LINKING INTRA-ORGANIZATIONAL STAKEHOLDERS:CIO PERSPECTIVES ON THE USE OF COORDINATIONMECHANISMS

Carol V. BrownVisiting ScholarCenter for Information Systems Research

V. SambamurthyAssociate Professor of MISFlorida State University

November 1998

CISR WP No. 304Sloan WP No. 4038

©1998 Massachusetts Institute of Technology. All rightsreserved.

Center for Information Systems ResearchSloan School of Management

Massachusetts Institute of Technology77 Massachusetts Avenue, E40-193

Cambridge, MA 02139-4307

CISR Working Paper No. 304

TitlE: Linking Intra-Organizational Stakeholders:CIO Perspectives on tfle Use of Coordination Mechanisms

Authors: Carol V. Brown, Visiting Scholar

Date: November 1998

Abstract: IT-based innovation requires the attention, involvement, and commitment ofboth business and IS stakeholders. Coordination mechanisms are an organization designsolution for linking these stakeholders via formal structures and processes as well as viainformal mechanisms that provide opportunities for voluntary collaboration. This paperreports the findings from semi-structured interviews with 38 senior IS executives fromprimarily Fortune 500 companies representing a broad range of manufacturing and serviceindustries. The data analysis yields an expanded scheme of 21 specific mechanisms within sixa priori categories (integrator roles, groups, processes, informal-relationship building, humanresource practices, IT-based systems) and an emergent scheme of three coordination goals(strategic alignment, partnering, learning). The coordination goals and design challengesassociated with specific mechanisms are described from a CIO perspective, and insightsrelated to implementing a portfolio of mechanisms are shared. Our intent is to provide both aroadmap for the IS practitioner and some empirical data on a topic of increasing importancefor the IS researcher.

27pages

This paper is also available as Sloan Working Paper No. 4038

INTRODUCTION

The delivery of innovative IT solutions has become a fundamental business competency inmost contemporary firms (Feeny and Wilcocks, 1998). Yet, IT-based innovation requiresthe attention, involvement, and commitment not only of information systems (IS)management, but also of senior management and business (line) management (Mata,Fuerst, and Barney, 1995; Ross, Beath, and Goodhue, 1996). Thus, one of the challengesfacing CIOs is how to build and sustain effective intra-organizational relationships amongthese key stakeholders (Rockart, Earl, and Ross, 1996).

Many organizations are attempting to link business and IS stakeholders by implementinggovernance structures that gives business management significant authority for ITapplications. For example, under a federal governance design, corporate IS has authorityfor the IT infrastructure and other enterprise-wide IS issues, whereas business units haveauthority for local IT application decisions (Rockart et al., 1996).

However, in most mid-sized and large organizations, there are three reasons whygovernance structures alone might be insufficient for linking together key stakeholders forIT innovation. First, when individual business units assume authority for specific ISactivities, business-level objectives may be in conflict with enterprise-wide objectives,including attention to synergistic opportunities (Brown and Magill, 1998; Kayworth andSambamurthy, 1998). Second, many IS activities require a joint consideration of businessand IS issues. In particular, IT applications decisions require a consideration of the currentIT infrastructure, and IT infrastructure investment decisions should include judgmentsabout the current and future business needs for IT (Rockart, 1988; Zmud, 1988). Yet, ISmanagement may not be aware of emerging strategic thrusts or business management maynot have the necessary expertise to make the appropriate IT-related assumptions. Third,gaining the trust of other stakeholders is a prerequisite for cross-unit partnering and suchtrust is built on interpersonal relationships fostered through people-to-people contacts(Gambetta, 1988; Henderson, 1990).

Coordination mechanisms represent an additional way for CIOs to link intra-organizationalstakeholders. While governance arrangements vest decision authority with specificstakeholders, coordination mechanisms laterally link together people from different parts ofan organization (Galbraith, 1993). In particular, horizontal mechanisms bind togetherstakeholders through deliberately orchestrated interactions via formal roles, groups, orprocesses, or via opportunities for voluntary informal exchanges (Mohrman, 1993).Contemporary firms are using coordination mechanisms to build and nurture linkagesamong senior, business, and IS management (Brown and Ross, 1996; Rockart, 1988;Zmud, 1988) under different IS governance arrangements (Brown, 1998).

Most prior empirical research on coordination mechanisms for linking IS stakeholdersaddresses the merits, pitfalls, and implementation challenges associated with individualmechanisms (e.g., Hufnagel and Birnberg, 1989; Torkzadeh and Xia, 1992). A few recentstudies also provide frameworks for categorizing coordination mechanisms for linking ISstakeholders (Brown and Ross, 1996; Brown, 1998). Missing from this empirical research

I

is an investigative survey of the way different types of coordination mechanisms are beingimplemented by CIOs to link intra-organizational stakeholders and an understanding of thespecific coordination goals that they seek to achieve with such mechanisms.

This paper draws upon semi-structured interviews with CIOs from 38 primarily Fortune500 companies in order to develop insights into the usage of coordination mechanisms forlinking intra-organizational stakeholders. The next section reviews prior organizationtheory and IS literature to establish our existing knowledge. This literature also providesthe basis for an initial taxonomy of six categories of IT coordination mechanisms that arethe focus of this study.

PRIOR LITERATURE AND RESEARCH QUESTIONS

Coordination theory has its roots in the basic logic of organizing. As described byThompson (1967):

"By delimiting responsibilities, control over resources, and other matters, organizationsprovide their participating members with boundaries within which efficiency may be areasonable expectation. But if structure affords numerous spheres of boundedrationality, it must also facilitate the coordinated action of those interdependentelements (p. 54)."

In other words, differentiation gives rise to organizational interdependencies -- "situationsin which what happens to one organizational actor affects what happens to others" (Pfeffer,1981, p. 68). Further, interdependencies require coordination for multiple reasons. First,interdependencies necessitate greater amounts of information sharing. Rapid changes inenvironments and complex internal resource dependencies strain the ability of formalgovernance structures to quickly and effectively direct the flow of information (Mackenzie,1986). Coordination mechanisms create lateral linkages across individuals or units tofacilitate information sharing. Second, interdependencies require collaborative problem-solving across different organizational units. Coordination mechanisms facilitateinteractions between individuals from different units in order to pool knowledge anddevelop a common language and shared understanding that are needed for cross-unit sense-making (Galbraith, 1993; Weick and Daft, 1984). Finally, information sharing and sense-making require trust among the collaborating actors (Gambetta, 1988; Henderson, 1990).Trust is needed for individuals to share accurate information, to tolerate task-relatedambiguities, and to be influenced by others. Again, differentiation impedes thedevelopment of trust between different individuals responsible for different tasks(Mohrman, 1993), but coordination mechanisms can be used to build trustful andharmonious relationships across unit boundaries.

Specific Coordination Mechanisms in the Organization Theory LiteratureOrganization theorists have identified a variety of formal and informal mechanisms for themanagement of organizational interdependencies. For example, formal groups andintegrator roles are considered to be common mechanisms for managing complexinterdependencies (Galbraith, 1994; Mohrman, 1993). Formal groups bind together

2

individuals from two or more units and vest them with authority and responsibility formanaging interdependencies. Integrator roles vest individuals with formal authority andresponsibility for managing dependencies between designated units. Formal processes,including planning, budgeting, and transfer pricing mechanisms, can also be used as formalmechanisms to manage interdependencies (Mohrman, 1993). These mechanisms alsofoster the development of trustful relationships and enable more concentrated attention toinformation transfer and collaborative problem-solving. On the other hand, these types offormal mechanisms require substantial personal commitment and time.

In addition to groups, roles, and processes that orchestrate formal interactions, voluntarycoordination behaviors can be facilitated by "informal" mechanisms (Mohrman, 1993).For example, mechanisms as varied as job rotation, physical co-location, and IT-basedcollaboration systems can create a context in which the cross-unit integration of work is a"natural" or "spontaneous" response (Galbraith, 1993).

Coordination Mechanisms in the IS LiteratureTwo recent studies by IS researchers have developed frameworks for categorizingcoordination mechanisms for linking IS stakeholders. Based on data from twelve largeorganizations, Brown and Ross (1996) document the usage of individual roles, group roles,and process enhancement mechanisms for achieving two differing objectives: building IS-business partnerships and supporting IT infrastructure development. These includemechanisms for linking corporate IS with both business management and decentralized ISunits. Brown (1998) has proposed a scheme of four mechanism categories based on asynthesis of prior organization theory and IS literature: formal groups, formal roles,informal networking practices, and cross-unit human resource practices. Empiricalevidence for the usage of all four categories of mechanisms to coordinate across IS andbusiness units, as well as across corporate IS and dispersed IS units, is reported on the basisof a theoretical sampling: one case site with a centralized IS governance but geographicallydispersed IS units, and a second case site with a federal IS governance mode. Initialsupport is also provided for the propositions that the most highly valued mechanisms inorganizations with centralized IS governance will be those directed at IS-business linkages,whereas the most highly valued mechanisms in organizations with federal IS governancewill be those directed at IS-IS linkages.

Looking next at the literature on specific mechanisms, steering committees are a type offormal group mechanism and one of the most frequently researched IT coordinationmechanisms (Brown, 1998). Steering committees have been commonly used by CIOs tomanage resource allocation interdependencies: committee members representing differentbusiness units allocate limited IS resources based on project requests from differentstakeholder groups (Drury, 1984). According to Rockart et al. (1996), CIOs in leadingedge firms have steering committees that focus on strategic IS and business alignment andthis formal group mechanism helps to develop organization-wide perspectives. Earl andFeeny (1994) found that effective CIOs use this type of mechanism to educate their seniorbusiness stakeholders about IS issues, "seed" them with creative ideas for IT use, and tohelp interpret external IT success stories to stimulate executives' thinking about ITinnovations. However, a steering committee mechanism is not universally appropriate

3

(Brown, 1998): context factors, such as the CIO's membership on a top management team,or the presence of another formal mechanism such as an integrator role, can influence theappropriateness of a steering committee mechanism.

Empirical IS research on integrator roles is scarce, although evidence of the use of this typeof role mechanism has been documented in the IS practitioner literature (e.g., Hildebrand,1994). Iaccono, Subramani, and Henderson (1995) report that account manager roles areused to coordinate the delivery of IS services in response to business clients' interests andneeds, to manage impressions about the responsiveness of the IS organization, and toestablish accountability for client satisfaction with IS services. Brown (1998) found thatintegrator role mechanisms are highly valued by IS and business managers for achievingbusiness and IS integration under both centralized and federal governance contexts.

Strategic planning processes are another highly researched coordination mechanism (Byrd,Sambamurthy, and Zmud, 1995). As defined by Henderson and Sifonis (1988), strategicplanning is a formal process for enhancing linkages between the strategic business plan andthe strategic IT plan and for defining ways in which IT products and services can betargeted at strategic priorities. Another type of formal process for managing IS-businessinterdependencies is a chargeback system (Ross, Beath, and Vitale). Chargebacks typicallyinclude data center costs, but the inclusion of other services widely varies. A relatedprocess mechanism, service level agreements, can be used for a variety of IS services,including application development and both network and data center operations. Theperceived fairness of the basis on which clients are charged for IS services has beenassociated with the quality of the IS-business relationships (Hufnagel and Birnberg, 1989;Ross et al).

