Communications of the ACM
-Understanding the Is--ser Divide in It Innovation
A senior executive participating in a 1992 study of user-IS relations in 10 multinational companies said: "IS people do not understand our requirements ... are 'techies' and don't understand business ... don't have good interpersonal skills ... have a sort of arrogance ... tend to zero in on the technical side of things and live in their own world ... are full of jargon ... It's taking too long to develop strategic systems ... The business isn't prepared to wait ... We want the Corvette, not the garage, and we don't care what the garage looks like" [10].
An IS manager in the same study responded: "The business people weren't fully aware ... The users don't know what they want or what is happening to them ... We've lost the feeling of control ... Someone else is pushing the buttons ... We try to make users happy but don't give them whatever they want" [10].
This exchange rings true more than 10 years after it took place. Commonly cited reasons for the poor relations between users and IS departments include delayed delivery of information technology (IT)1, limited user involvement in IT decision making, poorly prepared requirements, and user resistance to change. Users often complain that mistrust of the IS function (ISF)2 can escalate before or during implementation and extend to other projects, eventually to the overall IS operation.
Until recently, users, with little recourse other than reliance on their organizations' internal IS service providers, were more likely to tolerate this divide. But today, they're no longer mere recipients of passive service, demanding a stronger say, sometimes a leading role, in the adoption, development, and implementation of IT innovation. In many cases, users are prepared to go it alone without the IS department if they perceive they have the knowledge and development skills to identify and implement technical innovation. We should therefore want to understand the precise locations of the gaps between user and IS perceptions and skills and how to reduce the tension.
Users and IT Innovation
Traditionally, the adoption and implementation of IS and IT were viewed as complex tasks performed only by a formally constituted IS department. However, the proliferation of personal computing and individualized software, the fall of hardware costs, the rise of fourth-generation programming languages, and the mass-market adoption and popularity of the Internet, have elevated the user's role in directly driving IT implementation. Users have become increasingly confident about their ability to satisfy their own IT needs. Moreover, corporate management is increasingly inclined to delegate information and technology management to functional departmentsand away from the centralized IS function.
IS users today play an important part in developing and outsourcing many types of business applications. Functional and departmental managers increasingly ask their own people to take active roles in IT adoption and implementation. This trend was identified in [1], along with the "local heroes" in user departments taking on the responsibility of satisfying those departments' information needs, independent of ISF help. The ISF in response often complains that few such heroes consider planning, controls, standards, or documentation as important parts of IT innovation. Complaints about user-driven IT development projects involve sketchy requirement descriptions, crude programming, and practically no documentation. Making matters worse is the fact that companies are virtually unprotected against the turnover of local heroes.
User-driven IT innovation is subverting the traditional IS department as a driver of IT innovation.
At a higher organizational level, an IT user champion might look to identify a more senior manager from a functional department (such as marketing, engineering, or finance) with the organizational clout and savvy needed to see the implementation of a complicated IT project through to completion. IT user champions typically have enough technological experience and status to promote their personal visions for using IT, relentlessly pushing their projects over or around traditional approval mechanisms. Not surprisingly, IT user champions rarely need ISF help in rechanneling the organizational inertia to get their projects approved, budgeted, or implemented. Such independent spirit and user self-reliance can create additional conflict, as IS managers often have to give up something, possibly postponing other projects, in favor of a user-champion's projects.
The common thread in these observations is the trend toward user-driven IT innovation in organizations, moving the decision-making control over IT resources and budget authority to user departments; users increasingly make critical decisions regarding projects and resource allocations. The result is that user-driven IT innovation is subverting the traditional IS department as a driver of IT innovation. This trend is likely to continue in light of the twin factors of increasing user sophistication and the corporate need for rapid IT deployment to support e-business in competitive global markets where any information advantage might represent the difference between market share and also-ran [11].
Meanwhile, corporate managers do not adequately assess and manage the tensions between the two groups [5]. Yet, today, more than ever, there is a need for understanding between the ISF and user departments to determine the roles each should play in acquiring IT, controlling projects, training and supporting users, enforcing standards compliance, and maintaining IT once it is implemented.
Points of Contention
The tension between the ISF and users often stems from differences in background and cognitive style [4]; factors making it worse include: poor communication and imperfect information exchange between them; differing objectives and incentives; and a lack of knowledge or ability in either group [8].
Points of contention between users and the ISF may arise before, during, or after IT adoption. Before a new IT is adopted, the IS department and users may disagree on its relative advantage or may disagree on such issues as compatibility, complexity, visibility, and cost [9].
