99 the integration of manufacturing and marketingsales decisions

Journal of Operations Management 20 (2002) 221–240

The integration of manufacturing and marketing/sales decisions:impact on organizational performance

Scott W. O’Leary-Kellya,∗, Benito E. Floresb,1

a Information Systems Department, Sam M. Walton College of Business, University of Arkansas, Fayetteville, AR 72701, USAb Department of Information and Operations Management, Lowry Mays College of Business, Texas A&M University,

College Station, TX 77843, USA

Abstract

Research in the areas of both manufacturing and marketing/sales have advocated the integration of several importantinterrelated decisions between the two functions (i.e. product development, process development, marketing/sales planning,and manufacturing planning decisions). The process of managing the strategic alignment between a firm’s business strategy,external environment, and the integration of manufacturing and marketing/sales decisions is very complex phenomenon thatrequires a level of analysis that has not occurred previously. This study examined the moderating effects of business strategyand demand uncertainty on the relationship between the integration of manufacturing and marketing/sales-based decisionsand organizational performance. The study found general support for the proposed model, suggesting that the impact of theintegration of manufacturing and marketing/sales decision on organizational performance is moderated by a firm’s businessstrategy and demand uncertainty. © 2002 Elsevier Science B.V. All rights reserved.

Keywords:Marketing/operations integration; Operations strategy; Empirical research

1. Introduction

The integration of key decision areas between man-ufacturing and marketing/sales is widely cited as ameans for gaining a competitive advantage in the mar-ketplace (e.g.Shapiro, 1977; Wheelwright and Hayes,1985; Nemetz and Fry, 1988; Konijnendijk, 1994).Although there is anecdotal support that integrationof decisions between these two functions may leadto increased organizational performance, there is littleempirical research to support this claim. In addition,most anecdotal studies tend to ignore the substantial

∗ Corresponding author. Tel.:+1-501-575-4035;fax:+1-501-575-4168.E-mail addresses:[email protected](S.W. O’Leary-Kelly), [email protected] (B.E. Flores).

1 Tel.: +1-409-845-4248.

costs associated with decision integration, such as thecosts resulting from added structural and infrastruc-tural mechanisms necessary for high levels of integra-tion (Galbraith, 1973; McCann and Galbraith, 1981;Thompson, 1967; Adler, 1995).

Therefore, it currently is not clear whether the ben-efits of integration always will exceed the costs. Thebasic premise of this study is that it may not be ben-eficial to integrate decisions between manufacturingand marketing/sales under all circumstances. Instead,the benefits will depend on the manufacturing andmarketing/sales decision area (e.g. product develop-ment as opposed to manufacturing planning), and thestrategic and environmental context within which thefirm is competing. This proposition is supported byan organizational contingency-based principle thatorganizational performance is dependent on the “fit”between an organization’s strategy, structure (e.g.

0272-6963/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved.PII: S0272-6963(02)00005-0

222 S.W. O’Leary-Kelly, B.E. Flores / Journal of Operations Management 20 (2002) 221–240

integration), and environment (Lenz, 1980, 1981;Miller, 1988). Specifically, this study will examinethe moderating effects that both a firm’s businessstrategy and environmental uncertainty have on therelationship between the integration of manufactur-ing and marketing/sales decisions and organizationalperformance.

2. Contingency perspective and functionalintegration

A contingency perspective is based on the princi-ple that organizational performance is dependent onthe “fit” between an organization’s strategy, structure,and environment (Preston, 1977; Lenz, 1980, 1981;Miller, 1988; Venkatraman, 1989a). That is, in orderto achieve high levels of organizational performance,certain business strategies require specific structuralforms, or specific environmental conditions necessi-tate certain structural designs. The contingency per-spective rejects the premise of simple, unconditionalassociations between an organization’s strategy, struc-ture, or environment, and organizational performance(Lenz, 1980, 1981; Miller, 1988). Cross-functional in-tegration among different departments represents animportant aspect of organizational structure in termsof the types of lateral relationships and the degreeof collaboration and participation that exists betweenthe different functions (Lawrence and Lorsch, 1967;Galbraith, 1973; Khandwalla, 1973; Hrebiniak andJoyce, 1984). Previous research supports the premisethat the relationship between functional integrationand organizational performance is moderated by afirm’s strategy and environment.

For example, the interaction between the level ofintegration and a firm’s business strategy is thought toinfluence performance (Miles and Snow, 1978; Porter,1985; Miller, 1988). Miller (1988) argued that forfirms stressing product innovation as a means of com-peting, functional integration would be strongly andpositively associated with organizational performance.Alternatively, he hypothesized that for firms focusedon low cost as a means of competing in the mar-ket, functional integration would be negatively associ-ated with organizational performance. Miller’s study,which included 89 service and manufacturing firms,showed general support for these hypotheses.

Similarly, Lawrence and Lorsch (1967)found thatthe need to integrate different functions varied fromone competitive environment to another. For exam-ple, in organizations where on-time delivery wascritical, it was imperative that manufacturing andmarketing/sales be closely integrated as opposed, e.g.marketing and engineering. In their study, the compa-nies that were able to integrate specific functions, asnecessitated by their competitive environment, outper-formed those companies that did not. These findingsare consistent with other work that has suggested thatthe need for functional integration is contingent on anorganization’s strategy and environmental uncertainty(Miles and Snow, 1978; Lenz, 1981; Porter, 1985).

3. Key decision areas between manufacturing andmarketing/sales

Several decisions often are cited as benefiting froma high level of integration between manufacturing andmarketing/sales (seeTable 1). These decisions arehighly interrelated, in that the decisions made by onefunction will have a direct impact on the decisionsand actions of the other function (Shapiro, 1977; Hill,1989; Wheelwright and Clark, 1992).

Product development decisions pertain to the deve-lopment of new products, as well as to modificationsof existing product designs. These decisions oftenplace unique requirements on the capabilities of afirm’s production system. For example, when Gen-eral Electric’s dishwasher product line underwent achange from metal to plastic tub-liners, this simplechange required an entirely new type of productionprocess capability. Critical to the successful change inthis product’s design was the inclusion of the manu-facturing function early in the product design decisionprocess (Wheelwright and Clark, 1992). Similarly,process development decisions pertain to the develop-ment and/or acquisition of new production processes,as well as modifications to existing production sys-tems. Process development decisions impact the mar-keting/sales function, in that these decisions eithercan constrain or open new avenues of product design.

Marketing/sales planning decisions involve man-aging product demand. These decisions include long-term demand forecasting, determination of salestargets, and timing of product/sales promotions.

S.W. O’Leary-Kelly, B.E. Flores / Journal of Operations Management 20 (2002) 221–240 223

Table 1Key decision areas between marketing/sales and manufacturing

Decision area Functional domain

Manufacturing Marketing/sales

Process and product development Determining changes to existing productionprocess capabilities

Determining changes in product designspecifications

Development of new production processescapabilities

Developing new product designspecifications

Manufacturing and marketing/sales planning

Determining long-term capacity requirements(resource planning)

Developing long-range demand forecast

Developing long-term production plans(production planning)

Developing sales plansDetermining the timing of productpromotions

Manufacturing planning decisions, on the other hand,pertain to decisions involving capacity planning andproduction scheduling. Decisions in this area directlyimpact marketing/sales’ ability to carry out theirmarketing/sales plan. For example,Braham (1987)relates the story of a company that geared up for amajor sales promotion. However, the marketing/salesdepartment was uninformed that the production plantwas scheduled to be shutdown for the month prior tothe promotion and as a result the company incurredheavy losses.

Table 2The integration of manufacturing and marketing/sales decisions: previous research

Decision areas Type of research

Conceptual Analytical Empirical

Process and productdevelopment

Blois (1980) Kim et al. (1992)Fitzsimmons et al. (1991)Hayes and Wheelwright (1979)Hill (1988)Konijnendijk (1994)Nemetz and Fry (1988)Shapiro (1977)Stalk and Hout (1990)Utterback and Abernathy (1975)Wheelwright and Clark (1992)Wheelwright and Hayes (1985)

Manufacturing andmarketing/sales planning

Spencer and Cox (1994) Crittenden (1992) Van Dierdonck and Miller (1980)Konijnendijk (1994) Damon and Schramm (1972)Palmatier and Shull (1989) Leitch (1974)Powers et al. (1988)Shapiro (1977)Shapiro et al. (1992)Vollmann et al. (1997)

3.1. Literature review—the integration ofmanufacturing and marketing/sales decisions

This section examines the literature that directlyaddresses the integration of the four manufacturingand marketing/sales decisions identified in the previ-ous section (seeTable 2 for a summary). Althoughthis topic has drawn substantial attention over the pastcouple of decades, very little research has directly fo-cused on the integration of decision areas involvingthe manufacturing–marketing interface. In addition,


the majority of the work in this area focuses on theproject or department level of an organization.

