Business Marketing - ...

7/28/2019 Business Marketing - ...

1/19

Business Marketing:Present and Futureby Gary L. LilienPennsylvania State University, USA

1. Business Marketing: Nature and U niquenessIndustrial marketing or the new term, business marketing, is thema rketin g of goo ds and services to indus trial customers for use, in turn ,in their own production of goods and services. The marketing ofbusiness goods a nd consumer goods have two attributes in common: apurchase is the usual outcom e of the process an d the result derives fromsome decision-making activities. In spite of these superficialsimilarities, business or indu strial marketing pro blems m ust be handleddifferently from consumer marketing for several important reasonsou tlin ed below. (In wha t follows, we use the terms "business" an d"industrial" mark eting interchangeably.) This pap er highlights areas ofprogress in structuring industrial marketing problems and outlinesprospects f or impo rtant new developments in the near future.The reasons for the consumer market/industrial market differencesinclude the following:Derived Demand. One characteristic of the indus trial good s market istha t its demand is derived from the dem and fo r final consumer goodsand services. This implies that the direct demand for most industrialgo od s is fairly inelastic. A major increase in the price of steel cord inradial tyres, for example, will not reduce the demand for tyres by amo tor m anufacturer (although it may cause manu facturers to look forsub stitute materials). Thu s, products are purchased by organ isations tomeet the needs of their customers. Impulse buying is uncom mo n whileobjective criteria-such as meeting production needs an d scheduleswith a minimum-cost product - often drive the choice process.

BusinessMarkering / 3


2/19

Multiple Influences. More than one and, often, many individuals areinvolved in the purchasing decision process; as Wind (1976) points out,purchasing managers rarely make a buying decision independent of theinfluence of others in the organisation.Long Purchase Cycles. Due to (a) the often high cost volume involved,(b) the number of individuals involved and (c) the technical nature ofthe products under consideration, the purchasing process may take along time. This extended purchase process, taking months or years insome cases, makes it difficult to determine a direct relationship betweenmarketing effort and customer actions.Customer Heterogeneity. The number and heterogeneity of customersin industrial markets lead to special problems. There are more steelcompanies than car manufacturers, so the car market for steel has moresellers than buyers. A manufacturer of machine tools may numberamong his customers many small machine shops and General Motors aswell. These firms respond differently to marketing activities, havingdifferent purchasing requirements in terms of service needs, deliveryrequirements and the like.Decentralised Transactions. Products in industrial markets often movethrough difficult-to-measure channels. A driving force behind thesophistication of consumer marketing analysis methods is theavailability of services that regularly and frequently collect detailedproduct-class sales data as well as retail price, promotion and brand-switching information. Because of the importance of direct selling andthe independent nature of the industrial distributor, such da ta are rarelyavailable in industrial markets.Varying, Fragmented Market Structure. Industrial markets with fewbuyers and few sellers are structured differently from consumermarkets. When a market contains only three o r four customers, and acomplete census is taken of customer activities, statistical methods arenot relevant. In such markets, buyer-seller relationships are often long-term, and based on mutual respect as well as relative power (Corey,1983). This leads to purchasing arrangements that differ significantlyfrom those in consumer markets. In general, less data are available inindustrial markets, with fewer and harder-to-measure transactions, andrelevant data for decision making must incorporate measures of powerand long-term buyer-seller relationships.Systems Selling. As Hudson (1971) and others have pointed out,organisational buyers are interested i6 the satisfaction of a total need.The determining factors in the success of an offering may include suchitems as technical support, training, delivery dates and financial terms.

4 1 Industrial Markering and Purchasing 2.3


3/19

This is part of the trend towards systems selling in industrial markets(Mattsson, 1973).For the reasons given above, it is unrealistic to expect that approachesdeveloped in consumer markets should apply directly to industrialmarketing problems. As Webster (1978, 1984) has pointed out, the wordcornplexzty is used most appropriately to describe industrial markets.That complexity is often used by academics and practitioners alike asthe reason for not applying rigorous problem-solving methods toindustrial marketing problems. Yet industrial marketing activityrepresents over half of the economic activity in the industrialised world,and the very complexity of industrial marketing makes the potentialuour.reat for the introduction of analytical ri,In what follows, we will focus on a broad range of progress areas andprospects for development. The treatment here will be selective ratherthan exhaustive and, as in any list of research priorities, the selection oftopics is personal.

11. Buying BehaviourTo analyse marketing opportunities when customers are otherorganisations requires an understanding of organisational buyingbehaviour. A number of thorough reviews of the literature on theorganisational buying process are available and we will not try toduplicate them here (Webster, 1984; Sheth, 1977; Choffray, 1977;Bonorna et a/., 1977; Wind 1978a; Wind and Thomas, 1981; Moriarty,1983; Moeller, 1985). Much of this research has focused on three issues:the nature and the structure of the buying centre, the organisationalbuying process, and the factors affecting purchasing decisions.A number of empirical studies (Buckner, 1967; Lilien and Wong, 1984)have demonstrated that a single industrial decision does not exist. Thebuying centre concept (Wind, 1978a; Spekman and Stern, 1979) iswidely accepted, yet little is known about the dynamics of the buyingcentre, what determines that composition and how it changes. Windand Thomas (1981) review the relevant literature and find littleagreement in the findings of published studies.The buying process has been modelled in a number of ways, often as asequence of stages (Robinson and Faris, 1967; Ozanne and Churchill,1971; Webster and Wind, 1972; Kelley, 1974; Bradley, 1977; Wind,1978a). Variations in these sequences suggest that the process iscomplex and varies by product, market, company and buying situation.

