An exploration of enterprise technology selection and evaluation

Journal of Strategic Information Systems 13 (2004) 329–354 www.elsevier.com/locate/jsis

An exploration of enterprise technology selection and evaluation Peter Tingling*, Michael Parent1 Simon Fraser University, 8888 University Drive, Burnaby, BC V5A 1S6, USA Available online 26 January 2005

Abstract The evaluation-and-selection of enterprise technologies by firms has been said to be largely rational and deterministic. This paper challenges this notion, and puts forward the argument that substantial ceremonial aspects also play an important role. An in-depth, exploratory longitudinal case study of a bank selecting a ubiquitous and pervasive e-mail system was conducted using grounded theory and a hermeneutic [pre] understanding of institutional and decision making theories. Intuition, symbols, rituals, and ceremony all figured prominently in the decision process. However, rather than being in conflict with the rational processes, we found them to be in tension, leading to a more holistic social construction of decision processes. For researchers, this suggests that a focus on process rationality, not outcomes, might lead to a fuller understanding of these critical decisions. For managers, it underscores the importance of understanding the past in order to create the future. q 2004 Elsevier B.V. All rights reserved. Keywords: Institutional theory; Decision making; Evaluation; Selection; Rationality; Ceremony

1. Introduction The definition and study of Enterprise Systems (ES) has evolved from its initial resource planning focus to a broader approach that typically includes aspects of ‘integration’ and ‘best practice’. These concepts, however, are imprecise and may differ across contexts and perspectives. Not only do best practices frequently exist in the eye of

* Corresponding author. Tel.: C1 604 291 3473. E-mail addresses: [email protected] (P. Tingling), [email protected] (M. Parent). 1 Tel.: C1 604 291 5214. 0963-8687/$ - see front matter q 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jsis.2004.11.003

330

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

the beholder and vary within industries but integration is rarely a dichotomous condition and can be implemented at a variety of levels and in a number of ways. Furthermore, although the expanded definition allows us to increase the relevance of ES research, ES are frequently characterized by debate about their adoption, use, and efficacy and there is a growing recognition that they must be studied longitudinally and within their technological and organizational context. For example, Orlikowski and Iacono (2001) remind us to consider the specific aspects of technology that affect our research and Robey and Boudreau (1999) recommend that we consider ‘logic of opposition’ and recognise implicit contradictions and opportunities by focusing on theories that promote and oppose social change and explain a wider range of outcomes. This paper ameliorates understanding of organizational evaluation-and-selection between ES by describing longitudinal research conducted through institutional and decision theory lenses at Fidelis, a large financial institution. Believing that industry best practices required a reduction in the number of systems and wanting to integrate a new application, evaluative and decision processes were used to justify a move from multiple systems to a product in use at other organizations and compatible with a new application under consideration. Technology evaluation and decision making is usually considered a predominantly analytic or traditionally rational activity where options are compared against criteria such as efficacy, market share, and total cost of ownership (TCO). In this characterization, noninstrumental or ceremonial elements of the decision process are frequently ignored, omitted entirely, or at the very least discounted. However, following an extensive qualitative case oriented research methodology that focused on the technological and organizational context, ES evaluation-and-selection processes were found to consist not only of traditionally rational quantitative and qualitative elements but interwoven in a fabric made up of ceremony and ritual. The paper proceeds as follows. In Section 2 we describe the state of technology selection-and-evaluation research and introduce the methodological and theoretical perspectives applied: case study research and grounded theory within an umbrella of phenomenological hermeneutics bracketed by decision and institutional theory. This is followed by a description of the research site, data collection and analysis: longitudinal interviews with multiple respondents analyzed using Atlas/Ti. Section 4 presents the findings and interpretation, and the final section concludes with implications for future research and theory development.

2. Background and theoretical methodology This section briefly summarizes evaluation-and-selection research, institutional and decision theory, and grounded theory and hermeneutics. 2.1. Technology evaluation-and-selection The empirical study of research on technology evaluation-and-selection is a relatively new area, and work is predominantly exploratory (Riddle and Williams,

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

331

1987; Schilling, 2002). However, it is an important and worthy area of study for at least three reasons. 1. Technology accounts on average for more than one-third of all business capital spending (Bakos, 1998), is vital to competitive advantage, and has been identified as among the most important competitive decisions that managers must make (Clark, 1987; McRary, 1995). 2. Evaluation-and-selection processes often precede adoption and use (Fichman and Kemerer, 1999). Therefore understanding how organizations perceive the benefits and issues of a particular technology prior to acquisition can provide context and insight into subsequent evaluation or description of the technology (Langley et al., 1995). 3. Many technologies benefit from increasing returns and network effects—conditions under which the utility increases rather than decreases with the number of adoptions (Arthur, 1987; Beggs, 1989; Shapiro and Varian, 1999). Because this can result in path dependence and curtail choice, understanding how technologies are evaluated will therefore not only inform theorists and providers or consumers of technology but is important to policy makers and regulators (Ciborra and Hanseth, 1998). Information technology decisions are different from other forms of investment (Beggs, 1989; Powell, 1992), and frequently have high levels of intangibility that make them difficult to assess or quantify. In addition, technological uncertainty—or the degree to which it is difficult to determine the reliability, capability, and precision of the new technology or whether newer technology will soon appear to make it obsolete (Gerwin, 1988)—surrounds many new innovations. Evaluation processes, based mainly in decision theory, have typically been described in rational, objective, and systematic terms despite the recognition that notions of what is ‘best’ include socially constructed or politically defined ends (Garud and Ahlstrom, 1997). Formal processes occur in about thirty percent of firms and organizations (Lubbe and Remenyi, 1999), although when used they can be extremely complex (Noori, 1995). Evaluation has focused on rational elements, and mainly involved two-dimensional matrices or subjective expected utility (SEU) measures in multi-attribute decision making (MADM) models (Rosenhead, 1989; Khouha, 1995; Garud and Ahlstrom, 1997). Many methods have been proposed to help decision makers evaluate and select the best technologies (Yap and Souder, 1993). Decision makers have been encouraged to use complex quantitative tools such as multi-stage analytic decision aids (Khouha, 1995; Verter, 2002) or attributes and weights developed from a Q-Sort (Yap and Souder, 1993), and to focus on analytic rather than subjective elements (Torkkeli and Tuominen, 2002). Recent work has raised questions about the applicability and use of quantitative product evaluations (Riddle and Williams, 1987), and recognized that the processes are underresearched (Schilling, 2002). In addition, studies have questioned our knowledge of evaluation processes and the degree to which they are traditionally rational rather than ceremonial or ritualistic (Currie, 1989; Avgerou, 2000; Tingling and Parent, 2002).

332

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

2.2. Institutional theory The essence of institutional theory is that organizations operate in highly institutionalized contexts (Zucker, 1987; Wicks, 2001) in which processes can attain a state of being taken for granted (DiMaggio, 1988; Tolbert, 1988; Jepperson, 1991) where organizational survival and behaviour is driven by the need to conform to social norms of acceptable behaviour and legitimacy (Meyer and Rowan, 1977; DiMaggio and Powell, 1983; Scott, 1987; Zucker, 1987; Covaleski and Dirsmith, 1988; Avgerou, 2000; Wicks, 2001). Institutionalists posit that competition and the need for efficiency are the not the key drivers of structural change in organizations, but that an organization’s iterative and ongoing relationships with other organizations, quest for legitimacy, and the process of structuration makes them more similar but not necessarily more efficient (DiMaggio and Powell, 1983; Donaldson, 1995). Over time, practices can become infused with value beyond the technical requirements at hand and organizations constrain themselves with regulative, cognitive, and normative frameworks that provide meaning and stability to social life (Scott, 1994). Regulative elements are based upon rational actor models and involve conformity through sanctions; cognitive elements are widely held beliefs and taken-for-granted assumptions; and normative elements are those that incorporate traditional mores and informally sanctioned social obligations. Institutional theory encompasses many versions and varieties that are not completely reconcilable (Scott, 1987, 1994; Zucker, 1988; Oliver, 1991; Westfall, 1997). However, a frequent distinction is often made between old and new institutionalism (DiMaggio and Powell, 1991). Old institutionalism emphasizes issues of conflicting interests, coalitions and competing values, and power and influence at the community level (Clark, 1972; DiMaggio and Powell, 1991; Greenwood and Hinings, 1996). New institutionalism stresses organizational homogeneity, the stability of institutionalized components, legitimacy and the embeddedness of organizational fields, and cognitive processes where normative obligations are imposed upon actors (Meyer and Rowan, 1977; DiMaggio and Powell, 1983; 1991; Scott, 1995; Greenwood and Hinings, 1996; Williamson, 1996). Common to all versions is the idea that organizational activity cannot be simply explained by the rational action of actors and institutionalization occurs when social orders become accepted as social facts or become widely understood to be appropriate and necessary components of efficient, rational organizations (Zucker, 1987; Avgerou, 2000). 2.3. Decision theory Decision making, the act of choosing between alternatives (Barnard, 1938; Keeney, 1988) or a ‘commitment to action’ (Mintzberg et al., 1976), has a rich history within management literature (Langley et al., 1995; Tversky and Kahneman, 2000). Viewed by Simon (1960) as a cognitive sequence of ‘intelligence-design-choice’, decision making has been considered a bounded process in which rational approaches and formal analysis lead to systematic methods that result in better outcomes. Decision theory has been criticized because difficulties exist determining exactly when or where decisions occur, and frequently a loose coupling exists between a decision and action