The conceptual research in this area implies thatincreased integration of the key decisions in manufac-turing and marketing/sales leads to increased organi-zational performance (e.g.Shapiro, 1977; Stalk andHout, 1990; Spencer and Cox, 1994). This suggeststhe presence of a universal law governing the rela-tionship between integration and performance. How-ever, as previously noted, this perspective has ignoredthe costs associated with integration (e.g. coordinat-ing mechanisms, such as expensive computer infor-mation systems, administrative time and overhead)which, when taken into account, would suggest thatincreasing the level of integration may not always bebeneficial to overall performance (Adler, 1995).

Several studies have utilized analytical approachesfor examining the relationships between manufactur-ing and marketing/sales decision integration and or-ganizational performance. The findings in several ofthese studies cast doubt on the existence of a uni-versal law. Studies byDamon and Schramm (1972)and Leitch (1974)utilized linear program models toexamine the relationship between the level of integra-tion of production planning decisions and organiza-tional performance. In their studies, they found thatdecision integration was positively related to organi-zational performance. Although their findings supportthe notion of a universal relationship between inte-gration and performance, caution is warranted whengeneralizing their results, as they were based on de-terministic models in which both the demand and pro-duction requirements were known with certainty.

However,Crittenden (1992), utilizing a simulationapproach, demonstrated that the relationship betweenthe level of integration of manufacturing planningdecisions and organizational performance was mod-erated by production environmental conditions. Theresults of this study indicated that under conditionsof constrained long-term capacity, the level of deci-sion integration was positively related to organiza-tional performance. In contrast, firms operating underconditions of excess long-term capacity showed norelationship between the level of integration and or-ganizational performance. The general findings ofthis study support a contingency-based relationshipbetween the level of integration of manufacturingplanning decisions and organizational performance.

Another analytical study byKim et al. (1992)examined the moderating effects of a firm’s envi-ronment on the relationship between integration ofmanufacturing and marketing/sales product-processdecisions and the firm’s performance. The authorstested their model under stochastic conditions withrespect to product demand using mathematical pro-gramming and simulation. They explored three sepa-rate dimensions of a firm’s environment: complexity,uncertainty, and tightness. They modeled environmen-tal complexity as a function of the degree of productand process variety exhibited by an organization. En-vironmental uncertainty was depicted as the degree offorecasting accuracy, and environmental tightness wascharacterized as the average contribution margin fora firm’s products. In support of a contingent relation-ship, they found that a firm’s level of environmentalcomplexity moderated the relationship between thelevel of integration and firm performance. Firms op-erating in very complex environments and engaged inhighly integrated decision making with regard to prod-uct and process development decisions outperformedthose that were not integrating decisions. With regardto firms operating in environments characterized bylow levels of complexity, no significant performancedifferences were found between high and low levelsof decision integration. This study also provides evi-dence of a moderated relationship between the degreeof decision integration and firm performance.

One of the only empirical studies in this area ofresearch was byVan Dierdonck and Miller (1980). Intheir study, they examined the influence of a firm’senvironment on the level of integration of decisionsrelated to production planning. They hypothesizedthat the degree of environmental complexity and un-certainty surrounding the production planning processwould influence the need to integrate production plan-ning decisions. Their results indicated that the degreeof environmental uncertainty was positively associatedwith the level of integration and that environmentalcomplexity was not related to integration. It is impor-tant to note that in their study, the level of integrationof production planning decisions was measured acrossall functions and not just manufacturing and market-ing/sales. However, given the high degree of interde-pendence between manufacturing and marketing/salesfunctions with respect to this decision area, one couldsafely assume that to a large degree the level of


integration between these two functions was reflectedin their measure.

In summary, very little systematic (empirical oranalytical) research has examined the relationshipbetween the integration of manufacturing and market-ing/sales decisions and organizational performance.However, results of several of these studies raiseserious questions regarding the validity of an uncon-ditional relationship between integration and organi-zational performance (e.g.Van Dierdonck and Miller,1980; Crittenden, 1992; Kim et al., 1992). In addi-tion, unlike most of the previous studies, we exam-ine the impact of the level of integration regardingmanufacturing and marketing/sales decisions at theorganizational level. That is, the impact of integrationbetween manufacturing and marketing/sales func-tions is to be assessed across strategic business units(SBUs), which include independently operated firms,including those owned by a parent company and jointventures, and individual divisions of corporations forwhich independent financial measures were reported.

4. Theoretical framework

The theoretical framework depicted inFig. 1 rep-resents the contingent relationships that are examinedin this study. This framework is consistent with thefindings of research from a contingency perspectivein the areas of strategic management and organiza-tion theory. It indicates that the relationship between

Fig. 1. Decision integration contingency model.

the level of integration regarding manufacturing andmarketing/sales decisions and organizational perfor-mance is moderated by a firm’s business strategy andthe demand uncertainty faced in its environment. Theresearch variables and hypotheses relating to this con-ceptual framework are discussed in detail in the fol-lowing sections.

4.1. Manufacturing and marketing/sales decisionintegration

Integration has been characterized in terms of:collaboration (Lawrence and Lorsch, 1967), cohesive-ness (O’Reilly et al., 1989), cooperation (Thomas,1976, 1992; Ettlie and Reza, 1992), coordination(Wheelwright and Hayes, 1985; Sundstrom andAltman, 1989; Ettlie and Reza, 1992), communica-tion/information exchange/interaction (Thomas, 1976,1992; Bonoma and Slevin, 1978; O’Reilly et al.,1989; Sundstrom and Altman, 1989), and mutualagreement/supportive actions (Thomas, 1976, 1992;Bonoma and Slevin, 1978; Wheelwright and Clark,1992). Collectively, these terms characterize inte-gration as being comprised of both interaction andagreement between two entities.

Similarly, Thomas (1976, 1992)described inte-gration to be a function of the cooperation and as-sertiveness between two entities; whereas cooperationcorresponds to the agreement of both parties to sup-port each others’ objectives, assertiveness relates tothe amount of effort extended by each party to achieve


its own goals. The more assertive each party is, themore they tend to interact. In this sense, a high level ofintegration is characterized by a high degree of bothcooperation and interaction. Paralleling prior defini-tions, in this study, the level of integration refers tothe extent to which separate parties work together ina cooperative manner to arrive at mutually acceptableoutcomes. Accordingly, this definition encompassesconstructs pertaining to the degree of cooperation,coordination, interaction, and collaboration.

This study will investigate the contingency rela-tionship regarding the integration of these four man-ufacturing and marketing/sales decisions previouslydiscussed: manufacturing planning decisions, mar-keting/sales planning decisions, product developmentdecision, and process development decisions.

4.2. Business strategy

A business strategy represents a pattern of decisionsregarding how an organization will compete in itsmarket. Research has strongly supported the propo-sition that a firm’s business strategy is comprised ofmultiple dimensions. Numerous studies have assessedan organization’s business strategy by measuring thecompetitive emphasis placed on several dominantstrategic dimensions (e.g.Miller, 1988; Venkatraman,1989b; Zahra and Covin, 1993).

The following strategic variables have been identi-fied in the literature as comprising important compo-nents of a firm’s business strategy: differentiation viaproduct innovation, cost leadership, superior productquality, on-time delivery, and breadth of product-lines.With the exception of on-time delivery, these strate-gic variables/dimensions have been used extensivelyto gauge a firm’s overall business strategy (Hambrick,1983; Porter, 1985; Miller, 1988; Zahra and Covin,1993). These variables, along with their relationshipto the different manufacturing and marketing/sales de-cisions, are detailed below.

