ray et al./it & the customer service process...ray et al./it & the customer service process...

Ray et al./IT & the Customer Service Process

MIS Quarterly Vol. 29 No. 4, pp. 625-652/December 2005 625

RESEARCH NOTE

INFORMATION TECHNOLOGY AND THEPERFORMANCE OF THE CUSTOMERSERVICE PROCESS: A RESOURCE-BASED ANALYSIS1

By: Gautam RayDepartment of Management Science and

Information SystemsMcCombs School of BusinessUniversity of Texas, AustinAustin, TX [email protected]

Waleed A. MuhannaDepartment of Accounting & Management

Information SystemsFisher College of BusinessOhio State UniversityColumbus, OH [email protected]

Jay B. BarneyDepartment of Management and

Human ResourcesFisher College of BusinessOhio State UniversityColumbus, OH [email protected]

1V. Sambamurthy was the accepting senior editor for thispaper. Anandhi Bharadwaj was the associate editor.Paul Tallon, Weidong Xia, and Sanjay Gosain served asreviewers.

Abstract

Delivering quality customer service has emergedas a strategic imperative, one that is increasinglytied to a firm’s information technology resourcesand capabilities. This paper presents an empiricalstudy that examines the extent to which IT impactscustomer service. More specifically, this studyinvestigates the differential effects of various ITresources and capabilities on the performance ofthe customer service process across firms thatcompete in the North American life and healthinsurance industry. The paper builds on (1) infor-mation systems work that suggests that the effectsof IT are best documented at the level of pro-cesses within a firm, (2) information systems workthat suggests that the performance effects of ITare likely to be contingent in nature, and (3) devel-opments in the resource-based view, whichdescribes the kinds of IT resources and capa-bilities that are likely to enable a process in onefirm to outperform the same process in competingfirms. The findings suggest that tacit, sociallycomplex, firm-specific resources explain variationin process performance across firms and that ITresources and capabilities without these attributesdo not. Of particular interest to IS scholars, it isfound that shared knowledge between IT andcustomer service units—an important driver of howIT is implemented and used in the customer ser-vice process—is a key IT capability that affects


626 MIS Quarterly Vol. 29 No. 4/December 2005

customer service process performance andmoderates the impacts of explicit IT resourcessuch as the generic information technologies usedin the process and IT spending, which—consistentwith resource-based predictions—were not foundto be directly and positively associated with rela-tive process performance. The implications of thefindings for research and practice are discussed.

Keywords: IT resources and capabilities, sharedknowledge, resource-based theory, business pro-cesses, process performance, business value of IT

Introduction

During the past two decades, customer servicehas emerged as a strategic imperative for mostfirms (Reichheld and Sasser 1990; Rust et al.2000; Treacy and Wierseman 1995), and servicequality has been the subject of considerableinterest among marketing academics and practi-tioners, spurred by the original work of Parasur-aman et al. (1985). Today, there is general agree-ment among marketing scholars that quality cus-tomer service is not only the most important factorfor achieving the paramount marketing outcome,namely customer satisfaction, but it is the principalcriterion for measuring the competitiveness of thecustomer service process ( Szymanski and Henard2001; Zeithaml 2000). At the same time, the in-creased emphasis on customer service hasemerged as a key driver for IS priorities, reflectingthe general recognition of the essential role ITplays in support of this process (El Sawy andBowles 1997). However, while a number of casestudies do highlight the critical role of IT in cus-tomer service (Elam and Morrison 1993; El Sawyand Bowles 1997), empirical research examiningthe link between IT and customer service perfor-mance has been lacking. This article fills this voidby drawing on resource-based theory (Barney1986, 1991; Rumelt 1984; Wernerfelt 1984) toinvestigate the differential effects of five differentIT resources and capabilities on the performanceof the customer service process across firms thatcompete in the North American life and healthinsurance industry.

In examining how IT impacts customer service, ourapproach is consistent with the process perspec-tive to the question of IT business value, sug-gesting that enterprise level impact of IT can bemeasured only through their intermediate (i.e.,process) level contributions (Barua et al. 1995;Mooney et al. 1995; Mukhopadhyay et al. 1997;Sambamurthy 2001; Tallon et al. 2000). The argu-ment here is that IT is deployed in support ofspecific activities and processes, and, therefore,the impact of IT should be assessed where thefirst-order effects are expected to be realized. Ourapproach is also consistent with a second streamof research that takes a contingency approach,suggesting the need to take into considerationother variables that may mediate or moderate thepayoff from IT investments (Markus and Soh 1993;Weill 1992), as well as organizational investmentscomplementary to IT (Barua et al. 1996; Bryn-jolfsson et al. 1998).

Consistent with resource-based theory, thefindings suggest that valuable, rare, and costly toimitate IT capabilities such as shared knowledge—the level of shared knowledge and common under-standing between IT and the customer servicemanager regarding how IT can be used to improvethe performance of the customer service pro-cess—is a key IT capability that affects relativecustomer service process performance. Thisresult extends earlier empirical findings that linkshared knowledge (at the firm level) with increasedlevels of IT use (Boynton et al. 1994), with in-creased operational and service performance ofthe IS group (Nelson and Cooprider 1996), andincreased IT assimilation in value-chain activitiesand business strategies (Armstrong and Samba-murthy 1999). We also find that the effects ofexplicit IT resources such as technical skills of ITlabor, generic information technologies, and ITspending are contingent on the level of the sharedknowledge. In other words, shared knowledge hasboth direct and moderating effects on customerservice process performance.

The rest of the paper is organized as follows. Inthe next section, the theoretical framework andhypotheses are developed. The method used totest the hypotheses is then presented. The data


MIS Quarterly Vol. 29 No. 4/December 2005 627

and data collection are described and the resultsof the data analysis presented. The paper con-cludes with a discussion of the findings as well asdirections for future research.

Theory and Hypotheses

Resource-Based Explanationsof Performance Differentials

Historically, resource-based theory was developedto understand the conditions under which firms areable to gain and sustain a competitive advantage(Amit and Schoemaker 1993; Barney 1986, 1991;Rumelt 1984; Wernerfelt 1984). However, morerecently it has begun to be applied to understandwhy the performance of processes within a firmmay vary across a set of competitors (Hendersonand Cockburn 1994; Miller and Shamsie 1996;Schroeder et al. 2002). Ray et al. (2004) sum-marize the arguments that, in some cases, usingresource-based theory to examine the economicimplications of resources and capabilities at thefirm level can lead to misleading conclusions, andthat process-level analysis may be more appro-priate.2

Differences in performance—whether they are atthe firm or process level—are explained inresource-based logic in terms of the types ofresources and capabilities that different firmscontrol.3 The value, rarity, and imitability of these

resources have been shown to be important(Barney 1991). Resources are valuable when theyenable firms to increase the efficiency oreffectiveness of processes compared to whatwould be the case if these resources were notexploited in these processes. That is, valuableresources increase the absolute level of perfor-mance of a process. However, just becauseexploiting resources can improve the absoluteperformance of a particular process in a firm doesnot mean that investing in these resources willimprove the relative performance of this processacross competing firms. Whether valuable re-sources explain variance in the performance of aprocess across competing firms depends on howrare and costly to imitate these resources are.

Resource-based theory suggests that resourcesthat are held by large numbers of competingfirms—that is, resources that are not rare—cannotexplain variance in the performance of a processacross competing firms. It also suggests that evenwhen these resources are held by just a few com-peting firms, if they are also not costly to imitate,they will rapidly diffuse among competitors. In acompetitively mature industry, such resources willalso not explain variance in the performance ofprocesses across competing firms. Only whenvaluable resources are rare and costly to imitatecan they explain variance in performance of pro-cesses across competing firms (Barney 1991). Aresource is likely to be costly to imitate in thepresence of isolating mechanisms such as pathdependence, causal ambiguity, social complexity,and team-embodied skills (Barney 1986; 1991;Dierickx and Cool 1989; Rumelt, 1984).

IT Resources and Capabilitiesand Process Performance

As suggested earlier, the purpose of this paper isto apply this general theoretical framework toexamine the relationship between IT resourcesand capabilities and the relative performance ofthe customer service process across a group ofcompeting firms. To accomplish this objective, itwas first necessary to choose an industry withinwhich to study it.

2The arguments are about the attribution and appro-priation of economic impacts of resources, and thegrowing consensus in the strategy literature thatbusiness processes are the basic unit of competitiveadvantage.

3A variety of labels (inputs, assets, capabilities, compe-tencies) have been used to describe a firm’s resourceendowments. The label is not critical in this context;what is important is identifying resources that are likelyto be sources of competitive advantage. Following Grant(1991), Amit and Schoemaker (1993), and Makadok(2001), in this paper the label resource is used in thegeneral sense to refer indistinctly to all of these con-cepts. The term capability is defined as a special type ofresource, encompassing a firm’s capacity to coordinateand deploy other resources to effect a desired end (Amitand Schoemaker 1993).



