Wong

Making Sense of Information Technology Change: An Interpretive Approach to IT Implementation

C. Kym Wong Doctoral Student, Benedictine University, Illinois, USA [email protected]

Submission to Midwest Academy of Management 2004 Organization, Development & Change Track

Making Sense of Information Technology Change: An Interpretive Approach to IT Implementation C. Kym Wong, Benedictine University ([email protected])

ABSTRACT In recent years researchers have gradually revised their assumptions regarding information technology (IT) related change. It is now contended that this is best understood as a dynamic, reciprocal and interpretive process during which human actions and institutional structures are inextricably linked. These developments suggest a new view of the IT implementation process – one which underscores the cognitive, social and contextual nature of IT change. Reframing IT implementation as an interpretive process calls for a shift towards a different set of assumptions and an alternate change intervention framework. Implications for organizational development and change practitioners in terms of the purpose, focus and levels of intervention are identified. KEYWORDS: Information technology, organizational change, structuration theory

INTRODUCTION The topic and potential of information technology (IT) in organizations has been the subject of considerable interest and debate since Leavitt and Whisler first speculated on the effects of a new technology which they designated as information technology. The authors predicted radical changes – including the reorganization of middle-management levels, increased centralization, and “major psychological and social problems” – occurring with the introduction of IT. Much of the existing literature has continued this trend of focusing on the outcomes and impacts associated with IT. IT is purported to be both an ‘enabler’ and ‘driver’ of change : “demanding the fashioning and incorporation of new roles, responsibilities, relationships, lines of authority, control mechanisms, work processes and work flows – in short, new organizational designs” , “enabling the rationalization of work and the better functioning of teams and by the transformation of work practices” , and “increasing process efficiency, changing the locus of knowledge and power, forcing old organizational structures into new configurations” . In a review of IT articles from six leading management journals, Dewett and Jones summarized these findings into five distinct categories of potential organizational outcomes: improved ability to link and enable employees, improved ability to codify the organization’s knowledge base, improved boundary spanning capabilities, improved information processing .. and improved collaboration and coordination .. (p.316). Yet, despite high expectations and huge corporate investments, success remains elusive and IT failures remain a serious problem for practitioners and researchers . Forty percent of all corporate IT projects are abandoned before completion and unused or underused systems cost businesses millions of dollars each year . The enormous potential

of IT to transform the fundamental nature of organizations coupled with the high rate of project failures raise important questions regarding its implementation within organizations. Recent developments in our understanding of the nature and role of IT have led to a new view of the process of IT implementation as a social phenomenon. Over the past decade, researchers have developed Structuration models of technology which offer new insights into IT-enabled organizational change. This structurationist perspective focuses on looking beneath the surface of technology’s role in organizational change to uncover the layers of meaning brought to technology by an organization’s social systems . This paper is organized as follows. In the first section, I review the findings of these interpretive IT studies and present their key concepts. The next section offers a brief discussion of the cognitive perspective on organizations. These two sections provide the rationale for the next section where I propose that IT implementation should be reframed and viewed as an interpretive process. The final section of the paper proposes several practice implications for organization development and change.

IT CHANGE: AN OCCASION FOR SOCIAL CONSTRUCTION In studying the effects of the introduction of an identical new technology (CT scanners) into two community hospitals, Barley found that differing outcomes ensued. The author suggested that these variances in the institutionalized patterns of interaction resulted from the technology becoming a ‘social object whose meanings were defined by the context of its use’; proposing that technologies can be viewed as ‘occasions for structuring’ – acting as triggers which can potentially modify or maintain an organization’s existing social dynamics. Orlikowski and Robey have built on Barley’s treatment of technology by accentuating that IT (unlike CT scanning technology) can be physically shaped by the actions of users and by their social settings. The authors argued that IT should be conceptualized as both a product and a medium of human action: IT is the social product of subjective human action within specific structural and cultural contexts, and is simultaneously an objective set of rules and resources involved in mediating human action – hence contributing to the creating, recreation and transformation of these contexts . Orlikowski has used Gidden’s Theory of Structuration to propose a conceptualization of IT which underscores its unique duality as objective reality and as socially constructed product. The adaptation, appropriation and enactment of technology by organizational actors are viewed as a key factor in organizational change. While technology incorporates interpretive schemes, norms and resources , it is only through the activation or appropriation of IT that it comes to be utilized, and therefore play a meaningful role in organizational processes. The structurationist framework views the mutual adaptation of technology and organization as one involving reciprocal causation – it is a relationship where the specific institutional context and human actions are both mediators in the ongoing interaction . The interpretive dynamics involved in the process of IT-enabled change has been explicated – using the notions of symbols, spirit and technological frames – by Prasad, DeSanctis and Poole, and Orlikowski and Gash. Prasad’s symbolic interactionist study of