4.2.1. Differentiation via product innovationProduct innovation is widely recognized as a com-

mon approach that companies choose in order to com-pete in the marketplace. Companies that adopt thisapproach tend to sell products that are on the cuttingedge of technology. They are characterized by theirextensive involvement in new product research and de-

velopment, and tend to be the first to market withnew product designs (Capon et al., 1992). Conceptualresearch has identified the integration of both productdevelopment and process development decisions asan essential element for companies to succeed whenadopting a product innovation strategy (Wheelwrightand Hayes, 1985; Wheelwright and Clark, 1992;Gerwin, 1993). It is proposed that firms that pursue aproduct innovation-based strategy, and simultaneouslyachieve high degrees of decision integration betweenmanufacturing and marketing/sales with respect toproduct development and process development de-cisions, will outperform those that do not integratethese decisions. Therefore, it is hypothesized that:

H1. For firms that adopt a business strategy of dif-ferentiation via production innovation, there will be apositive association between the level of decision inte-gration with respect to both product development andprocess development decisions and firm performance.For firms that do not adopt a strategy of differentia-tion via production innovation, there will not be anassociation between integration and performance.

4.2.2. Cost leadershipSince the work ofPorter (1980), cost leadership

has been recognized as an important alternative in theset of strategic choices available to firms. Companiesthat pursue a cost leadership strategy will emphasizecost reduction as a primary means of competing inthe market place. Containing production costs is anessential ingredient for manufacturing firms pursu-ing a cost leadership strategy (Porter, 1985; Zahraand Covin, 1993). Effective manufacturing and mar-keting/sales planning is essential to maintaining an ef-ficient production system. Studies show that increasedintegration between manufacturing and marketing/sales regarding both manufacturing and marketing/sales planning decisions can improve production effi-ciency and, consequently, firm performance (Damonand Schramm, 1972; Leitch, 1974; Shapiro, 1977).Therefore, it is hypothesized that:

H2. For firms that adopt a business strategy of costleadership, there will be a positive association betweenthe level of decision integration with respect to bothmarketing/sales planning and manufacturing planningdecisions and firm performance. For firms that do not


adopt a strategy of cost leadership, there will not bean association between integration and performance.

4.2.3. On-time deliveryOn-time delivery relates to the ability of a firm

to meet its delivery date commitments. Since 1988,on-time delivery has been ranked among the top threecompetitive priorities in the manufacturing futures sur-vey, a comprehensive survey of senior executives inleading US manufacturing firms (Kim, 1994). Compa-nies that compete via on-time deliveries emphasize thetracking of on-time delivery performance and utilizethis information in the evaluation of their firm’s overallperformance. A highly integrated relationship regard-ing manufacturing and marketing/sales planning deci-sions should enhance a firm’s ability to meet promiseddelivery dates for their customers (Shapiro, 1977).Consequently, it is hypothesized that:

H3. For firms that adopt a business strategy of on-timedelivery, there will be a positive association betweenthe level of integration with respect to both market-ing/sales planning and manufacturing planning deci-sions and firm performance. For firms that do not adopta strategy of on-time delivery, there will not be an as-sociation between integration and performance.

4.2.4. Superior product qualityCompanies that compete via superior product qua-

lity must devote a great deal of resources toward to-tal quality management. This would involve a highlevel of participation from both the workers and man-agement, as well as significant interaction with sup-pliers and customers (Garvin, 1988; Saraph et al.,1989). In addition, investment in organization-wideemployee training involving quality concepts wouldbe characteristic of these companies (Garvin, 1988;Saraph et al., 1989).

Several studies have proposed that product qualityis directly enhanced through the development of closerelationships between manufacturing and marketing/sales functions with regard to both product develop-ment and process development decisions (Hamilton,1991; Wheelwright and Clark, 1992). Wheelwrightand Clark (1992)demonstrated that a close relation-ship between manufacturing and marketing/sales notonly leads to improved product designs, but also toincreased efficiency in the production of those prod-

ucts, both of which lead to increased organizationalperformance. Therefore, it is hypothesized that:

H4. For firms that adopt a business strategy of supe-rior product quality, there will be a positive associationbetween the level of decision integration with respectto both product development and process developmentdecisions and firm performance. For firms that donot adopt a strategy of superior product quality, therewill not be an association between integration andperformance.

4.2.5. Product breadthProduct breadth pertains to the number of different

products a firm sells in the marketplace. Firms makea conscious choice whether to produce a narrow orbroad range of products. Take, e.g. a company likeHarley–Davidson that produces and sells a very nar-row range of products, focusing almost exclusivelyon heavyweight on-road motorcycles. Compare this toone of their major competitors, Yamaha, that producesa much wider variety of products; they not only pro-duce both on- and off-road types of motorcycles, butalso four wheel recreation vehicles, water jet skis, aswell as snowmobiles.

A firm that competes by producing a wide varietyof products increases the complexity of the decisionsassociated with both marketing/sales and manufactur-ing planning decisions. It is hypothesized that as acompany increases the variety of products it markets,it will require greater integration of both manufactur-ing planning and marketing/sales planning decisions.In line with this reasoning, it is hypothesized that:

H5. For firms that adopt a business strategy of pro-ducing a wider variety of products, there will be a pos-itive association between the level of integration withrespect to both marketing/sales planning and manufac-turing planning decisions and firm performance. Forfirms that do not adopt a strategy of producing a widervariety of products, there will not be an associationbetween integration and performance.

4.3. Perceived demand uncertainty

This study also focuses specifically on manage-ment’s perceptions concerning their ability to accu-rately predict product demand in their market


environment. The greater the level of uncertaintyfaced by a company regarding product demand, themore frequent the need to alter both manufactur-ing and marketing/sales plans to accommodate un-expected changes. For example, if marketing/salesmakes changes to its sales plans without consultingmanufacturing, this may have an adverse effect onmanufacturing’s performance and quite possibly onoverall organizational performance. In turn, the sameadverse effects can arise given the reverse situation,where manufacturing is making changes to its pro-duction plans without consulting marketing/sales. It istherefore postulated that, under conditions of greateruncertainty, organizations that are highly integratedwith respect to either marketing/sales plans or man-ufacturing plans will outperform companies that arenot integrated.

H6. For firms that experience high levels of demanduncertainty, there will be a positive association be-tween the level of integration with respect to bothmarketing/sales planning and manufacturing plan-ning decisions and firm performance. For firms thatexperience low levels of demand uncertainty, therewill not be an association between integration andperformance.

5. Research method

5.1. Data collection

The firms included in this study were sampledacross multiple industries (i.e. five two-digit SICcodes). These industries included: primary metal,fabricated metal products, industrial machinery andequipment, electronic equipment, and transportationequipment. Firms were selected from the Ward’sBusiness Directory of US Private and Public Compa-nies, and included only those that were located in thecentral US and had more than 200 employees.

For each SBU, data was collected from three keyinformants (a CEO level executive, a marketing/salesexecutive, and a manufacturing executive) via a mailedsurvey following the guidelines suggested byDillman(1978). A two-stage process was used to collect thedata. First, companies were contacted by a letter sentto the head of each firm. Interested respondents were

asked to complete and return participation forms pro-viding the information required to mail the surveysto the appropriate individuals within the firm. Next,a survey, a pre-paid return envelope, and a brief de-scription of the study were mailed to the three infor-mants within each company. The initial mailing of thesurveys was followed-up by two reminder cards sentto non-respondents after the second and fourth weeks.After 3 weeks, the initial contact letter, a second mail-ing was made to all non-respondents.

A total of 846 firms were contacted, of which 35were unable to participate for restructuring reasons(e.g. they were recently bought or sold, or they werebeing consolidated with another division) and 22non-deliverable letters were returned, providing an ef-fective sample size of 789 firms. A total of 121 firmsreturned usable surveys from all three informants.This represents an effective response rate of 15.3%which is comparable to other studies in which mul-tiple high level respondents were used (e.g.Phillips,1981; Phillips and Bagozzi, 1986). The vast majorityof informants held top level positions in their respec-tive areas and had been with the firm on average forover 13 years.