BusinessMarkering 1 5


4/19

The buying situatio n has long been recognised as a determinant of thestructure of the buying centre as well as of the information needs andthe decision process associated with the purchase. Robinson and Faris(1967) have suggested a class ification of buying situations int o new buy,modified rebuy and straight rebuy situations. In addition, thecomp osition of t he buying centre an d its buying process is affected byfour sets of factors: personal, interpersonal, organisational andinterorganisational (W ind and T ho ma s, 1981).There have been several attempts to provide a structure for theinterorganisational decision process. Two models (Webster and Wind,1972; Sheth, 1977) provide checklists of the issues and interactionsimportant in organisational buying. A model by Choffray and Lilien(1980), while simpler, suggests operational models and associatedmeasurement metho ds for a mo re comprehensive approach.A somewhat different perspective o n the nature of the organ isation albuying process is what is known as the Interaction Approach(Hakansso n, 1982). This ap pro ach posits tha t the key to und erstandingindustrial product marketing and purchasing lies in the interactionprocess between individuals within functional departments andhierarchical levels in supplier and customer firms. That interactionprocess can be described in terms of: (1) the elements of interactions(product exchanges, information exchanges, etc); (2) the partiesinvolved (organisation, individuals, product technology, etc); (3) theenvironment (ma rket structure, cultural/geographica1 distance) an d (4)the atmosp here (the power/dependence relationships tha t characterisethe long-term interaction process). This approach has been extensivelyapplied by the International Marketing and Purchasing (IMP) groupvia an extensive programme of parallel, multi-country research studies(see Ford, 2980; Hak ansso n an d Wootz, 1979; Hak ansso n, 1982;Turnbu ll and V alla, 1986; Ford ef al., 1986). Several research methodsincluding the Network Approach (Easton and Araujo, 1986) andDecision System Analysis (Vyas an d W oodside, 1986) have been sho wnto be useful in understanding these relationships and the buyingprocess.On net, ou r current knowledge of organisational buying provides som euseful decision-making s upp ort f or the industrial marketing manager.Currently available are (1) general, conceptual models, identifyingimp ortant variables an d providing a checklist of im portan t issues, (2)much detailed, situation-specific research, suggesting the type anddirection of effects in certain situations, and (3) the beginning of thedevelopment of comprehensive methodologies to support industrialmarketing decision mak ing.

6 / Industrial Marketing and Purchasing 2,3


5/19

Our knowledge of organisational buying is clearly rudimentary at themoment, however, and some important areas of potential progressinclude: (1) How can the variables and the relationships in models suchas Webster and Wind's and Sheth's be put into equation form and used?( 2 )What does purchase influence mean? How can we establish, validlyand reliably, who is involved (or likely to be involved) in a decisionprocess and what influence these individuals are likely to exert? (3)How do individuals in organisations become aware and knowledgeableabout product information? (4) What product, market andenvironmental characteristics can predict the structure of the buyingprocess across industries? (5)What measurement instruments (decisionmatrices? buying panels? protocol studies? gaming/laboratoryprocedures? network analysis? decision systems analysis?) can bedeveloped for industrial buying situations?

111. SegmentationIndustrial markets are heterogeneous. Customers have differentconstraints, needs and incentives. Products compete imperfectly withone another in satisfying those needs. Market segmentation is thegrouping of potential customers into sets that are homogeneous inresponse to some elements of the marketing mix to allow the allocationof marketing resources among heterogeneous customer-requests.Empirical and conceptual work by Wind and Cardozo (1974) forms thebasis of current thinking about segmentation in industrial markets. Theresults of a survey they performed reveal that segmentation strategiesare used primarily after the fact, to assess a product's past performance,rather than to develop effective marketing programmes. Managersperceive organisational-characteristic bases as the easiest to use butonly modestly relevant, while they perceive DMU (decision-makingunit) characteristics as most relevant but also the most difficult toimplement. Wind and Cardozo recommend a two-stage approach. Thefirst stage calls for identifying macrosegmentsthrough the use of suchcharacteristics as (1) end-use market, (2) product application, (3 )customer size, (4) usage rate, and (5) geographical location. The secondstage calls for dividing macrosegments into microsegmenls through theuse of such characteristics as (1) job position, (2) personalcharacteristics, (3) perceived product importance, (4) attitudes towardsvendor, (5 ) buying decision criteria, and (6) stage in the buying process.Thomas and Wind (1982) provide a review of current literature andfindings in the area of industrial market segmentation (see alsoChoffray and Lilien, 1980; Cardozo, 1980; Plank, 1985; Wind, 1978b).Their review is structured around five basic decisions faced by industrial

Business Markering / 7


6/19

marketing managers: (1) the segmentation decision (to segment or not),(2) how to segment (identification), (3 ) resource allocation, (4) segmentselection (what segments should see resource allocations) and (5)implementation.In spite of recent progress in the area, several key issues face researchersand practitioners. If DMU composit ion is to be used as a segmentationbasis, how can such composition be measured reliably and validly? Thisproblem is critical. As Thomas and Wind (1982, p. 13) point out,"...more multiple person buying center studies with 'buying centerresponse' variables need to be conducted". Methods such as theAnalytic Hierarchy Process (Wind and Saaty, 1980) may be employedhere. In addition, market segmentation studies need to be linked toresource allocation. The normative theory of segmentation (Claycampand Massy, 1968), integrated segmentation and resource allocation andthe theory-practice gap has yet to be bridged (Wind, 1978b).

