This paper contrasts three management orientations relevant for exporters: export, technology and customer orientation. The general hypothesis is that all orientations co-variate positively with export performance. However, an alternative hypothesis regarding customer relations is propounded (negative impact on performance). Also, the interaction effect between technology and customer orientation is supposed to be positive and significant. The hypotheses were tested in a sample of 132 firms in the information technology industry in of Norway. Regression analysis was carried out based on 80 usable answers (61% response rate). The results support the hypotheses that export performance increases with export commitment. Technology orientation correlates positively with export performance. On the other hand, the much venerated customer orientation shows negative correlation with export performance. It is argued that customer orientation may turn into what we might call customer obsession without due attention to cost consequences and strategic orientation. Also, too much customer orientation may lead the firm away from its ability to innovate, leaving the company behind its competitors in the longer term. The interaction between customer and technology orientation gave no effect.
Over the years a number of articles have discussed the concept of successful exporting (for a review, see Aaby and Slater 1989 and Cavusgil and Zou 1994). The main conclusion of this strand of literature is that success in exporting hinges first and foremost on organisational variables such as proactive attitudes and management commitment, much more than on operational variables (like different aspects of marketing mix or market choice). Taking this discussion somewhat further, this article discusses the impact of different management orientations on export performance. Levitt (1960) was among the first to discuss management attitudes and orientations, contrasting the product oriented firm with the customer and market oriented one. Only 30 years later, the concept of market orientation was subject to scientific investigation (Kohli and Jaworski 1990, Narver and Slater 1990). After these seminal articles a string of research has followed on different aspects of market orientation, testing its effect on performance. As far as we know, no one to this day has attempted to see the emerging literature on market orientation in an export context. Indeed, Breman and Dalgic (1998) have examined the learning effects on market orientation in exporting firms, whereas the relations with performance remain unexplored.
The present study focuses on three management orientations: customer, technology and export orientation and explores their respective impact on export performance. In principle one could allege that success in exporting would not present any difference than business success in general, and that one therefore does not need specific studies in the subject matter. Yet, the specifics of the exporting situation involving new dimensions like an unknown cultural environment, unknown risks, both commercial and political, unknown strategic challenges would warrant the phenomenon per se to be studied.
What then about a study of the IT industry? This relatively young branch of modern industry is facing a number of critical challenges into the next millennium. The pace of technological innovation which is one of the specific features of this industry, together with the need for swift market penetration in an increasingly unfriendly competitive environment in a globalising market place tremendous requirements on the shoulders of top management of IT firms. Therefore, the technological and innovation aspects seem to justify a particular study of export performance of this specific industry. Introducing the technological variable, one is then inclined to include yet another factor to the discussion: market or customer orientation. In the present context, we will particularly investigate the firms’ customer orientation and contrast that to technology orientation. Export orientation or rather export commitment (Aaby and Slater 1989) is introduced as a third independent variable in order to explore its relative importance to the other two constructs.
The article is structured as follows: A literature review summarises the relevant literature and presents the hypotheses to be tested. In the methodology section the constructs are discussed and defined, and the method is presented. The findings of the survey are presented in the following section. Finally the paper concludes with implications for management and research.
This section will discuss contributions on the three management orientations in order to develop hypotheses to be tested in the study.
Export performance has been subject to investigation by several authours. For instance, Kamath, Rosson, Patton and Brooks (1989) - based on a study of Canadian exporters - conclude that virtually ‘any strategy will do’ when it comes to the composition of marketing mix, rifle or shot gun market strategy, product development expenditure (except in high-tech industries), planned or ad hoc marketing approaches, to mention but some of the most controversial findings. In other words it was difficult to identify any clear correlation between these independent variables and export performance.
Commitment of resources – financial and human – to the exporting venture was, however, recognised as critical by Hunt, Froggart and Hovell already in 1967. Later in the seventies and the eighties this relationship has been explored from different angles. The seminal work in the field of internationalisation was made by the "Uppsala School" with contributions from a long range of writers, among them Johanson and Wiedersheim-Paul (1975) and Johanson and Vahlne (1977). They suggest that there is a loop process between the market and the firm whereby market knowledge leads to commitment decisions in the firm, the ensuing marketing activities in their turn leading to increased market commitment and knowledge, and so on. The theory posits that the learning process comes about primarily through experience in the market. "Experiential knowledge generates business opportunities and is consequently a driving force in the internationalisation process" (Johanson and Vahlne, 1990, p.). The theory of incremental international involvement has received substantial support by a host of researchers (see for instance Bilkey and Tesar 1975, Cavusgil 1983, Czinkota and Johnston 1986), particularly in the early stages of the process Forsgren (1989).
For instance, Kirpilani and Macintosh (1980) note that organisational variables of the company play a much larger part than do situational, product and manufacturing policies. Factors like information for control reporting, top management effort, and the degree of structuring and maturity of the firm are all significantly linked to export success. Furthermore, they note that it is possible to penetrate world markets with "commonplace" (p. 90) products. Success hinges more on organisation than on products. Solberg (1988) supports this view and suggests a model of "beneficial export circle" of three factors: attitudes, skills and embodiment. Analysing 114 Norwegian exporters, he suggests that attitude factors like for instance marketing orientation, empathy and delegation of authority; skill factors like how to deal with representatives abroad, quality and price, use of market information; and management commitment and the role of the board of directors (embodiment) were key determinants for export success.