computerization revealed that technology can simultaneously hold different meanings for individuals and groups within an organization. She suggested that understanding the nature of these symbolic realities is important as it influences the deployment of technology: “the use and adoption of certain technologies have immense symbolic value for management which may even outweigh productivity and performance concerns” . In proposing their Adaptive Structuration Theory approach for studying IT and organizational change, DeSanctis and Poole (1994) suggested that the properties provided by IT can be viewed as being comprised of ‘structural features’ (rules, resources and capabilities of the system) and the ‘spirit’ of the technology. The ‘spirit’ of IT represents its values and goals – its general intent as presented to its users. The spirit of a technology functions as a means of signification in helping users interpret its meaning, contributes to processes of domination and provides legitimation by supplying normative frames with regard to appropriate behaviors. As the technology structures are applied in interaction, new forms of social structure emerge which may then be reproduced and institutionalized over time. Understanding the spirit and features of the technology as well as the characteristics of the organizational environment may help in predicting the degree of ‘fit’ as well as potential appropriation and organizational outcomes . An underlying premise of Orlikowki’s structuration model of technology is the concept of interpretive flexibility: “ Interpretive flexibility is an attribute of the relationship between humans and technology and is a function of the material artifact, characteristics of the human agents, and the institutional context in which technology is developed and used” (1992:409). According to Orlikowski and Gash, the interpretive flexibility of technology allows it to be open to different interpretations by multiple groups who construct different ‘technological frames’ or assumptions, meanings and cognitions used to understand the nature and role of technology . Since technological frames strongly influence the views held about the function, value and role and hence the choices made regarding technology, IT-enabled change can therefore be understood in terms of shifts in technological frames over time. One of the issues raised by the concept of technological frames is the relationship between intended and unintended change outcomes. In an extension of her structurationist model of technology, Orlikowski conducted a study which highlighted the ‘enactment’ of different ‘technologies-in-practice’ using Lotus Notes across different contexts (multiple user groups at 3 companies) . The sites studied varied based on three kinds of conditions: interpretive (“understandings and shared meanings that members of the community constructed for sense-making”), technological (“tool and data properties available to users”), and institutional (“social structures constituting part of the larger social system within which users work”). She found that people’s interactions with technology not only enacted emergent technology structures but also enacted other social structures simultaneously. Three kinds of consequences were identified: (1) processual – changes in work practices, (2) technological – changes in technological properties and (3) structural – changes in social structures and systems. Orlikowski proposed that the first two types of consequences were often intended outcomes, whereas structural consequences were more likely to be unintended consequences of actions. She found that the structural status quo was reinforced and reproduced despite variations in interpretive, technological and institutional conditions. Thus, users working within hierarchical and individualistic institutional conditions would enact social structures that reproduced or enhanced these hierarchical/individualistic practices, whereas users working in strong team and

collaborative cultures would reinforce these cooperative conditions. Since Lotus Notes embodied values related to group collaboration, team-work and information sharing, there was very minimal use in the former instance, in contrast to maximal technology use in the latter scenario. Orlikowski’s findings indicated that enacted technologies-in-practice were more aligned with the technological intentions and properties when user’s social practices were already compatible with these properties. Thus, in understanding the interpretive dynamics of IT- enabled change, the meanings that users assign to the technology (which can be understood via the notions of symbols, spirit, or technological frames), the technology’s interpretive flexibility, as well as the institutional context and properties within which the usage of technology is imbedded are all key factors, as outlined in the figure below:

