Business Process Re-engineering (BPR) represents a well-defined way to realize the potential benefits of information technology. Many U.S. corporations significantly improved their business performance in the 1990s after re-engineering their core processes. Hammer and Champy6 assert that re-engineering "capitalizes on the same characteristics that have traditionally made Americans such great business innovators: individualism, self-reliance, a willingness to accept risk, and propensity for change" and "fits America's penchant for innovation, change, and focusing on tomorrow rather than yesterday." The prevailing business culture in the U.S. undoubtedly encouraged the rapid and radical changes prescribed with BPR. However, America's egalitarianism and individualism also created a need for strong-armed leadership to complete BPR projects successfully and explicit incentives to facilitate teamwork.
The influence of the American culture on BPR raises questions about the worldwide application of re-engineering and other types of IT-enabled organizational change. Both the characteristics of re-engineering and the information systems (IS) that are implemented when processes are re-engineered may be affected by international and cross-cultural differences since transfers of both management technologies (like BPR) and information technologies are complicated by human factors.10 Technology transfers are more likely to succeed if it is understood how the donor and recipient contexts differ, and how those differences might influence the technology transfer.
This article develops such an understanding internationally by reporting on 12 BPR initiatives in six countries with differing cultural profiles. We first discuss the concept of national or societal culture and outline our comparative research approach. We then summarize how and why specific dimensions of culture influence both IT-enabled organizational change and the accompanying types of IS before considering the implications of our findings. In this way we address some of the issues identified in Martinsons and Davison.11
Culture is a "collective programming of the mind" that distinguishes not only societies (or nations) but also industries, professions, and organizations. At the national level, Hofstede8 delineated four dimensions of culture: power distance (PD), uncertainty avoidance (UA), masculinity (MAS), and individualism (IND). The Chinese Cultural Connection2 subsequently identified a fifth dimension which reflects the short versus long term orientation (STO/LTO) of a society. Societies can thus be clustered using the five dimensions that are described in Table 1 and discussed further in Hofstede.8
For example, Anglo societies (including the U.S., Canada and the UK) with their shared mother tongue and British colonial legacy form a tight cluster with low PD scores, fairly high MAS scores, very high IND, and a STO. Chinese societies (including mainland China, Hong Kong, Singapore and Taiwan) represent another distinct cluster, with much higher PD scores, lower MAS scores, much less IND, and a comparatively LTO.
Other dimensions of national culture include low/high context communications,7 monochronic/polychronic time,1 and universalism/particularism.12 However, including more dimensions in our study would have made it overly complex and unmanageable. Therefore, our research design was based on the five dimensions associated with Hofstede.
We deliberately chose to study BPR in six countries with various combinations of cultural values (see Table 2). Within each country, we controlled for industry differences by looking at BPR initiatives in two enterprises a telecom services provider and an IT product manufacturer. The dozen business-level cases enabled us to study BPR cross-culturally. Each case involved a genuine re-engineering project in that it aimed to drastically improve key measures of business performance; focused on cross-functional processes, such as product/service innovation and/or customer relationship management; intended to review these processes thoroughly; and used IT as the primary enabler.
We gathered qualitative data using interviews, observation, and document reviews. Triangulation was useful to increase reliability and avoid common source bias. For each case we conducted semi-structured interviews with at least one top management team member and another line manager or consultant who was directly involved with the project. Both probing questions (to establish or confirm project details) and open-ended questions (to focus on the how and why of each project) were asked to determine the: objectives of the project; involvement and formality of the planning process; planned change; pace and smoothness of implementation; resistance to change and how it was overcome; use and effectiveness of teams; enacted change from this project; and reasons for differences between the planned and enacted change. Our research focus is shown in Figure 1.
The evidence from each case was reviewed to ensure that it was both internally consistent and reflected a logical flow of events. Comparisons across cases relied on methods suggested by Miles and Huberman.13 Triangulation across the various data sources for each case revealed a very high degree of consistency. The few inconsistencies that emerged were reconciled by collecting additional data or verifying existing evidence. For example, we went back to a few interviewees and pointed out "you said this, but they said (or the documents indicate) that". This minimized the scope for subjective judgment and thus assured the validity and accuracy of our findings.
Each of the five cultural dimensions affected both the process of re-engineering (see Figure 2) and the IS accompanying it (see Figure 3). Key findings related to each dimension, supported by relevant interviewee quotes, are presented sequentially below.
Equal or Unequal? Consistent with Davenport and Stoddard,4 we found that an IS/IT functional manager often proposed re-engineering. However, the authorization for and initiation of large-scale BPR was typically done by a general manager. The centralized authority in high PD settings made it easier to authorize and initiate BPR. For example, Brazilian employees rarely questioned changes proposed by their bosses. Meanwhile, an international consultant commented that "in China, once the project is approved, things develop quickly...once the big bosses OK it, everyone else goes along."