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

333

(Butler, 1990; Mintzberg and Waters, 1990; Pettigrew, 1990); yet decision theory is applicable to technological evaluation where an organization is selecting between competing options. While there is not a single widely accepted decision theory (March, 1988), choice can be categorized as three modes: bargaining, judgment, and analysis (Mintzberg et al., 1976). Bargaining involves trade-offs between multiple decisions with conflicting goals and is often politically characterized (Carter, 1971a,b); judgment is an individual approach wherein decision makers select an option using a process that they might not consciously recognize or be able to articulate; and analysis is a discovery-oriented process in which alternatives are objectively evaluated according to a predetermined schema. Of these three, the analytic approach is one of the most emphasized in the normative literature and is itself dominated by three somewhat competing approaches: reason-based models, subjective expected utility, and prospect theory. Reason-based models list the advantages and disadvantages of each choice, discard the least attractive ones, and carefully review the remainder. Subjective expected utility (SEU) is a rational mathematical value model in which decision makers choose by assigning probabilities and utilities to choices, calculate the expected utility, and select the option with the highest outcome. Prospect theory is a value model originally intended for individual financial decisions and is based around arguments of asymmetric risk-seeking and risk-aversion behaviours. 2.4. Integration of decision models and institutional theory Rational behaviour, efficiency, and effectiveness are key elements of decision and institutional theories but are treated differently. In decision theory, despite recognition that complete rationality is unrealistic (Simon, 1957; Bell et al., 1988), rationality is usually considered a ‘self-evident truth’ (Tversky and Kahneman, 2000) or ‘intuitively appealing’ (Quattrone and Tversky, 2000). Institutional theory, on the other hand, elevates the pursuit of legitimacy over simple rational behaviour and draws a distinction between irrationality and nonrationality (Williamson, 1985). Decision and institutional theories overlap in their inclusion of rituals. March and Heath (1994), for example, note that acquisition decisions are not necessarily simply rational maximization technology selections but can be important organizational signalling opportunities. Brunnson (1990) notes that meeting minutes and internal and external correspondence often have more legitimating effects than rational usefulness. However, even though ceremony is recognized as significant in the development and communication of what is happening and why it is important (Bell et al., 1988; March, 1988; March and Heath, 1994), ceremony has largely been ignored by decision research. Symbols, myths, and rituals are, to varying degrees, interdependent and frequently intertwined (Pettigrew, 1979), but can be differentiated. Symbols are objects, practices, or signs that evoke something else by association and link an organization’s experience to deep feelings (Cohen, 1974); myths are real or fictional stories, recurring themes, or character types that embody an organization’s cultural ideals or establish what is legitimate and unacceptable in an organizational culture (Pettigrew, 1979); and rituals are

334

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

Fig. 1. Conceptual evaluation-and-selection framework.

sets of ceremonial forms or mechanisms by which traditions are preserved and meanings sustained (Bocock, 1974; March and Heath, 1994). We integrate institutional and decision theory in the conceptual framework illustrated in Fig. 1. In this framework, rationality and legitimation processes interact with organizational structures and decision processes to affect evaluation decisions within an overall technological and environmental context. Conflict can exist between rationality and legitimation but the two processes might just as easily support and reinforce each other, depending upon the organizational and technical context. For example, where a technology is entrenched within an organizational field or benefits significantly from increasing returns or network effects, the processes might include legitimation or coercive isomorphism and be characterized as largely ceremonial. On the other hand, where these considerations are less salient, evaluations could be more rational in the traditional sense of instrumentation and subjective expected utility. Furthermore, legitimation and rationality could be related to factors such as the level of organizational idiosyncrasy or a firm’s relative propensity to adopt. 2.5. Grounded theory, hermeneutics, and case study methodology The methodology for this research was grounded theory with a hermeneutic preunderstanding. Grounded theory is an inductive methodology that closely ties empirical observation with theory building to discover theory from data (Glaser and Strauss, 1967; Martin and Turner, 1986; Eisenhardt, 1989; Orlikowski, 1993). Hermeneutics is concerned with the interpretation of understanding (Arnold and Fischer, 1994). Although there are no detailed and explicit guidelines for the hermeneutic method (Gadamer, 1989) and various disparate and contrasting hermeneutic

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

335

theories exist (Thompson, 1981), broad streams of hermeneutics can be grouped as hermeneutical theory, philosophical hermeneutics, and critical hermeneutics (Prasad, 2002). Hermeneutics and grounded theory are similar in many respects (Rennie, 2001). However, hermeneutics has a different orientation to the treatment of prior knowledge. Where grounded theory suggests that a priori biases and prejudices are to be held in abeyance so that they do not influence the study (a process known as bracketing), hermeneutics does not view all biases as obstacles but necessary conditions of any understanding (Prasad, 2002). Pre-understanding differentiates between legitimate prejudices which make understanding possible and those that hinder or lead to misunderstanding; it follows from the recognition that the object and the researcher exist in advance of any reflection. Our pre-understanding of technology evaluation-andselection causes us to question the degree to which evaluation-and-selection processes are defined either by decision theory or legitimation, the two dominant pillars of organizational research (Goodstein, 1994). Rather than viewing evaluation processes as dichotomously rational or legitimating, our pre-understanding considers that the two may interact and either support or conflict the decision process, depending upon contextual or environmental considerations. Our objective, therefore, is not so much validation or hypothesis testing as it is description and hypothesis building. This research was conducted using a naturalistic case study methodology (Lincoln and Guba, 1985) and embodies the philosophy that, to truly understand how and why events play out over time, we must examine them directly (Mintzberg, 1979). An ‘empirical inquiry that investigates a contemporary phenomenon within its real-life context’ (Yin, 1994), case study research has been employed by both positivist and interpretive researchers (Bonoma, 1985; Benbasat et al., 1987; Myers, 1997). It is well suited to organizational rather than technical issues, particularly in the study of an under-theorized area (Eisenhardt, 1989; Yin, 1994).

3. Description of the case study 3.1. The research site The research site was Fidelis, a large Canadian Bank.2 Introductory discussions were first held with Jay Richards, director of technology architecture, in April of 2003. Already in the later stages of an extensive review and restructuring of their technological decisionmaking processes, Fidelis was quite interested in the research and agreed to provide a high level of access and support. Formal meetings began in July. In August, after observing several cases of ongoing technical evaluations (e.g. the corporate browser standard and an electronic corporate directory structuring tool) the researchers decided to focus on a longitudinal observation of the process to select a new email system. This research continued until December, 2003. 2

All names are disguised.