IV. Product Policy and the Marketing MixWe structure the area of industrial product policy into several sub-areas:communications, pricing, distribution, product design, interactionsand competitive response.IV. I CommunicationsIndustrial marketing communications covers the mix of personal andimpersonal communications activities aimed at the industrial buyer,including personal selling, space advertising, the use of catalogues andbrochures, direct mail, trade shows and the like. The effectiveness ofthese activities depends not only on their particular merits but on theway they interact with other communication activities.The progress we review here can be broken into sub-categories:descriptive analysis, providing a picture of current procedures for thesedecisions in industry, and normative analysis, aimed at saying whatdecisions are best.Descriptive Models for Communicotions Decisions. Two studies haveattempted to explain differences in marketing communication spendingbehaviour by studying a cross-section of business situations: theADVISOR models (Lilien, 1979) and the PIMS-based models (Farrisand Buzzell, 1979). Both of these efforts showed that a few product andmarket variables -such as product sales rate, stage in the life cycle of theproduct, number and concentration of customers, product complexityand the like-explain a major fraction of the variation in budgets foradvertising and sales promotion, salesforce expenditures, and total

8 1 Industrial Ma rketing and Purchasing 2,3


7/19

marketing spending as well as the split in the budget between personaland impersonal marketing communication spending.Farris and Buzzell (1979, pp. 119-20) conclude that "in spite of someevidence for instability of a few regression coefficients . . . the overallpattern seems to be one of consistent relationships across a wide varietyof industrial businesses". This is consistent with the findings of Lilienand Weinstein (1984), comparing a sample of 80 European businesseswith 131 US businesses.Normative Models for Communication Decisions. The literature isalmost devoid of studies of response to marketing communicationeffort in industrial markets. An exception is reported by Lilien andRuzdic (1982). They report on applying a common measurement andmodelling approach to six product markets, each of which had gonethrough a major disruption which permitted knowledge of causality ofmarketing sales effort response.Only two important studies appear concerning industrial sales responseto industrial advertising. Weinberg (1956) related changes in a firm'smarket share to its "advertising exchange rate" (the firm's advertisingexpenditures per dollar of sales divided by the corresponding ratio forits competitors), and used the model to find a profit-maximisingadvertising .level. Morrill (1970) carried ou t a large body of surveyresearch, relating magazine readership to purchase behaviour. Hiswork, while having methodological flaws (Lilien et al., 1976), providesevidence that advertising pays off by making selling effort moreeffective.The salesforce area is rich in normative models. Zoltners and Sinha(1980), Zoltners and Gardner (1980) and Cravens (1979) provideclassifications and reviews. Three observations seem appropriate: (1)good, normative salesforce models exist; (2) those models contribute tothe state of the art through methodology and model formulation ratherthan through generalisable findings; and (3) those models are weak ontheir ability to integrate behavioural science concepts into thecalibration of sales response curves and rarely consider salesforce-advertising interactions.Important areas for future development include the following: (1) Twoimportant areas where descriptive norms a re not currently available arefor new industrial products and for industrial services; (2) We need tounderstand how stable are these norms. Do they vary over time? Acrossmarkets? (3) The work by Lilien and Ruzdic (1982) suggests that thedifferences in response elasticities may eventually be related to marketand environmental variables. A larger study of this nature, then, mayprovide material for development of normative communications

Business Marketing 1 9


8/19

spending guidelines; (4) Sales response to selling effort needs to be tiedto sales effort data, using a consistent methodology. The work of Lucaset a/ . (1975) illustrates what is required here; (5) Finally, integrativemodel structures should be developed and implemented. These modelsrecognise interdependencies between territory design, salesforce timeallocations and salesforce size (Beswick and Cravens, 1977; Shanker eta/., 1975; Lodish, 1980; Glaze and Weinberg, 1979).1V 2 PricingPrice is an intimate element of the product offering of the industrialmarketing mix (Monroe, 1979). It is generally believed that mostindustrial marketers set prices for new products and adjust prices forold products on a cost-plus basis. A more contemporary view, one thatis clearly tied to segmentation theory, is value-based pricing (Gross,1978). Here price is set on the basis of the value to the customer and mayvary by market segment.Economic theory provides some superficial guidance for pricing inmonopolistic situations or in situations of price competition. Thesesituations are unrealistic for the typical industrial market.Important current work models pricing strategy dynamically inmarkets where economies of scale and experience curve cost declinesmay drive costs down. Jeuland and Dolan (1982) and Kalish (1983)review approaches in this area and develop some important insights intodynamic pricing policies. Two other important areas in pricing areproduct line pricing decisions (Monroe and Zoltners, 1979) andcompetitive bidding models (Monroe, 1979).Several important issues facing the contemporary industrial marketerinclude: (1) How should price-terms be set during negotiations? Whenthe number of buyers and sellers is similar, bargaining theory will haveto be used to replace market response analysis (Chatterjee and Lilien,1986). Economists have failed to provide operational models for pricingin an oligopoly. How should the manufacturer of a processed productor material manage price when the market is oligopolistic and graduallymoving towards commodity pricing (price competition)? (3 ) Someobservations (Gross, 1982) suggest that advertising can be used tosustain a premium price in the marketplace. How can this informationbe incorporated in a pricing model? (4) The dynamics of pricingstrategy in a monopoly is beginning to be structured and understood.Similar dynamic guidance is needed for a firm that is second or thirdinto the market with a new industrial product (Eliashberg and Jeuland,1986).