One of the "informal" mechanisms mentioned above in the organization theory literaturereview, a physical co-location mechanism, has been documented by several IS researchers.According to Ross et al (1996), physical co-location facilitates the building of stronger

relationships among IS and business managers. Brown (1998) found that the co-location ofsystems development managers who were serving integrator roles with their clients was aneffective mechanism combination under centralized governance.

The usage of human resource mechanisms to increase the likelihood of collaboration acrossstakeholders has received much less attention in the IS literature. Brown (1998) describesan IS organization with a federal design that was able to leverage interpersonal networksbetween corporate and decentralized IS managers due to the organization's multi-yearhistory of aggressive, cross-unit job rotation practices.

Finally, the usage of IT-based systems to link organizational members has been widelydocumented in the general IS literature; the overall growth of groupware products is atechnological success story of the past decade (Johansen, 1989; Mankin, Cohen, andBikson, 1996). DeSanctis and Jackson (1994) argue for the potential of advancedinformation technologies such as electronic meeting systems and discussion databases toenable the coordination of IS activities and planning initiatives across a dispersed ISworkforce.

4

CritiqueOur review of the existing literature suggests that coordination mechanisms are animportant element of contemporary organization design. Organization theorists haveargued for the importance of implementing multiple formal and informal coordinationmechanisms. Although most prior empirical research on IT coordination mechanisms hasbeen confined to specific mechanisms, this body of literature as a whole provides initialevidence that coordination mechanisms can be effective organization design tools for CIOsto link not only IS and business stakeholders, but also decentralized IS stakeholders.

Our literature review also suggests an initial coordination mechanism schema of sixcategories (see Table 1). Two of these mechanisms are structural overlays emphasized byorganization theorists over the past two decades: integrator roles, such as an accountmanager, and formal groups, such as a steering committee. Another category of formalmechanisms in the framework by Brown and Ross (1996) is formal processes such asstrategic planning and negotiated service contracts. The remaining three categories ofmechanisms are referred to as informal mechanisms by organization theorists (e.g.,Mohrman, 1993). Informal relationship-building practices such as physical co-locationand human resource practices such as job rotations into and out of corporate IS have beenfound to be useful mechanisms for counterbalancing the effects of a centralized or federalgovernance structure (Brown, 1998). The sixth category, IT-based systems, has recentlybeen conceptualized as an important design tool in combination with other mechanisms(e.g., Galbraith, 1993; DeSanctis and Jackson, 1994). Its inclusion here as a separatecategory is further supported by the recent growth in the usage of groupware products toincrease cross-unit information sharing and other collaborative activities (Mankin, Cohen,and Bikson, 1996).

In the absence of a systematic examination of the set of coordination mechanisms in TableI for linking IS stakeholders within a sample of IS organizations, we undertook adescriptive study that focused on the following research questions:

1. From the CIO's perspective, what are the most important coordinationmechanisms for linking intra-organizational stakeholders?

2. What coordination goals are CIOs seeking to achieve through the use of thesemechanisms?

Our specific objectives were to expand the taxonomy in Table I and develop an initialschema of coordination goals associated with mechanism usage. In addition, the researchmethodology adopted for this study allowed us to capture insights from CIOs in mostlyFortune 500 companies about implementing mechanisms in a variety of organizationalcontexts and industries. We present the findings from our survey after a brief descriptionof our research methods.

5

RESEARCH METHODOLOGY

A professional group of senior IS executives interested in identifying best practices oncritical IS management issues sponsored this field study. The findings reported here arebased on data collected from 38 firms representing nine different industries. Theseorganizations were part of an initial set of 75 companies noted for their IT management inthe trade press or by peer reputation (see Appendix for details). Most of the participantswere CIOs at Fortune 500 companies; in a few instances, CIOs designated a direct report toparticipate.1Table 2 profiles the 38 participating organizations in terms of industry represented, theirformal IS governance arrangements, and their overall organizational contexts. Centralizedand federal governance modes predominate, which appears to be characteristic of ISorganizations within large companies in the mid-1990s (Brown and Magill, 1998; Earl,1996). We have used the term customized for organizations in which there is not a uniformcentralized governance mode: under a customized design, some of the larger business unitshave significant authority for IT use decisions and therefore operate in a federal mode,while other business units have centralized IS governance.

Data gathering occurred via one-hour telephone interviews. Interviewees were provided inadvance with a one-page prospectus that stated the study's objectives and a workingdefinition of the six coordination mechanism categories in our a priori scheme (for specificwording, see the Appendix). Each interview began with an inquiry about the respondent'sorganizational context. An open-ended question was then asked to elicit three to fivecoordination mechanisms that the CIO considered important for achieving a cooperative ITmanagement climate. Follow-up questions were asked to better understand why a specificmechanism was being used and any implementation challenges that were encountered.Many respondents provided details on how the different mechanisms fit together as theydescribed them. The interview typically lasted about an hour, although some respondentschose to continue talking longer to us than the initially scheduled time period.

The responses obtained from each company were first mapped into the six categories in thea priori scheme. Specific mechanism types for each category were then identified based oncross-company analyses for each category. Similarly, the qualitative responses oncoordination goals were culled from all interview transcriptions and then categorized via aniterative process. Further details of the methodology are provided in the Appendix.

By using a simple taxonomy and anchoring discussions around mechanisms that the CIOsconsidered most important, we were able to develop a relatively rich understanding of theusage of coordination mechanisms in a given organizational context as well as the goalsand challenges associated with a specific mechanism. Given the profiles in Table 2, ourfindings are intended to be generalizable to large organizations with different IS

I This research project was sponsored by the Advanced Practice Council of the Society for InformationManagement.

6

governance structures in a broad range of manufacturing and service industries.

FINDINGS: COORDINATION MECHANISMS

As described earlier, the existing organization theory and IS management literatures yieldeda scheme of six categories of coordination mechanisms (see Table 1). Further, the priorliterature suggests that CIOs use these mechanisms for linking the corporate IS unit withboth business and dispersed IS stakeholders. Our analysis of the interview data yielded anexpanded taxonomy of coordination mechanisms, with multiple types of mechanisms ineach of the six initial categories. Below we describe our specific findings for eachmechanism category, including any salient design challenges.

Integrator RoleWithin an IT coordination context, an integrator role is an individual who is formallyassigned with the responsibility for linking activities between a corporate IS unit and one ormore business units2. This individual is the primary point of contact for the business unitand is accountable for responding to business management's needs. Individuals who playthis role are also in a position to informally educate and "seed" business managers withinnovative ideas for IT use, as well as sensitize corporate IS managers about their clients'emergent IT issues. The establishment of this linking role theoretically increases thelikelihood that either a corporate IS initiative or a business unit initiative will jointly takeinto account both IS and business stakeholder perspectives.

Table 3 illustrates the different types of integrator roles and their governance contexts thatemerged from our data analysis. Twenty-nine of the 38 organizations in our study reportedsome type of integrator role to be among their most important coordination mechanisms 3.Four different types were identified within centralized or customized IS governancecontexts. The fifth type, which we refer to as a divisional information officer role, wasfound in almost three-quarters of the organizations with a highly decentralized or federal ISgovernance context.

Patterned on the model of a consulting organization, an account manager role with nodirect responsibilities for a systems development staff was found in all four organizationswith a customized design and five organizations with a centralized design. The accountmanager usually reports to corporate IS managers, typically the CIO, but is often physicallyco-located with one or more business clients. Few account managers had any formalreporting relationship to the business unit, although they were typically a part of businessmanagement team meetings and were expected to be knowledgeable about their businessclient's strategy and major processes. The process manager role is similar to an account

2 By this definition, the CIO role is also an integrator role. However, since our study was from theperspective of the CIO as an architect, this role is outside the scope of the study.

2 One manufacturing company with a customized governance design reported the usage of twointegrator roles: a DIO role played by IS heads reporting to business units and an account managerrole reporting to corporate IT. The total number of integrator role examples is therefore 30 rather than29.

7

manager, except this role links with business process owners. Both process managerexamples were in process-oriented organizations with major re-engineering initiativesunderway or responsibilities for ongoing integrated IT solutions for multiple businessclients.

Three companies reported the use of systems directors who had significant IS application orservice delivery responsibilities as integrators responsible for managing the business clientrelationship. We labeled this type of integrator an "enhanced" systems director becausethese directors played the role of account manager as managed an IS unit (usually systemsdevelopment staff) with a solid-line reporting relationship to corporate IS.

Four companies with centralized IS governance also reported a type of integrator roleplayed not by IS managers, but by business managers. Referred to here as client liaisons,these integrators were typically IT-literate business managers responsible for coordinatinginitiatives with corporate IS, such as serving as the "owner" of a key business application.These integrator role positions were typically implemented without any dotted-linereporting relationship with corporate IS. In one organization, a former IS manager hadbeen rotated into this role for an 18-month stint with the specific intention of broadeninghis business knowledge.

The divisional information officer (DIO) role describes those senior IS managers who wereresponsible for major application development staffs and also had significant IT operationsresponsibilities for a business unit, or business sector, in addition to their integrator roleresponsibilities. Two-thirds of the companies with federal IS governance designsmentioned this type of integrator role. Although not all DIOs were in officer-levelpositions within their business unit, all of our DIO examples had a solid-line reportingrelationship to a divisional business head and were typically a recognized member of thedivision's management team. Three of the eight DIOs under federal IS designs had matrixreporting (solid-line report to corporate IS as well as the business); two other companiesreported the use of a dotted-line DIO report to corporate IS.

Design Challenges. Our interview data also uncovered several significant designchallenges. One persistent challenge was the achievement of desired "allegiance" to bothIS and business management. For example, account managers report to central ISmanagers, but they need to be active participants in the strategic and tactical meetings oftheir business client as well as corporate IS. How can the person in the integrator roleavoid being perceived as only representing corporate IS interests in meetings with seniorbusiness managers? Similarly, how can such an individual avoid being perceived as onlyrepresenting the interests of one or more business clients when attending meetings of thecorporate IS leadership? As mentioned by one CIO, overzealous advocacy of a specificbusiness unit's interests can run the risk of the integrator being perceived as "taking sides"and becoming alienated from IS colleagues.

CIOs reported two design solutions to help integrators achieve the "right balance." Thefirst solution was to implement a dual reporting structure. In a few instances, integrators incentralized contexts who had a solid-line reporting to corporate IS were also given a dotted-

8

line reporting relationship to the business client. More common was the implementation ofa dotted- or solid-line reporting relationship to corporate IS for integrators in not totallycentralized contexts. However, only 3 of the 29 organizations reported implementing adual solid-line reporting (matrix) relationship. A second solution was to designperformance appraisal and incentive systems that included input from both IS and businessmanagement. For those integrators with dual reporting relationships, both IS and businessinput was already ensured; for those integrators without dual reporting relationships, multi-unit input provided an institutional incentive to achieve an appropriately balanced"allegiance" without a formal dual report.