336

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

The evaluation-and-selection of an email system was considered an appropriate research decision for five reasons. 1. Email is an enterprise-wide information technology artefact (Orlikowski and Iacono, 2001) with significant organizational implications and cumulative consequences on the actions of managers and departments (Swanson, 1994). Estimates of the fiscal impact of an email decision ranged from a low of $12 million to a high of $80 million over 5 years, costs that are directly comparable to ES conversion costs routinely cited in the literature (O’Leary, 2000; Ragowsky and Somers, 2002) as well as those of Fidelis’s own ERP implementation. Email selection could affect almost 30,000 employees and was widely integrated into hundreds of other applications and systems at the bank including seven that were considered core or essential. 2. Evaluation of the email system would be comprehensive yet time bound and involve both quantitative and qualitative aspects. Evaluations that focused mainly upon ‘quantitative criteria’3 were rejected because usually these decisions are largely structured and most of the decision process has already occurred in determining the measure (Siha, 1993; Hsee, 2000; Torkkeli and Tuominen, 2002). 3. The options being considered were supported, developed, and provided by relatively similar organizations making the comparison comparable and credible. 4. The process involved widespread access to artefacts and staff. Richards arranged for meetings with Fidelis staff as well as representatives from IBM and Microsoft, and access to archive documents of prior evaluations was still available. 5. Email was a technology that could be easily discussed at a variety of levels. Rather than trying to decipher esoteric technical minutia, the researchers were able to easily converse with both technical and nontechnical decision participants. 3.2. Case summary Fidelis was organized into five divisions: retail banking, wealth management Canada, wealth management United States, securities, and operations. Accounting for slightly more than half of the 50,000 employees, the retail banking division provided traditional end-customer branch banking services. The two wealth management divisions served Canadian and US investors through a separate national network of offices and a combination of call centers and internet services. The securities division provided services to institutional investors and other divisions of the bank, while the operations division provided common administrative services and much of the information technology. The operations division was essentially an administrative cost recovery unit, but the four other divisions were strategic business units operated on a profit-and-loss basis. Historically an early adopter, Fidelis was the first North American bank to offer fullservice automated banking and had a good reputation within the North American financial services community. Externally benchmarked against their competitors, Fidelis was 3

The issue if any criterion can be strictly quantitative and objective is debatable. Here the term is used to indicate a situation in which the measure has been identified or imposed exogenously; for example, processor selection on the basis of software benchmarks.

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

337

generally regarded as an efficient organization. However, two changes in information technology strategy were made under their new CEO, who had joined Fidelis in a recent merger. The first, an outcome of the CEO’s belief that technology was more of a strategic necessity than a source of long-term competitive advantage, resulted in a conscious effort to be less aggressive in terms of technological innovation and to avoid what was internally referred to as the ‘bleeding edge’. Instead, by preferring to be ‘a close second’, watching their competitors and being agile, Fidelis believed that it could benefit from the experience and mistakes of the earliest adopters and ultimately surpass them through superior execution. The second change reflected the executive’s thoughts on the integrated nature of technology; he believed that an opportunity for synergy and sharing existed across divisions, and had implemented a federal model for decision making. In stark contrast to the prior structure, where each division had been run autonomously and given free rein in technology decisions, current accountability was shared between the divisions and central head office. Vertical functions, or activities that did not affect other areas such as application design and development, were retained within each division. Horizontal functions, on the other hand, or those that involved areas such as architecture and standards, cross-functional systems, security, and infrastructure, were centralized and provided on a cost-recovery basis by the operations division. Fidelis spent approximately $500 million per annum on information technology, including more than $120 million in capital expenditures. The size of the information technology function within each division varied from a low of between 50 and 200 employees in one division to more than 1000 in the operations division. Information technology for each division reported directly a CIO who had a dual reporting relationship that consisted of a ‘solid line’ to the divisional business executive and a ‘dotted line’ to the CIO of the operations division. The operations CIO therefore was considered first among equals in terms of overall technical accountability. 3.3. Evaluation-and-selection The technology evaluation-and-selection processes at Fidelis had been revised earlier in the year following a review by an external consulting firm. Where previously Fidelis was concerned that their processes were ‘ad hoc, one-off, and potentially incomplete’, the new method for decision making consisted of both structure and process with the objective of making the processes more ‘complete, rigorous, repeatable, and formal’. In place of the spontaneous or haphazard interaction that had characterized the former processes, an ongoing, four-level structure of meetings was developed. These are summarized in Table 1. The most senior group was the technology leadership council or the TLC, which was chaired by the CIO of operations division and attended by the divisional CIOs. Organized as a governing body, the TLC had a number of working subcommittees that were a routine part of the ongoing agenda. Two of these were the standing subcommittee on architecture (chaired by the director and vice president for technology of the securities division) and

338

Table 1 Fidelis information technology evaluation-and-selection meetings Chair

Attendees

Meeting frequency

Purpose

Reports to

Level of formality and structure

Technology leadership committee (TLC)

CIO operations

Divisional CIOs

Bimonthly

Discuss strategy, issues, and areas for cooperation

–

High

Secretary-corporate vice president Director and VP Securities Technology

VP emessaging

Monthly

Standing Committee to review emessaging

TLC

Medium

Alternating thursdays

Discuss strategy and issues. Charter working groups and review and ratify their recommendations

TLC

Medium

Maximum of two participants or subject matter experts from each division

As required

Temporary ad hoc team to research and recommend technical direction— usually in the form of a statement of direction

ARB

Low

Senior Directors or IT managers from Operations Division

Alternate Thursdays from the ARB

Discuss Operations Division issues and ARB policies

Operations Division CIO and Operations Executives

Low

Messaging subcommittee of the TLC

Architecture Review Board (ARB)

Architecture Working Groups (AWG)

Examples include Email Working Group, Active Directory Working Group, Browser Working Group Architecture Management Team (AMT)

Director Technology architecture

Scribe-Architect Specified by the ARB-Usually assigned from the lead or most affected division

Director Technology Architecture

Messaging director Other divisional IT directors Senior IT directors from all business units

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

Name

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

339

the standing committee on electronic messaging (chaired by the messaging vice president of the operations division). The next senior group was the architecture review board or the ARB, which was chaired by the director of technology architecture for the operations division, and attended by senior information technology representatives from each of the five divisions (typically directors). The ARB met on alternate Thursday mornings for 2 hours, and was less formal and structured than the TLC but not spontaneous. The ARB provided a forum for general discussion of technological issues and concerns. If an issue was identified either by the ARB collectively or individually by a division, then an ad hoc architecture working group (AWG) was convened to research or review the issue and to report back to the ARB. In addition to these groups, an architecture management team (AMT), made up of seven to ten senior operations division managers from the ARB, also met on alternate Thursdays. The AMT was chaired by the director of technology architecture, operations division, and had as its objective the review and discussion of upcoming ARB items, and the review and resolution of issues with AWGs that had been commissioned by the operations division. Minutes and an agenda to these meetings were usually produced although they were frequently unavailable or late. These meetings were generally egalitarian and collegial, with members freely offering advice and participants often speaking openly and personally about concerns. 3.4. Email at Fidelis The main decision process observed during this research was an evaluation to determine which of two systems should be used. Email had a long history at Fidelis and, in various forms, had been in use for more than 15 years. However, reflecting Fidelis’s previous highly autonomous nature and acquisitive strategy, several different and somewhat incompatible systems were in use. Specifically, the operations division used and promoted Lotus Notes, both wealth divisions used Microsoft Exchange, and the retail division used a primitive form of mainframe-based email. Although the systems could exchange messages, advanced features, such as scheduling of meetings across the enterprise, could not be used. As a result a constant undercurrent of criticism about the email configuration existed. Sarcasm and off-hand comments regarding the poor level of interoperability, and disparaging comments about the number of email products (though latent or disguised as humour) were frequently heard across all levels of the organization. In the preceding 3 years the email configuration had twice been extensively reviewed, with the most recent evaluation completed just 6 months earlier in December of 2002. Conducted by the operations division, that analysis produced a 37-page strategy document that recommended retaining Lotus Notes and concluded that moving to a single email platform was not economical. This analysis, however, was not accepted by all members of the bank, and was hotly contested and given little credence outside of the operations division. In particular, the wealth and the securities divisions rejected its findings and recommendations, not only because they did not participate in the analysis but because they believed that the operations division, historically predisposed to solutions from particular vendors, was incapable of unbiased research.