Processual Consequences

User Interpretive Conditions

Technological Conditions

Institutional Conditions

User interaction with IT

Technological Consequences

Organizational Outcomes

Structural Consequences

Figure 1 Interpretive dynamics of IT-enabled change

COGNITION, SENSE-MAKING AND CHANGE IN ORGANIZATIONS The cognitive perspective views organizations as networks of subjective meanings or shared frames of references. Organizations are conceived as socially sustained cognitive enterprises where thought and action are linked or conceptualized as ‘interpretation systems’, where shared cognitive maps amongst top management formulate the organization’s interpretation (its process of translating events, developing models for understanding, bringing out meaning and assembling conceptual schemes). While Daft and Weick suggest that organizational interpretation is performed by a ‘small group at the top of the hierarchy’ (p.285), other scholars such as Dougherty, Boland, and Tenkasi have viewed organizational sense-making as a process of distributed cognition whereby multiple ‘communities of knowing’ with specialized knowledge interact to create holistic patterns of meaning and action . According to Boland and Tenkasi, distinct communities develop unique social and cognitive repertoires which guide their interpretations of the world; these ‘thought worlds have different ‘funds of knowledge’ and ‘systems of meaning’ which can inhibit knowledge sharing (p.351).

Dougherty in fact, found that different functional groups (such as manufacturing, engineering and planning) within the same organization have differing departmental “thought worlds”. These differing ‘systems of meaning’ through which members interpret issues were found to inhibit the development of new knowledge and new social forms as ideas that did not ‘fit’ existing ‘funds of knowledge’ were rejected. Dougherty’s findings can be viewed as a failure in ‘perspective-taking’ by these departmental thought worlds . The ability to surface, access and examine each other’s differing interpretive schemes – to take each other’s perspective into account in a self-reflexive way – represents the core of the perspective taking process (p.362). Issues arise in perspective taking because knowledge and meaning systems are often taken for granted and because there is a tendency to assume others worldviews are similar to one’s own . According to Tenkasi, the application of soft technologies create additional problems due to their abstract nature, the lack of clarity about the developers’ implicit assumptions, and their dependence on the recipient’s specific context and structurationist conventions for the determination of relevance (1999, p.129).

REFRAMING IT IMPLEMENTATION In 1991, Moore and Benbasat asserted that understanding how to implement IT successfully was one of the more challenging issues facing the IS field. Now, more than a decade later, their declaration is more relevant than ever. According to Markus, the failure of many large scale projects involving new IT for reasons unrelated to technical feasibility and reliability is well known and is usually attributable to its implementation . Current IT implementation methodologies1 contain implicit assumptions regarding the implications of these technologies on organizations – with respect to the content, context and nature of the change process. Results from the interpretive research studies outlined in the previous sections provides the rationale for the proposal of an alternate set of assumptions which reframes IT implementation as an interpretive process – underscoring the cognitive, contextual and social nature of IT change. I propose that inattentiveness to the dynamics of this interpretive process results in a higher probability of IT implementation failure – such as unused or underused systems as well as unintended organizational change outcomes. Conventional IT implementation methodologies focus primarily on the content of change – especially as it relates to the functions and features of the technology being implemented. In the case of enterprise applications such as Enterprise Resource Planning (ERP) systems for example, there is typically a great deal of effort devoted to business or work process improvement (or ‘re-engineering’). In most cases knowledge is ‘disseminated’ by software or ‘best practice’ experts to users. A ‘conduit model of communication’ is used – one which does not address the interpretive character of the messages transmitted. For the most part, the organization’s social context (its institutional tendencies and structural properties) is ignored with the emphasis being placed instead on normative (“one size fits all”) and deterministic approaches. Conventional IT methodologies rely on linear, staged models which do not take into account the dynamic 1

The implementation methodologies referred to here are approaches used by Tier 1 and 2 IT consulting and systems integrator firms (such as the ‘Big 4’) as well as the professional service divisions of enterprise software firms.

interplay and reciprocal interaction between technology and the social processes involving its use. Change is seen as Lewinian: “inertial, linear, progressive, goal seeking, motivated by disequilibrium and requiring outside intervention” and the change manager’s role revolves around “creating and influencing change” – finding points of leverage and communicating alternate schemas – with a focus on inertia .