Another significant challenge was related to the range of skills necessary for effectiveperformance in the integrator role positions. CIOs mentioned three types of skills as beingimportant: (i) strategic thinking, or the ability to understand business drivers and keybusiness processes and to be able to envision opportunities for strategic IT use; (ii)negotiation and collaboration skills, or the ability to influence people through interpersonalinteractions; and, (iii) IS service brokering skills, or the ability to diagnose client problemswith IS services and to identify appropriate IS experts who could efficiently and effectivelyresolve those problems. Several CIOs mentioned the difficulty of filling their integratorrole positions due to the scarcity of individuals that possessed this range of skills. Someindividuals selected for these positions were internal hires with well-establishedinterpersonal relationships with particular business clients; other individuals were externalhires from consulting organizations with proven technical and interpersonal skills. SeveralCIOs reported individually mentoring individuals in these positions or providing specialtraining programs in order to develop the needed skills. One organization with severalyears of experience with different types of integrator role positions in both centralized andfederal IS contexts reported that resources had recently been committed to developing acorporate IS training and development program specifically for integrator role skills.Another organization reported that a formal linkage across their account managers wasestablished using a "centers of excellence" team mechanism in order to facilitate learningfrom each other.

GroupsGroups are formally established councils or teams of either business and IS stakeholders, orcentral and dispersed IS stakeholders, that are given specific linking or oversightresponsibilities for IS-related activities4 . In contrast to integrator role mechanisms that vestspecific individuals with a coordination responsibility and authority, group mechanismsvest multiple individuals with coordination responsibilities. Further, while thecoordination activities of integrator roles are bilateral (the individual integrator and specificorganizational units), groups typically have multilateral responsibilities (i.e., coordinationacross more than two organizational units). Thirty-six of the 38 companies reported the useof a group mechanism and from these responses we identified six different types of groups

4 Task forces for specific systems development or acquisition projects (i.e., systems project teams) wereconsidered outside of the scope of this study. As described below, the task force classification wasused here for ad hoc groups typically chartered by an executive council to develop recommendationsrather than develop applications.

9

(see Table 4).We classified half of the instances of groups as an executive council. The typicalmembership of an executive council includes the chief operating officer and/or the chieffinancial officer, and senior business executives from the dominant lines of business, alongwith the CIO. Executive councils deal with enterprise-wide IS issues such as capitalappropriation for large IT projects, IT policy endorsements, and articulations of IT visionand business-IS strategy. They also as serve as educational and communication forums onIT-related issues. More than half of the CIOs in organizations with centralized governancemodes mentioned executive councils, but nearly half of the CIOs in federal governancecontexts also considered this mechanism to be significant. Divisional steering councilswere mentioned by CIOs in large, often global, multi-divisional companies, usually withfederal governance modes. In federal contexts, the council members are senior businessmanagers of the division and the divisional information officers.

One-third of the CIOs in centralized contexts mentioned the use of a project steering teamto oversee a process or one or more large application projects. The use of such a steeringteam ensured ongoing involvement of key business executives in major initiatives.

Two types of group mechanisms to link across IS stakeholders were also frequentlymentioned as significant coordination mechanisms: councils of IS managers (12 mentions)and IT standing teams (14 mentions). In two-thirds of the organizations with a federalgovernance mode, an IT management council was used to formally link divisional IS headswith the CIO and other corporate IS managers who had IT strategy, infrastructure, andcorporate (or enterprise-wide) IT application responsibilities. More specifically, thesecouncils direct attention to the "IS business" of the organization and serve as a coordinationforum to:

1. Advise (counsel) the CIO on strategic IS organization issues2 Surface potential impacts of corporate IS policies3. Share IT-related best practices, solutions, and problems4. Nurture and grow IS human resources

IT standing teams typically are groups of IS managers formed to address specific IT issueson an ongoing basis, such as IT standards issues or the identification of IT skill gaps. ITstanding teams are often chartered by one of the above mentioned councils and the standingteam's recommendations are typically submitted to this council for approval. A commonexample was an IT architecture group of IT experts to develop recommendations on high-level IT infrastructure issues. IT standing teams were prevalent across all four governancedesigns.

Six of the IT standing team mechanisms mentioned involved experimentation with aspecific form referred to as "centers of excellence" (CoE). Some firms used this groupmechanism as a "virtual homeroom" for IS employees with specific skillsets. These CoEsfocused on developing groups of people with organizationally valued technical skills,particularly emerging or underdeveloped IS skillsets (e.g., object-oriented development,client-server applications, project management). Another set of firms regarded their CoEsas an adhoc source of internal technological expertise: experts on specific IS topics were

10

designated as organizational "gurus" and responded to problem-solving requests from theentire organizational IS community. These groups also could play key knowledgedissemination roles, often using IT-based tools for communication across physicalboundaries.

Similar to IT standing teams, task forces are ad hoc groups of managers created to addressa specific task that requires cross-unit coordination for a limited period of time, such asproblem-solving related to a new technology, enterprise-wide applications of technologies,or enhancement of specific processes or applications. They differ from project teams forsystems development projects in that they are typically chartered by a council to developrecommendations on specific IT issues.

Design challenges. Several challenges for designing group mechanisms were identified.First, our respondents identified some of the traditional challenges with managing groups,such as identifying appropriate group members, choosing the basis of group decision-making (consensus vs. majority vote), promoting executive participation, and energizinggroup members with a sense of purpose and value to the organization (Goodman andAssociates, 1986). Another key challenge related to how to ensure communication ofgroup decisions and their acceptance by relevant organizational constituencies. Finally,some respondents pointed to the necessity of monitoring the continuing relevance of agiven group over time. For example, one CIO reported that he had recently disbanded anexecutive council since not many strategic issues were rising to the attention of thiscouncil: as business divisions in this company grew more differentiated, enterprise-levelissues decreased and divisional information officers were capably handling strategicalignment issues at the division level. IS issues that deserved enterprise-level discussionwere instead now being addressed in the CIO's IT management council, which included theDIOs.

ProcessesProcesses are formal organizational routines that achieve coordination through mandated ornegotiated behaviors by individuals in interdependent units. Fourteen firms mentioned aprocess mechanism in use. However, in retrospect, we think that this number may havebeen influenced by our methodology: that is, we collected data by telephone interview, andwe believe it might be more difficult to describe a process than a formal role or groupmechanism in this medium.

We have grouped our findings into three different process types (see Table 5): strategicplanning (including IT capital investments), chargeback and service level agreements(SLAs), and software acquisition processes. Similar to the IS literature, strategic planningwas the process mentioned most frequently; the nine mentions were from companies in sixdifferent industries. In centralized governance modes, corporate IS typically initiated theplanning process with an assessment of strategic thrusts at the enterprise level. Indecentralized governance modes, planning initially originated within the business units andcorporate IS typically played a review role in order to identify overlapping initiatives andcoordinate them at the enterprise level.

11

Four firms reported the use of chargeback and/or service level agreement processesbetween IS and the business unit in centralized or federal contexts. An IS controller orother senior IS manager was typically given the responsibility for overseeing theseprocesses.

Six organizations representing five different industries mentioned software acquisition anddevelopment processes as significant coordination mechanisms: specific methodologies toguide actions by business and IS managers to evaluate packaged software or deliver in-house applications. These processes were emphasized in organizations experiencingsignificant usage of packaged software or increased pressures for application delivery.

Design challenges. Two design challenges were noteworthy. The first involved designingfeedback to those responsible for overseeing a process. For example, one IT capitalinvestment approval process was designed to link with executive council decision-making:a post-implementation review date was set for each project, at which the businesschampion for the project was expected to identify factors associated with any scheduledelays or cost overruns as well as the realized business benefit(s). Similar to groupmechanisms, a related design challenge is the continual assessment of mechanismeffectiveness and appropriateness. For example, several companies reported a concernwith ensuring that a given process continued to direct attention to cross-unit coordinationand did not degenerate into a "ritual."

Informal Relationship-buildingInformal relationship-building mechanisms are intentional activities or practices that linkmanagers from two or more organizational units to facilitate voluntary interactions forcross-unit problem-solving. Three different types of informal relationship-buildingmechanisms were among the coordination mechanisms mentioned by our respondents:one-on-one management contacts, periodic briefings or conferences, and physicalco-location (see Table 5).

One-on-one contacts are scheduled and unscheduled interactions initiated by senior ISexecutives, most commonly with business peers or top managers but also with IS headsreporting to business managers. Our participants reported using one-on-one contacts togain senior executive buy-in to key IT initiatives as well as to increase IT-relatedknowledge and awareness of senior business executives, sometimes with the CIO acting asa "personal coach." Some CIOs reported using this mechanism as a substitute for anexecutive council in companies where the prevailing organizational culture precluded theuse of group mechanisms to link with high-level business managers. No pattern bygovernance mode was apparent, but one pattern by industry type was observed: four of thefive healthcare organizations in our sample mentioned this mechanism. Our findings thussuggest that the importance of one-on-one contacts for linking interorganizationalstakeholders may be highly influenced by context factors, such as the number ofprofessional employees in the organization and the overall organizational culture.

Periodic briefings and conferences entail structured information-sharing, but are alsooccasions for facilitating informal networking. Briefings occur in small group settings.

12

Conferences are highly interactive settings for larger groups, often with outside speakers orvendors as presenters. All the five companies that mentioned this mechanism as highlyimportant had governance designs that were not highly centralized.

In contrast, the CIOs that reported the use of a co-location mechanism were primarily incompanies with centralized governance. Physically locating an IS manager or unit in closeproximity to a business client was a mechanism that was implemented with and without adual reporting structure for the individual integrator or unit head. Some CIOs reported thattheir co-located individuals acted as if they had a reporting relationship to businessmanagement, even though they formally reported to an IS manager only.

Design challenges. Our interviews surfaced two design challenges specific to theimplementation of informal relationship-building mechanisms. The first is related to thevoluntariness associated with these mechanisms: their usage is dependent on thewillingness of the participants to engage in voluntary cross-unit interactions. In thoseorganizations where informal interactions were a key characteristic of the culture,participation was assured-i.e., cross-unit decision-making was conducted via one-on-onecontacts at the executive level in these organizations in general. The second challenge wasdetermining circumstances under which to continue to rely on informal mechanisms versusinstitute a formal group, role, or process mechanism. For example, in one organizationwhere informal relationship-building mechanisms were reported to be of great importance,the CIO mentioned plans for introducing an executive council and account manager rolesafter there was a greater appreciation of IT coordination needs on the part of businessmanagement.

Human Resource PracticesWe defined human resource practices as actions related to the human resource managementof IS personnel that are also intentional activities or practices to facilitate voluntary cross-unit problem-solving across two or more individuals reporting to different organizationalunits. Our participants mentioned three types of practices that we classified in thiscategory: (i) training and development initiatives specifically directed at increasing cross-unit linkages, (ii) job rotations of a temporary nature that were implemented in order toprovide cross-unit experiences to improve IT stakeholder linkages, and (iii) rewards andappraisal systems designed to incent cross-unit collaboration (see Table 5). No pattern bygovernance mode was apparent. One CIO reported delegating the "championship" ofspecific HR-related initiatives to his corporate IS directors in order to ensure that HRmanagement tasks became a priority.

Eight organizations mentioned training and development mechanisms. A variety oftraining approaches to "grow" IS managers' soft skills, especially in conjunction withintegrator role positions, and to "stretch" the technical skills of business personnel werereported.