340

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

At the beginning of June 2003, after purchase of a customer relationship management system that integrated only with Microsoft Exchange, the CIO of the Canadian wealth division formally requested that the TLC again review the email configuration. Rather than develop an interface to Lotus Notes, wealth instead wanted to deploy four hundred additional Microsoft Exchange email users while remaining consistent with Fidelis’s email strategy. The TLC therefore directed the vice president of messaging to conduct a review. The vice president then requested the director of messaging (Elliot Ralph) form a project team and present a recommendation to the TLC by October 31, 2003. Ralph seconded a project manager from outside operations division to act as an unbiased team leader and, after developing a formal charter that outlined the scope of work for the team, authorized a $75,000 project budget. During this project, Fidelis provided access to meetings, artefacts, and participants. All team meetings were observed by the primary researcher, and all team members were interviewed at least once, as were senior representatives from Fidelis’s systems vendors and consulting. Transcription of the audio interviews was completed by third parties and verified by the researchers against the original audio file. As a further member check, interviewees reviewed and corrected the transcripts. Meetings and informal meetings were not recorded but extensive in vivo notes (Martin and Turner, 1986) were taken and reviewed either by another person in attendance or the chair or scribe of the meeting. In total, during the two months of focused data gathering, the primary researcher spent more than 21 full days on Fidelis premises and 40 h in telephone conversations or audio conferences, and conducted 33 formal interviews of 25 participants with more than 26 h of recorded interviews (more than 75 h spent with individual participants). A detailed case write-up and teaching note was prepared and reviewed by Fidelis (Tingling et al., 2004a,b) as were draft versions of this paper. Interviews with the participants were scheduled throughout the decision-making process using Spradley and McCurdy’s (1988) approach of grand-tour and mini-tour questions. Grand-tour questions lay out an in-depth descriptive sequence were formulated to begin the discussion of evaluation-and-selection of competing products and to set the context within which specific mini-tour questions could be asked. Following the grandtour questions, mini-tour questions, which arose from the earlier answers but were focused on specific scenarios and dealt with a much smaller unit of analysis, were also posed. 3.5. Data analysis The researchers followed Martin and Turner (1986), in identifying concepts and concept definition with theoretical memoranda. All notes and interviews were repeatedly read and assigned to primary categories in an open-coding technique using Atlas/Ti version 5 qualitative software (Strauss and Corbin, 1990). Once primary categories were complete, axial coding or relating of the categories to main categories was conducted. Theoretical memoranda began after the first interview and continued as a constant process until the very end as additional concepts or themes emerged or took shape and new incidents modified the meaning of existing concepts (Glaser, 1978). The concepts and

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

341

themes were discussed with several of the main Fidelis informants and either further refined or probed as the research continued.

4. Findings and interpretation Our approach was to look for themes or affirmations that control behaviour and are important to the character, structure, or direction of a culture (Opler, 1945), and then to consider them in the context of the framework presented earlier. In total, 10 themes were identified. Two themes—rational decision choice and ceremony—came directly from our pre-understanding, were the focus of this research, and are illustrated in Table 2. The rational choice theme described a belief that a thorough review process can determine the most appropriate technological option, and was characterized by instrumental collection and weighting of information about potential choices. Conversely the ceremonial theme described a situation where processes were ritualistic and had little instrumental utility in terms of evaluation-and-selection. At Fidelis, the email evaluation was a ceremonial process performed by convention and the resulting recommendation was a symbol. The outcome was known and defined in advance of the evaluation-and-selection process. The rational choice and ceremonial themes are described in the context of Meyer’s (1984) decision-making structure of clinical, fiscal, political, and strategic decision making. In this structure, instrumental behaviours are characterized by objective evaluations, evidence that technical influence or financial forecasts inform choices, the shaping of outcomes through power struggles, and intended strategies. Symbolic modalities, on the other hand, are characterized by the imputation of technological or business goals that justify prior choices, financial rituals that rationalize choices, pluralistic ceremonies that legitimize outcomes, and strategies that are discovered in retrospect. During the study, evaluations were generally portrayed as rational decisions to be solved with analytic tools and approached with objectivity. Objectivity, however, was not Table 2 Rational choice, analysis, and ceremony Categories

Concepts

Examples

Rational choice and analysis

A thorough analytic evaluation and review process can determine the most appropriate option. Objective information about choices is collected and weighted.

Ceremony

Evaluation processes are prescribed by ritual and have little instrumental value beyond symbolism. They are mainly performed by convention or as a ritual— identification of the preferred option—is known prior to the evaluation-and-selection process.

The project team has diverse representation and objectively discusses options and capabilities of each choice. SEU and weighting factors are used to create a decision matrix and the choice selected has the highest reported utility. The process is described as ‘due diligence’ and designed to justify future questions should the decision prove to be suboptimal. The evaluation process is repeated until the desired answer is obtained. Decisions may be nested within a higher order decision that has already been made thereby making the process moot.

342

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

necessarily viewed as a social product of open interaction and a variety of subjective judgments (Ackoff, 1979); rather it was contrasted against subjectivity and qualitative measures. At Fidelis, objectivity in evaluations was given the consistency of an absolute, God-like, single and consistent quantum (Ormerod, 1995). Comments by Keith Wood, a technical director with had spent more than 25 years at Fidelis, typically describe how evaluation-and-selection processes were conducted. “The approach to making a decision is setting up some criteria. You have already agreed these are the criteria, so you’re making the decision these are what you are going to measure against. You’re going to then put some measurement into them. Some of the measurements may be subjective. But you’re going to agree these are the scores we’re putting down against these things and you’re going to come with an end result and an end result is your decision.” These views were echoed by both Candice Summer, a systems director with 20 years of experience, and Jay Richards, who joined Fidelis following a recent merger. (Candice Summer): Those are very conscious decisions. Also, when it comes to technology decisions, generally speaking, if it’s a strategic kind of investment, we do have what I think is a fairly rigorous process that helps us evaluate the options, weight the options, and score them against a certain set of criteria, and make a decision or a recommendation on that purchase. (Jay Richards): What I’m learning is that a degree of rigour is required [at Fidelis]. I really sincerely believe that. To get work done and get decisions made. Now it’s not perfect but at least it’s got formality and process around it. The technical decisions observed at Fidelis displayed varying degrees of process, from an informal and low level that accompanied a decision to standardize on a particular Internet browser decision to a higher level during the email decision and other more highly visible projects. For example, Fidelis exerted ongoing effort to make the email decision objective and to address previous criticisms that the staff was biased. An example of this was the selection of a project manager, John Jones, new to Fidelis and outside of the main operations division. At times symbolic aspects of the process become more visible. For example, attendance on the project team was limited to two representatives from each affected unit in order to increase the democratization of the decision process and to avoid suppression of smaller voices. However, the opposite result occurred. Team composition of the email AWG was intended to be largely self-selected yet, in the case of operations division, was predominantly assigned. Operations division did limit itself to two attendees from each area but, because the email team had business units from each of the other divisions and the attendees were advocates of Microsoft Exchange rather than Lotus Notes, the overall effect was that Exchange tended to dominate discussion. In addition to the numeric overrepresentation of Exchange advocates, representatives from operations division (with one or two exceptions) had lower organizational status than other representatives, as indicated by titles and reporting relationships. As a result of this makeup, the team was predisposed towards choosing Exchange even though it had been asked to consider an

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

343

ideal outcome regardless of the existing configuration. While John Jones, the project manager, was believed by all members of the project team to be unbiased and objective (many members holding the incorrect belief that he was an external consultant), other participants were labelled and identified as being in either the Lotus Notes or Microsoft Exchange camp. During all evaluation meetings, discussion was dominated by proponents of the Exchange system and many of those who advocated Notes marginalized themselves, disconnecting from the process out of frustration. All members of the working group team believed throughout the process that the decision to standardize on Exchange as a single product was the desired outcome. Underlying this belief was the recognition that the email configuration was being reviewed for the third time in 3 years and that reviews were now occurring more frequently. Members believed that, if the outcome of the decision process was not different from previous recommendations, the decision would be revisited repeatedly until it was. Project members indicated they believed that a decision had already been made to go to Exchange, and that their job was to support it by writing a document that provided justification for a single Exchange configuration. The belief that the outcome was a fait accompli was muted at the (nonparticipating) more senior levels, where some managers indicated they did not have a predisposition to a particular system or that, rather than determining which email system was to be used, the objective of the project was to determine the optimum technical configuration—a definition that left substantially more room for manoeuvring. Project members identified an unarticulated process objective was to provide ‘due diligence’—a phase that meant evidence of analytic rigor and that decision processes would not be viewed as capricious and could be used to redirect future questions or criticisms of the decision (Starbuck, 1983). Ceremonial decision making was first identified during an early interview with a senior director prior to the start of the email evaluation. Technology director (Rick Neal): We are revisiting email—there it is—I think that there has been a decision made at Matt’s level (COO) and David (CIO) to go with Exchange and we are going to get off of Notes. [Pause] Elliot has been asked to do another review of email [Pause] even though we just did one eight months ago and we did it ad nauseum you know three or four years ago when we first picked the product. [Pause] The executives have already made up their minds. Again, [Pause]—somewhat off the record—we are being told that this is the way that we are going—we need to get on board—be team players—make the decision work out to be Exchange. Similarly, project members echoed sentiments that the decision had already been made and that the process was aimed at producing a document to support a decision to standardize on Exchange, as illustrated by one member’s comment about the Project, following the initial project kick-off. Project member (Elaine Winter): I think again that a decision has been made, and you know I don’t like wasting time and to me it’s like, you know, we know we wanna go here so let’s just get with it.

