Organizational Processes For B2B Services IMC Data Quality

Organizational processes for B2B services IMC data qualityJournal of Business & Industrial MarketingDebra Zahay, James Peltier, Anjala S. Krishen and Don E. SchultzVolume 29 · Number 1 · 2014 · 63 –74scholars and practitioners are increasingly extolling the virtuesof having cross-functional organizations (Peltier et al., 2013),specifically with regards to developing and implementinghorizontal communication systems (Hadjikhani andThilenius, 2005; Schultz and Kitchen, 2000; Schultz andSchultz, 2003). From the given literature, we identify thefollowing key dimensions for improving the quality of IMCdata available to marketers:.IMC data vision;.marketing/IT integration;.marketing/IT cooperation;.marketing/IT conflict;.marketing manager support; and.data sharing.Customer data quality is likely to be contingent on having acommunication network in place that is cooperative,collaborative, open, and allows users to interact on afrequent basis to set project priorities and generate newproject ideas (Cooper et al., 2008). The process of putting thecustomer database together in a systematic, strategic andquality-focused manner will be even more important incommunication flow and the process of learning how theinformation held in the customer database will benefit allstakeholder groups.In contrast, inter-functional conflict is expected to have anegative impact on IMC data quality due to the fear ofchanging the status quo (Keramati et al., 2010), havingdisparate and even incongruent goals between functionalareas, and a host of personality and cultural differences thatnaturally arise when bringing functional areas together withdifferent ways of thinking and operationalizing (Pascale andSternin, 2005; Peltier et al., 2002). At issue is the concern thatdisparate functional areas often differ in how they valuespecific types of IMC data, particularly with regard tobehavioral as compared to more relational/attitudinal data(Zahay et al., 2004). Strategically, cross-functional conflictthat surfaces through data quality implementation practicescan only be resolved through a process of communication andcollaboration that identifies how customer data can be utilizedby various internal stakeholder groups (Lindgreen et al.,2009). Ultimately, we expect that the success of IMC dataquality initiatives will be a result of how well organizationalmembers are persuaded that these practices have value, andthat the time and energy to implement this new way ofthinking merits consideration. Based on the precedingdiscussion, we posit that:Consistent with emerging IMC research, data qualityenhancement occurs through the merging of organizationalcommitment, cross-functional interactions, and effectivedepartmental utilization of cross-platform data (Zahay et al.,2012).IMC data visionThe trend towards more sophisticated cross-platform datarepositories is linked in part to marketing-oriented cultures inorganizations and how they place emphasis on developinginformation-driven and personalized IMC customer contactstrategies. In this regard, overcoming poor IMC data requiresan organizational culture that places priority on personalizingrelationships and building higher levels of customerengagement (Rubinson, 2009). This corporate vision can beimplemented with a collective mindset within the organizationthat recognizes the benefits of cross-platform data to fosterand enhance customer value. Supporting this idea, Homburget al. (2007) found that data quality enhancements are“anchored in an organization’s values, belief structures, andnorms” (p. 20). Closing the information divide will require ahorizontal organizational culture committed to customercentricity, one that begins with upper management andpermeates within and across all functional entities responsiblefor achieving this focus (Swain, 2004; Rappaport, 2010;Rubinson, 2009). Logically, firms with a top-down andorganization wide commitment to customer informationsystems would place high priority on developing andcontinually improving the quality of their customer database(Salojärvi et al., 2010). We thus posit:H1.H2.H3.H4.Marketing/IT integration is positively related to IMCdata quality.Marketing/IT cooperation is positively related to IMCdata quality.Marketing conflict/IT is negatively related to IMC dataquality.Marketing manager support and data SharingAs the link between an organization and its markets, themarketing/advertising and sales functions are responsible forcommunicating with and managing IMC data (Rust et al.,2010). Importantly, the proliferation of customer touchpointshas changed how IMC data are captured and shared(Richards and Jones, 2008; Voss and Voss, 2008). Thistransformation