The insurance industry was chosen for threereasons.4 First, the financial services industry,because of the digital nature of its products andservices, has historically been among the largestinvestors in IT. One recent survey suggests thatinsurance companies on average, spend about 3.5percent of their gross premium on IT. This putsthe insurance industry in the top tier of industriesin terms of IT investment. Moreover, IT is a criticaltool for providing customer service personnel withthe information they need to deliver quality service(Elam and Morrison 1993). Second, in the highlycompetitive insurance industry, customer serviceis widely seen as being strategically important(Berry 1995; Griffith 1993). For example, theability of USAA (one of the largest insurancecarriers in the United States) to consistently pro-vide superior customer service is often credited forthe company’s stellar reputation and overallperformance (Garvin1995; Teal 1991). Finally,there is a high level of variance in the reportedability of firms in this industry to satisfy theircustomers (LOMA 1993), suggesting that firmsdiffer in their ability to execute this process.5

IT Resources and Capabilities and Their DirectImpact on the Customer Service Process

A wide variety of IT resources and capabilities arerelevant to the execution of the customer serviceprocess. An examination of the IS literature andinterviews with customer service and IT managersin insurance companies led to the identification oftwo general categories of resources that are asso-ciated with the planning, conception, implemen-tation, and use of IT applications. The first class of

these IT-related resources encompasses raw ITspending and two pure technology resources,namely, technical IT skills and generic informationtechnologies used in the customer serviceprocess. The second class of these resourcesincludes two capabilities that influence how thefirst class of resources is used: shared knowledge(knowledge IT managers have about the customerservice process linked with the knowledge thatcustomer service managers have about IT) and ITinfrastructure flexibility.6 Based on the review ofthe literature and the interviews, it was initiallyassumed that all of these five IT resources andcapabilities are valuable in the sense that theyhave the potential to improve the absoluteperformance of a customer service process. Thefocus of this research was to explore whetherthese resources and capabilities could explainvariance in performance of the customer serviceprocess across competing firms. We will examineeach next, drawing on resource-based logic.

Technical IT Skills. Technical IT skills refer togeneral, explicit skills (e.g., programming), pos-sessed by the firm’s IT staff that are needed todevelop IT applications. While these skills can bevery valuable, since they are widely available tofirms (either through hiring employees or con-sultants with these skills) they are usually not rareor costly to imitate, and thus such skills, by them-selves, are not likely to explain variance in theperformance of the customer service processacross competing firms. Moreover, as Mata et al.(1995) note, even when such skills are hetero-geneously distributed across firms, they aretypically mobile as it is not difficult for competitorsto hire away this value-creating resource from theircompetitors at their market price.

4In this industry, customer service is defined as activitiesthat involve episodes of interaction between customers(and agents acting on the behalf of customers) andcompany employees when customers make inquiries,request changes to a policy, or conduct financial trans-actions (LOMA 1993). Virtually all of the firms in thissample had a separate customer service process at thetime of the study.

5The competitive maturity of the North American life andhealth insurance industry also plays a role in the empi-rical analysis reported below.

6Of course, other IT resources besides these, such astop-management support or end-user environment, couldalso be relevant for customer service processes withinfirms. However, interviews and our analysis suggestedthat most of these other IT resources were either specialcases of, or closely related to, these resources. More-over, in order to bound the study and make data collec-tion feasible, it was decided to limit the analysis to thesefive resources.

adatta

Highlight

adatta

Highlight

adatta

Highlight



Generic Information Technologies. Genericinformation technologies refer to the set of well-known hardware and software technologies thatcan be purchased from outside suppliers. In thecontext of the customer service process, generictechnologies include scanning and imaging tech-nology, computer networks with agents andbrokers, Web-enabled customer interfaces, call-tracking and customer relationship managementsoftware, computer and telephony integration, andcustomer service expert systems. These technol-ogies are generic in the sense that they are avail-able from multiple suppliers for those who wish tosource them and are lacking in asset specificity inthe sense that, although aimed at the customerservice process, they are not specific for anyparticular firm. While much of this technologymust ultimately be integrated into a firm’s customerservice process, these applications and tech-nologies, per se, are available to all of the firms inthe insurance industry. Such technologies may bevaluable in an absolute sense: investing in thesetechnologies can enhance the performance of thecustomer service process compared to that pro-cess without these technologies. However, sincemost firms have access to the same technology,such generic technologies are, per se, unlikely toexplain variance in customer service performanceacross competing firms. Research on the diffusionof generic information technologies in differentindustries is consistent with these expectations.For example, Powell and Dent-Micallef (1997) intheir study of the retail industry found that genericinformation technologies like point-of-sale termi-nals and electronic data interchange with suppliersdid not explain variance in firm performance.

IT Spending. The level of raw dollar spending onIT is an important resource for the customerservice process. For example, it has already beensuggested that generic information technologiescan be valuable for the customer service process.Failure to invest in IT resources and capabilities,by sourcing them internally or externally, can put afirm at a competitive disadvantage in terms of theperformance of its customer service process. Forthis reason, firms in this industry have a strongincentive to invest in the IT assets necessary tomaintain a competitive level of service. While in agiven year a particular firm may face budget

constraints, over time, investing in IT has almostbecome a competitive necessity in this industry.7

As such, IT spending, per se, is not likely toexplain variance in customer service processperformance across firms in this industry.

Hitt and Brynjolfsson (1996) argue that to theextent that IT assets are equally available to all ofthe participants, in a competitive market all of thefirms will make optimal IT investments in equili-brium, and no firm will gain an advantage fromtheir spending per se. Of course, some firms mayhave access to capital at lower cost relative toothers. However, as Mata et al. (1995) note, it isthe special resources and capabilities of thesefirms that enable them to manage the technicaland market risks associated with IT investmentsmore efficiently, and not access to capital per sethat is the source of distinctive advantage. Inshort, there is no reason to expect that ITspending per se, will explain variation in customerservice process performance across competingfirms. The above observations lead to thefollowing hypotheses:

Hypothesis 1a: Technical IT skills, perse, will not explain variance in the perfor-mance of the customer service processacross firms in the North American insur-ance industry.

Hypothesis 1b: Generic informationtechnologies used to support the custo-mer service process, per se, will notexplain variance in the performance ofthe customer service process acrossfirms in the North American insuranceindustry.

Hypothesis 1c: The level of IT spend-ing, per se, will not explain variance inthe performance of the customer serviceprocess across firms in the NorthAmerican insurance industry.

7Indeed, commitments to IT spending in this industrytend to vary with the financial conditions of the industry:When profits are on the rise, IT spending across firms inthe industry tends to rise, and vice versa (LOMA 1993).

adatta

Highlight



While it is hypothesized that the first three ITresources, per se, may not explain variance in theperformance of the customer service process, theremaining two IT capabilities may.

Shared Knowledge. Rockart (1988) and Hender-son (1990) argued that it is the shared knowledgebetween line and IT managers that determines thestrategic use of IT. Similarly, Boynton et al. (1994)use absorptive capacity theory (Cohen and Levin-thal 1990) to show that an organization’s IT use isinfluenced by the presence of a mosaic of IT-related knowledge that binds the firm’s IT and linemanagers. A major component of a firm’s absorp-tive capacity regarding IT is represented by theconjunction of IT and business-related knowledgepossessed by and exchanged among the IT andline managers. In the context of the customerservice process, it is the knowledge that the ITmanager possesses about the customer serviceprocess, the knowledge that the customer servicemanager possesses about the potential oppor-tunities to apply IT to improve customer service,and the common understanding between the ITand the line manager regarding how IT can beused to improve customer service process per-formance that constitute the construct we refer toas shared knowledge. Shared knowledge is,therefore, an important capability that enables theorganization to conceive, effectively implement,and use IT applications to improve customerservice process performance. In this regard,Nelson and Cooprider (1996) found that increasinglevels of shared knowledge between IS and linegroups are linked with increased operational andservice performance of the IS group.8 Similarly,Armstrong and Sambamurthy (1999) found thatshared knowledge influences IT assimilation.Additionally, recent research (Reich and Benbasat2000) suggests that shared domain knowledgebetween IT and business executives influences thelevel of IT-business alignment, a key successfactor long emphasized in the IS literature.

Drawing on resource-based logic, Mata et al.(1995) argue that, among the commonly discussedIT resources, only managerial IT skills (a constructsimilar to shared knowledge, although theseauthors place more emphasis on the IT manager’sside of the dyad) can be a source of sustainablecompetitive advantage. Shared knowledge isdeveloped over long periods of time. The trust,interpersonal relationship, and a shared body offirm-specific knowledge between the IT and thecustomer service managers at a level where theyare able to effectively work together to conceivenovel IT applications can take years and numerousjoint development projects to evolve. Thus thedevelopment of shared knowledge is often a path-dependent and socially complex process. To theextent that this knowledge is valuable and hetero-geneously distributed across firms, it can be a keyIT-related differentiator as it is not subject to lowcost imitation. Thus, variance in the extent towhich there is shared knowledge between IT andcustomer service managers about how IT can beused to improve the customer service process canexplain differences in the performance of thisprocess across competing firms.

Flexible IT Infrastructure. In recent years, ITinfrastructure has been identified as anothercapability that can influence a firm’s ability to useIT strategically (Allen and Boynton 1991; Arm-strong and Sambamurthy 1999; Broadbent et al.1999; Davenport and Linder 1994; Duncan 1995;Ross et al. 1996; Sambamurthy et al. 2003; Weill1993). IT infrastructure is defined as a shared setof capital resources that provide the foundation onwhich specific IT applications are built (Broadbentand Weill 1997; Duncan 1995). The primary con-stituents of IT infrastructure are (1) computing plat-form (hardware and operating systems), (2) com-munications network, (3) critical shared data, and(4) core data processing applications (Byrd andTurner 2000).

The differential flexibility of firms’ IT infrastructuremakes the cost, pace, and value of IT-enabledinnovation different for different firms (Broadbentand Weill 1997; Duncan 1995). A flexible IT infra-structure facilitates rapid development andimplementation of IT applications that enhance

8Their shared knowledge construct shares similaritieswith our conceptualization, although their operation-alization, level of analysis, and dependent measurediffer.



customer service process performance byenabling the organization to respond swiftly to takeadvantage of emerging opportunities or to neutral-ize competitive threats. On the other hand, aninflexible IT infrastructure may get in the way ofsome important initiatives, limiting the freedom ofthe company to respond to market forces andinnovate (Davenport and Linder 1994). In thissense, flexible IT infrastructure is seen as likely tobe a valuable capability for the customer serviceprocess. The flexibility and enabling nature of anIT infrastructure is manifested in the extent towhich a firm adopts standards for the componentsof its IT infrastructure. Standards for hardware,operating systems, communications network, data,and applications imply that data and applicationscan be shared and accessed throughout theorganization (Broadbent and Weill 1997).