10 Industrial Marketing and Purchasing 2,3


9/19

It? 3 DistributionThe distribution decision is the determination of the most profitableways to reach customers. The seller may distribute his product directlyor indirectly through agents or distributors. This decision hasimplications for the size and type of the salesforce, the size of theadvertising budget and the price strategy.In a survey of industrial purchasing managers, Perreult and Russ (1976)found that physical distribution services were second only to productquality and more important than price in influencing industrial

, purchase decisions.Distribution problems can be broken into four components: (1)strategy, how to sell products to end markets; (2) location, how manyand where to locate warehouses and other outlets; (3) logistics, the bestway to supply product to intermediate sellers or final buyers, and (4)management, the development, management and control of margins,allowances and services that encourage the network to operate at peakcapacity.Hu tt and Speh (1984) view the channel strategy question in five stages:(1) Channel objectives, integrating the channel strategy into the overallmarketing plan; (2) Channel constraints, recognising limitations inmiddleman availability, traditional patterns, etc; (3) Channel tasks,assigning tasks to each element in a level in a channel; (4)Evaluation ofalternatives, in terms of number of levels, number and types ofintermediaries and the number of distinct channels, and (5) Channelselection, choosing the best alternative.There have been few quantitative studies related to the distributivechannel strategy decision. Lilien (1979) provides a descriptive model,relating product and market factors -such as size of firm, order size,stage in the life cycle- to the selection of internal versus externalchannels. Webster (1984) provides a discussion of the role ofdistribution in marketing strategy. The other areas of distribution-outlet selection, logistics and management - ridge the rnarketing-operations-management gap (see Geoffrion, 1975, for a discussion ofsome of the important issues).Distribution planning and strategy has seen little formal research inindustrial marketing. Two important issues are: (1) How do advertisingselling effort and distribution channel choice interact? i.e. what are theimportant synergies and the important trade-offs? (2) How can theelements of distribution strategy- location, logistics and management-be related in an integrative structure (for a related, ambitious attemptat this problem, see Corstjens and Doyle, 1979).

BusinessMarketing 1 7 1


10/19

IV: 4 Product DesignThe physical (as well as psychological) aspects of an industrial productcritically affect that product's market potential. The role of productspecifications is well established in industrial purchasing behaviour. Theproduct design question addresses two important issues:What percentage of the potential buying population will find a givenproduct design feasible?

a How do potential customers trade off design features? (How muchmore can be charged for a product with a 15-year life than for onewith a 10-year life?)The most widely used method for industrial product design is conjointanalysis: Cattin and Wittink (1982) and Wind et al. (1978) refer tohundreds of industrial applications. An alternative approach specificallydeveloped for industrial markets is called "designer" (Choffray andLilien, 1982).One important research issue in the product design area is measurementreliability and validity testing: the data for conjoint analysis and designerprocedures are derived from surveys or other artificial choice situations.Wittink and Montgomery (1979) and Cattin and Wittink (1982) pointout that the main problems with conjoint analysis involveexternal/predictive validity, the lack of reliable estimation methods andthe lack of ability to handle large numbers of attributes. A secondimportant research area relates to handling market dynamics: mostproduct design models are static and do not incorporate market learning.The time-path of the new product is a critical piece of managementinformation and associated methodology would be an importantaddition to the field.IT/:5 Interactions and Competitive ResponseThe effects of elements of the marketing mix interact and competitorsrespond to marketer-actions. Both of these effects complicate themarketing mix decision. Dolan (1981) examined a number of industriesto determine the extent to which structural variables determine the modeof competition and concluded: (1) High fixed costs promote competitiveresponse to share gain attempts; (2) High storage costs reduce competitivereactions; (3 ) Growing primary demand reduces competitive reactions;(4) Large firms avoid price competition.As noted earlier, studies have suggested that advertising can supportprice and that advertising and selling interact. In addition, advertisingand other elements of the marketing mix may show non-linearalitiesas well as interactions. An approach to the competitive response issueis that of the "reaction matrix",* which permits the econometric

12 1 Industrial Marketing & Purchasing 2,3


11/19

measurement of contemporaneous or lagged responses to change inei ther the same or a marketing mix element different from the onechanged. (A price change this period "causes7' com petitor "A7' o raiseadvertising next period.) Work by Bensoussan et 01. (1978), Lambin(1976), Schu ltz an d H anssen s (1976) an d Ha nsse ns (1980) suggest thatthe react ion matrix approach, while seeing some conceptual andempirical difficulties, is a step in the direction of incorporatingcompeti t ive effects into ma rketing mix m odels.Research in this area should involve the following: (1) We need an

, understanding of the way marketing mix elements interact . There islack of agreement about whether advertising increases or decreasesprice elasticity. We must und ers tan d how elem ents of the marketing mixinteract and model those effects bet ter; (2) Competitive response iscri t ical for decision making, yet most marketing models ignore thatresponse. A s with interactions, th e nat ur e of com petitive response mu stbe bet ter understood before it can be mo delled adequately (Chatterjeeand Lilien, 1986).

V. New Product ForecastingIndustrial f i rms recognise that new product development andintr od uc tion is accom panied by high c osts a n d high risks. Yet suchinnov ation is necessary to the long -term survival of the firms. Risks canbe controlled through a well conceived and professionally managedprog ram me for new product de velopment. Key ingredients for such aprog ram me include effective ma nag em ent of new product research an ddevelopment as well as sound, explicit models for planning andforecasting new product sales.Two types of model are generally available for con sume r products: pre-test market models and market forecast ing models. The former typeattempt to project market sales from laboratory data while the lat terproject sales from early market data. Currently there is no industrialana logy to pre-test m arket m odels (see Urb an an d Hauser, 1980, for areview).An approach called substitution analysis has been extensively appliedto indu strial products. This work, derived fr om th at of Mansfield (1961,1968) an d Fisher and P ry (1971), pos tulates tha t when a new industrialproduct or process replaces an old one, the rate of adoption isprop ort ion al to the fract ion of the older o ne st i ll in use t imes the currentlevel of penetration. Mansfield (1968) showed that the rate ofsubstitution was higher when the relative profitability from adoptingth e innovation was greater, an d w hen the initial investment was lower.Blackman et d. 1973) and Blackman (1974) show that a number of

Business Marketing / 13


12/19

buying industry factors-such as current and planned R&Dexpenditures, new product sales as a percentage of total sales, etc -canbe used to create an industry coefficient index that can be used to helppredict the relative diffusion rate.

