is important to see how eople adjust their actions in response both to subtle and dramatic changes in their environment – to see how change emerges from the day-to-day actions of organizational members (Feldman, 2000). Ethnography is well suited to reveal the everydayness of change as enacted by organizational members.

In our performative accounts we looked for breakdowns that focused our attention on actions taken to align work with the new model and that reflected the accommodations made to get on with the work at hand. Below we provide a few examples of how elements of the SDS model were modified in practice to accommodate the demands of the work.

Pooling resources As mentioned earlier a central feature of the SDS model was creating pools of system administrators who provided IT services to multiple clients. The benefit of this feature was that it provided better utilization of skills and balanced the work across clients (e.g. highly skilled workers supported multiple clients and resources could be more easily allocated across clients depending on work load). While there was evidence that pooling was helping to achieve these goals, there was also a downside. System administrators were finding it difficult to develop deep knowledge of the technical and business infrastructure for the multiple clients served by their pool. In the old model they were dedicated to a single client and as such became very familiar with the particular technical infrastructure and operational policies of the client. Attaining this depth of knowledge was very difficult, if not impossible, when serving multiple clients. The pain was felt not only by the system administrators, but also by some clients who expected a certain level of familiarity with their technical environment and preferred ways of interacting. To respond pool managers began to assign system administrators to particular clients where they focused on tickets from “their” client, while still available to serve other clients as time and expertise allowed. Also as new clients were added to existing pools specific system administrators would be asked to focus on the new account making sure that someone in the pool was developing client specific knowledge which later would become a resource for others in the pool.

Dispatch The SDS model originally specified that the role of the dispatcher would to be filled by a senior system administrator who had the ability to assess the skill level required to respond to particular problem and change tickets. While technical knowledge was useful in deciphering tickets, in practice the work of dispatch turned out to be more administrative than technical. Dispatchers found themselves monitoring incoming tickets from multiple clients and entering ticket information into the SDS ticketing tool so the work could be properly assigned and tracked. Senior system administrators began to avoid taking on the dispatcher role, some even refusing the job. In response some pools began to make the role of dispatcher a rotating assignment so no one person had to stay in the job for too long and others relaxed the requirement that the dispatcher had to be a senior system administrator.

Work segmentation Another important characteristic of the SDS model was the ability to segment the work according to skill levels of system administrators and the type of service provided. This resulted in pools focusing on only one service line (e.g. server management, network, application management, mainframe, help desk). While segmentation created efficiencies, the model did not adequately account for the work of integration across service lines as not all problems or change requests involved only one service line. This problem was exacerbated when the pools responsible for different service lines (for the same client) were located in different delivery centers, making coordinate and integration even more challenging. A new role not originally specified in the model was emerging to help address this problem. Although the particulars differed from one delivery center to the next, it became apparent that there needed to be someone in the delivery center who was responsible for ensuring that pools worked together as required to integrate across service lines. In addition at the global level fragmentation was being addressed by making an effort to locate the pools associated with the service lines for any given client in same delivery center.

As these examples show employees were improvising and experimenting in response to the requirements of the work and in so doing were adjusting their practices to better enable the overall transformation. The SDS model was not simply reproduced at each of the delivery centers, but through its enactment the model was changing in both subtle and more dramatic ways.


The results of our research resulted in five follow-on projects three of which are relevant to the issues explored in this paper. The first involved developing a “shared ownership model” for SDS where local delivery centers would have a say in changes to SDS. New versions of SDS would now be developed with the active participation of the local delivery centers. In this way local innovations could be integrated into the model as appropriate. In addition, assessments that before were focused mainly on compliance with the model would now be treated as opportunities to learn about what was and was not working. In this way the emergent quality of enterprise transformation was being acknowledged and the achievement of corporate goals was being enabled by employee involvement in policy making and implementation (Mack and Kaplan, 2009).

The second project involved developing more clearly articulated career paths for the three system administrator job levels and for the specialty roles of quality analyst and dispatcher. This was to acknowledge that the new model of work translated differently depending upon the background, experiences, and employment context of workers in the different global delivery centers. The meaning of roles and the paths to advancement needed to be put in a larger enterprise context so employees could more clearly understand where they fit and what possibilities were available for advancement.

The third project addressed some of the inherent tensions between the model’s focus on delivery center and pool performance and the fact that at the end of the day clients cared more about the quality of their service not the overall operational performance of delivery center pools. This project is examining the new model of work from the perspective of clients to see where adjustments to the model or its implementation might be made to ensure better connection between delivery center performance and client value.