Four companies mentioned temporary job rotations for IS personnel that were shortassignments, typically within a business division, to increase an individual's businessknowledge or sensitivity to special unit needs. For example, a CIO with centralized

13

governance reported temporarily assigning IS personnel to do "real business work" for afew weeks in order to enhance their business knowledge and learn the department's "hotbuttons." A CIO with federal IS governance reported using 6-month assignments in an ISunit within manufacturing plants for corporate IS personnel in order to increase theirunderstanding about differences in local versus enterprise-wide IS objectives or priorities.Redesigned appraisal systems and special rewards for cross-unit collaboration werereported by four companies. For example, one company had established discretionaryrewards to recognize contributions to a unit for which the employee had no direct reportingrelationship.

Design challenges. The primary challenge in implementing human resource practices ascoordination mechanisms is that the implementation of such practices typically requireseither formal approval by the corporate HR department or an executive-level agreementabout the IS department's freedom to "work around" institutionalized policies andprocedures. Some CIOs reported having corporate HR buy-in to their implementation of anew HR mechanism as a pilot initiative for the overall organization.

IT-based systemsIT-based systems were mentioned as an important mechanism for linking intra-organizational stakeholders by only eight firms. Most of the examples involved using IT-based systems to link IS personnel in different organizational units. Examples included an"ask the CIO" bulletin board, videoconferencing among dispersed IS managers, databaseson technology topics designed to pool together "thinly spread" knowledge or expertise, andother collaborative groupware applications. Electronic media were especially useful fordisseminating time-sensitive information related to technical problems or solutions. Nopattern by governance mode or industry was apparent.

Design challenges. Consistent with the knowledge management literature (Quinn,Anderson, and Finkelstein, 1996), we found that one of the major challenges in the use ofIT-based systems for cross-unit collaboration is related to motivating individuals to becomeactive contributors, not just users of information provided by others. For example, oneorganization mentioned experimenting with monetary incentives to motivate contributionsto a Lotus Notes discussion database.

FINDINGS: COORDINATION GOALSOur second research question for this study is: What coordination goals are CIOs seeking toachieve through the use of these mechanisms? Whereas our data collection on importantcoordination mechanisms involved the use of an initial schema, our data collection oncoordination goals was a "blank slate" approach: the participants were not provided withany a priori categories or variables. As described in Appendix A, our analysis of theresponses to our open-ended inquiry initially yielded nine types of coordination goals.These were subsequently clustered, via an iterative process, into a schema of threecategories of coordination goals: Strategic Alignment, Partnering, and Learning.

Table 6 presents our emergent schema along with a sample of CIO responses for each ofthe specific goals. In the following subsections we first briefly present our emergent

14

schema along with references to contemporary literature to ground our findings. Next weshare our findings on types of goals by governance form (see Table 7). Finally, we closethis section with a discussion of our findings on the associations between mechanism usageand our emergent three-goal schema (see Table 8).

Emergent Schema of Three Coordination GoalsStrategic Alignment refers to a synchronization of current business strategy and emergentstrategic thrusts with IT strategy and competencies (Henderson and Venkatraman, 1992;Rockart, et al., 1996). Alignment is achieved when key IT investments are directed at thefirm's strategic priorities and the current abilities of the IT infrastructure proactively shapethe emerging business strategy. Successful firms therefore not only design IT capabilitiesto support current business strategies, but also to implement new business strategies basedon emerging and proven technologies. Thirty-four of the 38 firms emphasized the use ofcoordination mechanisms to accomplish three specific Strategic Alignment goals (seeTable 6): (i) fusion of IT and business strategies, goals, and resources, (ii) market valuethrough IT, and (iii) business ownership of IT and endorsement of IT policies.

Partnering between business and IS is a goal that requires long-term commitment, mutualcooperation, and both risk-sharing and benefit-sharing on the part of each stakeholdergroup (Henderson, 1990). According to Henderson, partnership relationships are sustainedwhen business executives understand the value created by the IS function and havedeveloped trust with their executive counterparts. Ross et al. (1996) view such a"relationship asset" as one of three IT assets requisite for the effective use of IT insustaining competitive advantage. Twenty-six of the 38 firms associated the use ofcoordination mechanisms with achieving four specific partnering goals: (i) ongoing, two-way communication among business and IS management, (ii) trust, credibility, andresponsiveness of the IS function, (iii) valuing standards for the IT infrastructure, and (iv)an understanding of, and satisfaction with, the value of overall IT investments.

Learning reflects the need for people-to-people interactions to leverage the organization'sIT and business know-how. People-to-people knowledge sharing is required for knowledgecreation and sense-making (Nonaka and Takeuchi, 1995). As businesses experiencepressures for globalization, customization, and cycle time reduction, locating andintegrating webs of dispersed expertise within an organization becomes an imperative(Fruin, 1997; Grant, 1996; Leonard-Barton, 1995). As the range of valued IT skills andcosts associated with their acquisition rise, the pooling of IT expertise across organizationalboundaries in order to transfer ideas, best practices, and IT solutions also becomes animperative. The CIOs in our study indicated the following goals as being important in thecontext of learning: (i) leveraging IT expertise, (ii) promoting a collaborative workenvironment between IS and business, and (iii) promoting a collaborative workenvironment among IS professionals. Twenty-six of the 38 organizations also associatedthe use of coordination mechanisms to achieve specific learning goals.

Goal Findings by Governance FormTable 7 plots our findings on specific coordination goals by IS governance form of therespondent organizations. CIOs in all governance modes mentioned coordination goals of

15

all three categories. However, two of the most significant differences provide support forthe theoretical notion that coordination mechanisms are used by CIOs for linking acrossorganizational boundaries that are established or reinforced by an IS governance design(Brown, 1998).

First, IS managers under a centralized governance design are already linked with each othervia their reporting relationships, but a centralized structure provides no formal linkagesbetween IS and business management. Not unexpectedly, then, CIOs with a centralizedgovernance mode mentioned one specific strategic alignment goal almost twice as often asany other specific goal: fusion of IT and business strategies, goals, and resources. Second,federal and decentralized governance designs formally align business and IS management,but erect organizational boundaries between divisional IS staff and corporate IS staff. Notunexpectedly, then, all but one of the CIOs organizations with a federal or decentralizedgovernance mode mentioned learning goals, especially the specific goal we labeledleveraging of IT expertise. In contrast, the leveraging of IT expertise goal was mentionedby only two CIOs among the centralized governance firms.

Goal Findings by Coordination Mechanism TypeTable 8 presents our findings on associations between the three overall coordination goalsand the specific mechanism types discussed above and presented in Tables 3, 4 and 5.Shading is used in Table 8 to emphasize patterns where more than half of the instances of aspecific mechanism type were associated with an overall coordination goal.

As can be seen from Table 8, all five types of integrator role mechanisms were associatedwith achieving strategic alignment goals. Further, the three group mechanisms that linksenior business managers with IS leaders (executive councils, divisional steering councils,and project steering teams) were also predominantly associated with strategic alignmentgoals. Not unexpectedly, all nine organizations that mentioned a strategic planning processassociated this mechanism with strategic alignment goals as well. Only one "informal"type of mechanism was associated with this goal cluster: periodic briefings or conferences.

In contrast, only one type of coordination mechanism was strongly associated withpartnering goals: the IT management council. Recall that this type of council, which wasmentioned by two-thirds of the organizations with a federal governance context, links theCIO and other corporate IS managers with senior IS managers reporting to and/or co-located in a business unit. This mechanism is therefore considered to be critical forachieving corporate IS-business partnering, including goals related to valuing ITinfrastructure standards (see Table 7). One half of the instances of two informalrelationship-building mechanisms were also associated with partnering goals: 1-on-Icontacts and physical co-location. Both of these mechanisms foster ongoing two-waycommunication and the building of trust across people in different units.

Looking at the learning goals, all instances of IT-based system mechanisms were associatedwith this goal cluster, which included specific leveraging and collaboration goals. Twogroup mechanisms that typically link IS managers for cross-unit initiatives--IT standingteams and task forces--were also associated with learning goals. Although the total

16

numbers are small, three of the four instances of temporary job rotation mechanisms werealso associated with learning.

FINDINGS: DESIGNING A PORTFOLIO OF COORDINATION MECHANISMS

In the section on findings for specific coordination mechanisms we reported the keychallenges for designing specific mechanisms reported by-our respondents. In addition, ourinterviews surfaced some key insights related to an organization's entire portfolio ofcoordination mechanisms. We present these insights in two parts below.

Inter-linking Logic for a Mechanism PortfolioOur interviews uncovered three distinct explanations for why CIOs select and design specificmechanisms that extend beyond associations with coordination goals. We have named thesemirror-image, continuous improvement, and visionary. Each approach is described below,along with a case example.

Mirror-image logic describes an approach where the mechanism portfolio mirrors the firm'sinternal practices; this logic taps into the milieu of the firm's management practices andcoordination mechanisms in use for the overall business. This approach is found inorganizations where, for example, mechanisms that enable team-oriented management mayalready be well accepted within the organization. CIOs draw upon the existing institutionallegitimacy of specific mechanisms to create a portfolio of coordination mechanisms, in orderto minimize the difficulties associated with implementing new types of formal mechanisms.

For example, business success through continuous and rapid product innovation in one of ourparticipating firms was internally attributed to the use of multidisciplinary teams with eitheroften overlapping or exclusive responsibilities. The firm's portfolio of coordinationmechanisms tapped into the company's recognized success with the use of groups with cross-unit membership: the firm uses an executive council, divisional steering councils, projectsteering teams, and a variety of IT standing teams, mostly with overlapping roles andresponsibilities, for achieving goals in all three categories described above.

Continuous improvement logic describes an approach where the portfolio is periodicallyassessed; individual mechanisms are either retained, modified, or abandoned, depending uponassessments of their value-adding contributions; and new mechanisms are added as the CIOsand other senior executives learn about them. This logic is suitable in firms where a spirit ofcontinuous improvement and measurement exists, or where the firm has had a history ofsuccess with a portfolio of coordination mechanisms and anticipates only incrementaladjustments in the near term.

For example, in one organization that has been externally recognized for success with strategicIT applications, an executive council was associated with achieving fusion of between IT andbusiness strategies. However, while this council was able to surface ideas for high-levelstrategic opportunities, difficulties surfaced in moving them forward to implementation at thetactical level. Therefore, the CIO changed the mechanism portfolio in two ways. First, aproject steering team mechanism was introduced to ensure continued business management

17

oversight of specific initiatives. Second, systems designers were invited to participate as"backbenchers" in the executive council meetings to increase their knowledge about emergingstrategic thrusts.

Visionary logic describes the design and implementation of a portfolio of coordinationmechanisms as part of a fundamental IS organizational transformation. This approach isfound in organizations where CIOs are repositioning their IS organizations in anticipation of,or concurrent with, discontinuous changes in their business or IT environments.