In many cases of IT innovation, users are so confident in their ability to drive the adoption and implementation of IT that they delegate the ISF to a secondary role. But being a backseat driver may not sit well with ISF managers and professionals, given the dimensions of the egos of some CIOs and among their IS staffs. To a large extent, deciding who actually drives IT adoption depends on the power and influence users and the ISF have over IT planning and resource allocation. Depending which group actually drives IT adoption, the other may feel its participation was inadequate, creating yet another source of conflict. One result may be lower user satisfaction from the new technology after its implementation.
Our study of user and IS perceptions of IT innovation across a variety of technologies sought to improve our understanding of the gaps between user and ISF perceptions of IT innovation and ultimately where and how the IT adoption process should be improved to establish a better user-ISF working relationship. We examined the implementation of four types of ITlocal-area networks (LANs), multiuser database management systems (MDBMSs), email, and spreadsheet softwarein 181 medium- and large-size U.S. service and manufacturing companies. We investigated the gaps between users and the ISF in terms of technical and organizational factors, including: level of need; understanding of innovation characteristics for each IT adopted; user technical expertise with each technology; user leadership in driving IT adoption; and user satisfaction with IT implementation.
Empirical Research Approach
We designed a survey instrument to collect the perceptual data needed to improve our understanding of the gaps. We mailed two versions of our questionnaire (one modified for the ISF, one modified for the user department) to 1,799 randomly selected companies from the "Compact Disclosure" database of U.S. companies with a minimum of 500 employees and a maximum of 10,000 employees.3 A CEO or senior member of each organization was asked to function as an initial contact to identify targeted departments, distribute questionnaires, and subsequently solicit completed questionnaires from both users and ISF professionals. Before we would include a company's answers, however, we required respondents from both the user department and the ISF be involved with each IT innovation from inception through implementation. After three months of telephone and mail follow-up, we had collected a total of 181 matched-pair responses (each pair representing an ISF response and user response). This 10.1% overall response rate should be viewed as fairly high, given the study's matched-survey approach.
IT Innovation Process
Figure 1 outlines the IT innovation process, breaking it into three distinct stages: initiation (pre-adoption); adoption; and implementation. Initiation involves an idea or proposal for a new IT. Adoption of an innovation reflects an organizational mandate for change. Implementation aims to fully integrate the technology within organizational behaviors and processes. We used survey data measuring ISF and user perceptions of the innovation process throughout these stages, as shown in the figure.
During the initiation stage, as reported by a majority of survey respondents, innovative ideas surface when business managers learn of a technological means of supporting a recognized need within their organizations. Critical IT attributes influencing the rate of the technology's adoption include: compatibility, or the extent it functions within existing values and the existing information infrastructure while satisfying existing needs; ease of use; trialability, or how relatively easy it is to experiment with the technology on a trial basis; cost; visibility, or how noticeable the technology is in operations; and demonstrability [7].
Technical expertise is another important factor influencing IT adoption and implementation. Its presence or absence enhances an organization's overall ability to implement a technology and may ultimately determine the kind of leadership needed for an IT project. It is measured in terms of people's confidence in their personal and collective technical abilities to see an IT implementation through to completion.
Users may be gaining influence in driving IT innovation at the expense of the IS group. During an IT project's adoption stage, we measured the degree of project leadership, or the extent the project was user- or ISF-driven, and users' decision-making control over the purchase, development, and installation of IT at the time of adoption. Finally, to gain insight into the overall IT innovation process, we measured perceived user satisfaction from both the user and ISF perspectives for each technology included in the study.
Our conceptual model (in Figure 1) emphasizes the importance of the differences of perceptions across the stages of the IT innovation process. We also concur with previous research that found that the type of IT being implemented can be another important issue in the IT innovation process [2]. For example, there are distinct differences between taking charge of development of the innovation for personal and inexpensive IT, such as, say, spreadsheet software, and the installation of a costly organizationwide LAN. However, please note that just because a technology like enterprise resource planning is more complex and has greater organizational effect than a personal technology like a spreadsheet does not necessarily mean users want to acquiesce control to the ISF.
Selection of the four technologies for the survey was based on our judgment concerning such tangible criteria as a technology's potential business effect on an organization, the nature of the task being performed with it (individual vs. organizational), and ownership (see Figure 2). We placed the technologies on a continuum to symbolize a range from primarily individually oriented applications (in terms of data, hardware, or software use) to organizationwide applications; the figure shows their relative positioning along this spectrum. We generally distinguish spreadsheets as the most purely individual technology of the four, while LANs and MDBMSs have the greatest potential for delivering organizationwide effects. Email is in the middle because it shares some of the characteristics of both personal and organizationwide effect.