In a comprehensive review of the literature on export performance Aaby and Slater (1989) conclude that export success is strongly related to top management commitment in some form or another. This commitment is necessary in order to build the distribution network and information channels indispensable for the firm to engage in the export learning process (Johanson and Wiedersheim-Paul 1975, Johanson and Vahlne 1977). They also refute the generally accepted belief that larger firms are more successful than smaller ones. Cavusgil and Zou (1994) develop a model of export marketing performance involving the four traditional marketing mix factors: product adaptation, promotion adaptation, support to distributors/ subsidiary and price competitiveness. Product adaptation and distributor support were found to be positively linked with export performance, whereas promotion adaptation correlates negatively, and price competitiveness did not show any significant correlation. Two other factors, the firm‘s international competence and its commitment to the export venture have both a direct and indirect positive impact on performance, lending support to the review presented by Aaby and Slater (1989).
The above discussion leads us to the following hypothesis:
Seringhaus and Rosson (1990) argue that:
"The company that is knowledgeable about exporting will be able to determine what information to collect and how to use it, to a greater extent than their less knowledgeable counterparts. While based on information then, knowledge is clearly a much broader concept, guiding the company in all its endeavours. In a sense knowledge is a special resource that is present to varying degrees in companies. Like other resources, we should recognize that, without husbanding and replenishment, export knowledge will be depleted over time". (pp. 154-55).
Thus, they make indirectly the link between the internationalisation process (Johanson and Vahlne 1977 and 1990), market information/market knowledge and market orientation. Kotler (1994) discusses different management orientations in his standard text-book on marketing management, where market orientation departs from the more sales, product and production oriented management styles in its emphasis on customer value and customer satisfaction. Kohli and Jaworski (1990) suggest a model of market orientation, with the underlying hypothesis that the more market oriented firms will perform better than product oriented firms. Their definition of market orientation including elements like information gathering, dissemination and response was among the first attempts to identify a measure of this construct. Narver and Slater (1990) define market orientation by three behavioural ingredients (customer orientation, competitor orientation and inter-functional orientation) and by two decision variables (long term perspective and profitability). Gatignon and Xuereb (1997) define customer orientation as the capability to identify, analyse, understand and meet customer requirements. Deshpandé, Farley and Webster (1993, p. 27) define customer orientation as "the set of beliefs that puts the customer interest first". Thus customer orientation may be defined as a dedication of the firm to bring about customer satisfaction. The close overlap between the customer orientation and market orientation concepts leads us to conclude that:
Customer orientation is distinct from market orientation in that it does not take into account the profitability and long term aspects suggested by Narver and Slater (1990). Their contention that the market oriented firms perform better than other firms (Slater and Narver 1994) may not necessarily be true for firms that are more "narrowly" customer oriented. Therefore one could maintain that firms focussing obsessively on customer needs, not necessarily paying attention to the two decision variables introduced by Narver and Slater (1990) - long term focus and profitability, will not perform well. The argument is here that customer orientation may turn into what we might call customer obsession without due attention to cost consequences and strategic orientation. One risks therefore ending up pleasing a lot of different customers with different kinds of requirements having dramatic repercussions on the profitability of the operations. Or in other words, the risk of too much customer adaptation is that "every product will be a new product development project" as one Norwegian managing director of an IT firm puts it. Also, concentration on customer orientation may lead the firm away from its ability to innovate (Bennet and Cooper 1981, Hayes and Abernathy 1980) and may therefore in the longer term leave the company behind its competitors. This is of course particularly critical in a high technology environment as is the case of the IT industry. Should the industry be customer oriented in its product development process it is questionable that products like the PC would ever have seen daylight. Finally, the IT industry still being in its infancy, it is conceivable that the industry has not yet reached the stage where marketing and customer orientation is called for in order to operate profitably. Rather the focus should be at product innovations. Therefore, alternatively:
The IT industry is living through a period of extremely rapid pace of innovation, with product cycle times shorter than ever before (Solberg, Berg and Veie 1992). This competitive environment calls for technology orientation of the firm, encompassing continuous research and development activities and development of relations and active collaboration with state-of-the-art external technological milieus such as leading universities and research institutes.
Does then technology orientation pay off? This question has been investigated from several angles. Many studies have for instance sought to establish a positive relation between "first mover" and "advantage". One of these, Buzzell and Ferris (1977), found that late entrants have to spend almost fifty percent (relative to sales) more to promote their products than pioneers. First mover advantage may be linked to the concept of technology orientation of the firm in that first movers are the first ones to introduce technological innovations to the market. However, technology orientation has received only scant attention in the marketing and export literature. Yet, for many products, technological innovations are paramount to secure competitiveness (Kamath et al 1989, Gatignon and Xuereb 1997).