For example, at an organization facing dramatic regulatory changes for its industry, the CIOanticipated the need for a change-ready IS organization to facilitate the organizational agilityrequired by the company in the future. One large systems development unit served as a pilotfor transforming a staff with mainframe legacy support skills to one with new client-serverskills that could deliver applications within short cycle times. An entirely new portfolio ofcoordination mechanisms was designed, including skills-based IT standing teams (centers ofexcellence), account manager roles, a new resource planning process, and new HR practicesfor job moves and reward systems. The entire portfolio was implemented by the CIO with a"jump in" approach; in the words of the senior IS executive, "you must trust in your visionand drive that vision forward."

Portfolios of Formal and Informal Coordination MechanismsRoles, groups, and processes are all formal mechanisms that have received the mostattention in the prior literature (e.g., Mohrman, 1994; Galbraith, 1993). Our findings inTable 8 suggest that many specific formal mechanisms, including all five types ofintegrator roles, three group mechanisms that particularly link senior business managerswith IS leaders, and the strategic planning process, are predominantly associated with thecoordination goal of strategic alignment. In contrast, the specific types of mechanisms inthe remaining three "informal" categories were more likely to be associated with learningor partnering objectives

Figure 1 captures these ideas through what we have called an "iceberg" metaphor.Integrator roles, groups, and processes are the more visible mechanisms for achieving theprimary goal of strategic IT alignment. However, informal relationship-building, humanresource practices, and IT-based systems exist "below the waterline" and are more likely tobe used to achieve the less obvious (and more difficult to measure) goals of partnership andlearning. In many instances these informal mechanisms may serve as supplements within aportfolio that includes formal roles, groups, and processes. For example, physical co-location and cross-unit human resource appraisal processes can be used to help achieve thegoals associated with an account manager role. Similarly, IT-based systems can be used tonurture the coordination behaviors of members of an IT management council or IT standingteam by providing a supplementary means of communication and information exchange.However, in other contexts these below-the-waterline mechanisms may be used assubstitutes for the use of formal mechanisms, perhaps even "paving the way" for theintroduction of more formal mechanisms in the future.

One of our executive sponsors suggested a tree metaphor for conveying the same idea:

18

informal relationship-building, human resource practices, and IT-based systems could beviewed as the roots, whereas roles, groups, and processes could be viewed as the more visiblelimbs of the tree. Whatever the metaphor, Figure 1 emphasizes the potential utility of all sixcategories within a portfolio of coordination mechanisms for linking intra-organizationalstakeholders.

DISCUSSION OF FINDINGS

The landscape for IT use and governance has changed significantly in recent years: IT hasemerged as a strategic differentiator in many contemporary firms' competitive strategies and avariety of IS governance modes have emerged in response to these new competitive pressures.This study began with an assumption that coordination mechanisms are an important

organization design solution for linking intra-organizational stakeholders. Our objective wasto extend the existing knowledge by: (i) expanding an initial schema of coordinationmechanisms that builds on prior literature, (ii) developing an initial schema of coordinationgoals associated with these mechanisms, and (iii) identifying some patterns of usage,including insights into the logic underlying the design of portfolios of coordinationmechanisms.

Our findings appear to provide extensive evidence for the utility of coordination mechanismsfor linking IS stakeholders within a variety of organizational contexts. As suggested byorganization theorists, multiple categories of formal and informal mechanisms can be used tolink people for cross-unit coordination tasks. Our findings also suggest that CIOs are usingthese mechanisms to achieve three coordination goals: strategic alignment, partnering, andlearning. The face validity of our emergent schema of three coordination goals for today's ISorganizations also appears to be supported by prior IS literature (e.g., see Boynton, Zmud, andJacobs, 1994; Cooprider and Victor, 1993; Henderson, 1990; Rockart, et al., 1996; Ross, etal., 1996; Weill and Broadbent, 1997). Further, our findings support prior literature whichsuggests that coordination mechanisms are used by CIOs to link corporate IS managers withboth business stakeholders and dispersed IS stakeholders, under different IS governance forms(Brown and Ross, 1996; Brown, 1998).

Our study also revealed that several organizational context factors other than IS governanceform can influence a CIO's selection and design of specific mechanisms. Figure 2 graphicallyportrays our contextual findings. First, as shown in Figure 2, the organization's cultureinfluences IT coordination mechanism selection. In some organizations certain mechanismshave already been institutionally legitimized. In other organizations, the prevailing cultureprecludes the use of a specific mechanism. Second, other governance designs within theoverall organization, such as the CIO's membership on a company's top management team,may influence mechanism selection. Both of these findings appear to support the conclusionsof prior researchers about important factors for predicting the CIO's ability to buildrelationships with their business peers (Armstrong and Sambamurthy, 1998; Brown, 1998;Earl and Feeny, 1994).

Third, in some organizations, the introduction of new coordination mechanisms is stronglyassociated with initiatives to respond to major changes in business imperatives. For example,

19

our respondents mentioned implementing mechanisms to develop fundamentally new ISdelivery capabilities or to enable the development of new business competencies.

Before discussing additional implications for research and practice, an assessment of thestudy's limitations is useful for interpreting its contributions. First, as stated earlier, wefocussed our attention upon capturing the CIO's perspective. Although this ensured that wecaptured the perspective of the chief architect of an organization's coordination mechanisms(Earl, 1996; Sambamurthy and Zmud, 1996), our study did not directly tap into viewpointsoutside of corporate IS. Second, in order to capture data from a large number of CIOs, weused a telephone interview data collection method and asked the participants to focus on thediscussion of three to five important IT coordination mechanisms. Therefore, while our studyhas provided empirical evidence for a variety of mechanisms that our respondents consideredto be especially important for nurturing their desired IT coordination goals, we did not attemptto collect data on the full range of mechanisms in their portfolio. Third, our work isdescriptive rather than evaluative: no attempt was made to systematically capture effectivenessmeasures for either the specific mechanisms or IS organizational performance.

Implications for PracticeThis study was supported by sponsors interested in identifying best practices for linking ISstakeholders. Our model in Figure 2 supports the notion that the answer is not a universalone: mechanisms for linking stakeholders should be selected and designed with an attention toboth overall organizational and IS contexts. However, the 21 types of specific mechanismsidentified in this study for the six a priori mechanism categories can serve as a catalog ofpotential mechanisms. CIOs can use our schema of mechanisms and coordination goals, alongwith the three inter-linking logic examples that we have described (mirror-image, continuousimprovement, and visionary), to re-examine their own company's portfolio of IT coordinationmechanisms and coordination goals as well as to make future mechanism decisions.

In addition, although specific mechanisms from all categories were considered significant inorganizations with different governance designs, some specific types of coordinationmechanisms within our six categories were associated with particular governance modes. Forexample, account managers were the most common integrator role in centralized contexts;divisional information officers most commonly played integrator roles in federal contexts.Some mechanisms were also observed to be more strongly associated with a particularcoordination goal. For example, all integrator roles were strongly associated with strategicalignment goals; all IT-based system examples were associated with learning goals; and themajority of IT management councils were implemented in federal contexts to achievepartnering goals.

Our study also yielded several insights about addressing specific design challenges associatedwith different mechanism categories. Table 9 summarizes our findings on mechanism designchallenges in the form of questions to consider for each mechanism category. Along with theother tables in this paper, this listing can be used by senior IS managers as a roadmap forbuilding linkages with top management, business management, and other IS stakeholders.Our final implication for practitioners is based on not only this individual study's findings, butalso the authors' research on related IT organization design topics over the past decade: our

20

coordination mechanism schema can be useful for designing a new kind of value-adding ISorganization. For example, this schema could be used to analyze IT transformation initiativesranging from a centers of excellence design for a systems development unit (Clark, 1997) tothe design of an IS function heavily outsourced to a customized set of vendors (Cross, Earland Sampler, 1997).

Implications for ResearchersWe began this paper with a brief discussion of organization interdependencies. Crowston(1997) has argued that the utility of coordination theory lies in its ability to facilitate anunderstanding of how an organization's task interdependencies could be managed throughalternative mechanisms. Three types of intra-organizational interdependencies have beenidentified by Malone and Crowston (1994): resource allocation, producer/consumer, and task-to-subtask. The primary coordination goal that emerged in this study, strategic alignment, canbe argued to be a goal most associated with a producer/consumer type of interdependency.However, the secondary coordination goals found in this study, partnering and learning,appear to be less associated with achieving a particular task interdependency than withestablishing an overall environment conducive to cross-unit problem-solving and innovation.Taken together, our goal findings suggest that the interdependencies associated with the ISstaff function involve complex cross-unit relationships and "sticky" management issues.Further, in order to achieve IT-based innovation, effective relationships are required not onlyacross IS and business personnel, but also across corporate and dispersed IS personnel. Futureresearch that specifically applies prior theoretical work on task interdependencies tomechanism usage at the senior management level, as well as at other IS managerial levels(Brown and Ross, 1996), may prove to be a fruitful extension of prior coordination theory tothe study of IT coordination mechanisms.

Figure 2 and our specific findings in Tables 3-8 also suggest an enlarged research agenda.Studies that ferret out factors associated with the effective use of each of the differentcategories of coordination mechanisms are required. These investigations need to includeresearch designs that gather data from multiple respondents (top management, linemanagement, and IS management) to enhance our understanding about the effectiveness andimpact of alternative coordination mechanisms on IT innovation and business performance.

Further, studies to examine the effectiveness of alternative portfolios of coordinationmechanisms under different organizational contexts are required. We have offered ourpreliminary insights on the inter-linking logic underlying a portfolio of coordinationmechanisms and have suggested some appropriate organizational contexts for theseapproaches. However, systematic investigations are required to provide answers to questionslike the following: What mechanism portfolio (selection of specific mechanisms, their inter-linking logic, and specific design characteristics) is associated with a given pattern of overallorganizational structure, business strategies, and IS governance arrangements?Configurational theories at the organization level could provide a useful theoreticalperspective for examining these research issues (e.g., Doty, Glick, and Huber, 1993).

ConclusionThe primary objective of this study was to refine our a priori schema of six types of

21

coordination mechanism for linking IT stakeholders from the CIO's perspective. Our findingson mechanism types, coordination goals, and their usage within organizations of different ISgovernance types have yielded a roadmap for practitioners. We have also shared insightsgleaned from more than three dozen CIOs about challenges encountered when designingspecific mechanisms, as well as some approaches to designing a portfolio of coordinationmechanisms.

Although much additional research is needed to develop a theory of IT coordinationeffectiveness, this descriptive study has also yielded empirical evidence and insights from top-level managers in primarily Fortune 500 companies that greatly expands our knowledge aboutan emerging research topic. The usage of coordination mechanisms for linkingintraorganizational stakeholders appears to be an IS capability well worth the attention ofpractitioners faced with demands for IT innovation within complex, highly dynamicorganizations and IS researchers interested in research for practice.

22

REFERENCES

Armstrong, C.P. and Sambamurthy, V., The influence of top management team and chiefinformation officer characteristics on business use of information technologies, Working Paper,Florida State University, 1998.

Beath, C.M., Supporting the information technology champion, MIS Quarterly, 15(4), 1991: 355-374.

Boynton, A.C. and Zmud, R.W. Information technology planning in the 1990's: Directions for research, MISQuarterly, 11(1), March 1987: 59-71.