The new industrial product area provides a number of excitingopportunities for development: (1) How, before the preliminary sales ofa product are recorded, can diffusion coefficients be estimated?Mansfield's study is 15 years old; an update is required on a largedatabase incorporating more product and marketing variables(Choffray and Lilien, 1986);(2) How should the new product budget beallocated? How large should the budget be? How much should go forresearch? For development? For each of the various stages of marketdevelopment? (3 ) What sources should be tapped, in a particularmarket, for successful new products? Von Hippel (1978) suggests thatcustomers are often good sources of new products. In whatindustries/markets is this likely to be the case? (4) Many industrialmarkets (for new communications equipment, for energy systems, etc)may take 10-15 years to develop. How can we forecast market potentialmany years into the future? (5) Can the ideas that lead to successful pre-test product models in the consumer goods area be adapted to theindustrial environment? Is a laboratory-measurement procedureapplicable for predicting penetration of industrial products? (6) Howcan the market be assessed for an "original product", one that does notshare its market with existing products? Such a product may create anew market; how can that market be evaluated?

VI. Market D efinitionWe have assumed throughout that the market is well defined or wellunderstood. This is rarely the case. Key strategic issues in industrialmarketing, such as the basic business definition, opportunityassessment, threat analysis and the like are closely tied to the breadth ornarrowness of the market. When attainment of market share representsa desirable objective, market boundaries must be defined to determinethe extent to which that objective is met.Day et a!. (1979) define a product market as a group of physicalproducts perceived to be substitutes by a particular group of customersfor a specific use. They classify methods for identifying productmarkets by whether they rely on behavioural data (cross-elasticity ofdemand, behaviour similarities, brand switching) or judgemental data,perceptual mapping and similarity assessments. The area of market

14 / Industrial Marketing and Purchasing 2,3


13/19

structure analysis is in its infancy and has yet to yield a "best method''to describe an industrial market.

VII. Industrial Market Strategy DevelopmentThe tools for industrial market strategy development and analysis canbe classified as:(1) Shared experience models (the P IMS approach);(2) Product portfolio models:(i) standardised (BCG/GE)(ii) customised (conjoint analysis, analytic hierarchy process)(iii) financial (efficient portfolio/risk-return models);(3) Normative resource allocation models.

All of these approaches, explicitly or implicitly, incorporate life cycleanalysis, experience curve effects, market definition and marketstructure approaches in their development.The PIMS approach (Schoeffler et a[., 1974) used the concept that theshared experience of a large number of successful and unsuccessfulbusinesses will yield strategy insights. Product portfolio models (Windand Mahajan, 1980) deal with product interactions in a number ofways, some specific to the product-market (Wind and Saaty, 1980),some standard, to be applied to all markets (the BCG approach).Normative product portfolio models (Larrechk and Srinivasan, 1981,1982; Corstjens and Weinstein, 1982)provide an analystic framework toproduce normative guidelines from these product portfolio concepts.This field -model development for industrial marketing strategy- is inits infancy. The next decade should permit synthesis, empirical testingand improvement of all these proposed approaches.

VIII. Industrial Marketing Databases and Dec ision Support SystemsLittle (1979) provides a picture of a marketing decision support system(MDSS) as an integrated system of data, models, tools and computerhardware and software aimed at the problems of marketingmanagement. As this paper has suggested, many of the models in theindustrial area are nearly as well developed as consumer marketingmodels. Why then are industrial MDSSs in a state ofunderdevelopment?The problem lies mainly with data sources. Consumer marketers havetraditionally had store audit data (Nielsen), warehouse shipment data(SAMI), or panel data (MRCA) to analyse. The advent of the

Business Marketing / 15


14/19

automated check-out system is providing even cleaner and richer datasources for the consumer market analysts.Can comparable developments be expected in the industrial marketplace?The lack of accurate, inexpensive, frequently collected industrial marketdata will continue to be the weak link in development in this area fora number of years to come. However, as telemarketing begins to substitutefor personal sales in some markets, and as entrepreneurs learn tocentralise the collection of data in some industrial market segments,we expect this to change. Some sources (TRINET and Dunn and

' Bradstreet, in the USA) are providing more and better quality data aboutmarkets, but even these data are not as directly usable for marketingdecisions as are their consumer counterparts.The salesforce represents an untapped yet rich source of marketinformation. Salesforce-based data systems combined with computerisedcollection of transactions data from other sources promises to bring theMDSS revolution to the industrial marketplace, although a bit after theconsumer revolution.IX . ConclusionThis brief review of progress and-prospects for development is by nomeans exhaustive. It portrays the industrial market as one with analyticaland conceptual problems but one with large opportunity as well. The"complexity" of the industrial market must be exploited in successfulmethodological applications; consumer-based approaches will fail inmost cases due to simplifying and naive assumptions.The areas of significant progress identified above suggest that realprogress in measurement and modelling is possible. Webster (1984)suggests that the small sample, heterogeneous nature of industrialmarkets may require more use of experienced managers' judgement inplace of larger scale of data collection.Of over-riding significance is the fact that rewards await those who investthe time and creativity to address industrial marketing problems as theyexist, who adapt and develop the tools required to treat industrialmarketing problems as they are.ReferencesBensoussan, A., Bultez, A. an d Naert, P. (1978) "Leader's Dynamic Mark eting Behaviorin Oligopoly", TIMS Studies in the Management Sciences, Vol. 9 , pp , 123-45.Beswick, C.A. and Cravens, D.W. (1977) "A Multistage Decision Model f or SalesforceManagementy7, ournal of Marketing Research, Vol. 14, May, pp. 135-44.Blackman, A.W., Jr. (1974) "The Market D ynamics of Technological Substitutions",Technological Fbrecasting and Social Change, Vol. 6 No. 1, pp. 41-63.