The new model of work was motivated by a desire to deliver greater service quality and reliability at a lower cost and to facilitate the transfer of work to sites around the global best able provide the services to clients. However, the model’s design and implementation underestimated the work of integration and the need to adapt the model to local contexts, including differences in the employees and the clients they served. Kimberly and Bouchikhi (1995:9) note that, “Transformation can not simply be mandated. To be effective, it must be undertaken in a way which builds on rather than runs over the past.” It is important to consider the particular histories of delivery center sites and experiences of the employees who work there.

Synoptic views of change help us recognize and to some degree anticipate change trajectories, but equally important is understanding change as ongoing and emergent. Employees enact change and by so doing iteratively transform their organizations (Tsoukas and Chia, 2002) as they react to unanticipated contingencies and improvise in response to problems and unforeseen constraints (Olikowski, 1996). Success of transformation initiatives depends on employees adjusting to the particular, contingent, and ever changing requirements of the work. Employees, change agents, and clients can avoid some of the pain often experienced when new process, policy, or technology standards are introduced by viewing change as an iterative process that grows from the inside out, even when the transformation is defined from the top down. The challenge is in reconciling adaptations and adjustments that are the inevitable response to local conditions and circumstances with the desire to create standard practices across the enterprise.

The experiences with the SDS enterprise transformation also agues for adopting flexible and loosely coupled standards (Hanseth et al., 2006) that allow for variation in response to differences in the service lines supported, the size of client contracts, and employee skills and employment history. Variation actually enables the standard to evolve and be taken up in particular contexts. Local innovations might be better looked upon as opportunities to improve the standard through the process of iteration and enactment in everyday practice.

The priorities in many executive suites continue to stress centralized control, including monitoring performance and compliance with enterprise standards. While there is some acknowledgement of the emergent nature of change, the organizational impulse is to attempt to control change through mandates and compliance assessments. Regimes of control make it difficult to resist change initiatives even when they are not creating the desired outcomes and can lead to behaviors where for example passing assessments becomes the goal rather than integrating the change into everyday practice. This can result in additional work with little value to the enterprise other than to demonstrate compliance. Pressure to strictly adhere to a standard even when ill-suited to local contingencies can result in adjustments occurring only after “breakdowns” which negatively affects such things as client satisfaction, employee retention, and service quality.


The ethnographic approach that guided our research allowed us to explore how globalization is produced in the everyday actions of workers. As Buroway argues (2001:157) “The first step made by global ethnography is to restore history and agency to the reception and contestation of the global in “local”, to give life to the local.” Buroway (ibid:158) goes on to propose that the agenda of global ethnography should be “to replace abstract globalization with a grounded globalization that tries to understand not only the experiences of globalization but also how that experience is produced in specific locations and how that productive process is a contested and thus a political accomplishment.” On this view attempts at global enterprise transformation should not be seen as leading to convergence, but to organizational diversity arising from local contingencies (Guillén 2001).

The initiative to design and implement a global enterprise-wide standard of work has offered us the opportunity to explore how standards are variously adapted, resisted and integrated into diverse contexts. A major challenge for us is to find ways to disrupt the accepted view of top down, control driven change and offer alternative framings that are “actionable.” It is not enough to show how standard models of work are variously confronted and modified or even to argue for the inevitability of these responses. We must be able to show how alternative strategies will be more successful in achieving the desired outcomes of improved productivity, increased service quality, and a motivated and committed workforce.

Jeanette Blomberg is a research scientist at IBM Research whose research explores social aspects of technology production and se, ethnographically-informed organizational interventions, participatory design, and service innovation. Previously Jeanette held positions at Xerox PARC, Sapient Corporation, and the Blekinge Institute of Technology in Sweden. Jeanette received her Ph.D. in Anthropology from the University of California, Davis.


I wish to acknowledge the sponsors of our research who showed a genuine interest in the well being of the employees who were asked to adopt a new model of work and to the employees who participated in this study for sharing their experiences with the hope that by so doing the new model of work would evolve in ways that responded to the realities of their work. I also would like to thank Sara Moulton Reger who has been a thought partner over the years in reflecting on the importance of practice in organizational change agendas.

1 IT outsourcing services involve a third party service provider assuming responsibility for all or part of the IT operations and maintenance for another company. As a rule IT outsourcing services are arranged into technology function areas, sometimes called service towers, and these include such services as managing a client’s portfolio of applications, their mainframe and server environments, and help desk services.

2 This is a pseudo-name for the new model of work.

3 Comparisons have been made to the social and economic forces at work in the diffusion of global standards with what occurred during colonialism where western ideas and ways of doing things were imposed on the colonies.