A flexible IT infrastructure is a complex set oftechnological resources carefully planned for anddeveloped over time. Because of its path depen-dent nature, there can be significant differencesacross firms in how their infrastructure is con-stituted. Moreover, these differences can be longlasting, since disassembling one infrastructure anderecting a new one can be both costly and timeconsuming. To the extent that the flexibility of ITinfrastructure varies across firms in the insuranceindustry, and to the extent that a flexible infra-structure enables firms to implement IT applica-tions to support customer service more efficientlyand effectively, the variance in infrastructureflexibility could explain differences in the perfor-mance of the customer service process acrossthese firms. These observations lead to thefollowing hypotheses:

Hypothesis 2a: The level of sharedknowledge will explain variance in theperformance of the customer service pro-cess across firms in the North Americaninsurance industry.

Hypothesis 2b: The flexibility of ITinfrastructure will explain variance in theperformance of the customer service pro-cess across firms in the North Americaninsurance industry.

IT Complementarities

IT can affect customer service process perfor-mance differences across firms in at least twodifferent ways. First, if a firm possesses valuable,rare, and costly to imitate IT capabilities likeshared knowledge or a flexible IT infrastructure,then, as suggested by hypotheses 2, the appli-cation of such capabilities to the customer serviceprocess can lead to relative gains in process per-formance. In this sense, the valuable, rare, andcostly to imitate IT capability, per se, can explainvariance in customer service process per-formance.

Alternatively, a firm with valuable, rare, and costlyto imitate IT capability may be able to leverage thiscapability to realize the full competitive potential ofIT resources like technical IT skills, generic infor-mation technologies, and IT spending, that are, assuggested by hypotheses 1, by themselves,unlikely to explain variance in customer serviceprocess performance across firms. This is espe-cially the case with respect to process-specific ITcapabilities, such as shared knowledge, asopposed to an organizational-level capability suchas a flexible IT infrastructure. The ability to gener-ate superior customer service process perfor-mance compared to competitors, from explicit ITresources, is contingent upon the level of sharedknowledge. After all, it is the shared knowledgethat drives how IT resources like generic tech-nologies, technical skills, and IT spending aredeployed and used to improve customer serviceprocess performance. Therefore, it is hypothe-sized that shared knowledge has a direct and amoderating effect on the ability of a firm to achieverelative gains in customer service process perfor-mance through IT. If these moderating effects arenot taken into account, the impact of explicitresources (technical IT skills, generic informationtechnologies, and IT spending) on customer ser-vice process performance may be obscured.While these explicit IT resources, by themselves,may not explain variance in the performance of thecustomer service process across firms, they mayhelp explain this variance in settings in which firmshave high levels of shared knowledge. In thissense, the shared knowledge may also be an ex-

adatta

Highlight



Relative Process Performance

Technical IT Skills

Generic Technologies

IT Spending

Shared Knowledge

(H2a) +

(H2b) +

(H3a) + (H3b) + (H3c) +

(H1c) = 0

(H1b) = 0

(H1a) = 0

Flexible IT Infrastructure

Relative Process Performance

Technical IT Skills

Generic Technologies

IT Spending

Shared Knowledge

(H2a) +

(H2b) +

(H3a) + (H3b) + (H3c) +

(H1c) = 0

(H1b) = 0

(H1a) = 0

Flexible IT Infrastructure

Figure 1. The Research Model

ample of a complementary organizational IT assetas discussed by Barua et al. (1996) and byBrynjolfsson et al. (1998). Thus, it is hypothe-sized that the interaction of the explicit resourceswith shared knowledge will explain variance incustomer service process performance. Theseobservations lead to the following hypotheses:

Hypothesis 3a: Technical IT skills, inthe presence of a high level of sharedknowledge, will explain variance in theperformance of the customer service pro-cess across firms in the North Americaninsurance industry.

Hypothesis 3b: Generic informationtechnologies used to support the custo-mer service process, in the presence ofa high level of shared knowledge, willexplain variance in the performance ofthe customer service process acrossfirms in the North American insuranceindustry.

Hypothesis 3c: The level of IT spend-ing, in the presence of a high level ofshared knowledge, will explain variancein the performance of the customerservice process across firms in the NorthAmerican insurance industry.

The research model developed in the previousdiscussion is presented in Figure 1.

Research Methodology

Measures

A two-part survey instrument was designed to elicitinformation about all of the variables. Whereverpossible, existing scales were used. However,some new scales were also developed. A copy ofthe surveys used to measure these variables isincluded in Appendix A.

adatta

Highlight

adatta

Highlight



Independent Variables

The scale developed by Leitheiser (1992) and byLee et al. (1995), focusing on programming skills,was used to assess the technical skills of the ISunit.9 A new scale, similar to the scale developedby Powell and Dent-Micallef (1997), was devel-oped to measure the generic information technol-ogies variable. Interviews with IT and customerservice managers were used to identify the rangeof well-known information technologies that areused to support the customer service process inthe insurance industry. The scale identifies whichof these technologies a particular firm has and theextent to which these technologies have beenimplemented. The primary measure for ITspending is the standard benchmark metric forenterprise IT spending, namely, IT budget10 peremployee.11

Shared knowledge was measured using aninstrument adapted from Boynton et al. (1994).The instrument consists of two 3-item scales fortwo informants in each firm: the manager of thecustomer service unit and the senior IT executive.Obtaining perceptions about shared knowledgefrom both sides of the customer service and ITrelationship provides a more accurate measure ofthis capability than relying on firm-level perceptualmeasures or from a single informant. Answers to

these questions were averaged.12 The variable“flexible IT infrastructure” was measured using aset of questions adapted from Duncan (1995).This scale assessed the degree to which the firmhad standardized on the platform and data suchthat information could be easily shared acrosssystems and business units.

Dependent Variable

The dependent variable in this study is the relativeperformance of the customer service process.Two approaches could have been taken inmeasuring the level of customer service perfor-mance across firms. First, productivity measuressuch as the level of throughput and process cycletime could be used. Second, the quality of theoutput of this process could be assessed. In thisstudy, the latter approach was adopted.

This choice of measure for the dependent variablereflects both the specific process and the industrystudied here and the theoretical interest of thisstudy: understanding the impact of IT on therelative performance of this process across com-peting firms. Several authors have argued thatwhile productivity measures may be appropriate inthe manufacturing context where the output(product) can be defined and measured easily andunambiguously, they are less appropriate forservice processes (Brynjolfsson 1993; Parasur-aman 2002). In contexts like customer service ininsurance companies, the experience of thecustomer is the critical competitive performancecriteria, not the productivity of the process, per se.Indeed, some very inefficient customer serviceprocesses—defined in productivity terms—areroutinely cited as sources of superior customerservice in the marketing and strategy literatures(e.g., Collins and Porras 1994). Moreover, it isunderstanding variance in this customer experi-ence that is most consistent with the notion of

9The results did not change when, instead of focusing onprogramming skills, we used a four-item scale withemphasis on other technical skills (systems analysis,database design, operating systems, systemsintegration).

10The IT budget is inclusive of outsourcing costs.

11Additional measures of IT spending were alsoexplored, including the firm’s total IT budget and theproportion of that budget used for customer service(customer service IT budget). The results were not sub-stantially different. However, because these measuresof spending turned out to be almost perfectly correlatedwith firm size and highly correlated with genericinformation technologies, we only present the regressionresults using IT spending per employee as a measure ofIT spending, since this allows us to better partial out thepossible effects of firm size on process performance.

12A measure of shared knowledge calculated by multi-plying the two sides of the dyad was also explored.However, it generated the same results as the simpleradditive approach , and thus this additive approach wasadopted.

adatta

Highlight



relative process performance that is central to thisstudy. As Parasuraman (2002) notes, productivitymeasures fail

to take into consideration customers’input into the process (e.g., waiting timeand emotional energy due to frustration),as well as the output experience by thecustomers (service performance, satis-faction).

Service quality takes the customer view intoaccount and is, therefore, believed to be a moreappropriate indicator of process performance inthe context of this study.

The specific measure of customer service perfor-mance used in this study is a widely used scaleadapted from Parasuraman et al. (1988). Thisinstrument is normally administered to a firm’scustomers to assess its customer service quality.In order to calculate the variance in the perfor-mance of the customer service process acrossfirms in this industry, it was necessary to measurethe quality of customer service for a large numberof companies. Thus, it was not feasible to admin-ister this instrument directly to the customers ofthese companies. Therefore, a modified version ofthis instrument was administered to the customerservice managers to assess the customer serviceperformance of their units. This measure of custo-mer service performance is called PZB in the restof the paper.