Boynton, A.C., Zmud, R.W., and Jacobs, G.C., The influence of IT management practice on IT use in largeorganizations, MIS Quarterly, Volume 18, Number 3, 1994: 299-318.

Brancheau, J.C., Janz, B.D., and Wetherbe, J.C., Key issues in information systems management: 1994-95 SIMDelphi results, MIS Quarterly, Volume 20, Number 2, June, 1996: 225-242.

Brown, C.V., Examining the emergence of hybrid IS governance solutions: Evidence from a single case site,Information Systems Research, Volume 8, Number 1, March 1997: 69-94.

Brown, C.V., Horizontal mechanisms under differing IS organization contexts, Working Paper # 303, Center forInformation Systems Research, MIT, 1998.

Brown, C.V. and Magill, S.L., Alignment of the IS functions with the enterprise: Towards a model ofantecedents, MIS Quarterly, 18 (4), December, 1994: 371-404.

Brown, C.V. and Magill, S.L., Reconceptualizing the context-design issue for the information systems function,Organization Science, Volume 9, Number 2, March-April, 1998: 176-194.

Brown, C.V. and Ross, J.W., The information systems balancing act: Building partnerships and infrastructure,Information Technology and People, Volume 9, Number I, 1996: 49-62.

Byrd, T.A., Sambamurthy, V., and Zmud, R.W., An examination of IT planning in a large, diversified publicorganization, Decision Sciences, Volume 26, Number 1, January/February, 1995: 49-73.

Cameron, K.S., A study of organizational effectiveness and its predictors, Management Science, Volume 32,Number 1, 1986: 87-112.

Clark, C.E., Cavanaugh, N.C., Brown, C.V., and Sambamurthy, V., Building change-readiness capabilities in theIS organization: Insights from the Bell Atlantic experience, MIS Quarterly, 21 (4), December, 1997: 425-455.

Cooprider, J.G., and Victor, K.M., The contribution of shared knowledge to IS group performance, in J.I.DeGross, R.P. Bostrom, and D. Robey (Eds.) Proceedings of the Fourteenth International Conference onInformation Systems, ACM Press, 1993: 285-300.

Cross, J., Earl, M., and Sampler, J., Transformation of the IT function at British Petroleum, MIS Quarterly,21(4), December, 1997: 401-424.

Crowston, K., A coordination theory approach to organizational process design, Organization Science, 8(2),March-April, 1997: 157-175.

Daft, R.L. and Weick, K.E., Toward a model of organizations as interpretation systems, Academy of ManagementReview, 1984, 9: 284-295.

23

D! LAveni, R.A., Hypercompetition: Managing the dynanmics of strategic maneuvering, New York, NY: The FreePress, 1994.

DeSanctis, G.L., and Jackson, B., Coordination of information technology management: Team-based structuresand computer-based communication systems, Journal of Management Information Systems, Volume 10, Number4, 1994: 85-110.

Doll, W., and Torkzadeh, G., The relationship of MIS steering committees to size of fimn and formalization ofMIS planning, Communications of the ACAM, 30 (11), November 1987: 972-978.

Doty, D.H., Glick, W.H., and Huber, G.P., Fit, equifinality, and organizational effectiveness: A test of twoconfigurational theories, Academy of Management Journal, Volume 36, Number 6, December, 1993: 1196-1250.

Drury, D.H., An evaluation of data processing steering cormnittees, MIS Quarterly, 8:4, December, 1984: 257-265.

Earl, M.J., Experiences in strategic information systems planning, MIS Quarterly, 17(1), 1993, 1-24.

Earl, M.J., Information management: The organizational dimension, New York, NY: Oxford University Press,1996.

Earl, M.J. and Feeny, D.F., Is your CIO adding value?, Sloan Management Review, Spring 1994: 11-20.

Feeny, D., and Wilcocks, L., Core IS capabilities for exploiting information technology, Sloan ManagementReview, Volume 39(3), Spring, 1998.

Florida, R. and Kenney, M., The breakthrough illusion. Corporate America's failure to move fom innovation tomass production, Basic Books, 1990.

Francett, B. 1990. Marketers dig for openings in a blizzard of raw data, Computerworld; October 15: 93-97

Fruin, M., Knowledge Works: Managing Intellectual Capital at Toshiba, New York, NY: Oxford UniversityPress, June, 1997.

Galbraith, J.R., Competing with flexible lateral organizations, Reading, MA: Addison-Wesley PublishingCompany, 1993.

Gambetta, D., Can we trust trust? In D. Gambetta (Ed.) Trust: Making and breaking cooperative relations, NewYork: Basil Blackwell, 1998: 213-237.

Goodman, P.S. and Associates, Designing effective work groups, San Francisco, CA: Jossey-Bass, 1986.

Goodman, P.S., Exchanging best practices through computer-aided systems, Academy of Management Executive,Volume X, Number 2, May, 1996:7-19.

R.M. Grant, Prospering in dynamically-competitive environments: Organizational capability as knowledgeintegration, Organization Science, 7(4), July-August, 1996: 375-387.

Henderson, J.C., Plugging into strategic partnerships: The critical IS connection, Sloan Management Review,Spring, 1990: 7-18.

Henderson, J. and Sifonis, J.J., The value of strategic IS planning: Understandign consistency, validity and ISmarkets, MIS Quarterly, 12(2), 1998: 187-202.

24

Henderson, J. and Venkatraman, N., Strategic alignment: A framework for strategic informationtechnology management, in T. Kochan and M. Useem (Eds.) Transforming Organizations, NewYork, NY: Oxford Press, 1992: 97-117.

Hildebrand, C., The PC payoff, CIO, June 15, 1994: 30-38.

Horowitz, A.S., Rotate!, Computerworld, November 4, 1996: 90.

Hufnagel, E.M. and Birnberg, J.G., Perceived chargeback system fairness in decentralized organizations: Anexamination of the issues, MIS Quarterly, 13:4 (December 1989), pp. 415-429.

Iacono, S.C., Subramani, M., and Henderson, J.C., Entrepreneur or intermediary: The nature of the relationshipmanagerli s job, in J.I. DeGross, G. Ariav, C. Beath, R. Hoyer, and C. Kemerer (Eds.) Proceedings of theSixteenth International Conference on Information Systems, ACM Press, 1995: 289-300.

Kayworth, T. and Sambamurthy, V., Facilitating localized exploitation and globalized integration in the use of ITinfrastructures: The role of IT infrastructure standards, Florida State University Working Paper, 1998.

Keen, P.G.W., Shaping the Future: Business Design through Information Technology, Boston, MA: HarvardBusiness School Press, 1991.

Lasher, D.R., Ives, B., and Jarvenpaa, S.L., USAA-IBM partnerships in information technology: Managing theimage project, MIS Quarterly, 15:4, December 1991: 551-566.

Lawler, E.E., From the ground up: Six principles for building the new logic corporation, San Francisco, CA:Jossey-Bass, 1996.

Leitheiser, R.L., and Wetherbe, J.C., Service support levels: An organizational approach to end-user computing,MIS Quarterly, 10: 4, December 1986: 337-350.

Leonard-Barton, D., Wellsprings of Knowledge: Building and Sustaining the Sources of Innovation, Boston, MA:Harvard Business School Press, 1995.

Mackenzie, K., Virtual Positions and Power, Management Science, 32(5), 1986, 622-642.

Malone, T. and Crowston, K., The interdisciplinary study of coordination, Computing Surveys, 26(1), 1994: 87-119.

Mankin, D., Cohen, S. and Bikson, T., (Eds.) Teams and Technology: Fulfilling the Promise of the NewOrganization, Boston, MA: Harvard Business School Press, 1996.

Mata, F.J., Fuerst, W.L., and Barney, J.B., Information technology and sustained competitive advantage: Aresource-based analysis, MIS Quarterly, 19(4), December, 1995: 487-506.

Mohrman, S.A., Integrating roles and structure in the lateral organization, in Galbraith, J.R., Lawler, E.E., andAssociates (Eds.), Organizing for the Future: The New Logic for Managing Complex Organizations, JosseyBass, 1993: 109-140.

Nadler, D.A., Gerstein, M.S., Shaw, R.B., and Associates, Organizational architecture: Designs for changingorganizations, San Francisco, CA: Jossey-Bass, 1992.

Nadler, D.A., Shaw, R.B., Walton, A.E., Discontinuous change: Leading organizational transformation, SanFrancisco, CA: Jossey Bass, 1995.

25

Nonaka, I. and Takeuchi, II., The Knowledge Creating Company: How Japanese Companies Cretae theDynamics of Innovation, New York, NY: Oxford University Press, 1995.

Pfeffer, J., Power in Organizations, Marshfield, MA: Pittman, 1981.

Quinn, R.E., and Cameron, K.S., Organizational life cycles and shifting criteria and effectiveness, ManagementScience, Volume 27, Number 1, 1983: 33-51.

Quinn, J., Anderson, P., and Finkelstein, S., Managing professional intellect: Making the most of the best,Harvard Business Review, 74(2), March-April, 1996: 71 -83.

Raghunathan, T.S., CEO participation on IS steering committees, Journal of Management Information Systems,8:4, Spring 1992: 83-96.

Rockart, J., The line takes the leadership: IS management in a wired society, Sloan Management Review,Summer 1988: 57-64.

Rockart, J.F., Earl, M.J., and Ross, J.W., Eight imperatives for the new IT organization, Sloan ManagementReview, 38: 1, Fall, 1996: 43-56.

Rockart, J.F. and Short, J.E., IT in the 1990s: Managing organizational interdependence, Sloan ManagementReview, 30(2), 1989.

Rockness, H. and Zmud, R.W., Information technology management: Evolving managerial roles. ResearchReport for Financial Executives Research Foundation, Morristown, NJ, 1989.

Ross, J.W., Beath, C.M., and Goodhue, D.L., Develop long-term competitiveness through IT assets, SloanManagement Review, 38: 1, 1996: 3 1-42.

Ross, J.W, Beath, C.M., and Vitale, M., The untapped potential of IT chargebacks, MIS Quarterly, in press.

Saaksjarvi, M., The roles and success of IS steering committees: An empirical evaluation, Proceedings of theSecond European Conference on IS, May 30-31, 1994.

Sambamurthy, V., Venkataraman, S., and DeSanctis, G., The design of information technology planning systemsfor varying organizational contexts, European Journal of Information Systems, 1993, 2(1): 23-35.

Sambamurthy, V. and Zmud, R.W., Information technology and innovation: Strategies for success, Morristown,NJ: Financial Executives Research Foundation, 1996.

Smith, K.G., Grimn, C.M., and Gannon, M.J., Dynamics of competitive strategy, Newbury Park, CA: Sage Press,1992.

Souder, W.E., Managing New Product Innovations, Lexington, MA: Lexington Books, 1987.

Tapscott, D. and Caston, A., Paradigm Shift. The New Promise of Information Technology, New York: McGraw-Hill, 1993.

Thompson, J.D., Organizations in Action, McGraw-Hill, New York, 1967.

Torkzadeh, G. and Xia, W., Managing telecommunications by steering committee, MIS Quarterly, 16:2, June1992: 187-199.

Violino, B., Gaining the upper hand, Information Week, 1995.

26

Von Simson, E.M., The 'centrally decentralized' IS organization, Harvard Business Review, July-August 1990:158-162.