4 The six issue areas were (1) vision and potential of the model, (2) implementation of the model, (3) tools and data to support model, (4) relation of delivery centers to customers, (5) career paths, and (6) overall work environment

5 Pool meet-ups were short, nominally half hour meetings, where members of a pool provided status updates, reviewed outstanding issues, and if needed asked for help.

6 At the time of the study the US economy was in the middle of a recession and unemployment was high leading people to make career changes some of which required relocating.


Anderson, H., Moe, G. L., and Wong, L.
2011 A data center goes lean. McKinsey on Business Technology, Number 22:24-25, Spring 2011.

Barley, S. A.
1990 The alignment of technology and structure through roles and networks. Administrative Science Quarterly, 35:61-103.

Barrett, R., Kandogan, E., Maglio, P. P., Haber, E. M., Prabaker, M., and Takayama, L. A.
2004 Field studies of computer system administrators: analysis of system management tools and practices. InProceedings of Conference on Computer-Supported Collaborative Work.

Burawoy, M.
2001 Manufacturing the Global. Ethnography, Volume 2(2): 147-159.

Cefkin, M., Thomas, J., and Blomberg, J.
2007 The Implication of Enterprise-Wide Pipeline Management Tools for Organizational Relations and Exchange. In Proceedings of Group, ACM press: 61-68.

Churchill, E. and Whalen, J.
2005 Ethnography and Process Change in Organizations: Methodological Challenges in a Cross-cultural, Bilingual, Geographically Distributed Corporate Project. Proceedings of EPIC: 179-187.

Feldman, M.
2000 Organizational routines as a source of continuous change. Organization Science. Volume, 11(6): 611-629.

Guillén, M. F.
2001 The limits of convergence: Globalization and organizational change in Argentina, South Korea, and Spain. Princeton University Press.

Hanseth, O., Jacucci, E., Grisot, M., and Aanestad, M.
2006 Reflexive Standardization: Side Effects and Complexity in Standard Making. MIS Quarterly, Vol. 30:2:563-581.

Holme, Mads
2010 Turn and Face the Strange: An Ethnographic Approach to Change Management. In Proceedings of EPIC. American Anthropological Association. Pp. 179-187.

Kimberly, J. R. and Hamid Bouchikhi, H.
1995 The Dynamics of Organizational Development and Change: How the Past Shapes the Present and Constrains the Future. Organization Science, Volume 6(1): 9-18

Kling, R. and Lamb, R.
1999 IT and Organizational Change in Digital Economies: A Socio-Technical Approach. Computers and Society, 17-25.

Mack, A. and Kaplan
2009 Policy Change Inside the Enterprise: The Role of Anthropology. Proceedings of EPIC, American Anthropological Association. 59-71.

Morgan, G.
2001 The multinationals firm: organizing across institutional and national divides. In Morgan G., Kristensen, P.H. and Whitley, R (eds). The multinational firm: organizing across institutional and national divides. Oxford: Oxford University Press.

Orlikowski, W. J.
1996 Improvising organizational transformation over time: A situated change perspective. Information Systems Research, Volume 7(1): 63-92.
2000 Using technology and constituting structures: A practice lens for studying technology in organizations.Organization Science, Volume 11:4:404-428.

Palmisano, S.
2006 The Globally Integrated Enterprise. Foreign Affairs, Volume 85:3:127-136.

Rouse, W. B.
2005 A Theory of Enterprise Transformation, Systems Engineering, Vol. 8, No. 4.

Rouse, W. B. and Baba, M. L.
2006 Enterprise Transformation. Communications of the ACM, Volume 49(7):76-72.

Sharpe, D. R.
2007 On Doing Ethnography: Relevance, Practice and Future Directions: Examples from the study of Multinational Organizations. Paper presented at Society for Industrial and Organizational Psychology Annual Conference, April 27-29, New York.

Tempel, A. and Walgenbach, P.
2007 Global Standardization of Organizational Forms and Management Practices? What New Institutionalism and Business-Systems Approach Can Learn from Each Other. Journal of Management Studies, Vol. 44:1.

Tsoukas, H. and Chia, R.
2002 On Organizational Becoming: Rethinking Organizational Change. Organization Science, Vol. 13:5:567-582.

Weick, K.E. and Quinn, R.E.
1999 Organizational change and development, Annual Review of Psychology, Volume 50, 361-386.

Womack, James P. and Jones, Daniel T.
1996 Lean Thinking: Banish Waste and Create Wealth in Your Corporation, Simon and Schuster.