Subordinates in high PD cultures tend to "accept directions from their superiors and do as they are told." In contrast, Japanese employees were highly involved in BPR projects. Planning was slower and the redesign was "revised repeatedly," but implementation was "very quick and quite smooth." Employees and line managers in Sweden (another low PD culture), also "expected to participate in planning the change." Decisions made without thorough consultation triggered "many complaints" and one particular process redesign "was rejected almost totally" by those responsible for implementing it.
The most dramatic improvements in completion times, customized services and other key performance indicators occurred when those "closest to the action were deeply involved". Process teams and individual workers in low PD cultures tend to accept responsibility and "take action into their own hands". With access to timely and complete data, they "made changes as necessary" to meet changing customer needs and/or improve product quality.
Employees in the U.S. and Sweden were also "ready and willing to take charge." Their attitude towards re-engineering was "generally positive" although "most opposed at least a few of the specifics." Efforts to empower employees tended to be more difficult in high PD cultures like Brazil and China. Going from a hierarchical structure with management control to a flat structure with management coaching was typically resisted by both managers and employees. Managers did not want to give up their positional power and control. Meanwhile, employees had became "comfortable in following instructions" and, resembling a case in Hong Kong,5 they did not see the value of taking on "new responsibilities which naturally had risks". For example, factory workers in Brazil repeatedly failed to stop a re-engineered assembly process even when something was wrong. Passive resistance was also very evident in both France and China.
The American and Swedish firms most readily adopted IT to enhance information access. In contrast, the Chinese managers who we interviewed admitted to being "anxious" and "un-enthusiastic" about formal enterprise-wide information systems. Centralized authority in their organizations, both before and after BPR, tended to reduce the need for both formalization and diffused information.12
New IS were installed in each BPR case, but the ones in China and France tended to primarily monitor the operations in an expanding business. More generally, enterprises in high PD cultures relied comparatively more on informal intra-organizational communication. For example, French bosses commonly used conditional and ambiguous statements (with qualifiers such as 'if', 'perhaps' and 'probably') to avoid overcommitment. Accusations of failure were rare in France because performance expectations were "rarely communicated."
The high PD cultures were less likely to develop formal IS plans or process models. IS were used primarily for vertical communications to reinforce the hierarchical control of business activities. Those in authority tightly controlled information and selectively issued directives to subordinates. One manager in China said that "only I need to know this information because only I use it for decision making." There was relatively little peer-to-peer (horizontal) communication to share ideas or co-ordinate actions.
Extensive knowledge sharing in low PD cultures encouraged IS development to support employee empowerment and informed decisions "close to the customer." E-mail and instant messaging were used commonly in the U.S. and Sweden to co-ordinate peer-to-peer activities and were helpful to improve teamwork.
Rigid or flexible? In high UA cultures like Japan and Brazil, IT was used typically as part of a concerted effort to reduce uncertainty by making forecasts and plans. BPR efforts were "planned in great detail" and implemented "very systematically." Both Brazilian and Japanese firms tended to involve many people in a structured change process that undoubtedly reduced project risks but raised concerns about possible "paralysis by analysis."
In contrast, businesses in low UA societies typically maintained more flexibility. Their plans tended to be directional rather than detailed while both their organizations and project teams had comparatively looser structures. One Swedish manager justified this by saying "we don't know exactly what will happen, so we prepare for different possibilities."
The ability to standardize IS across enterprises or even throughout a business network was more difficult in low UA societies. The rules and algorithms accompanying standardization tended to be "too rigid" and "insensitive to changing conditions." Enterprise resource planning (ERP) systems were also resisted because they enforced one specific way of doing business.9
Tough or Tender? Re-engineering proponents commonly advise managers to act like battlefield generals - "blowing up the old" and "shooting the stragglers." This military language together with an emphasis on material rewards reflects a high MAS culture. It downplays the co-operation and employment security that are valued in low MAS societies like Sweden and France.
Widespread acceptance of BPR's underlying philosophy underpinned the successful change projects. Radical forms of IT-enabled change experienced comparatively "manageable levels of resistance" in high MAS cultures. Given that "a blank sheet of paper in design usually requires a blank check for implementation,"4 we were not surprised to find only two firms authorizing the financial and human resources for a "clean slate" design - at "green-field" sites in China and Sweden.
Businesses in the U.S. and China, particularly the telecom service firms, had the most ambitious BPR objectives. They typically identified 57 core processes for concurrent re-engineering. This echoed the initial plans of BPR pioneers like IBM and Xerox.3 After encountering project management difficulties, they tended to refocus on a few key processes.
Significant resistance to BPR was a universal phenomenon. However, the firms in the high MAS settings tended to use strong leadership and financial incentives to overcome resistance. The Chinese firms offered "bonuses equal to between one and six months salary." Basing rewards on improving objective performance measures of performance required IS to collect and process quantitative performance data.