In order to assess whether a response bias waspresent, a Chi-square goodness-of-fit test was usedto compare the study sample to the original mail-ing sample with regard to: (1) the type of businessunit, (2) sales levels, and (3) number of employ-ees. All three tests failed to reject the null hy-pothesis of equal distributions at the 0.10 level. AKolmogorov–Smirnov one-sample test was also usedto compare the study and mailing sample regardingboth annual sales and the number of employees. Bothtests failed to reject the null hypothesis of equal dis-tributions at the 0.05 level. Although these resultshelp establish our confidence in a lack of responsebias, they do not preclude biases related to other fac-tors (e.g. firm performance) that were not tested dueto the lack of available data in the original mailingsample.

5.2. Sample

The general profile of the study sample is outlinedin Table 3. In order to limit the influence of factorsexternal to our study, we examined several factorsthat might influence a firm’s financial performance:


Table 3Organizational characteristics (n = 121)

Type of business unitPrivate firm (%) 34.1Public firm (%) 4.7Subsidiary of firm (%) 31.8Division of a firm (%) 28.7Joint venture (%) <0.1

Business product categoryConsumer goods (%) 19.4Capital goods (%) 43.4Raw or semi-finished goods (%) 5.4Components for other finished goods (%) 28.7Supplies (%) 3.1

Sales range$ 4–29 million (%) 17.8$ 30–50 million (%) 23.2$ 51–90 million (%) 20.2$ 91–158 million (%) 20.2$ 159–850 million (%) 18.6

Number of employees200–300 (%) 28.6400–500 (%) 16.3600–700 (%) 16.3800–1000 (%) 18.61100–4700 (%) 20.2

organizational size, type of industry, type of productmarket, and type of firm.

The effect of firm size on the financial performanceof an organization is not well established. The mostcompelling evidence against the existence of a rela-tionship came in a meta-analytic study byCapon et al.(1990). Their results indicated that firm size had nosignificant impact on a firm’s financial performance.In the present study, the association between firmsize (as measured by the reported number of em-ployees) and performance (as measured by perceivedprofitability) was assessed using correlation analysis.The results indicated that there was no significantassociation between firm size and performance in thissample.

The type of industry in which a firm competes alsomay have an impact on a firm’s performance. In thisstudy, firms were drawn from five different two-digitSIC codes. The effect of industry type, as character-ized by two-digit SIC codes, on a firm’s financial per-formance was analyzed using an unbalanced ANOVAmodel. The results indicate that perceived profitability

is not influenced by industry type. In addition, the ef-fects related to both the type of products marketed byfirms and the type of firm also were examined usingan unbalanced ANOVA model. The results also indi-cate that perceived profitability is not influenced byeither product type or firm type.

5.3. The measures

The measures (except the self-reported ROI) usedto operationalize the constructs in the study were com-prised of multiple items, each based on a seven-pointLikert scale. In addition, all constructs were measuredacross multiple key informants. Key informants wereselected on the basis of their specialized knowledgerelevant to the study (Kumar and Stern, 1993). Theuse of key informants to collect data is an effectivemeans for measuring organizational constructs and hasbeen used extensively in business strategy research(Phillips, 1981; Kumar and Stern, 1993). In additionto relying on previous research to ensure the contentvalidity of the measures, they also were reviewed byseveral “expert” judges (e.g. current and previous man-agers with manufacturing experience) and adjustmentswere made based on their recommendations.

5.3.1. Business strategyBusiness strategy was operationalized via five di-

mensions: differentiation via product innovation, costleadership, superior product quality, on-time delivery,and product breadth. The items used to measure afirm’s business strategy are based on past research.Specifically, differentiation via product innovationcaptures the extent to which a firm competes on boththe number and rate of product innovations they in-troduce into the marketplace. The measurement itemsused for this construct are based on the work ofCooper (1987), Miller (1988), Capon et al. (1992),andZahra and Covin (1993).

The cost leadership variable stems from the workof Porter (1980, 1985), which highlighted low cost asa central strategic position that could be adopted bya firm. The items used in this study were primarilyadopted fromZahra and Covin’s (1993)cost leader-ship scale. The items used to measure the superiorproduct quality variable are based on eight differentsub-dimensions associated with quality managementas identified bySaraph et al. (1989). The items used


to measure on-time delivery draw upon the work ofKumar and Sharman (1992)andBerry et al. (1990),in which they identify several factors that are impor-tant for differentiating firms with regard to on-timedelivery. Finally, the composite measure for productbreadth was based on scales developed by bothMiller(1988)andZahra and Covin (1993).

5.3.2. Perceived demand uncertaintyThe perceived demand uncertainty measure is

largely based on work byGerloff et al. (1991). Specif-ically, their items pertaining to the perceived state ofenvironmental uncertainty were adopted. They de-fined “state of uncertainty” as management’s percep-tion regarding the unpredictability of some elementin the environment, such as demand.

5.3.3. Decision integration between manufacturingand marketing/sales

Several factors that are indicative of the level of in-tegration between two groups are the level of coop-eration (Ettlie and Reza, 1992), open communication(O’Reilly et al., 1989; Sundstrom and Altman, 1989),collaboration (Lawrence and Lorsch, 1967), and mu-tual agreement (Thomas, 1976, 1992). These indica-tors have been incorporated into the items used tooperationalize the level of integration between man-ufacturing and marketing/sales with respect to eachof the following decision areas: product development,marketing planning, process development, and manu-facturing planning.

5.3.4. Organizational performanceIn this study, we collected both perceptual and

objective self-reported financial performance data.For the objective ROI data there was a much lowerresponse rate across all three informants and thesample size was significantly smaller. Therefore, weused only the perceptual measure of profitability inthis study (analysis involving the smaller sample ROIdataset did not support our hypotheses, perhaps due toreduced levels of power). Past studies have shown thatperceptual measures are reliable and strongly correlatewith objective measures (Dess and Robinson, 1984;Venkatraman and Ramanujam, 1987). The correlationbetween the perceptual and objective ROI measuresfor this study was 0.43 (significant at the 0.001 level)and was comparable to those found in other studies

by Dess and Davis (1984)and Venkatraman andRamanujam (1987).

The items used to measure the perceived profitabil-ity of a firm were based onVenkatraman’s (1989b)profitability measure. The informants were asked torate their organization’s performance along several di-mensions relative to the performance of their majorcompetitors.Venkatraman (1989b)concluded that theitems used in his perceived profitability measure sat-isfied the general measurement property requirementsat the mono-method level of analysis.

As in many studies, not all informants within afirm were willing to provide performance data (Dessand Robinson, 1984; Venkatraman and Ramanujam,1987). This presented a problem in that if only firmsfor which all three informants provided data were in-cluded, the sample size would have been substantiallyreduced. On the other hand, using the ratings of onlya single informant might lead to measures that are bi-ased. In order to avoid using measures based on a sin-gle informant and to limit the reduction of the samplesize, firms for which two or more informants reportedperceptual performance data were used in this study.

5.4. Construct refinement and validation

Although the use of both multiple items and multi-ple informants increases the reliability of the measu-res, it is necessary to assess both the convergentvalidity and discriminant validity of the measures(Malhotra and Grover, 1998; O’Leary-Kelly andVokurka, 1998). When measures are comprised ofmultiple informants (or methods), convergent anddiscriminant validity should be assessed at both themono- and multi-method levels (Phillips and Bagozzi,1986). The mono-method level of analysis assessesproperties within informants, whereas the multi-method level of analysis assesses properties acrossinformants.

5.4.1. Convergent validityConvergent validity first must be established before

the discriminant validity of the measures may be as-sessed. At the mono-method level, convergent valid-ity pertains to the degree to which measures are bothunidimensional and internally consistent (Phillips andBagozzi, 1986). Unidimensionality, which determinesthat items are associated with only a single variable,


must first be established before the internal consis-tency of a measure may be assessed. Because theconstructs were measured across multiple informantswithin each firm, the added burden of establishing thatthe measures are the same (i.e. have the same factorweights) across all informants also must be met.

Unidimensionality was assessed using a confirma-tory factor analysis (CFA) model. In order to establishstandard-unidimensional measures across all threeinformants, it was necessary to “re-specify” severalof the measures. In order for a construct to be unidi-mensional, the factor loadings for all items associatedwith the construct must be significant and the modelmust be correctly specified, as indicated by acceptablegoodness-of-fit values (Gerbing and Anderson, 1988;O’Leary-Kelly and Vokurka, 1998). Following thesuggestions ofAnderson and Gerbing (1982, 1988),items that did not have significant factor loadingsacross all three informants were eliminated. Althoughthis approach yields a set of standard measures foranalysis, the limitation is that the set of measures maybe less generalizable.