344

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

The project manager candidly noted that he had been asked to produce analysis supporting a position that had been considered less than a year earlier and rejected. Project manager (John Jones): Rick [technology director] told me to just take last year’s document and to change Notes to Exchange throughout the entire document and to republish—no one would notice. Project processes that usually have instrumental utility, such as formal management controls and a project charter, were created ceremonially in that none of the project members indicated that they had read or reviewed them. Furthermore, not only was the final recommendation document heavily based upon analysis that came to an entirely opposite conclusion less than 1 year earlier, but many believed that the document would be largely unread. Fiscally, Fidelis initially expected its evaluation process would instrumentally inform the choices and that the option selected would be the one that had the highest net present value or standardized costs. At the first meeting with the project sponsor, for example, the importance of financial disclosure was illustrated by the following discussion: Project sponsor (messaging director, Elliot Ralph): [In response to a question of cost] There’s already been some high level ones [estimates] ventured out from the more senior people before like myself. And those have come around using various research and those in ranges and trying to guesstimate. Like between two and five hundred bucks a user to convert. So we’re talking millions of dollars. Researcher: Millions? Elliot: The estimate to convert the whole seventeen thousand Notes users to Exchange is twelve million dollars, twelve to fifteen million [not including application costs]. Researcher: And do you see that number coming up in the presentations to TLC? Elliot: Sure. Researcher: So, on a slide or discussion? Elliot: Should be on a slide. So I want them [members of the TLC] to say when I make this decision I know these are the implications. So it’s sort of “I can make this decision” and then they don’t look surprised at me later on when I come back and say I want funding for this However, within the AWG, the fiscal process was less instrumental in that it rationalized rather than informed the choice. There was no continuous or formal financial representation on the team, conversion costs were not considered, and no financial projections were included in the final 24-page messaging document. Retail division of the bank, a group that accounted for more than 50% of the bank’s staff but did not attend any meetings, advised the messaging team that their main criterion was cost and was told that it could have a new, full-function email system for $179 per person per year, a figure that not only implied precision but was similar to the existing $15-per-person monthly charge. This figure, however, was derived with little analysis and rigour, and had little credibility.

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

345

Project members did not believe that it was possible to deliver a full-function email system at the same price as a basic system and that the actual cost when delivered could be more than twice as high. The request to review the email decision was similarly ceremonial. Ostensibly wealth division had asked for permission to roll out Exchange, but many wealth staff members were less diplomatic and indicated that wealth intended to proceed with Exchange irrespective of the project’s outcome. For example, during an email AWG meeting, the wealth division representative indicated that wealth had ‘already begun to implement Exchange’ and that he required everybody ‘onside’. In a follow-on interview, he commented: Project Member (Ian Goddard): I mean we’re already going hell-bent. We’re already going, moving in that direction. Like, this decision really is [irrelevant], from our perspective; we’re not quite sure why this effort is being undertaken in terms of a decision because a decision has been made to move to Exchange. One of the technical members of the project team recounted a similar response: Senior Technician (Sam Toronto): Right, and now we’re saying, “Okay, now jump through the normal hoops that you should have jumped through to get to this decision even though it’s already made.” Decision making frequently involves political aspects, particularly concerning capital and strategic issues (Carter, 1971a,b; Mintzberg et al., 1976). Politics and power, however, are difficult areas to research and their presence is often inferred (Mintzberg et al., 1976; Pfeffer, 1981, 1992; Starbuck, 1983). At Fidelis the political posturing was largely invisible, although several information systems managers suggested that the CIO of the operations division would be in an untenable situation if the evaluation did not confirm or legitimate the direction of his indirect subordinate, the CIO of wealth division. Because wealth division operated as a profit-and-loss business unit and operations division was simply an administrative unit, many within operations division believed they were in a dilemma. Even though many of the operations staff did not believe that a conversion to Exchange was the optimum choice, they believed that, if they recommended Notes, wealth would either refuse to comply (reducing the credibility of operations division) or would comply but would forever use the resulting infrastructure as a convenient excuse for business and technology problems. In effect, operations division would own all present and future wealth technology problems. One senior member of operations division recounted, with a level of animosity towards the posturing and political activities, that, although technically the CIO of wealth division reported to the CIO of operations, it was at best an ornamental relationship supporting the federated model of information technology rather than an authoritative relationship. The real reporting relationship for the CIO of wealth was to the head of the business unit. Four months after the conclusion of the data gathering, this relationship was formalized when the matrix reporting structure was dismantled and the CIO’s reporting relationship was isolated to the business unit. Strategically, Fidelis’s architecture group had a mandate to provide technology direction and governance to the divisions, characteristics that are consistent with an instrumental orientation. However, in practice many of the standards that were being

346

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

developed by the group were considered by the head of technology architecture to be ‘nobrainers that we can turn out quickly’, and reflected the existing configuration rather than an idealized or future state. For example, a statement of direction, a formal document that identified which technologies could be used with the bank, was being developed for Internet Explorer. This document, which reflected that Explorer rather than Netscape Navigator would be the browser of choice, was considered axiomatic by many because of the dominance of the Windows operating system. Commenting on Jay Richards’ desire to complete the statement of direction, a systems director noted: “He sort of has to do something visible and this is a quick and easy one. He has been in that department for 2 years and has produced absolutely nothing while other areas have created several. This lets him look like he is doing something that he can point to as output.” Other activities conducted by AWGs and discussed by the ARB, however, were considered strategic. Discussing a particular technology that was integral to Fidelis’s strategy, the architecture director bemoaned the group’s inability to obtain agreement or compliance with its recommendation on a key specification—a position that he considered ‘vitally important’ and a tenet of the computer architecture. The most visible decision symbols at Fidelis, aside from the meetings themselves, were the project charter and decision documents. The formal charter and structuring of the meetings were considered very important, yet seemed to have little effect or impact on the group and the outcome. At Fidelis project charters were used to initiate authorized or sanctioned projects by describing and circumscribing them. Existing in both hardcopy paper as well as electronic format, project charters were meant to be authorized by signature of the sponsoring executive yet, despite five revisions, the final charter for the email project was never physically signed or authorized. As indicated in these conversations with a key member of the project team, project charters were frequently unread despite their perceived importance. Researcher: Had you read the charter by then [when joining the project]? Project analyst (Margaret Soames): No Researcher: Have you read the charter yet [six weeks after project start, two weeks before scheduled end]? Margaret: I have read a number of documents but I don’t remember the charter specifically, have I read the charter. I read something a couple of days ago but I can’t remember what it was [laughs]. Researcher: How much has been written of the document so far? Any idea? Margaret: I think some of it will be pulled from the charter. I’m assuming that a lot of it will be pulled from that. Researcher: Is that an important document? Margaret: I’d say it’s uh, you have to have a document to have common understanding.