To check for possible common method varianceproblems, this subjective measure of processperformance was triangulated with three objectivemeasures. These measures were (1) customerretention rate, (2) self-assessment from surveysdistributed by a firm to its customers to assess thequality of its customer service, and (3) the com-plaints ratio—defined as the ratio of the number ofcomplaints to regulators relating to an insurer, tothe premiums written by that insurer—collected bythe National Association of Insurance Com-missioners. In general, it is assumed that firmswith high quality customer service (PZB) will havehigh retention rates, high evaluations from custo-mer surveys, and low complaints ratios. In fact,

the correlation between PZB and these threeobjective measures is consistent with these expec-tations. The positive and significant correlationsbetween PZB and retention (r = 0.355, p = 0.005),self-assessment (r = 0.489, p = 0.000), and com-plaints ratio (reverse coded) (r = 0.292, p = 0.042)provide confidence in PZB as a reliable measureof customer service performance. Moreover,examining the hypotheses with these alternativedependent variables led to results similar to thosepresented here.13

Control Variables

The literature in Strategy and Human ResourceManagement identifies service climate as a criticalnon-IT resource that determines customer serviceperformance (Hansen and Wernerfelt 1989;Schneider et al. 1992; Schneider et al. 1998).Service climate refers to the employee’s percep-tions of the practices, procedures, and behaviorsthat are expected, supported, and rewarded, withregard to customer service (Schneider et al. 1992;Schneider et al. 1998). Service climate wasincluded as a control variable and assessed usinga scale developed and validated by Schneider andhis colleagues (1992, 1998). To the extent thatfirm size might influence customer service processperformance and to the extent that firm size tendsto be associated with IT resources, firm size asmeasured using the number of employees in thefirm was also included as a control variable.

Data Collection and Analysis

Sample

The questionnaire was administered to managersin the life and health insurance industry during thefirst quarter of the year 2000. Data was collectedfor the year 1999. The target respondent list,

13The only substantive difference was that infrastructureflexibility had a significant and positive impact oncustomer retention.



which was compiled from two different sources,namely, the membership list of Life Office Manage-ment Association (LOMA), an industry organiza-tion, and the Dun & Bradstreet database, con-sisted of 800 companies comprising the life andhealth insurance companies operating in NorthAmerica with over 100 employees. The ques-tionnaire was divided into two parts: (1) customerservice and (2) information systems. Componentswere to be completed and returned independentlyby the customer service and the IT managerrespectively.14 Responses from 104 distinct firmswere obtained. Responses from both the custo-mer service and IT manager were obtained from72 firms. This generated a matched sampleresponse rate of 9 percent. This response rate iscomparable to other studies with matched surveysfrom senior executives (Sabherwal and Chan2001). A variety of analyses indicated that respon-dents were representative of the target population,and thus nonresponse bias does not appear to bea problem in the data.15 Descriptive statistics forthe firms in the sample are shown in Table 1.16

Data Analysis

The research model and the hypotheses pre-sented earlier were examined using a manifestvariable method. All of the multivariate variableswere estimated by averaging the item scores forthe variable. Table 2 summarizes for each vari-able the number of items in the scale, the mean ofitems in the scale, the standard deviation of themean, and the reliability of the scale. All the

Cronbach's alpha values were found to be greaterthan 0.70, the threshold recommended by theliterature, except for the scale for generic informa-tion technologies in customer service. This scale,in the spirit of a scale developed by Powell andDent-Micallef (1997), is intended to assess thetypes and scope of the generic technologies imple-mented to support customer service. As such,variances among the items in this scale may notbe as homogeneous as would be expected in theother scales, leading to a marginally lower alpha.Nevertheless, the alpha coefficient for this scale(0.65) is close to the widely accepted cutoff valueof 0.70 and greater than the minimum recom-mended (0.60) for newly developed scales(Nunnally 1988).

In order to assess the convergent and divergentvalidity of the scales, a factor analysis was con-ducted on all of the items used in the study. Theitems loaded highly on the right constructs and hadlow loadings on the other constructs indicatinghigh convergent and divergent validity (seeAppendix B). Table 3 shows the bivariate correla-tions between the dependent variable customerservice (PZB), the independent variables, and thecontrol variables. The correlation matrix showsthat larger firms make higher IT investments incustomer service. They also implement a largernumber of generic information technologies.

Ordinary least squares (OLS) regression analysiswas used to test the hypotheses. The residuals forall of the models satisfied distributional assump-tions. The normal probability plot of the stan-dardized residuals suggested that the residualsare normally distributed. The plot of standardizedresiduals against the standardized predictedvalues indicated linearity and equality of variance.Multicollinearity, as indicated by variance inflationfactors, was also consistently low. Table 4 showsthe results of the regression analyses.17

14For the LOMA sample, surveys were sent directly to ISmanagers (typically, the CIO) with instructions tocomplete and return the information systems componentand to forward the customer service component to thecustomer service manager in their company. For theDun & Bradstreet sample, surveys were sent to the chiefexecutive, with instructions to pass the surveys to the ISand customer service managers in their organization.

15Details of these analyses, omitted to conserve space,are available from the authors.

16Surveys from both the CIO and customer servicemanagers were received from 72 different firms.However, missing data on some of these surveysreduced the sample to 62.

17As noted earlier, rerunning the same set of modelsusing customer service IT budget as a measure of ITspending produced similar results. No control for firmsize was included in these models because customerservice IT budget is almost perfectly correlated with firmsize. However, Chow tests suggest that regressioncoefficients do not vary when the sample is split basedon firm size.



Table 1. Descriptive Statistics of Firms Included in the Sample (N = 62)

VariableSample

MinimumSample

MaximumSampleMean

SampleStandardDeviation

Total Assets (million $) 17.3 191,536.5 11,931.9 33,541.8

Total Premiums (million $) 8.7 9,896.5 1,129.9 2,282.2

Annual CS Budget (million $) 0.2 180.0 13.9 34.7

Annual IT Budget (million $) 0.3 1000.0 48.3 147.8

Annual CS IT Budget (million $) 0.03 300.0 17.8 46.7

Company Age (years) 3 175 58.9 40.8

Number of Customer Service Employees 3 3,600 210 574

Number of IS Employees 3 5,194 234 711

Total Employees 38 60,000 3,124 8,629

Number of Products Sold 1 13 3.9 2.5

Table 2. Independent and Dependent Variables’ Mean, Standard Deviation, andReliability

Variable TypeVariable

NameNumberof Items Mean

StandardDeviation

Reliability(Cronbach’s

Alpha)

DependentVariable

Customer ServicePerformance (PZB)

7 3.95 0.50 0.84

IndependentVariables

Technical Skills of ITLabor

3 2.84 0.77 0.79

Generic InformationTechnologies

6 2.09 0.81 0.65

Shared Knowledge 6 3.77 0.59 0.83

Flexibility of ITInfrastructure

5 3.36 0.83 0.80

Control Variable Service Climate 6 3.79 0.61 0.84

Table 3. Correlation Coefficients

Variables 1 2 3 4 5 6 7 8 9

1. Customer Service (PZB) 1.00

2. Technical IT Skills 0.048 1.00

3. Generic Technologies 0.018 0.253** 1.00

4. IT Budget Per Employee –0.032 –0.026 0.166 1.00

5. Customer Service IT Budget 0.199 0.168 0.430*** 0.158 1.00

6. Shared Knowledge 0.359*** 0.211* 0.303** –0.023 0.213* 1.00

7. Flexibility of IT Infrastructure –0.054 0.256** 0.243* –0.252** 0.076 0.170 1.00

8. Service Climate 0.395*** 0.224* 0.357*** 0.016 0.327*** 0.133 0.138 1.00

9. Number of Employees 0.214* 0.201 0.441*** 0.064 0.957*** 0.254* 0.130 0.293** 1.00

*p < .10, **p < .05, ***p < .01 (All tests are two-tailed)



Table 4. Results of Regression Analysis

Hypothesis/Variable

CustomerService

(Model 1)

CustomerService

(Model 2)

CustomerService

(Model 3)

CustomerService

(Model 4)

CustomerService

(Model 5)

Std. Estimatep – value

Std. Estimatep – value

Std. Estimatep– value



Intercept 4.125*** 4.094*** 4.089*** 4.129*** 4.078***

Technical Skills (H1a) –0.0580.634

–0.0450.714

–0.0680.569

–0.0570.640

–0.0560.643

Generic CSTechnologies (H1b)

–0.1720.253

–0.2000.185

–0.2300.125

–0.1740.251

–0.2390.117

Annual IT Budget perEmployee (H1c)

–0.0420.732

–0.0380.757

–0.0540.653

–0.0480.698

–0.0520.667

Shared Knowledge(H2a)

0.395***0.002

0.388***0.002

0.403***0.001

0.419***0.001

0.411***0.001

Flexibility of ITInfrastructure (H2b)

–0.1450.265

–0.1540.234

–0.2080.114

–0.1480.257

–0.2030.126

Service Climate 0.468***0.000

0.475***0.000

0.488***0.000

0.452***0.001

0.479***0.000

Log of Number ofEmployees

–0.0790.570

–0.0720.599

–0.0790.559

–0.0810.564

–0.0760.580

Shared Knowledge ×Technical IT Skills (H3a)

0.1570.175

0.1070.377

Shared Knowledge ×Generic Technologies(H3b)

0.241**0.047

0.1930.145

Shared Knowledge × ITBudget / Employee(H3c)

0.0940.431

0.0580.630

R2 0.326 0.350 0.376 0.334 0.387

Adjusted R2 0.237 0.250 0.280 0.232 0.264

F – Modelp – value

3.6630.003

3.4960.003

3.9100.001

3.2620.004

3.1550.003

Power at α = 0.05 87.65 87.31 92.59 83.26 85.05

Max VIF 1.745 1.779 1.812 1.745 1.826

*p < .10, **p < .05, ***p < .01 (All tests are two-tailed)



Model 1 tests the direct effects of IT resources andcapabilities on process performance (hypotheses1 and 2). Consistent with hypotheses 1a, 1b, and1c, technical IT skills, generic information technol-ogies, and IT spending did not explain significantvariance in customer service performance. Thedirect effect of shared knowledge is positive andsignificant at the .01 level. Hypothesis 2(a) is,therefore, supported by the data. The effect of theflexibility of the IT infrastructure is not significant.Thus, there is no support for hypothesis 2(b).Overall, the analysis suggests that shared knowl-edge is a critical IT capability that explainsvariance in customer service process performanceacross firms.