Ward, J., Griffiths, P., Whitmore, P., Strategic Planning for Information Systems, New York: John Wiley andSons, NY, 1990.

Weill, P., Broadbent, M., and St. Clair, D., IT value and the role of IT infrastructure investments, in J.N. Luftman(Ed.) Competing in the Information Age. Strategic Alignment in Practice, New York, NY: Oxford UniversityPress, 1996: 361-384.

Williamson, M., Getting on the same page, CIO, September 15, 1993.

Zmud, R.W., Building relationships throughout the corporate entity, in J. Elam, M. Ginzberg, P. Keen, and R.W.Zmud (eds.) Transforming the IS Organization: The Mission, The Framewlork, The Transition, ICIT Press:Washington, DC, 1988.

Zmud, R.W., Boynton, A.C., and Jacobs, G.C., The information economy: A new perspective for effectiveinformation systems management, Data Base, 18: 1, Fall 1986: 17-23.

27

AppendixDetails of the Research Methodology

Our research began with the identification of a list of organizations that have been noted for their IT management and use in the tradepress (e.g., Computerworld, CIO) and peer reputation (based on our own personal contacts and informal conversations with the CIOmembers of our sponsoring organization). A one-page prospectus was then mailed to senior IS executives at about seventy-five targetedcompanies. Forty-four executives indicated their willingness to participate in the research via follow-up phonecall and telephoneinterviews were scheduled. Each telephone interview typically lasted about an hour; all interviews took place during the first six monthsof 1995.

While the authors divided the interview task, a common semi-structured interview guide was used. Each interview began with areaffirmation of the purpose of the study and our interest in mechanisms that link business and IS units, as well as any mechanisms thatlink central and dispersed IS units. Open-ended questions were used to elicit the "three to five most important mechanisms" forachieving a cooperative IT management climate, as well as the participating firm's "climate goals." To help ensure shared meaning,participants were referred to the definitions and examples provided in the prospectus. These included examples from our a priorischeme of six categories, as follows:

Coordination mechanisms are structures, processes, and roles that link people and activities across functional boundaries.Examples include integrator roles such as liaisons or account managers, cross-functional groups such as steering committees ortask forces, informal relationships such as physical co-location, management processes such as standards setting or planningsystems, and human resource practices such as incentive systems that foster cross-functional networking. We are also interested inhow IT-based systems such as e-mail, bulletin boards, and document databases are being utilized as coordination mechanisms.

While respondents were given the above examples as aids in understanding the scope and nature of our inquiry, we did not ask them togive examples of each of the above mechanisms at their organization. Instead, they were encouraged to describe coordinationmechanisms that, in their opinion, were most important at their organization for managing intra-organizational relationships.

The principal investigator who conducted the interview then mapped the responses obtained from each firm and shared thesetranscriptions with the other researcher. The mechanisms were categorized as roles, groups, and processes, informal relationship-building, human resources practices, and IT-based systems. Next, each researcher took a lead role for three of the six categories ofmechanisms to understand variations in the types of specific mechanisms, organizational context issues, and implementation challenges.These analyses were then passed iteratively between the researchers until agreement was reached. For analysis of coordination goals,

the specific qualitative responses were culled from all interview transcriptions, and then sorted into common groupings -- firstseparately, then jointly by the researchers -- until agreement was reached on the ten goals shown in Table 6. Subsequently, these tengoals were grouped together into three higher-order coordination goals after discussions among the researchers and a consultation of theIT management literature. Finally, the two researchers worked together to examine the transcripts and understand how specificcoordination mechanisms were being used to accomplish the coordination goals.

Our initial findings were presented to the CIO members of our sponsoring organization. Feedback from these executives led to somefiner-grained mechanism analyses. A guidebook of IT coordination mechanisms was then produced and shared with our sponsors aswell as our participants.

The results reported in this paper are based on a re-analysis of the data collected from 38 of the original 44 participants. Sixorganizations were excluded due to unique industry or IS unit characteristics that were judged to constrain generalizability; one firmwith heavily outsourced IS functions, four firms in he consulting industry, and one firm that had responded in terms of a single businessunit rather than from the enterprise level were not included in our final analysis.

Ifcr:c

u

min

r,!

M3h* l

u

h

u

r r

v

Ll

WcU Vrs·

t¢Z

mh

A

v

r

h

rJ

t

u

01

-1

-c

.4:

*

* .�

-� - ~~~~~~~~~~ � ---

a vl

03C'n3 o\ o d

E c

t_~~~~~~~t

\O c3 [" = = ;

ctCtcl

'3vj` O\

C13aE O X U] '= a

5- Ct O

ms wd C

It

C,,

_ti

cn0C=1

03

C.U

e5

*;;

Qc 6

V4 ;o

U U *z

oa

a 0coF r

O Cd

00

O = C,,,1z :- C3\C,= 3 0 o

= P4¢ m~ dO

o =1

_ .n01 CC7 mV)U U) S: C'

O00 0i t -C, 0 0 ¢m : C/) $ 3 Uc-i

I~k = aE:Q

"C

01,oV)

O

0 M, 0;<^ UO! C

00o

O=¢,

rA

E

'EC%3

Q

04.

5:: = OC..U,4-0

E

fj

z

4.

*0

o~~~~~~~~~,

E "

>C a)

.: _Eb

E %:uz -°

*c) cn Uct.: sv

O k t

o °t~ Q._ .

6) W c:

° o

O 0.E

0

_C: E =

O O

oco,.a =

o0Ct d; :

O ed

c)as X

0 1S td

-C oCt 3.4: E

C'sCdCH 5-: O

.- (z)aO

E O

0O1 O

r S©, D~9

7r;oa)

4--·

O

00

©U4-40O

o,.2

0c)U

(D;>UctV)U

-C.

5 5:

L Q

c © ,--tv *

v, w

,c*> Z

~;

0vl*s mc t

Vs q

() O

;^oue0

O

._(

a)

,-U

0cc- ~

( L.O .

.D _:

E I,,oE .2

U -0:

U E:

U >oC1 =EO:

.2

rAQ)

O

tU

c4)

I

-S-4 Cz

;-4- 4-

a)0

4a)

cn;

75)

41.)

3-

3 - u

a)

a)3-

.5

3-a)0�0Ina)

In

In

0

U

SS01.)a)

3-0

-4-4

a)

S

3-a)

a)a)

4

-5Ina)

SIn0a).)In

Ina)a)�o�In

a1)N

3

a3)C,-

tS

a)

14

a)3-

In

.In

.0

(3-

.5

a3)

(3-

-J

Cl

a1)N

C,

x

en11

4�n

.,.owo-6--zelC�o

.,..4'.4u

W-4--;.4

w et�-4 r-4= w--I� 4

0Mu

.1.44�:14ft

�b0

w�z

-5C',

a)

3-00�3-40UI-

a)

a)U

0

In

0In

SInIn

4>4-

�a)

Ua)a)>

0b�

In 1 �a) 3 --4-4 a)

a)a)

0InUa)

3- --4

1 � a)

Lr) 0�-4 In

C)

-i

(a)

a2).5In

10

Ina)

3-

a1)

In

lz

.C3-j0

0t

0U

b.C

lb

(t

a3)N

(a)

a)

In

a).5

In

-0

0

a)

0

S(0

4

0

0

3-

lb

b.C3

,Ib

a1)

-1ba3)

a1)

a)

0Ina)

3-

a)

a1)0

Ina1)

a2)

-

(a)

bC~

'-4

ct

C)

In

0

3-

03--

0

N

ttb.C

0

r.

0U

-0b.2

a)b.C3-lb

"C)

Z-

a3)

Inla)Ina)u03-

Ina)

.In7

0

-

(a)

-i

0-L

~0

U)a)3-

~n'-4

a)

o-w

C)

InIna)

1.0

-0,In

a)a)3-

a)

0

In

l

0v

-NIn,

'-q

Cl4

a)

C)

-W-

0

01-4-

gd

wwzftz16.4

Q)

0

U 5cn ;.4

0-4 0

o W*

;.4.4.1�nO

lll�r-4

In

0

l

b.C

0

-40

~0lb

~0a)5

C)

-N

ft

a)C,-

In

a)

a1)

0v)

In

0

0

-~

4

-i

0

3-

(t

C)

(a)

LI)

U1

a)4>4-

3-

In

>4-

S0U

�03-

Ina)InIna)

In

-z

a)

34-

.5

3-a)0�0Ina)

0�SoSU 3-a) '-4-

1 � a)

H>Cl 0

a)

N

C)U

C--

In

Inj

U;~

0

lbN

b.C

0

.C

0

Q)

3-

a)

a)lb

0

lb4--

a1)N

C)u

In

lb

0U

CZ

.Z

In

(3-

a)

"I

3-

4

-3'-34

a)N

0

In

u

(A

In

a)

z

-0

a1)

lb

Cr.

CZ

a3)

a)

Ina)a)'A

-.0(Aa)lb

a1)

3-

In-

v)-il

0U

lb~z0

Cl

l

N

C)u

In

Ina)

*0In

0S0

0

In

a)

0Ua)

flb

N'--4

iq4

C)

a1)

M4'

4-

In

-3-

CU

Ina3)

0)

~0

~0a)

0

C)

S

0

(t

'---4

-i

Q),a

I

U

3--

a3)

0ul

a)

N

3-

a)C,-

©0lZ3-0

0U

.5~0

3-jlb

cl

3-

(a)a)

C-,l

-

--

-t

U=

t- I-)

t3-

U,�M-0�0

(:tCQ

.4 U

UOj

Z .a ) ) C :

00 ci

4-1 ~ ~ -.4.J 4-4 Es 0 a

* - ~ 0 * .

Lf)

a)

S

0*-¢ O

0

2(-)

04-

0Ua1)

04-

004a1)

a1).

OM

4.

IbO

. 5

4.4

4-4

a) \a)W

.5O

*o o

°t Q 1

- a)°

cJO a)=

W; b.

0 0 0

CN

a)

a)l

0)

(JO42.m

;:

I-U (o-

a,)

4

- -04C)4.0

00404 a); Ww r.a

-_ *_

0 _._ 0

(1) lc

0.U

4.-U ¢

a)-'

U3)

U VW 0

40_--'-4 a)

X -a)

0

0 .AU I

.4a.a)

4--

U

a)CL

a

a,uu

a)S

C

* 0 0

CA

a)

U 0

- a)

0

4.-a)

0*-

.- a)- -4-

4.40

(JO -I-,

0004-ccr1 4-'

�

rJ'j(10 �-4

a)a)

(JO 4-

�00�4-

<8

a) t.-a)

UO

a) -

y .S-*-J3

0 iD

-ta

O - .

Cf) a)0< :L S 0

a) S

0 S

0

a).

* ¢

u

a)

0

U

0 W

oSi

- a)cn .

Lfl.

_-

a)

0

~0a) .-

n3 r .-

U 4-

:(1 tt(n0H*t-Ev bl

a)a)

o

(jz

._t3

04-

a)'-

Sa)a)

0 0 0

a)

U

0

*p..-o: . -

6 g

U4-4

C

4-.j

00

F-I*,

U,tt

a)'-4-0U)-

)

U)

a).4 O_ 4

E= 0

4

CZ

F Q t

a)r,U

a)

0C)U

a)4-

a)N-i

r. -a)Ua)r-

75

N

U

'tRC

- 1 I

-0

-.N

01)U-

o

0

OUW4- U

. (a

o

-- 4 4.