Innovative firms have been described as extremely R&D-oriented and take a proactive stance to acquiring new technologies (Cooper 1994, Kanter 1988). One may, therefore, assert that successful exporters may embed characteristics of both technology and customer oriented management. In addition, several authors have found that product advantage and quality are the number one factor impacting on innovation performance (Calantone and di Benedetto 1988, Cooper 1979, Cooper and Kleinschmidt 1987). From this we can conclude that:
The two constructs, customer orientation and technology orientation, are not necessarily mutually exclusive. On the contrary, one may claim with Drucker (1954) that marketing and innovation are the two basic functions of the firm, and that the management culture ensuing from the one might reinforce the other. Kohli and Jaworski (1990) maintain that elements of innovation should be included in the definition of market orientation. Deshpandé et al (1993) allege that market orientation leads to successful innovation and higher performance. Nevertheless, Lawton and Parasuraman (1980) found no significant relationship between the two dimensions (market orientation and innovation activites). Atuahene-Gima (1996, p. 93) observed that "market orientation makes a significant contribution to the innovation project’s impact performance, as measured by intermediate benefits for the firm", but "that it has little impact on its market success, as measured by sales and profit performance". However, he did not analyse the two constructs, market orientation and technology (innovation) orientation as two independent variables at the same level. In this article we argue that the two constructs, customer orientation and technology orientation represent two independent dimensions at the same level of analysis and that they therefore should be treated separately.
Gatignon and Xuereb (1997) define technology-oriented firms as those "with the ability and will to acquire substantial technological background and use it in the development of new products. Technology orientation also means that the company can use its technical knowledge to build a new technical solution to answer and meet new needs of the users". This definition overlaps partly the definition of customer orientation in that the technology orientation is directed toward the needs of the customers. Furthermore Gatignon and Xuereb (1997) hold that, "technology orientation also means that the company can use its technical knowledge to build a new technical solution to answer and meet new needs of the users". This definition overlaps partly the definition of customer orientation in that the technology orientation is directed toward the needs of the customers. Indeed, Gatignon and Xuereb (1997) find that when demand is relatively uncertain (as may be the case of IT products), firms should be both customer and technology oriented.
Moreover, we have seen that customer oriented firms may deviate their attention from technological innovations so important in high technology industries, rather concentrating their resources on product adaptations and adjustments (Bennet and Cooper 1981, Hayes and Abernathy 1980). On the other hand, one may also hypothesise that technology orientation alone, without any corrections from the market place, risks to lead the company into interesting technological gains, but without any real demand from potential customers. It is therefore suggested that:
The model in figure 1 shows the proposed relationships between the different variables:
Operationalisation of variables
This section discusses the four variables in figure 1.
Export commitment
We have seen that most of the export literature corroborate the view that organisational commitment is among the most important predictors of export performance (Aaby and Slater 1989, Cavusgil and Zou 1984). This variable has been defined in many different ways (Aaby and Slater 1989, Cavusgil and Zou 1994, Singer and Czinkota, 1994, Walters and Samie 1990, Axinn et al. 1998). In the present research three items have been used to identify export commitment among firms:
Defined in this way, the construct captures the resource commitment to exporting, rather than the more attitudinal variables suggested by for instance Axinn et al (1998). These variables embody Cavusgil and Zou’s (1994) constructs on export commitment which correlated positively with performance.
Customer Orientation
Based on the market orientation literature, this construct is somewhat narrower and encapsulates only certain of the elements contained in the market orientation definition (Narver and Slater 1990). In the present context customer orientation reflects the dedication of the organisation to fulfil customer satisfaction, including information seeking and knowledge of customer needs, top management involvement with customers. This interpretation is similar in its tenet to that of the marketing concept as it has been defined by Kotler (1994) and King (1965), the latter writing that the marketing concept is
"a managerial philosophy concerned with the mobilisation, utilisation, and control of total corporate effort for the purpose of helping consumers solve selected problems in ways compatible with planned enhancement of the profit position of the firm" (p. 85).
It involves not only the sales and marketing department but the whole organisation - starting with top management. Customer orientation is therefore measured in a three of different ways:
- The degree to which relevant departments are well known with customer demands.
- Top management involvement in customer relations.
- Personal relations of key personnel between the firm and customer.
Kohli and Jaworski’s (1990) elements have partly been included, but also other dimensions are introduced. One may claim that many of the constructs described in the literature on networks and supplier-buyer interaction (Håkanson et al. 1982, Ford et al. 1990) and relationship marketing (Dwyer et al 1987, Morgan and Hunt 1994, Sheth and Parvatyar 1994) are paraphrasing the basic disposition of a customer oriented company.
Technology orientation
Gatignon and Xuereb (1997) define technology-oriented firms as those "with the ability and will to acquire substantial technological background and use it in the development of new products". In addition they maintain that the company must organise its capacity to exploit the technological background acquired. Also technology orientation is characterised by a concern of product features (Kotler 1994), contrasting the customer oriented firm’s inclination to satisfy customer demands. Finally technology oriented firms will seek to emphasise technological advance and development. We have therefore measured technology orientation using three variables:
Export performance.
In a review of the literature, Cavusgil and Zou (1994) identify eleven factors indicating export performance: 1) export sales level, 2) export sales growth, 3) export profits , 4) ratio of export sales to total sales, 5) ratio of export profits to total profits, 6) growth in export ratio, 7) overcoming barriers to export, 8) acceptance of product by foreign distributors, 9) export involvement, 10) internationalisation and 11), attitudes toward exports. The problem with many of these measures of export performance is that they reflect only part of the story. Many reservations to these measures may be put forth: Is export sales growth a performance measure if it is achieved at the expense of profitability? How does the growth compare with the industry in general? Is a large firm with a high export ratio more successful than a small firm with a low export ratio? In their own study, involving 202 export ventures of 76 firms, Cavusgil and Zou (1994) retain the following elements of export marketing performance: goal achievement, perceived success, average sales growth of the export venture and overall average profitability over the first five years of the venture. In the present study we have defined export success:
The general perception of export success captures probably the essence of export performance the best, in that it not only translates the perceived degree of economic success, but also includes the respondents’ opinion of strategic elements of success, like market expansion, competitive response, market penetration etc. It is akin to Cavusgil and Zou’s (1994) goal achievement and perceived success. The other two measures have been included for their somewhat more concrete expression of export performance.