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

347

Researcher: But forgive me here. Margaret: Yeah I know. I haven’t read it [laughs]. With the exception of the project manager who drafted and revised the charter, all project members and decision participants indicated that, though they had familiarity with the charter and recognized that it defined the project scope and objective, they had not in fact read it. A similar prognosis was made for the final analysis document. The project manager, other participants, and senior decision makers noted that the paper itself would ‘be largely unread’, ‘never be read’, or ‘sit on a shelf unread but could be pointed at’, and the fact that large sections of text had been transcribed from a document that recommended an opposite conclusion would be unnoticed. The final PowerPoint presentation made to the TLC was initially expected to consist of four slides but grew to thirteen including a title slide, agenda, and the last-minute addition of a financial analysis slide developed outside the main project meetings. At Fidelis, all material to be presented to the TLC had to be finalized at least one week in advance and sent to the attendees along with the agenda. As a result, few recommendations were presented to any level of management without having already been extensively reviewed, discussed, and analyzed. Few surprises existed. Senior members of the AWGs interacted with the ARB, and senior members of the ARB presented to the TLC. Authorization was ceremonial and attendees were already familiar with what was going to be presented. When the electronic messaging vice president began presenting the results of the analysis to the TLC, he was interrupted at the opening slide and asked to address only the closing financial slide. With all members of the TLC present, the chair of the meeting advised that wealth division should proceed with its use of Microsoft Exchange but that, given the large conversion expenses, no further activity would be undertaken and both email systems would continue to be used. At Fidelis, decisions were a complex combination of instrumental rational decision making and symbolic or ceremonial action. Although little evidence of conflict other than staff cynicism was detected between the two, in many ways the instrumental aspects operated mainly at the surface or veneer of the processes despite being woven into the structure of the organization. The symbolic processes consisted of two groups. (1) Visible symbols, such as a formal charter and structuring of the meetings, were considered very important yet seemed to have little effect or impact on the group and the outcome. (2) Latent or less visible symbols, such as what the desired outcome should be or what possibilities were acceptable, were harder to detect yet more important once they surfaced. For example, the use of an open standards-based email system was highly desired by several of the senior (but nonexecutive) directors yet was never considered despite a view that it was lower in cost and technically more flexible. During a preliminary meeting with Elliot Ralph, he opined that his choice would be an open standards-based configuration not tied to the proprietary choices. In his view, such a system would meet the technical and business requirements, cost several millions of dollars per year less to operate, and provide flexibility and avoid constraint. However, despite these advantages, an open standardsbased configuration was neither presented to the senior management team as an option nor internally discussed because the solution was not in use by a mainstream organization.

348

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

Several months later, in a final interview concluding the research, Richards suggested that to be considered at Fidelis, options had to have the support of a powerful network or actor and that the only way that Fidelis would consider an open email system would be if another bank announced that they were implementing it. Institutional theory, which suggests that legitimacy is more important for organizations than rational efficiency arguments, could explain why Fidelis conducted formal evaluations to substantiate decisions that had already been made, and did not consider options that were potentially more efficient and effective than legitimate or organizationally sanctioned ones. Our research is consistent with earlier research that has found ritualistic or ceremonial justification of high technology to senior managers in order to obtain approval and to tell them what they wanted to hear (Currie, 1989). However, rather than the expected condition of legitimacy provided by the endorsement of senior managers, our findings are that senior managers facilitate and enable the ceremony but legitimacy is provided by more junior staff. Writing about the legal system, the jurist Lon Fuller (1978) acknowledged that not all rituals were harmful and that they could be beneficial, yet suggested that the core of human institutions was rational and that to elevate rituals as the basis of social order was to abandon hope of fruitful analysis (1978:358). Our finding suggests that Fidelis’s decision-making processes were superficially and ostensibly teleological and, at their core, were not instrumentally rational. When senior management decisions had already been made but required legitimation and internalization of junior staff, the processes seem to have achieved the desired results. However, when subjective expected utility and the relative ranking of criteria were utilized, the processes clearly did not achieve the expected result.

5. Implications for future research and theory development Much of the technology evaluation-and-selection literature has tended to be quantitative and prescriptive-developing variance models and suggesting how processes ought to be carried out rather than considering how decision processes actually unfold (cf. Currie (1989) for an exception). This research, on the other hand, looked beyond the presumed traditional rationality of decision models and illustrated how intuition, symbols, rituals, and ceremony are elements of the process; it considered that symbolic and instrumental aspects of the process might be in tension rather than in conflict. Instead of a fixed and simple linear decision process infused with outcome rationality, decision processes at Fidelis were consistent with much of the literature suggesting the path from goals to choice is recursive and includes loops and cycles (Simon, 1957; March and Simon, 1958; Mintzberg, Raisinghani and Theoret 1976). However, inconsistent with earlier research, we found that technologists did not present factual analysis of various alternatives to managers who then made judgment-based decision (Mintzberg, Raisinghani and Theoret 1976). Rather, the norm was for executives to make a judgment-based decision that analysts then justified and rationalized, or for decisions to be revisited until a desirable outcome was produced, a finding that has been observed by others (Soelberg, 1967).

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

349

An alternative explanation for this finding is that ceremonial processes develops and internalizes acceptance of the decision by more junior staff. This explanation has two advantages in that it fits the facts and preserves instrumental and procedural rationality. Though we did not directly study this possibility, some of our findings suggest that the opposite effect was achieved. Instead of acceptance, participants expressed resignation and indifference towards the decision. This research suggests that questioning managerial decision making from the perspective of competing analytic and symbolic decision processes need not cause conflict; rather, rationality and ceremony might be complementary. Decision processes and, therefore, the decisions themselves, may be socially constructed. Allowing that organizational actors have varying levels of influence and power and that ‘correct’ and ‘incorrect’ choices are not necessarily opposites, in some cases the appropriateness of particular choices may only exist exogenously in the case of incontrovertible evidence. Future research, therefore, should de-emphasize outcome rationality which considers the attributes of the decision itself and instead focus on process rationality, which stresses the extent to which decisions find their sense in attributes of the decision process. Despite the recognition that decisions themselves are signs and symbols to other organizational members (Feldman and March, 1981), the role of ceremony and symbols is often discounted in research either because they are noninstrumental or perhaps hard to observe. Symbols, though usually used in a pejorative sense (Cohen, 1969), need not necessarily be viewed as perversions of the decision-making process. Instead, they could also be motivational aids and signs of competence that symbolize the ability of decision makers (Leach, 1954; Cohen, 1975; Simon, 1988). Rather than being viewed as in opposition to traditionally rational behaviour, symbols can be complementary to, and contextualizing of, rational efficiency arguments (Scott, 1987; Dacin, 1997). Ceremony can be both rational and instrumental (Meyer and Rowan, 1977; Jackall, 1988; Clarke, 1999) as it is based upon the supposition that it may provide support in the event of failure by providing proof that the procedures used were prudent and the decisions were made by rational means (Starbuck, 1983). Langley et al. (1995) noted difficulties and inconsistencies associated with decision research, and suggested that, rather than reconstructing a decision back to its starting point—a process metaphorically described as tracking a wave back into the ocean—future research should instead take a phenomenological approach and track decision processes in real time. Though our research did not find the genesis of the wave, our findings suggest not only that both rationality and institutional forces can exist in an organization but that researchers need to probe carefully beyond surface observations of rationality. Understanding how organizations complete initial evaluation and acquisition of technology is important not just in its own right but also because the past can cast a long shadow on the future: how and why a particular technology was selected and evaluated in the past can have ongoing implications as to how it is used and understood in the present. However, decision research is complicated by the fact that individual actors can have different points of view and assumptions are created and held in complex social processes (Schutz, 1943). The subtleties and implications of interactions need to be understood within the context of each decision maker (Ormerod, 1995). Therefore two additional perspectives that might also further our understanding of ES are Actor Network Theory which views societies as

350

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

an interwoven and equal network of social and technical actors (Hanseth and Monteiro, 1997; Latour, 1999; Atkinson, 2000) and Soft Systems Methodology (Checkland, 1989), which, although it has been criticized for failing to adequately theorize the IT artefact and privileging actors over technology, attempts to articulate the complex processes that are used to add meaning to myths and assumptions.

Acknowledgements The authors would like to thank the research participants and Fidelis management who supported the research. All were helpful, but in particular we thank John Jones, Rick Neal, Elliot Ralph, and Jay Richards for their candour and encouragement. This paper was greatly improved with the thorough, insightful, and critical yet supportive assistance of the editors and anonymous reviewers. This research was supported by an Initiative on the New Economy (INE) grant from the Social Sciences and Humanities Research Council (SSHRC) Canada.