The next step involved establishing that the con-structs were based on identically weighted items(same item factor loadings) across all informants. Thiswas accomplished by testing the hypothesis that factoritems were equal across informants, using LISREL’smulti-sample analysis (Jöreskog and Sorbom, 1989).The results of this analysis indicated strong supportfor equal factor weights across all informants and fur-ther demonstrates that the informants are providingratings of the same construct.

Cronbach’s (1951)coefficient α and Werts et al.(1974) composite coefficient of reliability (ρc) wereused to assess the reliability of the measures. Althoughthere are no definitive rules regarding reliability, scaleswith α of 0.70 or higher (Nunnally, 1978), andρc of0.50 or higher are generally considered to be reliable(Werts et al., 1974). Expect for the cost leadershipscales, all of the measures had aCronbach’s (1951)αand a composite reliability above 0.7 (values rangedfrom 0.71 to 0.92). For the cost leadership scales, bothof the reliability coefficients ranged from 0.49 to 0.55.Although these values are low,Van de Ven and Ferry(1980) have argued that for broader measures, thosecomprised of several conceptually distinct terms (asin the case of the cost leadership measure),α valuesaround 0.50 are acceptable.

Convergent validity at the multi-method level ofanalysis refers to the degree of agreement between in-formants with respect to a particular construct. Thiscan be assessed by partitioning the variance of the ob-served measures into the variance due to the traits (orthe underlying constructs being measured), the methodof measurement (or the different key informants), andrandom error. The assessment of convergent validityinvolves testing a series of nested CFA models to deter-mine whether informant bias significantly impacts theconstruct measures. It is beyond the scope of this paperto provide a detailed discussion of this methodology,but the interested reader may refer toO’Leary-Kellyand Vokurka (1998)for a thorough discussion of thisapproach.

Because of the limited sample size and consistentwith other studies (e.g.Phillips and Bagozzi, 1986),the individual informant’s weighted factor scores foreach construct were used as the indicators in the CFAmodel. In addition, the exogenous variables were splitinto two groups: moderating variables (business strat-egy variables and demand uncertainty) and predic-tor variables (decision integration variables) and wereevaluated separately. The performance (endogenous)variables also were evaluated separately against theROI data collected in the study.

The partitioned variance and the respective signif-icance levels are reported inTable 4. With regard tothe moderating variables, all trait variances across in-formants were significant at the 0.005 level, indicatingstrong support for convergent validity (Bagozzi et al.,1991). It is worth noting that several of the strategyvariables display relatively large method (informantbias) and error variances, however, there does not ap-pear to be any systematic pattern across variables thatwould lead to the conclusion that one informant is bet-ter than another. Under these circumstances, there areno clear guidelines for combining measures. What isclear, is that using the ratings of a single informantwill increase the degree of informant bias (Phillipsand Bagozzi, 1986). Therefore, given that all trait fac-tors were significant for the moderating variables (andperformance variables), and that no systematic biasor error patterns were discernable across informants,the measures were combined across informants for themoderating variables (Phillips and Bagozzi, 1986).

The results involving the integration measures pro-vide a very different picture (seeTable 4). For these


Table 4Partitioning of variance due to trait, method (informant), and error

Measure (informant) Variance components

Trait Method Error

Moderating variablesa

Product innovation (CEO) 0.42∗∗ 0.06∗ 0.52Product innovation (marketing) 0.61∗∗ 0.09∗∗ 0.30Product innovation (manufacturing) 0.40∗∗ 0.07∗ 0.53Cost leadership (CEO) 0.53∗∗ 0.01 0.46Cost leadership (marketing) 0.30∗∗ 0.00 0.70Cost leadership (manufacturing) 0.42∗∗ 0.00 0.58Superior quality (CEO) 0.28∗∗ 0.37∗∗ 0.35Superior quality (marketing) 0.14∗∗ 0.53∗∗ 0.33Superior quality (manufacturing) 0.35∗∗ 0.46∗∗ 0.19On-time delivery (CEO) 0.58∗∗ 0.22∗∗ 0.20On-time delivery (marketing) 0.35∗∗ 0.22∗∗ 0.43On-time delivery (manufacturing) 0.33∗∗ 0.18∗∗ 0.49Product breadth (CEO) 0.64∗∗ 0.00 0.36Product breadth (marketing) 0.25∗∗ 0.01 0.74Product breadth (manufacturing) 0.21∗∗ 0.01 0.78Demand uncertainty (CEO) 0.38∗∗ 0.12∗∗ 0.50Demand uncertainty (marketing) 0.51∗∗ 0.03 0.46

Integration variablesProduct development decisions (marketing) 0.03 0.48∗∗ 0.49Product development decisions (manufacturing) 0.76∗∗ 0.20∗∗ 0.04Marketing planning decisions (marketing) 0.07∗∗ 0.65∗∗ 0.28Marketing planning decisions (manufacturing) 0.61∗∗ 0.26∗∗ 0.13Process development decisions (marketing) 0.30∗ 0.45∗∗ 0.25Process development decisions (manufacturing) 0.02 0.70∗∗ 0.28Manufacturing planning decisions (marketing) 0.46∗∗ 0.40∗∗ 0.14Manufacturing planning decisions (manufacturing) 0.05∗ 0.66∗∗ 0.29

Performance variablesPerceived profitability (CEO) 0.51∗∗ 0.07∗ 0.42Perceived profitability (marketing) 0.86∗∗ 0.05∗ 0.09Perceived profitability (manufacturing) 0.54∗∗ 0.07∗ 0.39

a As noted byCampbell and Fiske (1959), only traits with the same informants should be tested together. Because the demanduncertainty variable was not measured by the manufacturing informant, a separate set of models were tested using only the CEO andmarketing/sales informant. The results reported for demand uncertainty are based on this separate set of tests (it should be noted that thevariance for the other constructs were all within one standard error of the models ran for the strategy variables using all three informants).

∗ Significant at the 0.05 level.∗∗ Significant at the 0.01 level.

measures, the criterion for convergent validity (signif-icant trait factors across informants for a given mea-sure) was not met, thus, indicating a significant lackof agreement between the two informants (manufac-turing and marketing/sales). The one exception is withrespect to the marketing/sales planning decision inte-gration measure, where the trait factors are significantfor both informants. However, the trait factor for themarketing/sales respondent is only marginally signifi-

cant and accounts for only 6% of the measure’s totalvariance; therefore, little support for convergent valid-ity is evident.

Based on the lack of convergent validity, there is nobasis for combining measures across the marketing/sales and manufacturing informants. Instead, it isnecessary to assess the measures for the two respon-dents separately. With regard to the manufacturingrespondent, the trait variances for the two marketing/


sales-based decisions (product development and mar-keting/sales planning) are significant. Similarly, withregard to the marketing/sales respondent, the traitvariances for the two manufacturing-based decisions(process development and manufacturing planning)are significant. Based on these strong results, we usedthe manufacturing-based responses to measure thelevel of integration regarding product developmentand marketing/sales planning decisions. Likewise, weused the marketing/sales-based responses to measurethe level of integration regarding process developmentand manufacturing planning decisions.

The results for the integration variables are not al-together unexpected. Other studies have found sim-ilar results for marketing/sales and R&D functions,leading researchers to conclude that members of thesefunctions often do not share the same perceptions re-garding the level of integration between the two func-tions (Gupta et al., 1985; Saghafi et al., 1990; Souderand Sherman, 1993). For example, because market-ing/sales tends to have greater control over the out-come of product development decisions, they mayhave more of a biased view regarding their own will-ingness to cooperate and involve (i.e. integrate) man-ufacturing into the decision process.

Finally, the analysis of the perceived performancemeasures strongly support the attainment of conver-gent validity. Specifically, all trait variances across in-formants were significant at the 0.005 level, while themethods variances were all near zero, strongly indi-cating a lack of informant bias. Based on these results,it was deemed appropriate for the perceived perfor-mance measures to be combined across informants.