Sample
Based on a list of 207 potential respondents, a total of 132 marketing managers were telephoned in order to announce the forthcoming dispatch of a questionnaire relating to export issues, and to ascertain the name of the key informant (in most cases the marketing manager). The remaining 75 were not considered relevant for an export survey as they were not registered as exporters. 80 usable questionnaires were returned, a response rate of 61%. These companies should be quite representative for the total population of Norwegian IT exporters (see table 1):
One can observe that the sample firms are slightly larger and have a slightly lower profitability than the total sample frame. Table 2 shows the composition of the respondent firms.
The figures indicate that the propensity to export increases with size, except for the largest firms that tend to export less than the average. One reason for reduced export dependence of large firms may be ascribed to the fact that some notorious among them (for instance Alcatel and Siemens) by and large cater to the Norwegian market - mostly as local units of multinational firms. It is also interesting to note that generally speaking large firms seem to perform far less than the average.
Measurements
The variables were measured on a seven point Likert scale. The measurement model for the independent latent variables was tested by a CFA-model with three correlated factors as shown in table 3. Only export commitment and technology orientation shows a significant correlation.
3 Brønnøysundregisteret is a public register of company annual reports in Norway
The model was estimated with ML, the standardised factor loadings, estimated correlations, error variance and reliability measures as depicted in table 4. The Chi-square measuring overall fit is 24.10 (df = 24) with a P-value equal to 0.46 indicating that the model is valid.
In table 4 we see that the factor loadings vary from 0.46 to 0.73, which is considered to be practical significant for this sample size (Dillon & Goldstein, 1984, p. 69). The average variance extracted r VC is 0.36, 0.46 and 0.49 for the three constructs, denoting that the variance due to measurement error is larger than the variance captured by the constructs. This is furthermore confirmed by the relatively low coefficient of a (0.61, 0.67 and 0.74). The validity of the individual indicators can therefore be subject to discussion. This is particularly true for the customer orientation construct, implying that the three items defining this construct only capture part of it.
Model
Based on these findings two different estimation approaches have been tested out: OLS-regression and Structural equation modelling. Given the limited number of respondents (80 valid answers), it was decided to focus on the OLS results knowing that the ML estimator in LISREL needs a relative big sample to give precise parameter estimates. On the other hand, the confirmatory approach in LISREL is considered so important and interesting that the estimates are presented in appendix 1. The notational conventions established by Jøreskog & Sørbom (1989) are used both for the OLS and the LISREL model. The constructs will have the same meaning in the two approaches although the operationalisations will be quite different. The OLS regression will be based on summed scales and the LISREL model on CFA models. The two approaches represent two different philosophical ideas: the formative- and reflective measurement models. Therefore, we find it interesting to compare the estimates given by the two techniques.
The theoretical model was run in two variants: Without and with the interaction term:
(1)
(2)
where
and
Technology orientation.
Results
Two different models were tested: with and without interaction effect between customer and technology orientation. The rationale for the interaction model is investigated by a two-group approach recommended by Johnsson (1998) using the LISREL method. The sample was split in two subgroups. The first subgroup consisted of those which were strongly customer orientated (mean value above 5 – on a seven point scale) the other subgroup was those which were less customer orientated (mean value below 5). The effect parameter from technology orientation on export performance for high and low customer orientation was tested, finding no significant difference between the two groups. This indicates that there is no evidence of interaction between technology orientation and customer orientation. This is outlined in more details in appendix 1.
Table 5 shows that hypotheses 1, 2b and 3 are supported in the model without the interaction effect. Also, accepting hypothesis H2b, implies rejecting the alternative hypothesis, H2a. Furthermore, the variables in the model explain 49,2% of the variance in export performance.
Including the interaction effect does not seem to add much to the explanatory power of the model (R2=0.495). Moreover, the t-values of the variables in model 2 are considerably reduced in the case of customer and technology orientation, these two not being statistically significant any more. It is therefore concluded that model 1 gives a better picture of the impacting variables than model 2.
It is also interesting to note that the two estimation techniques OLS and SEM (Maximum Likelihood) support the same model – see Appendix 1. Deviation of parameter estimates should be expected, since the different estimation approaches only will converge in the rare situation when models are correct in the population (Olsson Troye and Howell, 1999).
Discussion
The model explains 49% of the variance in export performance, underscoring the importance of different management orientations. On the other hand the results may be regarded as quite provoking by the proponents of the market orientation school of thought. Admittedly, customer orientation - and not market orientation - has been investigated. Still, customer orientation - in terms of market (and customer) knowledge is an important ingredient in the market orientation construct (Kohli and Jaworski 1990, Narver and Slater 1990). Furthermore, customer focus and orientation are being embraced by management as the standard solution to reach profitability, with expressions such as "close to the customer" having survived ever since it was first coined by Peters and Waterman (1982). In exports of IT products, this seems not to be the case. Reviewing the data base in more detail, at least two sets of behaviours that correlate negatively with export success have been identified: the firm’s propensity to accommodate dissatisfied customers (regardless the loss effects on the particular sales contract), and the predisposition to yield in contract negotiations (or rather, not use power even if in a position to do so). Furthermore customer orientation may also entail a willingness to overstate the supply of service to customers who really don’t need it, and who will not reward the exporter by being loyal or paying the costs of the service (Jackson 1985). Interestingly, customer orientation does not appear to lead to "overadapted" product strategy. In fact there is no difference between high and low performers on this measure in the present survey.