References Ackoff, R.L., 1979. The future of operational research is past. Journal of the Operational Research Society 30 (2), 93–104. Arnold, S.J., Fischer, E., 1994. Hermeneutics and consumer research. Journal of Consumer Research 21 (1), 55–70. Arthur, W.B., 1987. Competing technologies. Increasing returns and lock in by historical events. IIASA Paper WP-83-90, 116–131. Atkinson, C.J., 2000. The soft information systems and technologies methodology (SISTeM): an actor network contingency approach to integrated development. European Journal of Information Systems 9 (2), 104–123. Avgerou, C., 2000. IT and organizational change: an institutionalist perspective. Information Technology and People 13 (4), 234–262. Bakos, Y., 1998. The Productivity Payoff of Computers: a Review of the Computer Revolution: an Economic Perspective by Daniel E. Sichel. Science 1998;, 2001. Barnard, C.I., 1938. The Functions of The Executive. Harvard University Press, Cambridge, MA. Beggs, A., 1989. A note on switching costs and technology choice. The Journal or Industrial Economics 37 (4), 437–441. Bell, D.E., Raiffa, H., Tversky, A., 1988. in: Bell, D.E., Raiffa, H., Tversky, A. (Eds.), Decision Making: Descriptive. Normative and Prescriptive Interactions. Cambridge University Press, Cambridge, pp. 9–30. Benbasat, I., Goldstein, D., Mead, M., 1987. The case research strategy in studies of information systems. MIS Quarterly 1987;. Bocock, R., 1974. Ritual in Industrial Society. Allen and Unwin, London. Bonoma, T.V., 1985. Case research in marketing: opportunities. problems, and a process. Journal of Marketing Research 22 (2), 199–208. Brunnson, N., 1990. Deciding for responsibility and legitimation: alternative interpretations of organizational decision-making. Accounting Organizations and Society 15 (1–2), 47–60. Butler, R.J., 1990. Decision-making research: its uses and misuses. A comment on Mintzberg and waters does decision get in the way?. Organization Studies 11 (1), 11–16. Carter, E.E., 1971a. The behavioral theory of the firm and top level corporate decisions. Administrative Science Quarterly 16 (4), 413–429. Carter, E.E., 1971b. Project evaluations and firm decisions. The Journal of Management Studies 8 (3), 253–279.

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

351

Checkland, P., 1989. Soft systems methodology, in: Rosenhead, J. (Ed.), Rational Analysis for a Problematic World. Wiley, New York, pp. 71–100. Ciborra, C.U., Hanseth, O., 1998. From tool to Gestell: agendas for managing the information infrastructure. Information Technology and People 11 (4), 305–327. Clark, B.R., 1972. The organizational saga in higher education. Administrative Science Quarterly 17 (2), 178– 184. Clark, K., 1987. Investment in new technology and competitive advantage, in: Renewal, D.J. (Ed.), The Competitive Challenge—Strategies for Industrial Innovation and Renewal. Ballinger Publishing Company, Berkeley, CA. Clarke, L., 1999. Mission Improbable: Using Fantasy Documents to Tame Disaster. University of Chicago Press, Chicago. Cohen, P.S., 1969. Theories of Myth. Man 4, 337–353. Cohen, A., 1974. Two Dimensional Man: an Essay on the Anthropology of Power and Symbolism in Complex Society. Routledge and Kegan Paul, London. Cohen, A.P., 1975. The Management of Myths: the Politics of Legitimation in a Newfoundland Community. Manchester University Press, Manchester. Covaleski, M.A., Dirsmith, M.W., 1988. An institutional perspective on the rise. Social transformation, and fall of a university budget category. Administrative Science Quarterly 33 (4), 562–587. Currie, W.L., 1989. The Art of justifying new technology to top management. Omega 17 (5), 409–418. Dacin, M.T., 1997. Isomorphism in context: the power and prescription of institutional norms. Academy of Management Journal 40 (1), 46–81. DiMaggio, P.J., 1988. Interest and agency in institutional theory, in: Zucker, L. (Ed.), Institutional Patters and Organizations. Culture and Environment. Ballinger Publishing Company, Cambridge, MA, pp. 3–20. DiMaggio, P.J., Powell, W.W., 1983. The iron cage revisited: institutional isomorphism and collective rationality in organizational fields. American Sociological Review 48 (2), 147–160. DiMaggio, P.J., Powell, W.W., 1991. Introduction, in: Powell, W.W., DiMaggio, P.J. (Eds.), The New Institutionalism in Organization Analysis. University of Chicago Press, Chicago, IL, pp. 1–38. Donaldson, L., 1995. American Anti-Management Theories of Organization. Cambridge University Press, Cambridge; NY. Eisenhardt, K.M., 1989. Building theories from case study research. Academy of Management Review 14 (4), 532–550. Feldman, M.S., March, J.G., 1981. Information in organizations as signal and symbol. Administrative Science Quarterly 26 (2). Fichman, R.G., Kemerer, C.F., 1999. The illusory diffusion of innovation: an examination of assimilation gaps. Information Systems Research 10 (3), 123–139. Fuller, L., 1978. The forms and limits of adjudication. Harvard Law Review 92 (2), 353–409. Gadamer, H.-G., 1989. Truth and Method. Crossroad, New York. Garud, R., Ahlstrom, D., 1997. Technology assessment: a socio-cognitive perspective. Journal of Engineering and Technology Management 14 (1), 25–48. Gerwin, D., 1988. A theory of innovation process for computer-aided-manufacturing technology. IEEE Transactions on Engineering management 35 (2), 90–100. Glaser, B.G., 1978. Theoretical Sensitivity: Advances in the Methodology of Grounded Theory. Sociology Press, Mills Valley, CA. Glaser, B.G., Strauss, A.L., 1967. Discovery of Grounded Theory: Strategies for Qualitative Research. Walter de Gruyter. Goodstein, J.D., 1994. Institutional pressures and strategic responsiveness: employer involvement in work— family issues. Academy of Management Journal 37 (2), 350–382. Greenwood, R., Hinings, C.R., 1996. Understanding radical organizational change: bringing together the old and the new institutionalism. Academy of Management Review 21 (4), 1022–1054. Hanseth, O., Monteiro, E., 1997. Inscribing behaviour in information infrastructure standards. Accounting, Management, and Information Technology 7 (4), 183–211.

352

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

Hsee, C.K., 2000. in: Kahneman, D., Tversky, A. (Eds.), Attribute Evaluability: its Implications for JointSeparate Evaluation Reversals and Beyond Choices, Values and Frames. Cambridge University Press, Cambridge, pp. 543–563. Jackall, R., 1988. Moral Mazes. Oxford University Press, Oxford. Jepperson, R.L., 1991. Institutional effects, and institutionalism, in: Powell, W.W., DiMaggio, P.J. (Eds.), The New Institutionalism in Organizational Analysis. University of Chicago Press, Chicago, pp. 143–163. Keeney, R.L., 1988. Value-focused thinking and the study of values, in: Bell, D.E., Raiffa, H., Tversky, A. (Eds.), Decision Making: Descriptive, Normative and Prescriptive Interactions. Cambridge University Press, Cambridge, pp. 465–494. Khouha, M., 1995. The use of date envelopment analysis for technology selection. Computers and Industrial Engineering 28 (1), 123–132. Langley, A., Mintzberg, H., Pitcher, P., Posada, E., Saint-Macary, J., 1995. Opening up decision making: the view from the black stool. Organization Science 6 (3), 260–279. Latour, B., 1999. On Recalling ANT, in: Law, J., Hassard, J. (Eds.), Actor Network Theory and After. Blackwell Publishers/The Sociological Review, Oxford, pp. 15–25. Leach, E.R., 1954. Political Systems of Highland Burma. Bell, London. Lincoln, Y.S., Guba, E.G., 1985. Naturalistic Inquiry. Sage, Beverly Hills, CA. Lubbe, S., Remenyi, D., 1999. Management of information technology evaluation—the development of a managerial thesis. Logistics Information Management 12 (1/2), 145–156. March, J.G., 1988. Bounded rationality, ambiguity and the engineering of choice, in: Bell, D.E., Raiffa, H., Tversky, A. (Eds.), Decision Making: Descriptive, Normative and Prescriptive Interactions. Cambridge University Press, Cambridge, pp. 33–57. March, J.G., Heath, C., 1994. A Primer on Decision Making: How Decisions Happen. The Free Press, New York. March, J.G., Simon, H.A., 1958. Organizations. Wiley, New York. Martin, P.Y., Turner, B.A., 1986. Grounded theory and organizational research. The Journal of Applied Behavioral Science 22 (2), 141–157. McRary, J.W., 1995. Leveraging the power of information technology for sustained competitive success. Engineering Management Journal 7 (1), 3–5. Meyer, A.D., 1984. Mingling decision making metaphors. Academy of Management Review 9 (1), 6–17. Meyer, J.W., Rowan, B., 1977. Institutionalized organizations: formal structures as myth and ceremony. American Journal of Sociology 83, 340–363. Mintzberg, H., 1979. An emerging strategy of direct research. Administrative Science Quarterly 24 (4), 580–589. Mintzberg, H., Waters, J., 1990. Does decision get in the way?. Organization Studies 11 (1), 1–6. Mintzberg, H., Raisinghani, D., Theoret, A., 1976. The structure of unstructured decision process. Administrative Science Quarterly 21 (2), 246–275. Myers, M.D., 1997. Qualitative research in information systems. Management Information Systems Quarterly 21 (2), 241–242. Noori, H., 1995. The design of an integrated group decision support system for technology assessment. R & D Management 25 (3), 304–322. O’Leary, D., 2000. Enterprise Resource Planning Systems: Systems, Life Cycles, Electronic Commerce and Risk. Cambridge University Press, New York. Oliver, C., 1991. Strategic responses to institutional processes. Academy of Management Review 16 (1), 145– 179. Opler, M.E., 1945. Themes as dynamic forces in culture. American Journal of Sociology 51 (2), 198–206. Orlikowski, W.J., 1993. CASE tools as organizational change: investigating incremental and radical changes in systems development. Management Information Systems Quarterly 17 (3), 309–340. Orlikowski, W.J., Iacono, C.S., 2001. Research commentary: desperately seeking the IT in IT research—a call to theorizing the IT artifact. Information Systems Research 12 (2), 121–134. Ormerod, R., 1995. Putting soft OR methods to work: information systems strategy development at Sainsbury’s. Journal of the Operational Research Society 46 (3), 277–293. Pettigrew, A.M., 1979. On studying organizational cultures. Administrative Science Quarterly 24 (4). Pettigrew, A.M., 1990. Studying strategic choice and strategic change. A comment on Mintzberg and waters does decision get in the way?. Organization Studies 11 (1), 6–11.