5.4.2. Discriminant validityDiscriminant validity refers to the extent to which

the different measures are unique. Two methods wereused to assess discriminant validity at both the mono-and multi-method level. First, discriminant validitywas assessed by measuring the correlation betweenall constructs and evaluating whether they were dif-ferent from 1.0 (O’Leary-Kelly and Vokurka, 1998).All variables included in the study were determinedto be significantly different from 1.0. This model wasthen compared to one in which the correlation co-efficient (ϕij ) was constrained to 1.0 (representingunity between constructs). A Chi-square differencetest between the values obtained for the constrained

and unconstrained models was used to assess dis-criminant validity. A significantly lower Chi-squarevalue for the unconstrained model provides support fordiscriminant validity (Anderson and Gerbing, 1988).All Chi-square difference tests were significant at the0.005 level, indicating strong support for discriminantvalidity at the mono-method level. Second, a moreconservative test of discriminant validity requires thatthe average variance extracted for each construct ex-ceed the squared correlation between that constructand all others in the model (Fornell and Larcker, 1981).All of the constructs met this criteria, providing ad-ditional evidence of acceptable discriminant validityamong the constructs.

In summary, all construct measures were deter-mined to exhibit acceptable levels of convergent anddiscriminant validity at the mono-method level of anal-ysis. With the exception of the integration variables,all independent variable measures were determinedto have sufficient levels of convergent and discrim-inant validity at the multi-method level of analysis.Based on these results, the average summated scores(i.e. the average of the item scores for each variable)were used as a composite measure for each construct(Hair et al., 1992).

With regard to the integration variables, which didnot meet the convergent validity criteria at the multi-method level of analysis, the composite measures foreach integration variable were based on a single infor-mant. That is, the measures of process developmentand manufacturing planning decision integration werederived from the data provided by the marketing/salesinformant. Likewise, the measures of product devel-opment and marketing/sales planning decision inte-gration were derived from the data reported by themanufacturing informants. Summary statistics for thefinal variables are reported inTable 5.

5.5. Data analysis

Moderator median split regression analyses (MM-SRA) were used to analyze the data (Arnold, 1982).MMSRA involves splitting the sample into twosub-samples (high and low) with respect to the me-dian value of the moderating variable. The criterionvariable (perceived profitability) is then regressed onthe predictor variable (e.g. product development de-cision integration or manufacturing planning decision

234S

.W.

O’L

ea

ry-Ke

lly,B

.E.

Flo

res

/Jou

rna

lo

fO

pe

ratio

ns

Ma

nage

me

nt

20

(20

02

)2

21

–2

40


integration) for each sample. A significant differencebetween the slope of the predictor variable in each ofthe sub-samples indicates the existence of an interac-tion effect. This was tested using the differencet-testoutlined byJaccard et al. (1990).

To control for the influence of a firm’s businessstrategy and demand uncertainty on organizationalperformance, these variables also were included in thelinear regression model in addition to the predictorvariable.

6. Results

The results for the moderated median split regres-sion analyses are reported inTable 6. Included are theβ coefficients of the integration variable (i.e. predic-tor variable) pertaining to each hypothesis, the signifi-cance of theβ coefficient, and theP-values associatedwith the differencet-test of theβ coefficients in eachsub-sample. Support for a hypothesis exists when theanalysis provides evidence of both a significantβ co-efficient in the predicted direction and a significantdifferencet-test. Partial support for a moderated rela-tionship exists when one of theβ coefficients is signif-icant, but the difference between the two sub-samplesis non-significant.

6.1. Effects of integrating marketing/sales-baseddecisions

Marketing/sales-based decisions include both prod-uct development and marketing/sales planning (high-lighted by the shaded areas inTable 6). The resultsinvolving the marketing/sales based decisions mostlysupport the hypotheses, i.e. a moderated relationshipbetween the level of decision integration and perfor-mance. Specifically, four of the six hypotheses werestrongly supported, indicating that the effect of mar-keting/sales based decision integration on performanceis dependent on the business strategy and demand un-certainty faced by the firm.

For example, for firms pursuing a strategy of eitherproduct innovation (H1a) or superior quality (H4a),increasing the level of integration with regard toproduct development decisions yielded improved per-formance. Likewise, for firms that pursued a strategyof on-time delivery (H3a), higher levels of product

development decision integration were associatedwith higher performance as measured by perceivedprofitability. Higher performance also was associatedwith higher marketing/sales decision integration forthose firms facing higher levels of demand uncer-tainty (H6a). We found no support for a relationshipbetween marketing/sales decision integration and firmperformance for firms pursuing a strategy of costleadership (H2a) or product breadth (H5a).

6.2. Effects of integrating manufacturing-baseddecisions

Manufacturing-based decisions involve both pro-cess development decisions and manufacturing plan-ning decisions. These results also tend to supporta moderated relationship between the level of deci-sion integration and performance, however, they werenot in the predicted form. For example, the resultsdemonstrated that for firms emphasizing a strategyof product innovation (H1b), higher levels of processdevelopment decision integration were not associatedwith improved firm performance. Instead, we foundthat for the those firms that did not focus on productinnovation, a higher level of process development de-cision integration was associated with improved firmperformance. Similarly, for firms that competed viaon-time delivery (H3b), or that emphasize a strategyof product breadth (H5b), higher levels of integrationregarding manufacturing planning decisions were notassociated with higher firm performance; instead, weagain found that for those firms that did not focuson either on-time delivery or product breadth, higherlevels of integration were associated with higher firmperformance.

Finally, there was partial support in the oppositedirection regarding the moderating effects of demanduncertainty (H6b) on the relationship between theintegration of manufacturing planning decisions andfirm performance. In this case, for those firms thatfaced lower levels of demand uncertainty, the level ofintegration was significantly associated with higherfirm performance, but higher levels of integrationwere not associated with higher firm performancefor those firms facing higher levels of demand un-certainty. However, there was only partial supportfor an integration–performance relationship, in thatthe β coefficients across the two sub-samples (high

236S

.W.

O’L

ea

ry-Ke

lly,B

.E.

Flo

res

/Jou

rna

lo

fO

pe

ratio

ns

Ma

nage

me

nt

20

(20

02

)2

21

–2

40


and low demand uncertainty) were not significantlydifferent.

7. Discussion

The results of this study support the principal ar-gument guiding this paper, that is the relationshipbetween the integration of manufacturing and mar-keting/sales decisions and firm performance ismoderated by a firm’s business strategy and by envi-ronmental uncertainty. An interesting finding of thisstudy was the differentiated results regarding the formof the moderated relationship. Specifically, the re-sults were found to vary with respect to both the typeof decision being analyzed (marketing/sales-basedversus manufacturing-based decisions), and the keyinformant used to measure integration (manufacturingrespondent versus marking/sales respondent).

In regard to the former, the results supporting thehypothesized form of the moderating relationshipsall involved marketing/sales-based decisions (productdevelopment and marketing/sales planning), and thosethat were in the opposite form than hypothesized inthe study all involved manufacturing-based decisions(process development and manufacturing planning).One possible explanation is the “time differential” thatexists between marketing/sales and manufacturingdecisions, in that marketing/sales-based decisions aretypically a source of input for the manufacturing-baseddecisions. For example, in a typical marketing/sales—manufacturing planning cycle, the marketing/salesplanning decisions serve as a primary input for themanufacturing planning decisions which then follow(seeFig. 2) (Vollmann et al., 1997). Here, it is moreadvantageous for decision integration to occur at themarketing/sales planning stage. If manufacturing hasa greater opportunity to review and identify potential

Fig. 2. Marketing/sales–manufacturing planning cycle.

problems for the manufacturing plan, this should leadto more efficient and effective planning overall. Onthe other hand, if a firm waits to integrate at the man-ufacturing planning stage, then manufacturing hasless lead-time available to make changes in responseto the marketing/sales plan. In addition, it is likelythat the firm already would have started to implementthe marketing/sales plan, and this would leave little orno opportunity for implementing suggested improve-ments by manufacturing. Consequently, it is likely thatthis would lead to less efficient and effective plans.

This same framework applies to the decisions maderegarding product development and process develop-ment. In many situations, changes to the productionprocess are in response to changes made in the prod-uct design phase (Wheelwright and Clark, 1992). In-tegration at the product design stage offers a couple ofimportant advantages over integration at the pro-cess development stage. First, it gives manufacturinggreater lead-time to make the required changes inthe production process. Second, it enables manufac-turing to provide critical input at the initial phase ofnew product designs, thereby giving marketing/salesenough time to incorporate the changes into the finaldesign. Numerous studies have shown that it is bothcostly and time consuming to change the product de-sign at later points in the development process (e.g.Wheelwright and Clark, 1992). The implication is thatfor firms emphasizing product innovation or superiorquality, integrating product development decisions ismore effective than integrating process developmentdecisions.