The relatively strong negative impact of customer relations indicates that the alternative hypothesis should be seriously considered in development of the firm strategy. The results indicate that top management involvement with customers together with close relationships at several levels in the buying and selling firms may lead to a customer bias at the expense of the exporter’s own operations. The consequences of such bias may be as follows:
1) In the long term innovation is paramount to staying competitive in high tech markets. Too much attention to customer demands may reduce the firm’s capacity to innovate (Bennet and Cooper 1981, Hayes and Abernathy 1980). 2) Dedication to customer needs may also involve too much product adaptation resulting in interrupted or reduced production series. In other words, "overfocusing" on customer needs leads to poor economies of scale, which in globalising markets are deemed to be of great importance (Levitt 1983). 3) Finally, caring too much for customer demands may lead the exporter catering to a host of different customer segments, the final outcome being a lack of strategic focus and dwindling profitability.
The other main finding is that technology orientation correlates positively and significantly with export performance. The way in which technology orientation has been defined (investments in state of the art technology, new products development - rather than adaptation/further development of existing ones, and technological edge) may suggest a tautological relationship between performance and one of the items (technological edge). On the other hand, technological edge is not the same as competitive edge, although it is an important component of this construct in IT industries. It is therefore possible to conceive of product advantages and technological edge without succeeding in international markets.
Also, the findings indicate that commitment to the export venture - in terms of dedicating resources - both people and money - seems to override the other two management orientations in explaining export success. This may imply that a company with a somewhat more mundane technological base may still yield good results if it devotes its resources to the export venture. How should these resources be spent, if not on the negatively correlated customer relations which constitute an important element of the negatively loaded customer orientation construct? The study gives us limited insight into this dilemma. On one hand, a certain level of direct customer contact may be desirable in order to secure access to a wider information base than the one stemming from the partner only (Gripsrud, Solberg and Ulvnes 1999). Also, direct customer contact fosters customer loyalty to suppliers through economic, non-economic and social benefits (Biong et al 1996). On the other hand, the present study seems to indicate that too much of a good thing (customer relations) tends to deviate the attention of the exporter from other critical aspects of strategy (such as technology and profitability). One may also hypothesise that many contact points with customers increase the complexity of the relations, thereby giving the company contradictory signals as to the direction of product development, or adjustments etc.
The question then remains to define the optimal allocation of resources on customer contact between the exporter and its local representatives. One possible avenue is to strengthen the relations with the local representative, without spending excessive resources on direct customer relations. In fact, it has been found that support to distributors and agents is positively correlated with export performance (Cavusgil and Zou 1994, Koh and Robichaux 1988). Another strategy may be to constrain the customer relations to a limited number of people in the company, so as to economise with its resources.
Finally, the interaction effect between customer orientation and technology orientation was minimal, implying that the two constructs seem to live independently of one another. In other words, technology orientation may not offset the negative impact of customer orientation.
Implications for research
The present study has investigated relationships between performance and three different management orientations. One of these, customer orientation, is not well explored in the literature. Rather, most writings have concentrated on the concept of market orientation, which is broader. Our finding that customer focus leads to negative performance contradicts most of the conventional wisdom in marketing and calls for further exploration. Both the construct and its theoretical underpinnings should therefore be reexamined in future research projects. Also, the relatively low a shown for customer orientation suggests further exploration of this measure. The setting (Norwegian IT exporters) may be special, in that they are mostly small and are more driven by low scale production and customer adaptation in specific niches than other "mainstream" manufacturers of IT products. A replica in another setting (country) would therefore cast light on the generalisability of the conclusions from the present study. The unexplored issue of allocation of resources between the exporter and its representatives should be included in such a follow-up study. Such a study should also investigate the effects of the somewhat broader concept of market orientation (Kohli and Jaworski 1990, Narver and Slater 1990). The partial overlap between the two concepts (customer and market orientation) calls for particular attention when the constructs are to be (re)defined.
Furthermore, export commitment stands out as a determining element in explaining export performance. The many ways in which this factor has been defined in the literature (including the present article) and the support it receives as a factor explaining export performance, suggests that the commitment construct may reflect a host of different phenomena inside the firm – both attitudinal (such as proactive attitudes towards exporting) and more tangible variables (such as investments or number of people involved in exporting). On the other hand, it would be of interest to find a definition of export commitment which could be embraced by most researchers, so that the robustness of the relationship between this construct and performance could be tested in different settings and countries.
Finally, technology orientation has not been much debated in the export literature. The increasing importance of technology, the reduced time lags between innovation and market penetration, the demands placed upon firms by an increasingly globalised market call for this factor to be explored further, both in so-called hi-tech industries, but also in more traditional industries.