The contingency relationships suggested inhypotheses 3 were tested with moderated regres-sion analysis (Aiken and West 1991). Interactionterms were formed by first centering the mainvariables (subtracting the sample mean value fromeach observation) and then multiplying the (cen-tered) shared knowledge variable with the (cen-tered) measures for the three explicit IT resources(technical IT skills, generic information technol-ogies, and IT spending). Centering eases theinterpretation of the non-product terms andreduces multicollinearity between each interactionterm and its component multipliers withoutaffecting the coefficient of the interaction term itself(Aiken and West 1991, Chapter 3). However,since each interaction term shares the samemultiplier, including all of the interaction terms in asingle model would likely produce biased regres-sion coefficients due to multicollinearity betweenthe interaction terms. Therefore, Table 4 alsopresents models where each interaction term wasincluded in a separate regression to minimizemulticollinearity among the independent variables.Significant regression coefficients for the inter-action terms support the hypotheses about themoderating role of shared knowledge.

Models 2 through 5 show the results of themoderated regressions. When all three multi-plicative terms are included in the same model(model 5), their coefficients, although not signi-ficant, are in the hypothesized direction (positive).Examining the correlation between the interaction

terms suggests that the lack of significance mightbe due to multicollinearity between the interactionterms.18 It would, therefore, be premature to con-clude that shared knowledge has no moderatingrole before examining the three contingencyrelationships independently.

Hypothesis 3a suggested that shared knowledgewould interact with technical IT skills to explainvariance in customer service process perfor-mance. This is tested in model 2 by adding theinteraction term involving technical IT skills to themain effects model (model 1). The interactionterm is not significant, although it is in the hypothe-sized direction. Hypothesis 3a is, therefore, notsupported in the regression analyses. Hypothesis3b, which suggested a positive interaction effectbetween generic information technologies incustomer service and shared knowledge, is testedin model 3. The coefficient for the interaction term(β = 0.241, p = 0.047) is significant. Thus, there issome support for hypothesis 3b. Hypothesis 3c,suggesting a positive interaction effect betweenshared knowledge and IT spending, is tested inmodel 4. The coefficient for this interaction term isnot significant. Thus, we find no support forhypothesis 3c in the regression analyses. Withregard to the control variables, as expected,service climate had a significant impact on cus-tomer service. Also, firm size had no impact oncustomer service.

To further test for the moderating role of sharedknowledge, a subgroup analysis was also con-ducted. The median score of shared knowledgewas used to split the sample into two subsamples,labeled “high shared knowledge” and “low sharedknowledge.” ANOVA and Fisher’s Z tests werethen used to evaluate whether the two subsamplescan be considered random samples from a com-

18In Model 5, the correlation between the generictechnologies and IT skills interaction terms is significant(r = 0.325, p = .005). The interaction term involvinggeneric technologies is also correlated with the inter-action term involving IT budget (r = 0.207, p = .054). Theestimated standard errors for the regression coefficientsof the interaction terms are, therefore, likely to be large,and with large standard errors, it is unlikely that any ofthose coefficients would be statistically significant.



Table 5. Descriptive Statistics of the Subsamples and ANOVA Tests

Firms withHigh Shared Knowledge

(n = 31)

Firms withLow Shared Knowledge

(n = 31)

Variable Mean S.D. Mean S.D. )

Shared Knowledge 4.25 0.32 3.29 0.36 ***

Infrastructure Flexibility 3.50 0.77 3.22 0.87 ns

IT Technical Skills 2.98 0.76 2.71 0.77 ns

Generic Technologies inCustomer Service

2.25 0.81 1.93 0.79 ns

Customer ServicePerformance (PZB)

4.16 0.46 3.75 0.46 ***

Service Climate 3.89 0.53 3.68 0.67 ns

Annual IT Budget forCustomer Service (million $)

25.39 60.92 10.13 23.21 ns

Annual IT Budget PerEmployee ($)

14423.82 8435.21 13844.96 6969.07 ns

Total Assets (million $) 20578.80 44624.47 4009.35 15187.93 ns

Total Premiums (million $) 1529.01 2648.71 742.54 1805.80 ns

Annual Customer ServiceBudget (million $)

18.93 4.77 8.99 12.97 ns

Annual IT Budget (million $) 76.9 202.51 19.2 31.54 ns

Company Age (years) 60.84 42.23 56.87 39.84 ns

Number of Customer ServiceEmployees

272.25 780.40 148.45 228.29 ns

Number of IS Employees 354.79 995.88 120.65 152.29 ns

Total Employees 4817.26 11805.19 1373.97 1908.23 ns

Number of Products Sold 4.07 2.32 3.79 2.72 ns*p < .10, **p < .05, ***p < .01 (All tests are two-tailed)

mon population. Table 5 shows the descriptivestatistics for the two subsamples and the results ofthe ANOVA tests. The analysis shows that firmswith high shared knowledge had significantly bettercustomer service performance, consistent withhypothesis 2a and our findings regarding sharedknowledge discussed earlier. The sample meansfor all of the other independent and control vari-ables were not statistically different across the twosubsamples.

According to Baron and Kenny (1986, p. 1174),

a moderator is a qualitative (e.g., sex,race, class) or quantitative (e.g., level ofreward) variable that affects the directionand/or strength of the relation betweenan independent or predictor variable anda dependent or criterion variable. Speci-fically within a correlational analysisframework, a moderator is a third variable



Table 6. Correlation between IT Resources and Customer Service Performance(PZB) in the Two Subsamples

Firms withHigh SharedKnowledge

Firms withLow SharedKnowledge

Variable r r Fisher Z

IT Technical Skills 0.186 –0.240* 1.62*

Generic Technologies in CS 0.132 –0.278* 1.57*

CS IT Budget 0.268* –0.127 1.51*

IT Budget per Employee –0.039 –0.069 0.11*p < .10, **p < .05, ***p < .01

that affects the zero-order correlationbetween two other variables.

Thus, the significance of the difference betweentwo correlations of the same variables, assessedusing the Fisher’s Z test, is a direct test of modera-tion between two independent samples. Table 6shows the result of this analysis. The small andinsignificant correlation between customer serviceperformance (PZB) and each of the three ITresources, technical IT skills, generic informationtechnologies, and CS IT budget, disguises off-setting effects of consistently negative (and for twovariables, significant) correlations with PZB forfirms with low shared knowledge, and consistentlypositive (and significant, for one variable) correla-tions with PZB for firms with high shared knowl-edge. Comparing these correlation coefficientsacross the two groups (using appropriate direc-tional one-tailed tests), the differences in thecorrelations across the two subsamples were allsignificant at the 0.10 level. No significant differ-ences were found across the two groups whenusing IT Budget per employee as a measure of ITspending. The subsample analysis, therefore, pro-vided some support for all the three hypotheses re-garding the moderating role of shared knowledge.

In summary, the hypotheses regarding technical ITskills (H1a), generic information technologies usedin customer service (H1b), and IT investment(H1c), were supported by the data. The analysis

also strongly supported the hypothesis regardingthe role of shared knowledge (H2a). The data,however, did not provide support for the hypoth-esis regarding the flexibility of the IT infrastructure(H2b). The hypothesis regarding shared knowl-edge moderating the impact of generic technol-ogies (H3b) was supported in both the regressionand the subsample analyses. The hypothesesregarding shared knowledge moderating theimpact of technical IT skills (H3a) and IT spending(H3c) received some support in the subsampleanalysis, but not in the regression analyses.Therefore, overall there is weak support forhypothesis 3a and 3c.

Limitations

The present study has both strengths andlimitations. Prior to discussing the results, it isnecessary to acknowledge the limitations so thatappropriate implications of the findings can bedrawn. First, the seemingly modest sample sizemight be a concern. However, it may be noted thatall the firms in the sample are from the sameindustry, as is required for a process level analysis.Further, it is felt that certain loss of sample size isinevitable when collecting data from multiplesources in a single firm. It is believed that one ofthe distinguishing features of this study is that itcombines data from two different respondents from



each organization. Collecting data from both sidesof the dyad is essential in the operationalization ofthe shared knowledge construct. Power analysisalso indicated that the statistical power exceededthe recommended 80 percent threshold. Finally,analyses suggested that the respondents are re-presentative of the mailing sample and that a self-selection bias is unlikely to be present in the data.

Second, the possibility of common methodvariance with respect to the self-reported measure(PZB) of the dependent variable may also be aconcern. However, the study involves two keyinformants in each organization, and factoranalysis did not suggest that the variables had acommon source. Moreover, the study involves theuse of three other measures of customer serviceperformance, namely, retention rate, self-assess-ment, and complaints ratios, all of which havesignificant positive correlations with PZB, in-creasing our confidence in PZB as an unbiasedmeasure of customer service performance.

Third, the empirical analysis was conducted in thecontext of one specific process in the insuranceindustry; thus, the generalizability of the resultsmay be limited. Further, because the methodologyis cross-sectional, we can only show association,not causality. The data shows that variance inprocess performance across firms is associatedwith certain resources (e.g., shared knowledge).This suggests that this resource is valuable andrare. A longitudinal study examining the imitabilityof shared knowledge would be needed to showcausality and directly address the question ofsustainability.

Discussion and Conclusion

Organizations spend millions of dollars on IT toimprove business performance. However, empiri-cal studies examining the contribution of IT invest-ments to firm performance show mixed results.Therefore, a theory explaining how IT can affectperformance is a significant challenge to the Infor-mation Systems discipline. Our approach, synthe-sizing rich traditions from process-orientation and

the resource-based theory, represents one step inthat direction. The resulting analysis serves toinform the debate on business value of IT and alsohelps to explain why some firms are able toleverage IT better than others. The approach pre-sented in this paper suggests that the effects of ITwill most clearly appear at the process level, andthat resource-based logic can be used to identifythe IT resources and capabilities that are mostlikely to explain variation in performance of theseprocesses across firms.