Identifying Gaps
The survey results offer some interesting insights into the ISF-user divide, showing it emerges mainly during the pre-adoption stage. We found the widest gap in perceptions before adoption (see the table here). Moreover, the perceptions of users and the ISF concerning innovation attributes before adoption differed significantly for LANs and MDBMSs. Except for the ease-of-use variables for MDBMSs, all other innovation attributes during the initiation (pre-adoption) stage of an IT project show significant differences between the two groups' understanding of relative advantage, compatibility, ease of use, visibility, cost-benefit, and demonstrability. For example, users had higher expectations for trialability and cost estimation before adoption, while the ISF respondents had higher expectations for relative advantage, compatibility, and visibility. This expectation divide suggests that for organizationwide IT, users take a more cautious position than the ISF, hoping to try out a technology before consenting to its adoption. Meanwhile, the ISF seems to take more of an advocacy position than users, as reflected in a tendency to count on the relative benefits and feasibility of the more complex and organizationwide technologies before adopting them.
Fewer significant differences appeared between user and ISF perceptions concerning innovation attributes for individual/personal IT during the initiation (pre-adoption) stage, as in the table; users and the ISF alike acknowledged the benefits and feasibility of these relatively less complex and more personalized technologies. Please note that users felt they were less compatible with existing IT architectures and established organizational processes than the ISF. Study findings further indicate that user-ISF differences concerning assessment of simpler IT is not that great, with both groups sharing the same basic impressions as to their benefits and characteristics during the pre-adoption stage.
A somewhat surprising finding involves differences in each group's perceptions of its own expertise in leading a technology's implementation. For example, users believe they have relatively more "expert" knowledge and experience related to facilitating the adoption of LANs, MDBMSs, and email than the ISF believes it has. This finding may begin to explain the confidence, demonstrated by some users' eagerness to take on major organizationwide IT development projects. The ISF's relative lack of confidence in its own technical expertise may simply reflect the realities of adopting and implementing sophisticated systems with potentially sweeping organizational benefits, with potentially strong chances for failure. These findings point to the possibility that users may be underestimating the level of technical expertise required to implement complex multiuser systems, while IS groups have greater ambivalence concerning their own ability to evaluate and implement the technologies. Perhaps this lack of common ground between user confidence and ISF cautiousness is further dividing user-ISF perceptions, pushing users to take the reins and seek to drive IT project initiatives.
While users believe they have the expertise needed to implement LANs, MDBMSs, and email systems, users and the ISF agree the ISF actually drives adoption of these technologies in their organizations. For email and spreadsheet applications, there is significant disagreement between users and the ISF as to how much influence users actually have in driving technology adoption. In either case, the ISF believes users play a greater role than the users imagine they play.
After a technology is implemented, no significant gaps, as reflected in the table, were found between user and ISF perceptions of user satisfaction with particular technologies. This finding indicates that the IS group has a better read on user perceptions during this stage. However, users and the ISF agree that users were most satisfied with the implementation of individual/personal technologies and least satisfied with multiuser/organizationwide IT applications.4 For example, users were least satisfied with MDBMSs (at 4.61) and most satisfied with spreadsheets (at 5.7) on a seven-point scale.
Implications
The study found that when the user-ISF perception divide is less apparent during the early stages of IT adoption and the technology is intended for more individual use, there is likely to be greater user satisfaction after implementation. One problem for the ISF may be alleviated (at least partly) based on one of our other findings regarding email; that is, when users and the ISF agree on the need and characteristics of a technology during the pre-adoption stage, users are more likely to be satisfied later when it is implemented, even when the ISF leads the adoption and users have confidence in their own technical expertise. We learned that closing this perception gap during pre-adoption by establishing mutual understanding about IT appears to be an important factor in successful organizationwide IT implementations and subsequent user satisfaction.
The fact that we found variation in the gap between user and ISF perceptions among all four technologies in the study suggests that organizations should differentiate the way they introduce different technologies. For example, users of organizationwide technologies prefer a try-it-before-buying-it approach. Moreover, the ISF may need to tone down its overly optimistic expectations of the effects a proposed technology will have on user jobs.