These findings offer important insights for manage-ment. Specifically, integrating marketing/sales-baseddecisions (as opposed to manufacturing-based deci-sions) provided the clearest benefits when coupledwith specific strategies and degrees of demand un-certainty. For example, it was shown that for firms


engaged in a strategy based on product innovation,integrating product development decisions betweenmanufacturing and marketing/sales improved firmperformance. However, for these companies, integrat-ing process development decisions did not lead toimproved performance. Similarly, for firms that fo-cused on a strategy of on-time delivery, increasing thelevel of marketing/sales planning decision integrationled to improved firm performance, whereas increas-ing manufacturing planning decision integration didnot.

As previously noted, the results of the study alsosystematically varied with regard to the informant usedto assess integration. Specifically, the results support-ing the hypothesized form of the moderating rela-tionships all involved integration measures providedby the manufacturing respondent (regarding the inte-gration of marketing/sales-based decisions), whereasthose that were in the opposite form hypothesized inthe study all involved integration measures providedby the marketing/sales respondent (regarding the inte-gration of manufacturing-based decision).

The fact that there were significant differences inperceptions involving the level of integration acrossinformants is instructive regarding the systematicdifferences in the results reported in this study. Itmay be that these results are related to differences inperceptual anchors across the two groups of respon-dents. Specifically, the decision integration measuresrequired each respondent to rate a series of questionson a seven-point Likert scale anchored by “stronglydisagree” and “strongly agree”. As an example, oneof the items was “manufacturing and marketing/salesfrequently discuss the details surrounding productdevelopment decisions”. In response to this question,a respondent will have a preconceived notion as towhat constitutes “frequently”. This preconceived no-tion serves as a perceptual anchor. If the respondentsin each group (manufacturing and marketing/sales)have different conceptions as to what constitutes“frequently”, this could lead to systematic differencesin the results, such as those reported in this study.Unfortunately, we do not have the data to examinethis potential cause, but it is an area that should beexamined closely in future studies.

However, these findings may have important man-agerial implications with regard to the ability of firmsto effectively monitor the degree to which they have

successfully achieved decision integration. In tryingto assess the level of integration for the different deci-sion areas, managers should not expect to get similarresponses across the two functional areas. Rather,the marketing/sales and manufacturing functions arelikely to have very different opinions when it comesto evaluating the levels of achievement regarding de-cision integration. Perhaps the most reliable approachto monitoring the level of integration is to rely onthe “outside” function to gauge the level of decisionintegration achieved across the two functions. Thatis, manufacturing informants should be used to gaugethe level of integration regarding both marketing/salesplanning and product development decisions. Like-wise, marketing/sales informants should be reliedupon to provide an assessment of the level of integra-tion regarding manufacturing planning and processdevelopment decisions.

8. Concluding remarks

Overall, this study has challenged conventionalthinking regarding the impact of decision integrationbetween manufacturing and marketing/sales on orga-nizational performance. The results clearly demon-strate that the effect of decision integration betweenmanufacturing and marketing/sales on firm perfor-mance is much more complex than has been suggestedpreviously. Second, the results indicate that market-ing/sales and manufacturing respondents have verydifferent perceptions regarding the level of integrationfor decision areas that they traditionally control; thishas important implications for both researchers andmanagers who desire to assess the level of decisionintegration. Finally, this study found very differentforms of moderating effects associated with decisionintegration for the two functional areas. Although, itpresently is impossible to attribute these differencesto a specific factor, some suggestions for a potentialstarting point for future research into this interestingphenomenon were presented.

References

Adler, P.S., 1995. Interdepartmental interdependence and coordi-nation: the cast of the design/manufacturing interface. Organi-zational Science 6 (2), 147–167.


Anderson, J.C., Gerbing, D.W., 1982. Some methods for re-specifying measurement models to obtain unidimensional cons-truct measurement. Journal of Marketing Research 14, 453–460.

Anderson, J.C., Gerbing, D.W., 1988. Structural equation modelingin practice: a review and recommended two-step approach.Psychological Bulletin 103 (3), 411–423.

Arnold, H.J., 1982. Moderator variables: a clarification of concep-tual, analytical and psychometric issues. Organizational Beha-viour and Human Performance 29, 143–147.

Bagozzi, R.P., Yi, Y., Phillips, L.W., 1991. Assessing construct vali-dity in organizational research. Administrative Science Quar-terly 36, 421–458.

Berry, L.L., Zeithaml, V.A., Parasuraman, A., 1990. Five impera-tives for improving service quality. Sloan Management Journal31 (4), 29–38.

Blois, K.J., 1980. The manufacturing/marketing orientation and itsinformation needs. European Journal of Marketing 14, 354–364.

Bonoma, T.V., Slevin, D.P., 1978. Executive Survival Manual.Wadsworth, Belmont, CA.

Braham, J., 1987. The marriage of marketing and manufacturing.Industry Week 233, 41–44.

Capon, N., Farley, J.U., Hoenig, S., 1990. Determinants of financialperformance: a meta-analysis. Management Science 36 (10),1143–1159.

Campbell, D.T., Fiske, D.W., 1959. Convergent and discriminantvalidation by the multitrait–multimethod matrix. PsychologicalBulletin 58, 81–105.

Capon, N., Farley, J.U., Lehmann, D.R., Hulbert, J.M., 1992.Profiles of product innovators among large US manufactures.Management Science 38, 157–169.

Cooper, R., 1987. Defining the new product strategy. IEEE Transac-tions on Engineering Management 34, 184–193.

Crittenden, V.L., 1992. Close the marketing/manufacturing gap.Sloan Management Review 33, 41–52.

Cronbach, L.J., 1951. Coefficient alpha and internal structure oftests. Psychometrika 16, 297–334.

Damon, W.W., Schramm, R., 1972. A simultaneous decision modelfor production, marketing and finance. Management Science19, 161–172.

Dess, G., Davis, P., 1984. Porter’s generic strategies as determi-nants of strategic group membership and organizational perfor-mance. Academy of Management Journal 27 (3), 467–488.

Dess, G., Robinson Jr., R.B., 1984. Measuring organizational per-formance in the absence of objective measures: the case of theprivately held firm and conglomerate business unit. StrategicManagement Journal 5, 265–273.

Dillman, D.A., 1978. Mail and Telephone Surveys. Wiley, NewYork, NY.

Ettlie, J.E., Reza, E.M., 1992. Organizational integration and pro-cess innovation. Academy of Management Journal 35, 795–827.

Fitzsimmons, J.A., Kouvelis, P., Mallick, D.N., 1991. Designstrategy and its interface with manufacturing and marketing: aconceptual framework. Journal of Operations Management 10,398–415.

Fornell, C., Larcker, D.F., 1981. Evaluation structural equationmodels with unobservable variables and measurement error.Journal of Marketing Research 18, 39–50.

Galbraith, J., 1973. Designing Complex Organizations. Addison-Wesley, Reading, MA.

Garvin, D.A., 1988. Managing Quality: The Strategic and Compe-titive Edge. Free Press, New York, NY.

Gerbing, D.W., Anderson, J.C., 1988. An updated paradigm forscale development incorporating unidimensionality and its asse-ssment. Journal of Marketing Research 25, 186–192.

Gerloff, E.A., Muir, N.K., Bodensteiner, W.D., 1991. Three compo-nents of perceived environmental uncertainty: an exploratoryanalysis of the effects of aggregation. Journal of Management17 (4), 749–768.

Gerwin, D., 1993. Integrating manufacturing into the strategic pha-ses of new product development. California Management Re-view 35 (4), 123–136.

Gupta, A., Raj, S., Wilemon, D., 1985. The R&D/marketing inter-face in high technology firms. Journal of Product InnovationManagement 2, 12–24.

Hair, J.F., Anderson, R.E., Tatham, R.L., Black, W.C., 1992. Multi-variate Analysis. Macmillan, New York, NY.