16 1 Industrial Marketing & Purchasing 2,3


15/19

Blackman. A.W., Jr., Seiigman, E. J. and Solgliero, G.C. (1973) "An Innovation IndexBased upon Factor Analysis7', Technological Forecasting and Social Change, Vol.4, pp. 301-16.Bonoma, TY.,Zaltman, G . and Johnson, W. (1977) Industrial Buying Behavior,Marketing Science Institute, Cam bridge, Mass.Bradley, M.F. (1977) "Buying Behaviour in Ireland's Public Sector"', IndustrialMarkering Management, Vol. 6, August, pp. 25 1-8.Buckner, H. (1967)How British Industry Buys, Hutchinson, London.Cardozo, R. (1980) "Situational Segmentation of Industrial Markets", European, Journal of Marketing, Vol. 14, pp. 264-76.

Cattin, P. and Wittink, D.R. (1982) "Commercial Use of Conjoint Analysis: ASurvey7',Journal of Marketing, Vol. 46 No. 3, Summer, pp. 44-53.Chatterjee. K. and Lilien, G.L. (1986) "Game Theory in Marketing Science: Uses andLimitations7', International Journal of Research in Marketing, Vol. 3, Winter,pp. 79-93.Choffray, J.-M. (1977) 'RMethodology for Investigating the Structure of theIndustrial Adoption Process and the Differences in Perceptions and EvaluationCriteria among Potential Decision Participants7', unpublished PhD dissertation,Massachusetts Institute of Technology, Cambridge, Mass, April.Cho fray, J.-M. and Lilien, G.L. (1980) Market Planning for New IndustrialProducts, John Wiley, New York.Choffray, J.-M. and Lilien, G.L. (1982) "DESIGNOR: A Decision Support Procedurefor Industrial Product Designy7, ournal of Business Research, Vol. 10 No. 2,Septem ber, pp. 185-97.Choffray, J.-M. and LiIien, G.L. (1986) "A Decision Support System for EvaluatingSales Prospects and Launch Strategies for Industrial New Products7', IndustrialMarketing Management, Vol. 15, pp. 75-88.Claycamp, H.J. and Massy, W.F. (1968)"ATheory of Market Segmentation", Journal

of Marketing Research, Vol. 5, November, pp. 388-94.Corey, E.R. (1983)Industriaf Marketing: Cases and Concepts (3rd ed.), Prentice Hall,Englewood Cliffs, N.J.Corstjens, M. and Doyle, P. (1979) "Channel Optimization in Complex MarketingSystems", Management Science, VoI. 25 No. 10, October, pp. 1014-25.Gorstjens, M. an d Weinstein, D. (1982) "Optimal Strategic Business Unit PortfolioAnalysis7', n Zoltners, A. (Ed.) Marketing Planning Models, TIMS Studies in theManagement Sciences, Vol. 18, North-Holland, New York, pp. 141-60.Cravens, D.W. (1979) "Salesforce Decision Models: A Comparative Assessment", inBagozzi, R. (Ed.) Sales Management: New Developments from Behavioral andDecision Model Research, Marketing Science Institute, Cambridge, Mass., pp.310-24.Day, G.S., Shocker, A.D. and Srivastava, R.K. (1979) "Consumer-Oriented

Approaches to Identifying Product Markets7',Journal of Marketing, Vol. 43 No. 4,Fall, pp. 8-19.Dolan, R.J. (1981) "Modeis of Competition: A Review of Theory and EmpiricalEvidence", in Enis, B. and Roering, K. (Eds.) Review of Marketing, AmericanMarketing Association, Chicago, pp. 224-34.

Business Markering / 17


16/19

Easton , G. and Araujo, L. (1986) "Networks, Bond ing an d Relationships in IndustrialMarkets", Industria/ Marketing an d Purchasrng, Vol. 1 No, 1, pp. 8-2 5.Eliashberg, J. and Jeuland, A.P. (1986) "The Imp act of Competitive Entry in aDeveloping Market upon Dynamic Pricing Strategies", Marketing Science, Vol. 5