means that successful organizations placegreater priority on marketing managers and their boundaryspanning ability to acquire and use customer data and formeasuring IMC performance (Liu and Comer, 2007).Supporting this view, Zahay and Peltier (2008) findqualitative evidence that the support of marketing managersfor collecting and using customer data is a key successcriterion for creating high quality customer informationsystems. Thus we hypothesize:An organizational IMC vision committed to securingand using customer information is positively related toIMC data quality.Marketing and IT system integration, cooperation and conflictIn today’s complex environment, it is difficult for anorganization to be successful with a purely functionalstructure, since functional or divisional silos often createcoordination chasms that inhibit change and growth(Keramati et al., 2010; Peltier et al., 2003). Unfortunately,research shows that the information chasm is especially widebetween marketing and IT departments, the two functionalentities that have the greatest influence on the adoption,utilization and modification of data-intensive IMC andcustomer contact systems (Zahay and Peltier, 2008). Thisintegration is critically challenging since departments areoften limited by functional boundaries and have differentviews on customer data quality.H5.Marketing manager support is positively related toIMC data quality.Organizations with marketing and communication silos offerfew opportunities for sharing customer data (Schultz andSchultz, 2003). Such organizations commonly find it difficultto adhere to IMC principles such as having consistent65

Organizational processes for B2B services IMC data qualityJournal of Business & Industrial MarketingDebra Zahay, James Peltier, Anjala S. Krishen and Don E. SchultzVolume 29 · Number 1 · 2014 · 63 –74messages and cohesive customer contact strategies (Peltieret al., 2006). To overcome these limitations, marketers shouldfocus on sharing customer-level data between differentfunctional areas. Surprisingly, the sharing of customer datais often a concern even within functional units, and especiallybetween marketing and sales personnel (Liu and Comer,2007). Although little empirical research has examined datasharing and IMC data quality, Zahay and Peltier (2008)indicate that the sharing of customer information betweenmarketing and sales personnel is critical for improving dataproblems related to information integration, informationaccess, and CRM information quality. Therefore we positthat:remaining non-respondents, either speaking with them orleaving a voicemail message.In total, 128 questionnaires were returned. Of these 128responses, four were removed due to non-response, for a totalof 124. This approach resulted in an overall response rate of24.4 percent, which is comparable to studies that surveybusiness executives (Morrison and Haley, 2006).Respondents’ business was split almost equally betweenB2B and B2C, with any other business coming from traderelationships not asked for or reported in the survey. AsTable I shows, 55 percent of the respondents’ business isconducted at retail or branch banking locations. These firmsrelied on outside sales personnel for about 27 percent of theirbusiness and the remaining 7 percent of their business camefrom online sales. The rest of sales (11 percent) came fromother sources not reported in the survey. The majority of therespondents (64 percent) were over 45 years old, suggesting ahigh level of industry experience. Most of the firms (76percent) reported over 250m in sales/assets undermanagement.H6.Data sharing is positively related to IMC data quality.IMC data quality and performanceWithin the CRM literature, a small but growing stream ofresearch shows that superior CRM implementation leads tobetter marketing and business performance (Kim and Kim,2009; Zahay and Peltier, 2008). Consistent with Stein andSmith (2009), Peltier et al. (2002, 2006), and Zahay et al.(2004), we would thus expect that superior IMC data is astrong proprietary advantage, one that will lead to a betterunderstanding of customer needs, more satisfied and loyalcustomers, and greater profitability. Although empiricalresearch linking IMC data quality to organizationalperformance is sparse, we suggest that:H7.Independent variablesIndependent variables were conceptualized as follows, basedon a review of the organizational learning literature and priorwork in customer information management:.IMC data vision – team spirit, common purpose,organizational vision, rewards, and understanding ofquality customer information strategy;.marketing/IT integration – narketing and IT set projectpriorities together;.marketing/IT cooperation – narketing and