Content

Traditional Assumptions

NEW Assumptions

Technical & work process focus (based on

IT is socially constructed by users and

features/functions of IT)

interpretively flexible – focus is on cognitions (frames, interpretations, meanings)

One-way transfer of knowledge via

Knowledge exchange via collaborative learning

training

& mutual perspective taking amongst differing ‘thought worlds’ & communities

Context

Change management’s primary focus is on

Change agent’s focus is on facilitating shared

inertia, finding points of leverage and

understandings & development of receptive

communicating an alternative schema

social structures (Freeze – Rebalance –

(Unfreeze - transition – refreeze)

Unfreeze)

Largely ignored

Institutional properties/social structures are significant mediators of IT appropriation by users

Process

Staged/Linear progression

Reciprocal interaction

Planned/Episodic Change

Continuous Change

Figure 2 Comparison of traditional and reframed IT Implementation approaches

Reframing IT implementation as an interpretive process – one that is socially shaped and context specific – gives rise to an alternate set of assumptions regarding the content, context and process of change. Here, IT is viewed as being interpretively flexible – the functions and features of the technology are not taken as a given, or as prescribed by ‘experts’– but as being socially constructed by its users (through their interpretations, appropriations and manipulations). IT and its context are also mutually constitutive – institutional properties of the setting are drawn on in the appropriation and enactment of technology, and at the same time technology appropriation also reinforces or modifies these social structures. This process is viewed as one of reciprocal causation and interaction between human actors, technology and the specific institutional context. Based on the findings of the studies cited earlier, the mode of change is best described as continuous –emergent, improvisational, ongoing – “one of translation where ideas have impact through a combination of fit with purposes at hand, institutional salience and chance” . IMPLICATIONS FOR PRACTICE

This section discusses the implications that these new assumptions have for practice. From an organization development and change perspective, what is the purpose, the focus and the level of change intervention within this new view of IT implementation? Reframing IT implementation leads to an alternate change intervention framework; one which more appropriately matches the conditions and mode of change. According to Weick and Quinn, when change interventions fail, there is a mismatch between prevailing conditions and the kind of motor activated by the change intervention . When change is continuous, the more fitting framework is one of freezing, rebalancing and unfreezing. In this scenario, the purpose of intervention is explained by the authors as follows: To freeze is to make visible, show patterns and capture sequences (through cognitive mapping, schemas); to rebalance is to reinterpret, reframe, relabel and resequence patterns; and to unfreeze is to resume improvisation, translation and learning in ways that are now more mindful and more flexible . As previously discussed, the structurationist perspective views organizational change as the joint effect of the actions of human agents interacting with institutional/social structures and information technology. In linking IT to the form and evolution of social practices, the structurationist perspective provides a framework to explain the complex and dynamic way in which technology and social structures mutually shape one another over time. Although this perspective does not offer explicit guidance for change management, the concept of modalities –which explicates the inter-linkages between social structure and human action – points to potential focus areas for intervention (see Figure 3). Signification

Domination

Legitimation

Interpretive Scheme

Facility/ Resources

Norm

Power

Sanction

Communication

Structure

Modality

Interaction

Figure 3 Giddens Structuration Theory (1984)

According to Giddens , these modalities of structuration relate the realm of action to the realm of structure as they are drawn on in the interaction and reconstitution processes (p.28). Communication, for instance, involves the use of interpretive schemes, and actors draw on these in order to ‘make sense’. In addition to enabling shared meanings however, interpretive schemes are structural elements that constitute structures of signification thus acting as constraints to the communication process. Modalities are both enablers and constraints of human action – they determine how the institutional properties of social systems mediate human action and how human action constitutes social structure . Although these modalities are merely Gidden’s analytical devices, from an organizational development perspective they provide a useful way of thinking about

focus areas for intervention. These intervention focus areas can therefore be categorized into: (1) interpretive schemes, (2) facilities/resources and (3) norms. The combination of all of the insights outlined in this section can be illustrated by a three dimensional framework which incorporates these intervention dimensions – purpose, level, and focus – and emphasizes the interpretive, iterative and contextual nature of IT change (see Figure 4). The framework is not meant to be prescriptive; its aim is one of fostering appropriate conditions for IT change. Used as a conceptual device, it can help change agents reflect on potential areas in which barriers may exist as well as potential kinds of interventions which might be undertaken in guiding and facilitating shared understandings (see Figure 5). Appendix A provides an example which illustrates how this framework might be used in the context of an enterprise system implementation.