.4-404>

V~ r

U)

* ,*..

~.~ O2 t

- U

O a)

U 4

Za)

V 04

O(40a)

4.44

H)-) .-4

0 -

-

't

r--(

r--�

00

2 Y~a) a);

a °

.: ·t *; =(a;-4

a O a O

U) ._ 0X a)V) a) _ (U

(a

a) *m X X"" e S

a) Cl -m *

* 0 -

Cl

Lr)

0La)

a)X ox --LL1

a)

(a

(a

C

a

o .r

V1(a

S .o

a)

0

a1)

a)UE

a)

(a

C)a)

(a

-c

a

0

0 0

tO

C)

U:) r.> .zm8(--)

Xa)

O-0U)

C60

;(Aa)t~o

SE

U)

U)a)U)VIo0t4

.0

a)r.

a)U)

a)

a)

.

.

0(a

U(a0

.U UC)i. a

a

OU 4-.(a-e!4-4 a)U

0 0

r-

'IC

bj(e-I

a) -

4-r

o

T u

-U) 0 a

a): X

8 s

.S N 3 e ;-4 -X

00 0 a

Ct) 0v(

-U OU N

0

0U)

Cl)

a)

U)

c0

a)

(a

a)

0 0

(

,o U)U )

a)-

a

cc

cl

O)

E -a) "

HOr -nz V

a.;

(a

X ·..C) O

O O ti

tV 40S 5a

O a

) O

U 4-

- a

3t U

C

a)

E*

· ·

Cl

Cc)m

LO

t

r-4

E

a)

toCZ

H-4.

U)

U);-( ;�

(a(a.-

�0C)-) *�

Cl)0O�

- U)-� .55�U)

0a)

U) -�a)a)� .� a.; �

-� .� .� ��

�0 � ><

I 5H

U) � 0 > -

a) U Cl) .5 �-� � a)a) a) 0� 5

- 0 0

"t

L-CC)

U

0

l)

Hucn

ct

E-

U

.4a)

U(a4-

UU)

(a

o.,4

dz (a

4Fd

00u.-

I)

HO

taZI:$

c4

4-·

(abD0,Ut

U)ftkn

.)blcmna)

a)

a)

0

>F0a

a)ar.a)u

(tCl

P

3a)

UC

a)0< ,

a)

U3

a)-I IIC 1

a)o-H 11

U

a)a)

S

.4

0

0a)

U,o

a

0.4>

av4

a)O

(a

N

(aoa)

a

C-I.

a)

S540O

a

0i

UIa.0

0

-4a

Z

- � .

-- � �

0 P -Ca :j$:.,>1 ;-

E- U

.- ut

00

Li~

0p ItJ

001

MP 4

V0uV)

u(z

U

(a)U~XD

aN

a)'.

Uo)

-C

>w

C)

00

U

)

0 S

C)

ua)

(aC

bO bO00

X )

U)

C0

; 00L .t

0

Cl4

ON

u a-

Lna) ZN 00Ur zb O C C0 U*

CU ('~ C 5- Q 00

* a)

U. j'cq

a)C

-C ;

Qa)(a a)>

_ g

(a -CJVCCa)(a

.2 00 =

O 0U 00O

U V -2C CUe

ZICC

* -

-e

,t

-'-. - 4-C C)

ct 00 CL,

C-,. a()C u >

Ct) (a a) :

Cl4

Cr)

00

?C-4u

V

CL

a)

;-. Of)

50

(aC=

0

0

ClA

V ) r

0 u5

-E

C t)-

OCv o

0 (OU0U

- :(a XUga)_~ °~

>S 00':-= a)

0 ;

r-

CLO.

u CO

.

- n b4(t a

- C 1

I_a

(-C4

-Co.

a) ._0 0.O

00

710.2 uV)Q)> Ua4 )tcz (t;

CZ

.C

( >

O

,00

5 ~

(a)

00 u

a) O

- (a00r

a. 00>

N X )

Of) C; C

Cl.

Na.

H.94

C4;z

00

(a5

C-

0

Cl4

Cl

t a0U)

't e

a) 0Y

4.5b

.5

C5 a )

00 X

.X 0

bo a)

Ca)

UC°a);-

' a).2 r0 _)

Oz O

D ,O 02U U:CZS~ I

Cl4

Cl

a)

C

00

'T Vn

n V)

a)C

CoI00

a)S=Ca)

U

O

a)

Ct

(abO0

Ca)-C

'40

(aCa)

a)S

a)

C

S

-SC

CX00

Ct

(a

0

a)

z

0000a)

9

rC

0a1)a)C)

00

0C_O

U._

C0

15

z0(aC

-S

C~U.9

:0.

00CU

C(A2_

1O

a)04

V)00

a1)Na3)

(0C)(A00a)u0

O-

00a)u

u*.zCS'L

P4r--

TABLE 6Emergent Schema of IT Coordination Goals

Number ofCoordination Goals Mentions Specific Examples

Overall Coordination Goal: Strategic Alignment (n=34 *)

Fusion of IT and business 23 * Walk in-step with the businessstrategies, goals, and resources * Alignment with the business

Ensure meshing of IT and business

Market value through IT 16 * Enhance speed to market through IT* Create shareholder value through IT* Achieving organizational agility

through IT

Business ownership of IT and 9 * Joint process/project ownershipendorsement of IT policies * Business endorsement of IT policies

Greater championship of ITinitiatives

Overall Coordination Goal: Partnering (n=26)

Ongoing two-way communication 13 · Talk about dilemmas, differentamong business and IT problemsmanagement * Two-way communication

Advice about key events, boilingpoints

IS trust, credibility, and 12 * Mutual trust and partnershipResponsiveness * Enhanced credibility of IS

Building a customer service culturein IT

Value standards for IT infrastructure 9 * Valuing standardization of IT assetsand services

Gain standardization and it'sassociated benefits

Understanding of, and satisfaction 4 Prove that IT is creating corporatewith, the value of IT investments value

Better understanding of the returnon IT investments/IT value

Overall Coordination Goal: Learning (n=26)

Leveraging of IT expertise 15 * Become a learning organization* Nurture an IT-literate client base* Leverage global IT knowledge &

experience

Collaborative work environment 9 * Willingness of IS and business tobetween IT and business share ideas and information

* Collaborative attitudes

Collaborative work environment 4 * Building a global IS communityamong IT staff * Collaborative attitudes

* Numbers in parentheses reflect the total number of organizations that mentioned one or more specific goals of this category.

.4rA

oO

N C

¢ ¢

0

a),ZtN._

z '1

11CU0)

7;,r,N

(9

*0 t

r 11

2 n-In Cu

01)3N X4- IIl

U HV

r,

.4, .

S

o ra

o '

. r. mbb C r :

= E

C

C RII11

U0 )*a .- C)

U *>

r U:C) w

_

'U)4-4 U

;(J

C.

II

*ZU)

(: )

caa-U) >. -4- -(t Q)c

a, z .>3z 5

C]

C]

r.C)

za4-r.

*) -

4 II> (

a _

03 - aa EU (z

I =0It

rh(1

rirl

17

I Qu

Q

IQ

14P-4

CO

n

c"

N

E-

=t

,:0

a,9.-(3)

Eza)

u _4 x

co

C]

C]

L)

C)a

C).x

C)

to

C)

~-

CO4C]

u)U

-v)

aU)

Eb*-

U)

0r-qo

ClII

4b13z

C)4=..et

=L

Lt)r(

m

IIt9

Ebr3a

-4

C'-4

._~w4b1rz;-

V)

(iAV)C)

V)

-a

aU)::Ioor.

In4.1

._IacaU)14-0

C)

aU)I)._02)

U)p:En

11

E-_C)

r.:oWa

C)S

4-aC)

Q)

EvIlUEF-

H

4-

a,)

4-

11

CA

O

O1

(r0

O

O4Y-0I )rv

1

C)

4-~C)

4-

a.1 lr.C1)

4aC)(tL,

-C

(du

U)U)C)

-aFr.,C)C3)4.;z;'IC

r.

c-III

-4C;-4edwl

r.

. � . I� . _

r-Ir-I I

'IC

,---q

r--( \-O r--4

eq

--------- -- .-- .

©

I,!t©

0

oc,

LrO

0

0

0

©

0

0

00)0)©

Oi. 0

0L

O

U

O

o

ti

.4-I

aJ*.

.4 -ua

*P

*- .

Qa)Z (d(be-

S ·rU" =

*a ,

u

)

00U

r.

00llc08U

LI)

0o

o

b40)4-'rl

(A

0U$;Iok

C3

C)

.=

=Ca

o.,.

(a.

>~~

:z0~~S Oa); (a, t

o ,. .~ _ a) S) -C

*: ~ o a)~

~ = O °

~-'O o ob-OC.. ) a) C -

C-.

(a

A. out a)

G)

- m

u~ C

C a a)I- Q t

( (ao Coa)o

,. OOa) C-) u

a) -

b- Uo

a)~ (D

U.)

-a)_C ·O -C :. U a

* * 0

(Aa3)

uo

p_.

a)

(a

._

a;

u

o,-c

-CU

(ao.

*_ .a o

o:0aCONO

OC C

U)

~a)

'._

UU

Cav

0.

-Ci)Ua)

a)

a)..

a) .m C..

. ,,,a ,. U -

*- U

U C(a.

: .5 C

~ C ao -

·: : U 04 CI)

- -°

U. ~ :(.. ~

U ,. u 5

,5; .4_, .4 4

CJa a

O U U

U) D

-- a)

V) U)o -cn(i-,a).o o

m

CA u

cu 0 r .OU Oa) OQ)

Uc t CT,

wa C (a)* .

(aaoa S

.C.C) Uo~ -~~~-

o~-A S; aS

- ~bOu ~O O O

. C ; W * a; Q C-C @> -C s

a) o- -

* * * * ;

u

C a)~~~~~~~~~~~~~~~~~~~~~~~ g CA "IS= C.)

cn

.bOa)

CI)

-C

o

a)

©

o .= Ct

Co)a)oCU· eft ·(oa a)

CC C C~

fi C S $

C- ~O U a)

C - -~ -

uz

0ca

00.e;

0

i)*i)C.i5

0U.)r.

U)

-

et

V)

0-,.

u

o(a

Sa)

U

8

oaa)0.4CU)0-.5

a)

C

0.

0a)

Sa)

(a

b4

' .

;i a)

·n o

~oo -)U:

._ o *(a a °

Na) Ca)GiS

~ o CO4- aJ a)o o,.

o ~(a a ) >. m

3o °

o bo , c: S

° Ct

> *:n

U -t

C- U -*

o oCoX zSal)C2

2C

(a

ct

(nE(n

0(d4uai

unC�.;z0

u

LL�

- "''t r. ="E Iz ;;,(D L-, -I (1) 5z ;.., -0

-Cdl(O.) A,J) 5ct-0 2E�, En- >1(A