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

353

Pfeffer, J., 1981. Power In Organizations. Pitman Publishing, Inc., Boston, MA. Pfeffer, J., 1992. Managing with Power: Politics and Influence in Organizations. Harvard University Press, Boston, MA. Powell, P., 1992. Information technology evaluation: is it different?. The Journal of Operational Research Society 43 (1), 29–43. Prasad, A., 2002. The contest over meaning: hermeneutics as an interpretive methodology for understanding texts. Organizational Research Methods 5 (1), 12–33. Quattrone, G.A., Tversky, A., 2000. in: Kahneman, D., Tversky, A. (Eds.), Contrasting Rational and Psychological Analysis of Political Choice, Choices, Values and Frames. Cambridge University Press, Cambridge, pp. 451–472. Ragowsky, A., Somers, T.M.S., 2002. Enterprise resource planning. Journal of Management Information Systems 19 (1), 11–15. Rennie, D.I., 2001. Grounded theory methodology as methodological hermeneutics: reconciling realism and relativism. Psychology Science 43 (3), 32–49. Riddle, W.E., Williams, L.G., 1987. Technology selection: an educational approach. IEEE Transactions on Software Engineering SE13 (11), 1199–1207. Robey, D., Boudreau, M.C., 1999. Accounting for the contradictory organizational consequences of information technology: theoretical directions and methodological implications. Information Systems Research 10 (2), 167–185. Rosenhead, J., 1989. Introduction: old and new paradigms of analysis, in: Rosenhead, J. (Ed.), Rational Analysis for a Problematic World. Wiley, New York, pp. 1–20. Schilling, M., 2002. Technology success and failure in winner-take-all markets: the impact of learning, orientation, timing and network externalities. Academy of Management Journal 45 (2), 387–398. Schutz, A., 1943. The problem of rationality in the social world. Economica 10 (38), 130–149. Scott, W.R., 1987. The Adolescence of institutional theory. Administrative Science Quarterly 1987;, 32. Scott, W.R., 1994. Institutional analysis, in: Scott, W.R., Meyer, J.W. (Eds.), Institutional Environments and Organizations: Structural Complexity and Individualism. Sage, Thousand Oaks, CA, pp. 81–99. Scott, W.R., 1995. Institutions and Organizations (Foundations for Organizational Science). Sage Publications, Thousand Oaks, CA. Shapiro, C., Varian, H., 1999. The art of standard wars. California Management Review 41 (2), 8–32. Siha, S., 1993. A decision model for selecting mutually exclusive technologies. Computers and Industrial Engineering 24 (3), 459–463. Simon, H.A., 1957. Administrative Behavior: a Study of Decision-Making Processes in Administrative Organizations. MacMillan, New York. Simon, H.A., 1960. The New Science of Management. Harper and Row, New York. Simon, H.A., 1988. Rationality as process and product of thought, in: Bell, D.E., Raiffa, H., Tversky, A. (Eds.), Decision Making: Descriptive, Normative and Prescriptive Interactions. Cambridge University Press, Cambridge, pp. 58–77. Soelberg, P.O., 1967. Unprogrammed decision making. Industrial Management Review 8, 19–29. Spradley, J.P., McCurdy, D.W., 1988. The Cultural Experience: Ethnography in a Complex Society. Waveland Press, New York. Starbuck, W.H., 1983. Organizations as action generators. American Sociological Review 48, 91–101. Strauss, A.L., Corbin, J., 1990. Basics of Qualitative Research; Grounded Theory Procedures and Techniques. Sage, Newbury park, CA. Swanson, E.B., 1994. Information systems innovation among organizations. Management Science 40 (9), 1069– 1093. Thompson, J.B., 1981. Critical Hermeneutics: a Study in the Thought of Paul Ricoeur and Jurgen Habermas. Cambridge University Press, Cambridge. Tingling, P.M., Parent, M., 2002. Mimetic isomorphism and technology evaluation: does imitation transcend judgment?. Journal for the Association of Information Systems 3 (5), 113–143. Tingling, P.M., Parent, M., Meister, D., 2004a. Fidelis Bank Financial Group: Selection of an Email System. Ivey Publishing 9B04E001.

354

P. Tingling, M. Parent / Journal of Strategic Information Systems 13 (2004) 329–354

Tingling, P.M., Parent, M., Meister, D., 2004b. Teaching Note for Fidelis Bank Financial Group: Selection of an Email System. Ivey Publishing 89B04E01W. Tolbert, P., 1988. Institutional Sources of Organizational Culture in Major Law Firms, in: Zucker, L. (Ed.), Institutional Patterns and Organizations: Culture and Environment. Ballinger, Cambridge, MA. Torkkeli, M., Tuominen, M., 2002. The contribution of technology selection to core competencies. International Journal of Production Economics 77 (3), 271–284. Tversky, A., Kahneman, D., 2000. Rational choice and the framing of decisions, in: Kahneman, D., Tversky, A. (Eds.), Choices, Values and Frames. Cambridge University Press, Cambridge, pp. 209–223. Verter, V., 2002. An integrated model for facility location and technology acquisition. Computers and Operations Research 29 (6), 583–592. Westfall, R.D., 1997. The telecommuting paradox. Information Systems Management 14 (4), 15–20. Wicks, D., 2001. Institutionalized mindsets of invulnerability: differentiated institutional fields and the antecedents of organizational crisis. Organization Studies 22, 659–692. Williamson, O.E., 1985. The Economic Institutions of Capitalism. The Free Press, New York. Williamson, O.E., 1996. The Mechanisms of Governance. Oxford University Press, New York. Yap, C.M., Souder, W.E., 1993. A filter system for technology evaluation and selection. Technovation 13 (7), 449–460. Yin, R.K., 1994. Case Study Research: Design and Methods. Sage Publications, Thousand Oaks, CA. Zucker, L., 1987. Institutional theories of organization. Annual Review of Sociology 13, 443–464. Zucker, L., 1988. Introduction: institutional theories of organization—conceptual development and research agenda, in: Zucker, L. (Ed.), Institutional Patters and Organizations. Culture and Environment. Ballinger Publishing Company, Cambridge, MA, pp. xiii–xix.