Conclusion
The present research has shown that export commitment plays a powerful role in explaining export performance of IT firms. It has also questioned the conventional wisdom that customer orientation leads to higher performance. On the other hand, it has been shown that technology orientation co-variate positively with performance, whereas the interaction between customer and technology orientation gave no effect. These results give reason to rethink the generally accepted dogma of the importance of customer relations, at least in an export and IT context: in exporting, customer relations may be better catered to by the local representative. A follow-up study is suggested where this element is included, as well as the somewhat broader customer orientation construct.
Bibliography
Aaby, Nils Erik., and S. F. Slater. "Management Influences on Export Performance: A Review of the Empirical Literature 1978-88." International Marketing Review 6, no.4 (1989): 7 - 26.
Atuahene-Gima, K., "Market Orientation and Innovation", Journal of Business Research 35 (1996): 93-103.
Axinn, Catherine N., Rodney L. Stump and Gerard A. Athaide, "How Commitment Influences Performance: An Exploratory Study", Paper presented at the Conference on Globalization, the International Firm and Emerging Economies, Izmir, Turkey, 27-31 May 1998.
Bennet R. C. and R. C. Cooper, "The Misuse of Marketing: An American Tragedy", Business Horizons 25 (November-December 1981): 51-61.
Bilkey, W. J. and G. Tesar, "The Export Behavior of Smaller-Sized Wisconsin Based Manufacturing Firms", Journal of International Business Studies, Vol. 8, 1977.
Biong, Harald, Atul Parvatyar and Kenneth Wathne, "Are Customer Satisfaction Measurement Models Appropriate for Measuring Relationship Satisfaction?", in Parvatyar, Atul and Jagcish Sheth (eds), Proceedings of the Third Conference on Relasionship Marketing, Contemporary Knowledge of Relationship Marketing, Atlanta 1996, pp. 258-275.
Boomsma, A., "On the Robustness of LISREL (Maximum Likelihood Estimation) against Small Sample Sizes and Non-normality". Sociometric Research Foundation, Amsterdam, 1988.
Buzell, R. D., and P. Ferris, " Marketing Cots in Consumer Goods Industries", in Hans B. Thorelli (ed.), "Strategy + Structure = Performance", Indiana University Press, Bloomington, 1977.
Calantone, R. J. and A. di Benedetto, An Integrative Model of New Product Development Process; An Empirical Validation, Journal of Product Innovation 5 (1988): 7-22.
Cavusgil, S. T., "Success Factors in Export Marketing", Journal of International Marketing and Marketing Research, vol. 8 No 2, 1983.
Cavusgil, S.T. & S. Zou (1994) Marketing Strategy-Performance Relationship: An Investiagtion of the Empirical Link in Export Market Ventures. Journal of Marketing, Vol. 58 (January), 1-21
Cunningham, M. T. and R. I. Spigel (1971), "A Study in Succesful Exporting", British Journal of Marketing, Vol 5.
Cooper, R. G., "The Dimensions of Industrial New Product Succes and Failure", Journal of Marketing 43, (Summer 1979): 93-103.
Cooper R. G. and E. J. Kleinschmidt, "New Products: What Separates Winners from Loosers?", Journal of Product Innovation Management, 8, June (1987): 169-184.
Czinkota, M. and W. J. Johnston, 1983, "Does Sales Volume Make a Difference?", Journal of International Business Studies, Spring/Summer.
Drucker, P., The Practiceof Mangement, Harper and Row, Nyew York, 1954.
Forsgren, M., Managing the Internationalization Process, The Swedish Case", Routledge, London, 1989.
Deshpandé, R. J. O. Farley and F. E. Webster ""Corporate Culture, Customer Orientation and Innovativeenss in Japanese Firms: A Quadrad Analysis", Journal of Marketing 57, (January 1993): 23-37.
Gatignon H. and J-M Xuereb, "Strategic Orientation of the Firm New Product Performance", Journal of Marketing Research 34 (1), (February 1997): 77-90.
Håkansson, H. International Marketing and Purchasing of Goods: A Nettwork Approach, Chichester, Wiley (1982).
Hayes, R. H. and W. J. Abernathy, "Managing Our Way to Economic Decline", Harvard Business Review (July-August 1980): 67-77.
Jackson, Barbara, "Build customer relations that last", Harvard Business Review, 1985.
Jaworski, B. J. and A. K. Kohli, "Market Orientation: Antecedents and Consequences", Journal of Marketing 57 (July 1993): 53-70.
Johanson J. and J.-E. Vahlne, "The Internationalization Process of the Firm - A Model of Knowledge Development and Increasing Foreign Commitments", Journal of International Business Studies, 8 (2),1977, pp. 23 - 32.
Johanson, J. and J.-E. Vahlne, "The Mechanisms of Internationalization", International Marketing Review, Fall, 1990, pp. 11 -24.
Johanson J. and F. Wiedersheim-Paul (1975), "Internationalization of the Firm; Four Swedish Case Studies", The Journal of Management Studies, 12 (October), pp. 305 - 322.
Jonsson F.Y. "Modeling Interaction and Nonlinear Effects: A Step-by-Step
LISREL example". In R.E. Schumacker & G. A. Marcoulides (eds), Interactions and nonlinear effects in structural equation modeling. Lawrence Erlbaum Associates, Publishers, 1998.