IT cooperateand have compatible goals;.marketing/IT conflict – marketing and IT experienceproblems working together;.marketing manager support – narketing and uppermanagement work together; andIMC data quality is positively related to cross-sellingcapabilities, customer retention, customer satisfaction,and customer ROI.MethodBackground, questionnaire development, andprocedureAugmenting our review of the literature, qualitative researchwas conducted with 17 managers in five firms to betterunderstand how varied inter- and intra-organizational factorscould impact IMC data quality. A survey was then pretestedwith a sample of 43 business executives. Although the pre-testsample size was relatively small, a factor analysis revealed thatthe expected dimensionality existed with acceptableCronbach’s a values and provided confidence that thequestionnaire was understandable and offered face andcontent validity. As a result of the pre-test, a largerquantitative test in the financial services industry wasconducted. The financial services industry was selectedbecause many firms in this area, such as Royal Bank ofCanada, Charles Schwab, and others, have been cited forsuperior quality customer information use. A national mailinglist from Hoovers of 525 banking executives was used. Samplemembers received a postcard indicating that they wouldshortly be receiving a survey in the mail. The questionnairewas mailed one week later, and included a 2 participationincentive, and a self-addressed, stamped envelope.Respondents were given the option of mailing thequestionnaire back or completing the questionnaire onlinevia the attached URL. A second mailing was sent to nonrespondents approximately 14 days after the mailing wasdelivered. Finally, two graduate assistants called theTable I Demographic profile of respondents, firm characteristicsMeanPercentage of salesRetail salesExternal salesOnline salesOther sources5527711Sales/assets under management 50 million51-250 million250.1 million-1 billion1.1-5 billion 5 billion816233023Respondent age 3535-4445-5455 19273925Note: n ¼ 124; figures shown are percentages66

.Organizational processes for B2B services IMC data qualityJournal of Business & Industrial MarketingDebra Zahay, James Peltier, Anjala S. Krishen and Don E. SchultzVolume 29 · Number 1 · 2014 · 63 –74The reliability of the measures all reached satisfactory levelsfor theoretical model development; coefficient a (Hair et al.1998) exceeded the 0.7 benchmark (Nunnally, 1978) forreflective scales and as such established construct reliability.Table III provides the mean quality scores for each of theIMC data quality measures. Table IV shows the mean scoresfor each of items within the six organizational learning factors.Table V contains the mean customer metric scores.The means analysis by itself holds a number of importantimplications. First, although the mean IMC data qualityscores for all three dependent measures were above themidpoint, there is considerable room for improvement withrespect to IMC data quality metrics in the respondentorganizations. Second, regarding organizational learningfactors, only the marketing manager support dimension hada mean score of 4.0 or above, suggesting that although thereappears to be good intra-department communication andsupport in the organizations studied here, other areas need toimprove. There is room to develop more horizontalcommunication activities needed for developing qualityIMC data systems, particularly data sharing. Last, exceptfor customer satisfaction, the findings for the IMCperformance metrics suggest that current practices have notfully leveraged the value of having good data as a means ofcreating deep customer relationships. Finally, muchimprovement also seems to be needed in the area of sharingcustomer data.A regression analysis tested six of our hypotheses using thefactor scores for the six organizational dimensions asindependent variables and a summated data quality variableas the dependent variable. As can be seen from Table VI, theoverall model was highly significant, with five of the sixorganizational variables being significant predictors of overalldata quality (R2 ¼ 0.399, F ¼ 12.8, p , 0.001). IMC datavision (H1) had the largest impact on respondents’ perceptionof overall data quality (b ¼ 0.524, t ¼ 7.3, p , 0.001). Thenext most important organizational variables were datasharing (H6; b ¼ 0.203, t ¼ 2.8, p , 0.01) and marketing/IT cooperation (H3; b ¼ 0.185, t ¼ 2.6, p , 0.01). Theremaining two significant variables were marketing/IT systemintegration (H2; b ¼ 0.153, t ¼ 2.1, p , 0.05) and marketingmanager support (H5; b ¼ 0.149, t ¼ 2.1, p , 0.05).Marketing/IT conflict (H4) was not significant in our modelas relating