Historically, many IS professionals have believed that users lack the expertise to carry out large-scale, long-term, complex IT projects, including the implementation of LANs and MDBMSs. Many computer professionals have believed that adoption of these technologies is just about always spearheaded by the ISF with minimal user participation. Our findings indicate strong user confidence in their own technical expertise and that the ISF's mistrust of users' abilities generates a divide between users and the ISF that may result in lower user satisfaction than might otherwise have been if users and the ISF were more inclined to trust one another's abilities. Interestingly, the finding that users are confident in their technical expertise for evaluating, buying, and implementing these technologies may spotlight a somewhat myopic view of their expertise that may, in and of itself, worsen ISF-user tensions.
IS professionals justifiably harbor deep concerns regarding the transfer of IT projects to users. Many such projects have resulted in quick-and-dirty applications intended to benefit a certain department or unit, not the organization as a whole. Users often fail to treat planning, controls, standards, compatibility, and documentation as important parts of their efforts. To mitigate these concerns, the ISF should seek to educate users about IT management and themselves about business management and strategy, as well as about managing customer expectations.
Users may exaggerate their ability to adopt complicated technologies while IS group members downplay their ability to implement technology; the result is a dynamic mix of miscommunicated expectations and actions. Our findings indicate that mutual ISF-user understanding of IT at the pre-adoption stage is critical for system success, particularly when the technology is complex, expensive, and its implementation affects the entire organization.
IS departments should confirm that users have the technical skills and/or experience needed to develop or install a particular technologyand that they share the same notion of the technology's overall organizational benefits. Communication between users and ISF professionals is also critical during the adoption stage; the ISF should combine its supporting role with the role of quality inspector to prevent mishaps and misunderstandings. If a constructive dialogue cannot be maintained between the two groups, pre- and post-adoption, we can expect the user-ISF divide to keep widening with both groups losing out.
References
1. Beheshtian, M. and Van Wert, P. Strategies for managing user-developed systems. Info. Mgmt. 12 (Jan. 1987), 17.
2. Brancheau, J. and Wetherbe, J. The adoption of spreadsheet software: Testing innovation diffusion theory in the context of end-user computing. Info. Syst. Res. 1, 2 (June 1990), 115143.
3. Damanpour, F. Organizational innovation: A meta-analysis of effects of determinants and moderators. Acad. Mgmt. J. 34, 3 (Sept. 1991), 555590.
4. Kaiser, K. and King, W. The manager-analyst interface in systems development. Mgmt. Info. Syst. Quart. 6, 1 (Mar. 1982), 4959.
5. Keil, M., Cule, P., Lyytinen, K., and Schmidt, R. A framework for identifying software project risks. Commun. ACM 41, 11 (Nov. 1998), 7683.
6. Kettinger, W. and Lee, C. Perceived service quality and user satisfaction with the information services function. Decis. Sci. 25, 5/6 (Sept./Dec. 1994), 737766.
7. Moore, G. and Benbasat, I. Development of an instrument to measure the perceived characteristics of adopting an information technology innovation. Info. Syst. Res. 2, 3 (Sept. 1991), 192222.
8. Pingry, D. and Marsden, J. End-user-IS design professional interaction: Information exchange for firm profit or end-user satisfaction. Info. Mgmt. 14 (Feb. 1988), 7580.
9. Rogers, E. Diffusion of Innovations, 4th Ed. Free Press, New York, 1995.
10. Smith, H. and McKeen, J. Computerization and management: A study of conflict and change. Info. Mgmt. 22, 4 (Jan. 1992), 5364.
11. Truex, D., Baskerville, R., and Klein, H. Growing systems in emergent organizations. Commun. ACM 42, 8 (Aug. 1999), 117123
Footnotes
1IT is used here as an inclusive term representing many enabling tools providing automated storage, processing, and communication of information, including computer and network hardware and software. IS is a set of production and service activities, including people, procedures, and technology to collect, process, store, and distribute information to solve specific organizational problems or to support specific business decisions.
2To distinguish an IS department/group from an information system, ISF is used interchangeably with IS department or IS group to represent the organizational unit responsible for IS service delivery [6].
3Whenever possible, we used established measures of constructs from the research literature. The questionnaire items were subjected to factor analysis and reliability testing, finding all constructs to be valid and reliable.
4Spreadsheet>email>MDBMS=LAN in terms of user satisfaction, at 0.05 significance level.
Figures
Figure 1. IT innovation process stage model.
Figure 2. Relative positioning of sample technologies, from multiuser/organizationwide IT to individual/personal IT.
©2002 ACM 0002-0782/02/0200 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2002 ACM, Inc.