Intervention Level: Individual Group Social System Technology

Intervention Focus: Interpretive Schemes Facilities/Resources Norms

Intervention Purpose: Freeze Rebalance Refreeze

Figure 4 IT Implementation Intervention Cube (adapted from Freedman's ODIC)

User Interpretive Conditions

Technological Conditions

Institutional Conditions

Processual Consequences

User interaction with IT

Technological Consequences

Structural Consequences

Figure 5 Role of Intervention Cube in relation to interpretive dynamics of IT change

Organizational Outcomes

CONCLUSION In this paper, I have used the findings of several interpretive studies to propose an alternate view of IT implementation and change. Conventional models of IT implementation rely on universalistic, normative and deterministic approaches towards IT change. Reframing IT implementation as an interpretive process calls for a shift towards an alternate set of assumptions – one where IT is viewed as interpretively flexible and socially constructed, is inextricably linked to its context, and where IT, organizational actors and institutional context interact in a process of reciprocal causation. I also identified several implications for change practitioners in terms of the purpose, focus and levels for intervention. Given that transformations fueled by IT are likely to be the focus of OD attention in the future , a deeper understanding and attentiveness to the dynamics of this interpretive process holds timely and important implications for the design of robust change methodologies which underscore the cognitive, social and contextual nature of IT change.

Appendix A Example: An Enterprise Systems Implementation Enterprise resource planning (ERP) systems are a class of IT designed to integrate an organization’s computing infrastructure across multiple functional units. They support organizations through integration of different business activities and processes (financial accounting, HR, sales and distribution, manufacturing, purchasing, etc), serve as a common data repository and facilitate information flows across the entire organization. These systems have been heavily tied to the BPR movement of the 90s’ and are typically also seen as incorporating ‘best practice’ knowledge in terms of business processes and work practices. They are labeled enterprise systems because their functionality spans the entire breadth of the organization. Sales of these systems accounted for $10 billion and services (systems implementation, training, process engineering) in support of enterprise systems implementation totaled $90 billion in 1998 alone . Organizational investments in ERP systems are usually quite large – with the average implementation commanding resources of $50-100 million – and failures have been known to lead to organizational bankruptcies. The implementation of an ERP system is a large-scale change effort which involves multiple stakeholders and constituencies (see Figure 1).

Stage Internal Stakeholders

Adoption Top management (COO, CFO,

Implementation Top management sponsors (CIO/CFO)

CIO)

IT department

Representatives from IT/functional

Accounting

departments

HR Sales/Distribution Manufacturing Purchasing Inventory Other functional groups

External Stakeholders

Consulting firm – selection

Consulting firm(s) – implementation

Consulting firm(s) –

Software vendor – consulting

implementation

Hardware vendor – consulting

Software vendor – sales/consulting Hardware vendor – sales/consulting Figure 1 Stakeholders involved in an ERP Program Initiative

From an internal perspective, top management and each of the impacted functional groups are involved to varying degrees depending on the stage of the implementation. From the external consulting perspective, the software and hardware vendors, system integration firm (usually a ‘Big 4’ consulting firm) and any number of smaller niche services players may be involved. Using the proposed change intervention framework provided in the previous section, relevant potential questions may be asked by the change agent with the purpose of ‘freezing, rebalancing and unfreezing’. Within this framework the three identified intervention focus areas can be simultaneously examined to assess if enabling conditions or potential barriers exist for a successful implementation: Freezing  What are the divergent interpretive schemes, frames or meanings that the different social groups bring with regard to the nature and role of the new system?  How interpretively flexible is the new technology? What are the interpretive schemes, spirit, or symbolic messages provided or embodied by the technology? Does the new technology reinforce or alter existing technological frames?  What is the distribution of power/resources within the social groups and across the organization and do dominant groups/frames exist? Does the new technology support or undermine existing patterns of power or resource allocations?  What are the existing norms, culture or legitimate behaviors within the social groups and across the organization? Does the new technology reinforce or undermine these existing orders? Rebalance  Do we have the appropriate interpretive, technological and institutional conditions for the organizational outcomes we would like to achieve? If not – are there aspects of any of these areas that need to be changed or realigned?  Can shared understandings be created or frames changed?  Can the expression of alternate schemes and perspectives be legitimized?  Can the technology be modified to better reflect conditions or changes desired?  Can other structural conventions – domination, legitimation – be changed to create greater receptivity for the organizational changes desired? Unfreeze  Going forward, do we have the mechanisms in place to ensure that these (current and emergent) conditions (interpretive, technological, and institutional) remain balanced throughout the change process?  In appropriating the new technology, are users enacting social structures that reproduce the current status quo? Is this the desired outcome and if not do we need to ‘rebalance’ (as outlined above)?

REFERENCES