This study makes several theoretical and empiricalcontributions. First, it distinguishes between twoperformance effects of IT: an absolute perfor-mance effect and a relative performance effect. ITcan improve the efficiency and effectiveness ofprocesses in an absolute sense (i.e., the costand/or quality of processes can be superior with ITcompared to what was the case before IT wasused). In this paper, such IT assets and invest-ments are described as economically valuable.However, that IT is economically valuable does notnecessarily mean that IT, per se, will improve theperformance of a firm’s processes relative tocompeting firms. This depends on how widelydiffused (i.e., rare) a valuable technology is, andhow fast its diffusion will occur (i.e., how costly toimitate it is). More broadly, applying this distinctionshifts attention away from simply describing theconditions under which IT can create value for afirm to describing conditions under which IT cancreate sustained advantage for firms.

The second contribution of this paper is that itgoes beyond the singular focus on IT spendingand empirically examines the differential effects ofvarious IT resources and capabilities on relativeprocess performance. Although there is concep-tual work in the IS literature on how and which ITresources are most likely to affect performance(Jarvenpaa and Leidner 1998; Mata et al. 1995),and empirical work examining the relationshipbetween broad characterizations of IT capabilityand firm performance (Barua et al. 2004;Bharadwaj 2000; Bharadwaj et al. 1999), this is thefirst study to examine the impact of specific ITresources and capabilities on relative processperformance. The empirical findings are largely



consistent with resource-based expectations.Tacit, path dependent, and socially complex ITcapability (shared knowledge) explains variation inprocess performance. Explicit IT resources (tech-nical IT skills, generic information technologies,and IT spending) do not.

Of course, these results do not mean that firmsshould not invest in generic technologies and othertangible aspects of customer service. Clearly,these kinds of explicit resources are required if afirm is to provide a competitive level of service. Assuggested earlier, these IT resources may bevaluable in an absolute sense: investing in theseresources can reduce the cost or increase thequality of customer service compared to the pro-cess without these IT resources. However, sincethese resources are not rare or costly to imitate,most firms in mature industries such as theinsurance industry will already have them in placeor can acquire them from factor markets, and thusthese resources, by themselves, are unlikely toimprove the relative performance of this process.Instead, their impact on process performance isconditional on the firm possessing a high degree ofshared knowledge so that these technologies canbe appropriately deployed and used to generaterelative differentials in process performance. Theinteraction effects studied here demonstrate that inthe right setting (namely, high levels of sharedknowledge), investment in generic IT does improverelative performance. At low levels of sharedknowledge, investing in generic IT may even re-duce process performance. Thus, it is not genericinformation technologies, per se, that impactrelative process performance, but this investmentin the context of firm-specific ability to properlyimplement and use generic information tech-nologies that generates process performancedifferentials across firms.

Based on these results, we conclude that superiorrelative process performance from IT rests less onthe level of IT spending or on the technical skills ofthe IT staff and more on how these resources aredeployed in a firm-specific manner in general, andon creating effective partnerships between IT andbusiness managers in particular. This reaffirmsthe growing consensus that the context within

which IT is applied is as important as the IT itself.This contingency view of the relationship betweenIT investments and performance suggests that justthrowing technology at a process does notnecessarily improve that process. Indeed, suchindiscriminant applications of technology mayactually reduce process performance—an obser-vation consistent with the results of our subsampleanalysis and the large negative, although insigni-ficant, coefficient for generic information tech-nologies reported here and with research findingsreported elsewhere (Powell and Dent-Micallef1997; Richardson et al. Zmud 2004; Wade 2002).

We find no support for the hypothesized positiverelationship between the flexibility of the ITinfrastructure and relative process performance.Two explanations come to mind. First, a flexible ITinfrastructure is a firm-wide resource. It is likelythat, although a flexible IT infrastructure has nosignificant positive impact on the relative perfor-mance of the customer service process, it mayhave a positive impact on some other processeswithin the firm. Second, it may simply be the casethat, in an industry as mature as the NorthAmerican life and health insurance industry, theoptions available with a flexible infrastructure arejust not very valuable. Of course, this possibilitycan only be examined by studying the impact ofinfrastructure flexibility in an industry whereflexibility is likely to be important (e.g., a rapidlychanging industry). Each of these possibilities willneed to be examined in future research.

The study raises a number of other importantquestions as well. First, how is it that some firmsare able to develop shared knowledge, while otherfirms are apparently unable to develop thiscapability? Additional research is needed toexplore the prerequisite skills required to haveshared knowledge and how this capability can bedeveloped and nurtured. In this regard, the workof Bassellier et al. (2003) is important in definingthe constituents of a line managers’ IT knowledge.Similar work is required to define the constituentsof IT managers’ knowledge of the business. Workis also required to examine the institutional condi-tions that facilitate or deter the development ofsuch a capability.



Second, this study has examined shared knowl-edge between IT and the customer service pro-cess. Can such shared knowledge exist betweenIT and other processes within a firm? Do IT unitsthat have high levels of shared knowledge with oneprocess, such as customer service, tend to alsohave high levels of shared knowledge with otherprocesses (e.g., manufacturing, research anddevelopment, and so forth)? That is, are the skillsnecessary to develop high levels of shared knowl-edge between IT and a particular process generali-zable to other processes within a firm? All of thesequestions deserve additional attention.

Finally, this research has focused on examiningthe differential effects of IT resources and capa-bilities on the relative performance of the customerservice process across a set of firms competing ina single industry. Only one non-IT related capa-bility—service climate—was examined here, butonly as a control variable. An important un-answered research question implied by this studyis: How do non-IT capabilities interact with IT-resources and capabilities to affect the relativeperformance of processes in these firms? In theend, understanding how these investments affectthe relative performance of processes within a firmmay require a higher level of integration betweenIT research and other studies on the non-ITdeterminants of process performance.

Acknowledgements

The authors would like to thank Cynthia Beath,Sirkka Jarvenpaa, Kay Nelson, Huseyin Tanriverdi,and Michael Wade for helpful comments on earlierversions of this paper. The authors would also liketo thank the senior editor, V. Sambamurthy, theassociate editor, and the three anonymousreviewers for their helpful feedback and construc-tive comments.

References

Aiken, L. S., and West, S. G. Multiple Regression:Testing and Interpreting Interactions, SagePublications, Newbury Park, CA, 1991.

Allen, B. R., and Boynton, A. C. “InformationArchitecture: In Search of Efficient Flexibility,”MIS Quarterly (15:4), 1991, pp. 435-442.

Amit, R., and Schoemaker, P. J. “Strategic Assetsand Organizational Rent,” StrategicManagement Journal (14:1), 1993, pp. 33-46.

Armstrong, C., and Sambamurthy, V. “InformationTechnology Assimilation in Firms: TheInfluence of Senior Leadership and IT Infra-structures,” Information Systems Research(10:4), 1999, pp. 304-327.

Barney, J. B. “Firm Resources and SustainedCompetitive Advantage,” Journal of Manage-ment (17:1), 1991, pp. 99-120.

Barney, J. B. “Strategic Factor Markets: Expec-tation, Luck, and Business Strategy,” Manage-ment Science (32:10), 1986, pp. 1231-1241.

Baron, R. M., and Kenny, D. A. “The Moderator-Mediator Variable Distinction in Social Psychol-ogical Research: Conceptual, Strategic, andStatistical Considerations,” Journal of Person-ality and Social Psychology (521), 1986, pp.1173-1182.

Barua, A., Konana, P., Whinston, A. B., and Yin,F. “An Empirical Investigation of Net-EnabledBusiness Value,” MIS Quarterly (28:4), 2004,pp. 585-620.

Barua, A., Kriebel, C. H., and Mukhopadhyay, T.“Information Technology and Business Value:An Analytic and Empirical Investigation,” Infor-mation Systems Research (6:1), 1995, pp. 3-23.

Barua, A., Lee, C-H. S., and Whinston, A. B. “TheCalculus of Reengineering,” Information Syst-ems Research (7:4), 1996, pp. 409-428.

Bassellier, G., Benbasat, I., and Reich, B. H. “TheInfluence of Business Managers’ IT Compe-tence on Championing IT,” Information SystemsResearch (14:4), December 2003, pp. 317-336.

Berry, R. “Define Quality Service So You CanDeliver It,” Best’s Review (95:11), March 1995,p. 68.

Bharadwaj, A. S. “A Resource-Based Perspectiveon Information Technology Capability and FirmPerformance: An Empirical Investigation,” MISQuarterly (24:1), 2000, pp. 169-196.

Bharadwaj, A. S., Sambamurthy, V., and Zmud, R.W. “IT Capabilities: Theoretical Perspectivesand Empirical Operationalization,” in Pro-ceedings of the 20th International Conference on



Information Systems, P. De and J. I. DeGross(Eds.), Charlotte, NC, December 1999, pp. 379-385.

Boynton, A. C., Zmud, R. W., and Jacobs, G. C.“The Influence of IT Management Practice on ITUse in Large Organizations,” MIS Quarterly(18:3), 1994, pp. 299-318.

Broadbent, M., and Weill, P. “Management byMaxim: How Business and IT Managers CanCreate IT Infrastructures,” Sloan ManagementReview (38:3), 1997, pp. 77-92.

Broadbent, M., Weill, P., and St. Clair, D. “TheImplications of Information Technology Infra-structure for Business Process Redesign,” MISQuarterly (23:2), 1999, pp. 159-182.

Brynjolfsson, E. “The Productivity Paradox ofInformation Technology,” Communications ofthe ACM (36:12), December 1993, pp. 67-77.