Jøreskog K. G. & Y. Fan, " Nonlinear Structural Equation Models: The Kenny-Judd With Interaction Effects". In Marcoulides G.A & R.E. Schumacker (Eds), Advanced Structural Equation Modeling. Issues and Techniques, 1996
Lawton, L. and A. Parasuraman, "The Impact on the marketing Concept on New Product Planning", Journal of Marketing 44 (Winter 1980): 19-25.
Levitt, T., "Marketing Myopia", Harvard Busines Review, 1960.
Levitt, T., "Globalization of Markets", Harvard Business Review, 1983.
Kamath, S., P. J. Rosson, D. Patton and M. Brooks, "Research on Success in Exporting: Past, Present and Future", in P. J. Rosson and S. L. Reid, Manageing Export Entry and Expansion, Praeger, New York, 1987.
King, R. L., "The Marketing Concept". In G. Schwartz ed., Science in Marketing, New York: John Wiley, 1965, pp. 70-97.
Kirpilani, V. H. and N. B. Macintosh, "International Marketing Effecti-veness of Small Technology-Oriented Firms", Journal of International Busines Studies, Winter, 1980.
Koh, Anthony C. and Robert A. Robichhaux, "Variations in Export Performance Due to Differences in Export Marketing Strategy: Implications for Industrial Marketers", Journal of Business Research 17, 249-258, 1988.
Kohli, Ajay K. & Bernhard J. Jawoski, "Market Orientation: The Construct, Research Propositions and Managerial Implications", Journal of Marketing, Vol 54, 1990, pp. 1-18.
Kotler, Philip. Marketing management. Englewood Cliffs: Prentice Hall, eight edition, 1994.
Narver, J. C. and S. F. Slater, "The Effect of Market Orientation on Business Profitability", Journal of Marketing 54, October, 1990: 20-35.
Olsson U.H , S.V. Troye & R.D. Howell "Theoretical Fit and Empirical Fit: The Performance of Maximum Likelihood versus Generalized Least Squares Estimation in Structural Equation Models", Multivariate Behavioural Research (MBR) . 34,1,1999: p. 31-59.
Peters, Thomas and Robert Waterman, In Search of Excellence: Lessons from America’s Best Run Companies, New York, NY.: Harper and Row, 1982.
Seringhaus, F. H. and Philip J. Rosson, Government Export Promotion, Routledge, 1990.
Singer, Thomas O. and Michael R. Czinkota, "Factors Associated with Effective Use of Export Assistance", Journal of International Marketing 2 (1), 1994, 53-71.
S. F. Slater and Narver, J. C., "Does Competitive Environment Moderate the Market Orientation-Performance Relationship?" Journal of Marketing 58, January, 1994: 46-55.
Solberg, Carl Arthur. "Successful and Unsuccessful Exporters: An Empirical Study of 114 Norwegian Export Companies." Working Paper, Norwegian School of Management, Sandvika, Norway, 1988.
Walters, G. Peter and Saeed Samiee, A Model for Assessing Performance in Small US Exporting Firms", Entrepreneuship Theory and Practice, Vol. 15, No. 2, Winter 1990, 33-50.
Appendix 1
In this appendix we will outline the details of the OLS-regression analysis and the linear structural model analysis, and those of the interaction models in both LISREL and OLS-regression.
Estimating the regression model with OLS
The formative measurement models for the dependent and independent variables are as follows:
Legend:
Variable 12 |
Top management customer involvement |
Variable 13 |
Personal relations with customer |
Variable 16 |
Knowledge of customer needs |
Variable 6 |
Number of export employees |
Variable 24 |
Investments in export markets |
Variable 25 |
Competence in exporting |
Variable 2 |
Product features = competitive advantage |
Variable 10 |
Technological edge |
Variable 26 |
Investments in new technology |
Since the scale
has no metric meaning we measure the variables in deviation scores.
This will imply that the intercept term in the regression equation will be zero.
Results
The estimated OLS model with the interaction term:
t-ratio : -0.0282 7.26 1.088 -0.66
R2 = 0.495
The linear OLS model without the interaction effect
t-ratio : -2.44 7.27 2.44
R2 = 0.492
Testing for interaction effect between Technology orientation and Customer orientation.
Two subgroups were
formed by splitting the mean of CO at 5. High CO ³ 5 and low CO < 5.
The models for the two subgroups are given by and
. We want to test the hypothesis
that
. This is done by a two-group
analysis in LISREL. In the first run we fix the
-
parameter to be invariant across the two groups. The results are
=0.52
with a t-value of 2.82 and a Chi-square = 3.22 with 7 degrees of freedom. In
the second run the
- parameter was
estimated for the two groups. This gives
=0.50
(t = 2.27) and
=0.56 (t=1.84) and
a Chi-square = 3.19 with 6 degrees of freedom. The difference Chi-square = 0.03
with 1 degree of freedom. This gives support to the hypothesis that the two
’s are equal. The implication of this
is that the interaction effect is not present. This test for interaction effect
is not a rigorous statistical test, but it does give some indications. An obvious
problem is the threshold values of TO, the low sample size and that the grouping
is based on the observed variables and not on the latent variable. All of these
will reduce the power of the test.
Modelling interaction with LISREL
For curiosity we modelled the LISREL interaction model using a full information approach due to Jøreskog and Yang (1997) where they formulate the "Kenny & Judd" model with only one product variable and also adding a constant term
The theoretical model with interaction term for this study is given in the equations (1) and (2):
(1) The structural equation:
(2) The measurement equations:
Endogenous variables:
Exogenous variables:
The observed variable
, which is the "observed"
product variable for the interaction term x 1x 3. I.e.