to overall IMC data quality.Lastly, one-tailed correlations were calculated betweensummated data quality and each of the customer-performancemetrics. As shown in Table VII, H7, which predicts a positiverelationship between IMC data quality and varied customermetrics, is supported by this analysis. The highest correlationwas found between overall data quality and customersatisfaction (r ¼ 0.284, p , 0.001), followed by customerretention (r ¼ 0.24, p , 0.01), customer ROI (r ¼ 0.212,p , 0.01) and customer cross-selling (r ¼ 0.173, p , 0.05).Although not specifically hypothesized a priori, a summatedcustomer metric score was also calculated (a ¼ 0.70). Thisaggregated customer metric had a stronger positiveassociation with IMC data quality (r ¼ 0.301, p , 0.001)than the individual metrics.data sharing – data are integrated in a single database andis accessible by a single data query tool by those whorequire the information.To operationalize these variables, 26 questions related to ourIMC data quality framework were considered for inclusion asindependent variables in our survey across the dimensions ofIMC data vision, marketing/IT integration, marketing/ITcooperation, marketing-IT conflict, marketing managersupport, and data sharing. All questions for the scalesrelating to these variables were measured on a five pointLikert scale ranging from 1 ¼ strongly disagree to5 ¼ strongly agree. The specific questions used to developthese scales are noted in Table II.Dependent variablesData qualityData quality was conceptualized according to the work ofWang and Strong (1996) and based on the work of Zahay andGriffin (2004). In general, data are of high quality if they areperceived to be so by the users of data along relevantdimensions. Three questions were used to measure dataquality in this research: data accuracy, data consistency, andoverall data quality. Data quality perceptions were measuredon a five-point scale ranging from 1 ¼ poor to 5 ¼ excellent.Customer metricsUsing a five-point scale (“To the best of your knowledge,please rate your business units’ performance in the past 2-3years relative to the competition on the following: 1 ¼ lowerto 5 ¼ higher”), respondents rated their customer-basedperformance on four dimensions:1 cross-selling;2 customer retention;3 customer satisfaction; and4 ROI on a customer basis.A summated customer metric score across the four variableswas also calculated.ResultsTo assess dimensionality, the data were first subjected to anorthogonal principal components factor analysis. For theindependent variables, as shown in Table II and consistentwith expectations, six organizational factors loaded ashypothesized:1 marketing-IT integration (five items, a ¼ 0.93);2 IMC data vision (seven items, a ¼ 0.91);3 marketing manager support (four items, a ¼ 0.83);4 marketing/IT conflict (four items, a ¼ 0.73);5 data sharing (three items, a ¼ 0.88); and6 marketing/IT cooperation (two items, a ¼ 0.74).The three data quality items – data accuracy, dataconsistency, and overall data quality – were subjected to aprincipal components factor analysis to determinedimensionally. All three loaded on one factor (factorloadings of 0.93, 0.91, 0.96, variance explained ¼ 85percent). The three quality variables were summated tocreate a measure of overall customer data quality (a ¼ 0.88).The four customer metrics items – cross-selling, customerretention, customer satisfaction, and ROI on a customer basis– were analyzed individually.Discussion and conclusionsThis research suggests that one problem with themanagement and measurement of cross-channel IMC67

0.930.7410.7260.6690.773DatasharingCoefficient a reliability ting .714IMC datavisionMarketing is involved with IT in setting new project schedulesMarketing is involved with IT in generating new project ideasMarketing is involved with IT in setting new project goals/prioritiesMarketing provides customer requirements for new IT projectsMarketing/IT frequently discuss the quality of the customer dataMarketing is involved in developing new IT projectsThere is agreement of our “organizational vision” for managing customerinformationWe share our vision across the organization for how we manage customerinfoThere is a “common purpose” in the management of customer informationOur customer information management team is rewarded for goodperformanceWe use cross-functional teams when managing customer informationA team spirit pervades our ranks in the management of customerinformationThere is universal understanding of ho

Integrated Marketing Communications Department, Northwestern University, The Medill School, Evanston, Illinois, USA Abstract Purpose – The objective of this paper is to investigate IMC metrics in the lens of an inst