Brynjolfsson, E., Hitt, L., and Yang, S. K. "Intan-gible Assets: How the Interaction of InformationSystems and Organizational Structure AffectsStock Market Valuations," in Proceedings of the19th International Conference on InformationSystems, R. A. Hirschheim, M. Newman, and J.I. DeGross (Eds.), Helsinki, Finland, 1998, pp.8-29.

Byrd, T. A., and Turner, D. E. “Measuring theFlexibility of Information Technology Infra-structure: Exploratory Analysis of a Construct,”Journal of Management Information Systems(17:1), Summer 2000, pp. 167-208.

Collins, J. C., and Porras, J. I. Built to Last:Successful Habits of Visionary Companies,Harper Business, New York, 1994.

Cohen, J. Statistical Power Analysis for Behav-ioral Sciences (2nd ed.), Lawrence ErlbaumAssociates, Hillsdale, NJ, 1988.

Cohen, W. M. and Levinthal, D. A. “AbsorptiveCapacity: A New Perspective on Learning andInnovation," Administrative Science Quarterly(35), 1990, pp. 128-152.

Davenport, T. H., and Linder, J. “InformationManagement Infrastructure: The New Compe-titive Weapon,” in Proceedings of the 27th

Annual Hawaii International Conference onSystem Sciences, IEEE Computer SocietyPress, Los Alamitos, CA, 1994, pp. 885-896.

Dierickx, I., and Cool, K. “Asset Stock Accumu-lation and Sustainability of Competitive Advan-

tage,” Management Science (35:12), 1989, pp.1504-1513.

Duncan, N. B. “Capturing Flexibility of InformationTechnology Infrastructure: A Study of ResourceCharacteristics and Their Measure,” Journal ofManagement Information Systems (12:2), 1995,pp. 37-57.

Elam, J., and Morrison, J. E. P. “United ServicesAutomobile Association (USAA),” Harvard Busi-ness School Case, 9-188-102, Boston, 1993.

El Sawy, O. A., and Bowles, G. “Redesigning theCustomer Support Process for the ElectronicEconomy: Insights from Storage Dimensions,”MIS Quarterly (21:4), 1997, pp. 457-83.

Garvin, D. A. “Leveraging Processes for StrategicAdvantage: A Roundtable with Xerox's Allaire,USAA's Herres, SmithKline Beecham's Leschly,and Pepsi's Weatherup,” Harvard BusinessReview (73:5), 1995, pp. 77-90.

Grant, R. M. “The Resource-Based Theory ofCompetitive Advantage: Implications for Stra-tegy Formulation,” California ManagementReview (33:3), 1991, pp. 114-135.

Griffith, P. A. “Customer Service: Key to theIndustry’s Future,” National Underwriter (97:34),1993, pp. 15-16.

Henderson, J. C. “Plugging into Strategic Partner-ships: The Critical IS Connection,” SloanManagement Review (31:3), 1990, pp. 7-18.

Henderson, R., and Cockburn, I. “MeasuringCompetence? Exploring Firm Effects inPharmaceutical Research,” Strategic Manage-ment Journal (15), 1994, pp. 63-84.

Hansen, G., and Wernerfelt, B. “Determinants ofFirm Performance: The Relative Importance ofEconomic and Organizational Factors,” Stra-tegic Management Journal (10:5), 1989, pp.399-411.

Hitt, L., and Brynjolfsson, E. “Productivity, Busi-ness Profitability, and Computer Surplus: ThreeDifferent Measures of Information TechnologyValue,” MIS Quarterly (20:2), 1996, pp. 121-142.

Jarvenpaa, S. L., and Leidner, D. E. “An Infor-mation Company in Mexico: Extending theResource-Based View of the Firm to a Devel-oping Country Context,” Information SystemsResearch (9:4), 1998, pp. 342-361.

Lee, D. M. S., Trauth, E. M., and Farwell, D.“Critical Skills and Knowledge of IS Profes-



sionals: A Joint Academic/Industry Investiga-tion,” MIS Quarterly (19:3), 1995, pp. 313-340.

Leitheiser, R. L. “MIS Skills for the 1990s: ASurvey of MIS Managers’ Perceptions,” Journalof Management Information Systems (9:1),1992, pp. 69-91.

LOMA. Quality Service in the Life InsuranceIndustry—A Cooperative Research Project ofACLI, LIMRA, LOMA, and NALU, Life OfficeManagement Association, Atlanta, GA, 1993.

Makadok, R. “Towards a Synthesis of theResource-Based and Dynamic Capability Viewof Rent Creation,” Strategic ManagementJournal (22:5), May 2001, pp. 387-401

Markus, M. L., and Soh, C. “Banking on Infor-mation Technology: Converting IT Spendinginto Firm Performance,” in Strategic InformationTechnology Management: Perspectives onOrganizational Growth and Competitive Advan-tage, R. D. Banker, R. J. Kauffman, and M. A.Mahmood (Eds), Idea Group Publishing,Harrisburg, PA, 1993, pp. 364-392.

Mata, F. J., Fuerst, W. L., and Barney, J. B.“Information Technology and Sustained Compe-titive Advantage: A Resource-Based Analysis,”MIS Quarterly (19:4), 1995, pp. 487-504.

Miller, D., and Shamsie, J. “The Resource-BasedView of the Firm in Two Environments: TheHollywood Film Studios from 1936-1965,”Academy of Management Journal (39:3), 1996,pp. 519-543.

Mooney, J. G., Gurbaxani, V., and Kraemer, K. L.“A Process Oriented Framework for Assessingthe Business Value of Information Technology,”in Proceedings of the 16th International Con-ference on Information Systems, J. I. DeGross,G. Ariav, C. Beath, R. Hoyer, and C. Kemerer(Eds.), Amsterdam, The Netherlands, 1995,pp.17-27.

Mukhopadhyay, T., Surendra, R., and Srinivasan,K. “Information Technology Impact on Processand Output Quality,” Management Science(43:12), 1997, pp. 1645-1659.

Nelson, K. M., and Cooprider, J. G. “The Contri-bution of Shared Knowledge to IS Group Per-formance,” MIS Quarterly (20:4), 1996, pp. 409-429.

Nunnally, J. C. Psychometric Theory, McGrawHill, New York, 1988.

Parasuraman, A. “Service Quality and Produc-tivity: A Synergistic Perspective,” ManagingService Quality (12:1), 2002, pp. 6-9.

Parasuraman, A., Zeithaml, V. A., and Berry, L. L.“A Conceptual Model of Service Quality and itsImplications for Future Research,” Journal ofMarketing (49), 1985, pp. 41-50

Parasuraman, A., Zeithaml, V. A., and Berry, L. L.“SERVQUAL: A Multi-Item Scale for MeasuringConsumer Perceptions of Service Quality,”Journal of Retailing (64:1), 1988, pp. 12-40.

Powell, T. C., and Dent-Micallef, A. D. “Informa-tion Technology as Competitive Advantage:The Role of Human, Business, and TechnologyResources,” Strategic Management Journal(18:5), 1997, pp. 375-405.

Ray, G., Barney, J. B., and Muhanna, W. A.“Capabilities, Business, Processes, and Com-petitive Advantage: Choosing the DependentVariable in Empirical Tests of the Resource-Based View,” Strategic Management Journal(25), 2004, pp. 23-37.

Reichheld, F., and Sasser Jr., W. E. “Zero Defec-tions: Quality Comes to Services,” HarvardBusiness Review (68:5), 1990, pp. 105-111.

Reich, B. H., and Benbasat, I. "Factors thatInfluence the Social Dimension of Alignmentbetween Business and Information TechnologyObjectives,” MIS Quarterly (24:1), March 2000,pp. 81-111.

Richardson, V. J., Subramani, M., and Zmud, R.W. “Linking IT Investment and Firm Benefits:The Role of IT Conversion and Effectiveness,”unpublished manuscript, Carlson School ofManagement, University of Minnesota, 2004.

Rockart, J. F. “The Line Takes the Leadership—IS Management in a Wired Society,” SloanManagement Review (29:4), 1988, pp. 55-64.

Ross, J. W., Beath, C. M., and Goodhue, D. L.“Developing Long-Term Competitivenessthrough IT Assets,” Sloan Management Review(38:1), 1996, pp. 31-42.

Rumelt, R. P. “Towards a Strategic Theory of theFirm,” in Competitive Strategic Management, R.B. Lamb (Ed.), Prentice-Hall, Engelwood Cliffs,NJ, 1984, pp. 566-570.

Rust, R. T., Zeithaml, V. A., and Lemon, K. N.Driving Customer Equity: How Customer Life-time Value Is Reshaping Corporate Strategy,Free Press, New York, 2000.



Sabherwal, R., and Chan, Y. E. “Alignmentbetween Business and IS Strategies: A Studyof Prospectors, Analyzers, and Defenders,”Information Systems Research (12:1), 2001, pp.11-33.

Sambamurthy, V. “Editor’s Comments— Re-search in Information Systems: What WeHaven’t Learned,” MIS Quarterly (21:4), 2001,pp. v-xv.

Sambamurthy, V., Bharadwaj, A., and Grover, V.“Shaping Agility through Digital Options:Reconceptualizing the Role of InformationTechnology in Contemporary Firms,” MISQuarterly (27:2), June 2003, pp. 237-264.

Schneider, B., Wheeler, J. K., and Cox, J. F. “APassion for Service: Using Content Analysis toExplicate Service Climate Themes,” Journal ofApplied Psychology (77:5), 1992, pp. 705-716.

Schneider, B., While, S. W., Paul, M. C. “LinkingService Climate and Customer Perceptions ofService Quality: Test of a Causal Model,”Journal of Applied Psychology (83:2), 1998, pp.150-163.

Schroeder, R. G., Bates, K. A., and Junttila, M. A.“A Resource-Based View of ManufacturingStrategy and the Relationship to ManufacturingPerformance” Strategic Management Journal(23:2), 2002, pp. 105-117.