Hambrick, D.C., 1983. High profit strategies in mature capitalgoods industries: a contingency approach. Academy of Manage-ment Journal 26, 687–707.

Hamilton, S., 1991. New-product development and manufacturingcompetitiveness: a Hewlett Packard perspective. In: Blackburn,J.D. (Ed.), Time-Based Competition: The Next Battleground inAmerican Manufacturing. Irwin, Homewood, IL, pp. 191–208.

Hayes, R.H., Wheelwright, S.C., 1979. Link manufacturing processand product life cycles. Harvard Business Review 57, 133–140.

Hill, T., 1989. Manufacturing Strategy: Text and Cases. Irwin,Homewood, IL.

Hrebiniak, L.G., Joyce, W.F., 1984. Implementing Strategy. Mac-millan, New York, NY.

Jaccard, J., Turrisi, R., Choi, K.W., 1990. Interaction Effects inMultiple Regression. Sage, Newbury Park, CA.

Jöreskog, K., Sorbom, D., 1989. LISREL 7: A Guide to the Pro-gram and Applications, 2nd Edition. SPSS, Inc., Chicago, IL.

Khandwalla, P.N., 1973. Viable and effective organizational designof firms. Academy of Management Journal 16, 481–495.

Kim, J.S., 1994. Manufacturing roundtable research report series,Boston University.

Kim, J.S., Ritzman, L.P., Benton, W.C., Snyder, D.L., 1992. Link-ing product planning and process design. Decision Sciences 23,44–60.

Konijnendijk, P.A., 1994. Coordinating marketing and manufactur-ing in ETO companies. International Journal of ProductionEconomics 37 (1), 19–26.

Kumar, A., Sharman, G., 1992. We love your product, but whereis it? Sloan Management Review 33 (2), 93–99.

Kumar, N., Stern, L.W., 1993. Conducting interorganizational re-search using key informants. Academy of Management Journal36 (6), 1633–1651.

Lawrence, P.R., Lorsch, J.W., 1967. Organization and Environment.Harvard University Press, Boston, MA.

Leitch, R.A., 1974. Marketing strategy and the optimal productionschedule. Management Science 21, 302–312.

Lenz, R.T., 1980. Environment, strategy, organization structure andperformance: patterns in one industry. Strategic ManagementJournal 1, 209–226.


Lenz, R.T., 1981. Determinants of organizational performance: aninterdisciplinary review. Strategic Management Journal 2, 131–154.

Malhotra, M.K., Grover, V., 1998. An assessment of survey re-search in POM: from constructs to theory. Journal of OperationsManagement 16 (4), 407–425.

McCann, J., Galbraith, J.R., 1981. Interdepartmental relations. In:Nyustroms, P.C., Starbuck, W.H. (Eds.), Handbook of Organi-zational Design, Vol. 2. Oxford University Press, New York,NY, pp. 60–84.

Miles R.E., Snow, C.C., 1978. Organizational Strategy, Structureand Process. McGraw-Hill, New York, NY.

Miller, D., 1988. Relating porter’s business strategies to environ-ment and structure: analysis and performance implications. Aca-demy of Management Journal 31, 280–308.

Nemetz, P.L., Fry, L.W., 1988. Flexible manufacturing organiza-tions: implications for strategy formulation and organizationaldesign. Academy of Management Review 13, 627–638.

Nunnally, J.C., 1978. Psychometric Theory. McGraw-Hill, NewYork, NY.

O’Leary-Kelly, S.W., Vokurka, R.J., 1998. The empirical assess-ment of construct validity. Journal of Operations Management16 (4), 387–405.

O’Reilly, C.A., Caldwell, D.F., Barnett, W.P., 1989. Work groupdemography, social integration and turnover. AdministrativeScience Quarterly 34, 21–37.

Palmatier, G.E., Shull, J.S., 1989. The Marketing Edge. OliverWright Limited Publications, Essex Junction, VT.

Phillips, L.W., 1981. Assessing measurement error in key informantreports: a methodological note on organizational analysis inmarketing. Journal of Marketing Research 18, 395–415.

Phillips, L.W., Bagozzi, R.P., 1986. On measuring organizationalproperties of distribution channels: methodological issues in theuse of key informants. Research in Marketing 8, 313–369.

Porter, M.E., 1980. Competitive Strategy. Free Press, New York,NY.

Porter, M.E., 1985. Competitive Advantage. Free Press, New York,NY.

Powers, T.L., Sterling, J.U., Wolter, J.F., 1988. Marketing andmanufacturing conflict: sources and resolution. Production andInventory Management Journal 29, 56–60.

Preston, L.E., 1977. Strategy–structure–performance: a frameworkfor organization/environment analysis. In: Thorelli, H.B. (Ed.),Strategy+ Structure= Performance. Indiana University Press,Bloomington, IN, pp. 30–49.

Saghafi, M.M., Gupta, A., Sheth, J.N., 1990. R&D/marketing inter-faces in the telecommunications industry. Industrial MarketingManagement 19, 87–94.

Saraph, J.V., Benson, P.G., Schroeder, R.G., 1989. An instrumentfor measuring the critical factors of quality management. Deci-sion Sciences 20, 810–829.

Shapiro, B.P., 1977. Can marketing and manufacturing coexist?Harvard Business Review 55, 104–114.

Shapiro, B.P., Rangan, V.K., Sviokla, J.J., 1992. Staple yourselfto an order. Harvard Business Review 70, 113–122.

Souder, W.E., Sherman, J.D., 1993. Organizational design andorganizational development solutions to the problem of R&D-marketing integration. In: Woodman, R.W., Pasmore, W.A.(Eds.), Research in Organizational Change and Development.JAI Press, Greenwich, CT, pp. 181–215.

Spencer, M., Cox III, J.F., 1994. Sales and manufacturing coordi-nation in repetitive manufacturing: characteristics and prob-lems. International Journal of Production Economics 37 (1), 73–81.

Stalk Jr., G., Hout, T.M., 1990. Competing Against Time. FreePress, New York, NY.

Sundstrom, E., Altman, I., 1989. Physical environments and work-group effectiveness. Research in Organizational Behavior 11,175–209.

Thomas, K., 1976. Conflict and conflict management. In: Dunnette,M.D. (Ed.), Handbook of Industrial and Organizational Psycho-logy. Rand McNally, Chicago, IL, pp. 889–935.

Thomas, K., 1992. Conflict and negotiation processes in organ-izations. In: Dunnette, M.D. (Ed.), Handbook of Industrialand Organizational Psychology. Rand McNally, Chicago, IL,pp. 651–717.

Thompson, J.D., 1967. Organization in Action. McGraw-Hill, NewYork, NY.

Utterback, J.M., Abernathy, W.J., 1975. A dynamic model of pro-cess and product innovation. Omega 3, 639–656.

Van de Ven, A.H., Ferry, D.I., 1980. Measuring and AssessingOrganizations. Wiley, New York, NY.

Van Dierdonck, R., Miller, J.G., 1980. Designing production plan-ning and control systems. Journal of Operations Management1 (1), 37–46.

Venkatraman, N., 1989a. The concept of fit in strategy research:toward a verbal and statistical correspondence. Academy ofManagement Review 14 (3), 423–444.

Venkatraman, N., 1989b. Strategic orientation of business enter-prises: the construct, dimensionality, and measurement. Man-agement Science 35 (8), 942–960.

Venkatraman, N., Ramanujam, V., 1987. Measurement of businesseconomic performance: an examination of method convergence.Journal of Management 13, 109–122.

Vollmann, T.E., Berry, W.L., Whybark, D.C., 1997. ManufacturingPlanning and Control Systems. McGraw-Hill, New York, NY.

Werts, C.E., Linn, R.L., Jöreskog, K.G., 1974. Interclass reliabilityestimates: testing structural assumptions. Educational and Psy-chological Measurement 34, 25–33.

Wheelwright, S.C., Clark, K.B., 1992. Revolutionizing ProductDevelopment. Free Press, New York, NY.

Wheelwright, S.C., Hayes, R.H., 1985. Competing through manu-facturing. Harvard Business Review 63, 99–109.

Zahra, S.A., Covin, J.G., 1993. Business strategy, technologypolicy and firm performance. Strategic Management Journal 14,451–478.

99 the integration of manufacturing and marketingsales decisions

Documents