No. 1, Winter, pp. 20-36.Farris, P.W. and Buzzell, R.D. (1979) "Why Advertising an d Prom otional C osts Vary:Some Cross-Sectional Analyses", Journal of Marketrng, Vol. 43 No. 4, Fall, pp.112-22.ish her, J.C. and Pry, R.H. (1971) "A Simple Substitu tion Model of TechnologicalChange", Ethnological Forecasting and Soc ial Chang e, Vol. 3, pp. 75-88.Ford , D. (1980) "The Developm ent o f Buyer-Se ller Relationships in Industria1Markets", E uro pea n Jou rn al o f w k e t i n g , Vol. 14 No. 516, pp. 339-53.Ford, D., Hakan sson, H. and Johans on, J. (1986) "How Do Companies Interact?"Industrial Marketing and Purchasing, Vol. 1 No. 1, pp. 26-41.Geo ffrion, A.M. (1975) "A Guid e to Computer-Assisted Me thods of DistributionSystems Planning", Sloan Mana gement R eview, Vol. 16 No. 2, Win ter, pp. 17-38.Glaze, R. and Weinberg, C.B. (1979) "A Sales Territory Alignment Program andAccount Planning System", in Bagozzi, R. (Ed.), Sales Management: NewDev elopm ents from Behavioral and Decision Mo del Research, Marketing ScienceInstitute, Cambridge, M ass.Gros s, I. (1978) "Insights from P rici ng Research", in Bailey, E. (Ed.) Pricing Practicesand Strategies, Th e Conference Board, New York, pp. 34-9.Cross, 1. (1982) Priva te correspon dence .Hakansson , H. (Ed.) (1982) Interactional Marketing and Purchasing of IndustrialGoods: A n Interaction Approach, Jo hn Wiley an d Sons, Chichester .Hakansson, H. and Wootz, B, (1979) "A Framew ork of Industria l Buying andSelling", Industrial Marketing M anageme nt, Vol. 8, pp. 28-39.Hanssens, D.M. (1980) "Market Response, Com petitive Behavior, an d Tim e SeriesAnalysis", Journal u Marketing Research, Vol. 17, Novem ber, pp. 470-85.H ud son , C.L. (1971) "Buying-Selling: Gr eat er Int egr atio n in th e Seventies", IndustrialMarketing Management, Vol. 1 No . 1, pp. 5 9-79.H utt , M. an d Speh, T.W. (1984) Industrial Marketing M anagement (2nd ed.), DrydenPress, New York.Jeul and , A.P. and Dolan, R.J. (1982) "An Aspe ct of New Product PIanning: DynamicPricing", in Zoltners. A.A. (Ed.) Marketing P lanning M odels, TIM S Studies in theM ana gem ent Sciences, Vol. 18, No rth -H oll an d, New York, pp. 1-21.Kaiish, S. (1983) "Monopolist Pricing with Dynamic Demand and Production",Marketing Science, Vol. 2 No. 2, Sprin g, pp. 135-59.Kelley, P. (1974) "Functions Perform ed in Ind ustrial Purch ase Decisions withImplications for Marketing Strategy", Journal of Busrness Research, Vol. 2,Oc tob er, pp. 421-33.Larreche, J.-C. and Sriniv asan, V. (1981), "ST RA TP OR T A Decision Su ppo rt System

for Strategic Planhing", Journal of ikfarketing, Vol. 45, Fall, pp. 39-52.Larrkchk, J.-C. and Srinivasan, V. (1982) "STRA TPO RT A Model for the Evaluationand Formulation of Business Portfolio Strategies", Management Science, V01. 28No. 2, Septem ber, pp. 979-1000.

18 1 Industria! Marketing a nd Purch asing 2,3


17/19

Lambin, J.-J. (1976) Adverrising, Competition and Marker Conduct in OligopolyOver Time, North-Holland, Amsterdam.

Lilien, G.L. (1979) "Advisor 2: Modelin g the Ma rket ing Mix Decision for Ind ustr ialProducts", Management Science, Vol. 25 No. 2, February, pp. 191-204.Lilien, G.L., Silk, A.J., C hoffray , J.M. an d Rao. M. (1976) "Indu strial Advertising' Effects and Budgeting Practices", Journal of Marketing, Vol. 40 No. 1, January,pp. 16-24.Lilien, G.L. an d Ruzdic, A.A. (1982) "Analyzing Na tura l Experime nts in Indust rialMarkets", in Zoltners, A.A. (Ed.) Marketing Plannrng Models, TIM S Studies inthe Man age men t Sciences, Vol. 18, pp. 241-69.Lilien, G.L. an d Weinstein, D. (1984) "An Internatio nal C om pari son of t heDeterminants of Industrial Marketing Expenditures", Journal of Marketing, VoI.48, Winter, pp. 46-53.Lilien, G.L. and Wong, M.A. (1984) "An Explo ratory Investigation of th e Struct ure ofthe Buying Center in the Metalworking Industry", Journal of Marketing Research,Vol. 21, February, pp. 1-11.Little, J.D.C. (1979) "Decision Sup por t Systems for M arketing Managers", Journal of

Marketing, Vol. 43 No. 3, Summer, pp. 9-27.Lodish, L.M. (1980) "Applied Dyn amic Pricing an d Prod ucti on Models with SpecificApplication t o Broadcast S pot Pricing", Journal of Marketing Research, Vol. 17,May, pp. 203-11.Lucas, H.C., Weinberg, C.B. a nd Clowes, K. (1975) "Sales R esponse as a Funct ion ofTerritorial Potential and Sales Representative Workload7', Journal of MarkeringResearch, Vol. 12, Au gu st, pp. 298-305 .Mansfield, E. (1961) "TechnologicaI Ch ang e an d th e Rate of Imitation",

Econometrics, VoI. 29 No. 4, October, pp. 741-65.Ma ns field, E. (1968) lndustrial Research and 2chnological innovation, Norton, New~ o r k ,Mattsson, L.G. (1973) "Systems Selling as a Strategy in Industrial Markets",Industrial Marketing Management, Vol. 3 No. 2, pp. 107-20.Moeiler, K.E.K. (1 985) "Research Strateg ies in Analyzing the Organizational BuyingProcess", Journal of Business Research, No. 13, pp. 3-1 7.~Monroe,K.B. (1979) Pricing: Making Pro itable Decisions, McG raw-H ill, New Yor k.Monroe, K. and Zoltners, A.A. (1979) "Pricing th e Produ ce Line Durin g Periods ofScarcity7',Journal of Marketing, Vol. 43 No. 3, Sum mer, pp. 49-59.Mo riar ty, R.T. (1983)Industrial Buying Behavior, Lexington Books, Lexington, M ass.Morrill, J.E. (1970) "Industrial Advertising Pays O ff * ,Harvard Business Review, Vol.48, Marc h/Ap ril, p p. 4-14.Ozanne, U.B. and C hurchill, G.A. (1971) "Five D imens ions of the Indus trial Adop tionProcess", Journal of Marketing Research, Vol. 18, August, pp. 322-8.Perreult, W.D., Jr. and Russ, F.A. (1976) "Physical Distrib ution in Industrial Purcha seDecisions7', Journal of firkering, Vol. 40, Apri I, pp. 3-10.Plank, R.E. (1985) "A Critical Review of Ind ustrial M arke t Segmentation", lndustrialMarketing Management, Vol. 14 No. 2, May, pp. 79 -92.Robinson, P.J. and Faris, C.W. (1967) Indusrrial Buying and Creative Marketing,Allyn and Bacon, Bo ston.