Szymanski, D. M., and Henard, D. H. “CustomerSatisfaction: A Meta-Analysis of the EmpiricalEvidence,” Journal of the Academy of MarketingScience (29:1), 2001, pp. 16-35.

Tallon, P., Kraemer, K. L., and Gurbaxani, V.“Executives’ Perceptions of the Business Valueof Information Technology,” Journal of Manage-ment Information Systems (16:4), 2000, pp.145-173.

Teal, T. “Service Comes First: An Interview withUSAA's Robert F. McDermott,” Harvard Busi-ness Review (69:5), 1991, pp.116-127.

Treacy, M., and Wierseman, F. The Discipline ofMarket Leaders, Addison Wesley, Reading, MA,1995.

Wade, M. Exploring the Role of Information Sys-tems Resources in Online Success,Unpublished Doctoral Dissertation, The Univer-sity of Western Ontario, 2002.

Weill, P. “The Relationship between Investment inInformation Technology and Firm Performance:

A Study of the Valve Manufacturing Sector,”Information Systems Research (3:4), 1992, pp.307-333.

Weill, P. “The Role and Value of InformationTechnology Infrastructure: Some EmpiricalObservations,” in Strategic Information Tech-nology Management: Perspectives on Organi-zational Growth and Competitive Advantage, R.D. Banker, R. J. Kauffman, and M. A. Mahmood(Eds.), Idea Group Publishing, Harrisburg, PA,1993, pp. 547-572.

Wernerfelt, B. “A Resource-Based View of theFirm,” Strategic Management Journal (5:2),1984, pp. 171-180.

Zeithaml, V. A. “Service Quality, Profitability, andEconomic Worth of Customers: What We Knowand What We Need to Learn,” Journal of theAcademy of Marketing Science (28:1), 2000,pp. 67-86.

About the Authors

Gautam Ray is an assistant professor in theMcCombs School of Business at the University ofTexas at Austin. He received his Ph.D. from theOhio State University. His research interests arein the area of impact of information technology onfirms and markets, specifically about how firmsmake choices to take advantage of opportunitiesprovided by information technology. He has pub-lished in journals such as Management Science,Marketing Science, and Strategic ManagementJournal.

Waleed A. Muhanna is an associate professor ofInformation Systems at Max M. Fisher College ofBusiness, The Ohio State University. He receivedhis undergraduate degree from the University ofTulsa, and his Master’s and doctorate from theUniversity of Wisconsin, Madison. His currentresearch focuses on assessing the business valueof information technology and understanding theimpact of information technology, including theInternet, on organizations and markets. His otherresearch interests include model and databasemanagement systems, performance modeling andevaluation, and heuristic algorithms. Waleed has



published numerous articles in scholarly journals,including Management Science, ACM Trans-actions on Computer Systems, IEEE Transactionson Software Engineering, Decision Support Sys-tems, European Journal of Operational Research,Computers in Human Behavior, Strategic Manage-ment Journal, and Annals of Operations Research.Waleed serves on the editorial boards of DecisionSciences Journal, Information Technology andManagement, and International Journal of E-Business Research. He also currently serves asvice-chair of INFORMS' Information SystemsSociety.

Jay B. Barney is a professor of Management andholds the Bank One Chair for Excellence in Cor-porate Strategy at the Max M. Fisher College ofBusiness, The Ohio State University. He receivedhis undergraduate degree from Brigham YoungUniversity, and his Master's and doctorate fromYale University. Jay teaches organizational stra-

tegy and policy to MBA and Ph.D. students. Hisresearch focuses on the relationship between firmskills and capabilities and sustained competitiveadvantage. He has published 4 books and over 75articles in a variety of journals and books, includingAcademy of Management Review, StrategicManagement Journal, Management Science,Journal of Management, and Sloan ManagementReview. He has been on the editorial boards forAcademy of Management Review and StrategicManagement Journal, served as associate editorfor Journal of Management, and served as senioreditor for Organization Science. In 1997, Jay wasawarded an honorary doctoral degree from theUniversity of Lund, and in 2000 he became anHonorary Visiting Professor at Waikato Universityin Hamilton, New Zealand. Jay was elected aFellow of the Academy of Management in 2001and in 2005 received the Irwin OutstandingEducator Award for the Business Policy andStrategy Division of the Academy of Management.



Appendix A

Survey Measurement Scales

Customer Service Component

A. Service Climate (CLI)

Please indicate, on a scale of 1 to 5, the degree to which you agree or disagree with the followingstatements.

StronglyDisagree = 1 Disagree = 2

Neither Agree norDisagree = 3 Agree = 4

StronglyAgree = 5

CLI1. Customer service unit has established clear standards for the quality of service to be delivered.CLI2. The company measures and tracks the quality of service provided by the customer service unit.CLI3. Customer service representatives are informed about external customer evaluations of the quality

of service delivered by the customer service unit.CLI4. The company offers competitive salaries to customer service representatives.CLI5. Customer service representatives are recognized for delivering quality service.CLI6. Customer service representatives are rewarded for delivering quality service.

B. IT Manager Knowledge (ITMK)

ITMK1. Managers in the information systems unit understand the business operations of the customerservice unit.

ITMK2. Managers in the information systems unit understand the business strategies of the customerservice unit.

ITMK3. There is a common understanding between managers in customer service and informationsystems units regarding how to use information technology to improve customer service.

C. Customer Service Performance (PZB)

PZB1. The customer service unit gives customers prompt service.PZB2. Customer service representatives are never too busy to respond to customers.PZB3. Customer service representatives are empowered to solve customers’ problems.PZB4. When the customer service unit promises to do something for a customer by a certain time, it does

so.PZB5. When a customer has a problem, the customer service unit shows sincere interest in solving it.PZB6. The customer service unit performs the service accurately the first time.PZB7. Customer service representatives understand customers’ specific needs.



D. Customer Service (Self Assessment)

What was the overall rating in your last survey of customers to evaluate the unit’s customer service quality?(choose the scale closest to one used in your company)

Unsatisfied Neutral SatisfiedScale 1: G 1 G 2 G 2

Poor Fair GoodVeryGood Excellent

Scale 2: G 1 G 2 G 3 G 3 G 5

ExtremelyPoor

ExtremelyGood

Scale 3: G 1 G 2 G 3 G 4 G 5 G 6 G 7 G 8 G 9 G 10

E. Customer Service (Retention)

What is your policy retention/persistence rate (in percentage) over the most recent one-year period?

Individual Life: __________% Group Life: __________% Health/Disability: __________%

Information Systems Component

A. IT Infrastructure (INF)

StronglyDisagree = 1 Disagree = 2

Neither Agree norDisagree = 3 Agree = 4

StronglyAgree = 5

INF1. Our company has established corporate rules and standards for hardware and operating systemsto ensure platform compatibility.

INF2. Our company has identified and standardized data to be shared across systems and businessunits.

INF3. Our customer service representatives and agents can access all data pertinent to a customerthrough a single interface.

INF4. What percentage of corporate data is standardized?INF5. What percent of corporate data is currently sharable across systems and business units?

B. Line Manager Knowledge (CSMK)

CSMK1. Managers in the customer service unit recognize the potential of IT as a tool to increase theproductivity (efficiency) of the customer service representatives.

CSMK2. Managers in the customer service unit recognize the potential of IT as a tool to improve the qualityof service delivered by the customer service unit.

CSMK3. There is a common understanding between managers in the information systems and customerservice units regarding how to use IT to improve customer service.



C. Technical Skills (SKL)

How would you rate the information systems units skills in the following areas?

Poor = 1 Fair = 2 Good = 3 Very Good = 4 Excellent = 5

SKL1. Object-oriented languages and systemsSKL2. Fourth generation programming languagesSKL3. Client-Server application development

D. Generic Technologies in Customer Service (TCS)

Please indicate the extent to which you have implemented the following technologies to support customerservice.

Don’t Intend to Implement = 0 Not Yet Begun = 1 Standard/Common Implementation = 3Highly Advanced Implementation = 5

TCS1. Scanning/Imaging TechnologyTCS2. Network with Agents/BrokersTCS3. Web-enabled Customer InteractionTCS4. Call Tracking/Customer Relationship Management System TCS5. Computer Telephony Integration (CTI)TCS6. Customer-service Expert / Knowledge-based System

IT Investment

What is your annual IT budget (including outsourcing contracts) for 1999? $_____

1. What percentage of IT applications directly support the customer service unit? ______2. What percentage of computer hardware directly supports the customer service unit? ______3. What percentage of the IS unit support staff is dedicated to supporting the customer service unit?

______



Appendix B

Results of Factor Analysis

Customer Service Component

Climate Customer Service (PZB)IT Managers Customer

Service KnowledgeCLI1 0.764CLI2 0.798CLI3 0.602CLI4 0.635CLI5 0.759CLI6 0.765PZB1 0.565PZB2 0.595PZB3 0.605PZB4 0.604PZB5 0.759PZB6 0.703PZB7 0.799ITMK1 0.898ITMK2 0.899ITMK3 0.791

Information Systems Component

FlexibleInfrastructure

Customer ServiceIT Knowledge Technical Skills

Generic Technologies inCustomer Service

INF1 0.410INF2 0.800INF3 0.683INF4 0.846INF5 0.826CSMK1 0.896CSMK2 0.881CSMK3 0.668SKL1 0.710SKL2 0.776SKL3 0.569TCS1 0.691TCS2 0.541TCS3 0.633TCS4 0.534TCS5 0.515TCS6 0.470The maximum cross loading on INF1 was 0.257 and the maximum cross loading on TCS6 was 0.240.

ray et al./it & the customer service process...ray et al./it & the customer service process...

Documents