Variable explanation:
x 1 = Customer orientation
x 2 = Export commitment
x 3 = Technology orientation, and
x 1x 3 = the interaction term between Customer orientation and Technology orientation.
The dependent variable h is Export performance.
In our situation we will maintain that Maximum Likelihood is the optimal estimation method. Jøreskog & Yang (1997) argue that the WLSA method is the optimal estimation method for estimating models with interaction terms. Taking the small sample size (N=80) into consideration is ML probably the best method as it does not require a weight matrix. This is also consistent with Jøreskog & Yang (1997). Olsson et al. (1999) find in a recent study that the ML gives unbiased fit statistics and parameter estimates for models that are both misspecified with respect to the data (the data were highly non-normal) and the model structure. However in this study ML was not able to converge. The results below are GLS-estimates and they should be interpreted with caution (Olsson et all. 1999). Given the small sample and the convergence problems with ML, fit statistics, standard errors and t-values should not be taken as formal test statistics, but rather as guidelines.
Results
The estimated structure model:
st.error : (0.09) (0.08) (0.13) (0.10)
t-ratio : -2.26 5.10 3.36 -1.31
Fit statistics and parameter estimates and t-values for the measurement model are shown in table 1.1
Table 1.1A: Factor loadings (not standardized)
Parameter |
Estimate |
t-ratio |
sign at 0.05 |
l y (1,1) |
Fixed to 1 |
------ |
---------- |
l y (2,1) |
1.35 |
5.85 |
Yes |
l x (1,1) |
Fixed to 1 |
------ |
---------- |
l x (2,1) |
0.49 |
2.81 |
Yes |
l x (3,1) |
0.34 |
2.62 |
Yes |
l x (4,2) |
Fixed to 1 |
------ |
---------- |
l x (5,2) |
0.65 |
5.17 |
Yes |
l x (6,2) |
0.33 |
2.82 |
Yes |
l x (7,3) |
Fixed to 1 |
------ |
---------- |
l x (8,3) |
1.00 |
5.34 |
Yes |
l x (9,3) |
1.17 |
5.05 |
|
l x (10,1) |
5.66 |
56.96 |
Yes |
l x (10,3) |
5.49 |
43.14 |
Yes |
Table 1.1B Correlations between latent independent variables*.
x 1 |
x 2 |
x 3 |
|
x 1 |
1.0 |
||
x 2 |
-.04 |
1.0 |
|
x 3 |
.23 |
.58** |
1.0 |
*The correlations betweenx 1x 3 and the other x -variables are fixed to zero.
**Sign. at the 5% level
Table 1.1C Selected Fit Statistics
Fit Statistic |
Values |
Chi-square |
55.94 (df=50) |
RMSEA |
0.039 |
CFI |
0.99 |
CN |
108.54 |
Estimating the linear structural model with LISREL
Based on the results from the interaction model we re-specified the model. In the re-specified model the interaction term was excluded, while the other parts of the model were left unchanged. The estimation method here was ML. We therefore have more confidence in these results. The structural model is given by equation (3):
The constant intercept term was included in the interaction model for estimation reasons (Jøreskog & Yang, 1997). For the linear model this is not necessary since there are no theoretical reasons for estimating the intercept.
Results
The estimated structure model:
st.errors: (.139) (.282) (.185)
t-values: -2.08 3.08 2.32
Fit statistics and parameter estimates and t-values for the measurement model are shown in table 1.2. The variance of the constructs were fixed to 1.0 for scaling purposes.
Table 1.2A: Factor loadings (not standardized).
Parameter |
Estimate |
t-ratio |
sign at 0.05 |
l x (1,1) |
1.09 |
5.03 |
Yes |
l x (2,1) |
.46 |
3.15 |
Yes |
l x (3,1) |
.65 |
3.57 |
Yes |
l x (4,2) |
1.84 |
8.31 |
Yes |
l x (5,2) |
1.29 |
5.47 |
Yes |
l x (6,2) |
.70 |
2.94 |
Yes |
l x (7,3) |
0.90 |
6.76 |
Yes |
l x (8,3) |
0.95 |
6.18 |
Yes |
l x (9,3) |
1.11 |
5.56 |
Yes |
l y (1,1) |
.93 |
2.86 |
Yes |
l y (2,1) |
1.29 |
2.93 |
Yes |
Table 1.2B: Correlations between latent independent variables.
x 1 |
x 2 |
x 3 |
|
x 1 |
1.0 |
||
x 2 |
-.002 |
1.0 |
|
x 3 |
.13 |
.30** |
1.0 |
**Sign. at 1% level
Table 1.2C: Selected Fit Statistics
Fit Statistic |
Values |
Chi-square |
45.984 (df=38) |
RMSEA |
0.051 |
CFI |
0.96 |
CN |
106.41 |
The fit statistics all except the CN indicate acceptable fit. Even though we attach more confidence to the ML-estimates than to the GLS-estimates above, we still have to look upon these as guidelines and descriptive measures rather than formal test statistics. The reason for this is the small sample size which is below the recommended lower limit of 100 (Boomsma,1983). Again, the alternative proposition on customer orientation is the one being confirmed. The main difference between the results of the two runs is that export commitment now shows a much greater strength in the model.