Business Marketing 1 19


18/19

Schoeffier, S., B uzzell, R.D. an d Heany, D.F. (1974) "Im pact of Strategic Plan ning onProfit Performance", Harvard Bu siness Review, Vol. 52 No. 2, March/April, pp.137745.Schultz, R.L. and Hanssens, D. (1976) "Logical Implications of CompetitiveBehavior: A n App roach to Mode1 Specification", Krannert G radu ate School ofMan agemen t, working paper 561.Sha nke r, R.J., Turner, R.E. a nd Zoltn ers, A.A. (1975) "Sales Territory Design: AnIntegrated Approach", Management Science, Vol. 22 No. 3, November, pp. 309-20.Sh eth , J.N. (1977) "Recent Developments in Or gani zatio nal Buying Behavior3', inWoodside, A. et al. (Eds.) Consumer and Industrial Buying Behavior, Elsevier-.N or th Holland , New York.Spekman, R.E. and Stern, L.W. (1979) "Environmental Uncertainty and BuyingGr ou p Structure: A n Emp irical Investigation", Journal of Marketing, Vol. 43 No.2, Spring, pp. 54-64.Thomas, R.J. and Wind, Y. (1982) "Toward Empirical Generalizations in IndustrialM ark et Segmentation:' in Spe km an, R.E. and Wilso n, D.T. (Eds.) Issues inIndustrial Marketing: A View to the F uture, American Marketing Association,Chicago, pp. 1-18.Turnbull, P.W. and VaIla, J. P. (Eds.) (1986) Strategies for International Indu strialMarketing, Croom Helm, London.Urban, G.L. and Hauser, J.R. (1980) Design and Marketing of New Products,Prentice-Hall, E nglewood C liffs, NJ.Von Hippel, E.A. (1978) "Successful Industrial Products from Customer Ideas",Journal o f Marketing, Vol. 42 No. 1, Jan uar y, pp . 39-49.Vyas, N. an d W oodside, A.G. (1986) "Micro Analysis of S upp lier Choice Strategies:Industrial P acking Materials", in Industrial Marketing: An American Perspective,Sp rin ger Verlag, New Y ork, pp. 41-68.Webster, F.E., Jr. (1978) "Management Science in Industrial Marketing", Journal ofMarketing, Vol. 42 No. 1, January, pp. 21-7.Webster, F.E., Jr. (1984) Industrial Marketing Strategy (2nd ed.)Webster, F.E., Jr. an d Win d, Y. (1972) Organizationaf Buying Behavior, Prentice-Hall,Englewood Cliffs, NJ.Weinberg, R.S. (1956) "Mu ltiple Factor B reak-even Analysis: Th e Ap plica tion of O RTechniques to a Basic Problem of M anagem ent Plan ning and C ontrol", OperationsResearch, Vol. 4, April, pp. 152-86.Wind, Y. (1976) "Preference of R elevant Others an d In dividual Choice Models",Journal of Con sume r Research, Vol. 3, Aug ust, pp. 50-7.Wind, Y. (1978a) "Organ izational Buying Behavior", in Zaltma n, G. and B ono ma, T.(Eds.) Review of Marketing, American M arketing Association, pp, 160-93.Wind, Y. (1978b) "Issues and Advances in Segmentation Research", Journal ofMarketing Research, Vol. 15, August, pp. 317-37.Wind, Y. and Cardozo, R. (1974) "Industrial Market Segmentation", IndustriaiMarketing Management, Vol. 3, pp. 153-66.Wind, Y., Grashof, J.F. and Goldhar, J.D. (1978) "Market Based Guidelines forDesign of Industrial Products7*, ournai of Marketing, Vol. 4 2 No. 3, July, pp.27-37.

20 1 Industrial Marketing and Purchasing 2,3


19/19

Wind, Y. and Mahajan, V. (1980) "M easu rem ent Issues in Portfolio Analysis", inLeone, R. (Ed.)Proceedings, Special Interest Conereence on Market Measuremenand Analysis, The institu te of Manag emen t Sciences, Providence, RI, pp . 50-3.Wind, Y. and Saaty, T.L. (1980) "Marketing Applications of the Analytic HierarchyProcess", Management Science, Vol. 26 No . 7, July, pp. 641-58.Wind, Y. and Thomas, R. (1981) "Organizational Buying Behavior: Problems andOpportunities", working paper no. 8-81, Wh art on School, University ofPennsyivania.Wittink, D.R. and Montgomery, D.B. (1979) "Predictive Validity of Trade-OffAnalyses for Alternative Se gmentation Schemes", in Beckwith, N. et ai. (Eds.) 1979AMA Educators' Conference, American Marketing Association, Chicago, pp.68-73.Zoltners, A.A. and Gardner, K.S. (1980) "A Review of Sales Force Decision M odels",working paper, Northwestern University, July.Zoltners, A.A. a nd Sinha, P. (1980) "Integer Pro gram min g Models for Sales ResourceAllocation", Management Science, Vol. 26 No. 3, March, pp. 1197-1207.

Business Marketing - ...

Documents