The typical response to BPR resistance in Sweden and France was reversion to more gradual forms of change. One business used a "cautious progress" philosophy to prototype, test, and implement new IS. This clearly demonstrates the "culture of incrementalism" that Hammer and Champy6 warned against when they argued that re-engineering simply "can't be carried out in small and cautious steps." This raises the interesting question of when is re-engineering no longer re-engineering, which is considered by Davenport.3
Now or later? Both acceptance of the BPR philosophy and project implementation depended significantly on a society's time orientation. Firms in STO societies like the U.S. found it easiest to accept radical changes driven by strong or charismatic leaders. The rapid results promised by BPR were attractive to U.S. managers who are "legally required to report on our financial performance every three months." They used cost-benefit analyses to justify projects and commonly asked questions like "how quickly will this pay off?"
Managers in LTO societies like Japan preferred "smooth and steady progress" rather than "quick fixes that may be disruptive." They commonly resisted outside consultants, preferring to "build up ... internal capabilities." Japanese managers were more likely to authorize "strategic" IS which lacked tangible or immediate benefits. This suggests that a LTO may trump uncertainty avoidance to enable strategic IS development.
More generally, we found that LTO cultures were more likely to modify BPR principles and adopt new IS gradually. Significantly, a consultant in China commented that "almost everything is possible with BPR, but changes (here) are not made quickly or easily" while a Japanese company president suggested that "we would never follow Hammer's ideas to the extent that they did in America ... it would be too disruptive." However, managers in LTO cultures conducted many post-project reviews to determine what went wrong or could be improved.
Alone or Together? Managers driving BPR in all four U.S. and French firms found that they lacked either the "personal authority" and/or "key relationships" to make all the changes needed to implement BPR successfully. We attribute this problem to a high IND culture.
Those driving BPR in lower IND cultures tended to work together more effectively. Cohesive leadership was found in Brazil, China and Sweden, but was most evident in Japan. This enabled changes in organizational structures and policies as well as job designs and work incentives, and thus facilitated BPR success. A Japanese manager said that "reaching a consensus behind the scenes was critical" to avoid the "unnecessary disputes ... (that) harm some Western firms."
Process teams (essentially groups responsible for an entire re-engineered process) in low IND societies were typically more effective, although the team leader's role varied from benign autocrat to facilitator of equals. Cohesive and productive teams were more diffcult to create and maintain in high IND societies. However, performance-based rewards for personal contributions commonly sparked superior and often innovative responses to problems in American and French enterprises. We attribute this to a high IND culture.
Meanwhile, the need to conform and maintain group harmony in low IND cultures can impair BPR. The Brazilian and Chinese businesses replicated changes made by others without adequately considering their contextual suitability. It was also difficult to evaluate and reward performance in low IND cultures, partly because individuals were reluctant to make discretionary decisions. A manager in Brazil attributed this to both the social conditioning of workers and a lack of individual incentives based on comments such as "it is not my role," "it is not worthwhile to say anything" and "I can not influence your <the bosses'> decisions."
However, these same low IND societies were comparatively more receptive to IS aimed at enhancing collaboration. Group communication tools based on mobile phone messaging tended to be implemented smoother and used more in low IND societies while PC-based applications were more common and successful in high IND societies.
Managers select the appropriate type of organizational change by comparing alternatives in terms of their expected benefits and risks. Cultural factors complicate this comparative analysis. IT-enabled organizational changes like BPR are both more likely to succeed if they fit the implementation context. Meanwhile, IT applications that challenge existing norms or attempt to standardize multinational practices are risky.
Implementation risks can be reduced by getting the support and cooperation of stakeholders. However, there is no ideal culture for making organizational changes or applying IT. Our study shows that different types of both changes and IS fit particular contexts. For example, a high PD culture makes it easier to initiate BPR, but complicates its implementation, especially if it involves empowerment. Meanwhile, strategic types of IS are most likely to be developed in masculine and LTO cultures.
The critical issues for managing IT-enabled organizational change and implementing IS also differ across cultures. Managers must recognize the most critical issues in their specific implementation context as highlighted in Table 3. These issues must be addressed effectively in order to improve the prospects for successful change. For example, managers in high PD cultures need to encourage process formalization and employee participation while those in low PD cultures need to ensure sufficient project control and coordination.
Multinational firms operating in various cultures should plan and implement changes flexibly. Rather than applying a one-size-fits-all approach, they must exercise cultural sensitivity and glocalize their best practices. Meanwhile, domestic firms must avoid transferring technologies naively from other countries and cultures. Different approaches to organizational change and different types of IS are appropriate in different contexts to achieve a common goal business success.
Finally, we caution that the relationships shown in Figures 2 and 3 are based on our 12 case studies they have not been proved. They are tentative propositions useful to guide management practice and suitable for confirmation by more extensive research.
8. Hofstede, G. Culture's Consequences: International Differences in Work-Related Values. Sage, London, 1983. A second edition (with a revised subtitle: Comparing Values, Behaviors, Institutions, and Organizations across Nations) was published in 2001.
11. Martinsons, M. G., and Davison, R. M. Culture's consequences for IT application and business process change: Research agenda. International Journal of Enterprise and Internet Management 5, 2, (2007), 158177.
©2009 ACM 0001-0782/09/0400 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2009 ACM, Inc.
No entries found