Authors:
Ruby Roy Dholakia
Jean L. Johnson
Albert J.
Della Bitta
Nikhilesh Dholaki
Ruby Roy Dholakia, Albert J. Della Bitta, and Nikhilesh Dholakia are Professors of Marketing at the University of Rhode Island. The first author is the Director of RITIM. Jean L. Johnson is Assistant Professor of Marketing at Washington State University, Pullman, WA.
Sections within this document:
Antecedents of Decision Making Time
In past research on organizational buying behavior, decision time has been of indirect interest. In this study, we report an empirical investigation which focuses explicitly on decision making time (DMT) and examines its antecedents such as buyclass, firm size, decision making unit (DMU) size, information sources, and size of the evoked set. This study consists of a national sample of organizations involved in the purchase of telecommunications systems. Findings suggest that DMU size, information sources and size of evoked set determine, in part, DMT. Although buyclass and firm size did not affect DMT directly, they did affect DMU size, information sources and evoked set size positively, and therefore had an indirect effect on DMT.
Decision making processes in organizational buying have been examined in terms of the structure of the decision making unit (e.g., McCabe 1987; Spekman and Stern 1978), the nature of relationships within a decision making unit (e.g., Jackson, Keith and Burdick 1984; Ryan and Holbrook 1982) and information search behavior (e.g., Moriarty and Spekman 1984), buyclass and product type (e.g., Anderson, Chu and Weitz 1987; Bellizzi 1978; Lilien and Wong 1984). While associated with all these variables, the time involved in arriving at a purchase decision has not been examined directly. What indirect evidence exists tends to suggest that these variables associate positively with decision making time (DMT) (Anderson, Chu and Weitz 1987; Doyle, Woodside and Mitchell 1979).
Despite researchers' apparent neglect of DMT, it can be an important factor in the organizational buying process. DMT can be critical for marketing positions occupied by various "in" and "out" suppliers. Longer DMT offers many more suppliers the opportunity to influence the decision process. It could weaken the position of established or "in" suppliers and change the size and composition of the decision making unit (DMU) as well as the evoked set considered for a purchase decision. In addition, the longer the DMT the greater the chance of internal change and external turbulence which could alter the purchase decision drastically.
This research entails a direct examination of DMT and its antecedents. Specifically, we examine buyclass, size of the organization, DMU size, number of different information sources, and size of the evoked set as determinants of DMT.
This section contains a review of some of the constructs involved in the organizational purchase process. These variables have been isolated as possibly having an impact on DMT. Following this review, a set of research propositions concerning the relationships between these variables and their impact on DMT is given.
The purchase situation in organizational buying has often been addressed in terms of buyclass (Robinson, Faris and Wind 1967). The buyclass taxonomy classifies industrial purchase situations into three groups based on: 1) how novel or unfamiliar the purchase situation is perceived to be by the industrial buyer, 2) how much information the purchaser must gather to make the decision, and 3) the extent to which the purchaser seriously considers alternative solutions to the purchase problem (Robinson Farris and Wind 1967).
The resulting buyclasses are new task, modified rebuy, and straight rebuy. New tasks are unfamiliar, require extensive information, and extensive evaluation of alternatives. Organizational buyers regard new tasks as important and associate them with high risk. In such situations, Buying Centers or DMUs tend to be large. The straight rebuy situations falls at the other end of the spectrum. It represents a very common and familiar purchase situation. Most often it entails a routine purchase with no further information requirements and little effort in general. The modified rebuy involves a somewhat familiar purchase situation with some new information requirements and some further evaluation of alternatives. It can be an upgraded straight rebuy or a previously new task that has become more routine (Anderson, Chu and Weitz 1987).
The buyclass model has been researched widely. In general it has been validated and found to be a useful conceptual and analytical tool. Evidence suggests that buyclass influences both the composition and size of the DMU as well as information search behaviors. For example, Robinson, Faris and Wind (1967) suggested that organizational members' participation in the DMU varies according to buyclass or the novelty of the purchase situation. Cardozo (1980) found responsibility in the buying process to be affected by the purchasing task situation. Weigand (1968) also suggested that a variety of individuals from outside the purchasing department participate and influence complex buying situations (i.e., new tasks). Other researchers concur (McQuiston 1989; Spekman and Gronhaug 1986; Spekman and Stern 1979).
Critics of the buyclass framework have contended that it oversimplifies a complex phenomenon, overstates the role of newness in the process, and neglects central issues such as the importance of the acquisition (Choffray and Lilien 1978; Johnston 1981). Several researchers failed to find a relationship between buyclass and DMU, lending some credence to the aforementioned criticisms (Bellizzi and McVey 1983; Jackson, Keith and Burdick 1984). These studies suggested that relative influence on decisions within the DMU is affected less by buyclass than by product type. Most recently, Anderson, Chu and Weitz (1987) and McQuiston (1989) validated the buyclass typology. Anderson, Chu and Weitz (1987) concluded that in new task buying situations the buying center is likely to be large, slow to make a decision, and engage in extensive information search. McQuiston (1989) found that dimensions of the buyclass framework such as novelty and complexity relate to participation and influence in the acquisition process. Firm Size The effects of organizational size have been researched widely in areas such as organizational theory (e.g., Blau 1970; Cullen, Anderson and Baker 1986). With few exceptions, the effects of firm size on the organizational purchase process has not received much attention.
The recent exception to this lack of attention is the study by Moriarty and Spekman (1984). In their study of firms in the process of acquiring "dumb" data terminals, they found that size related to the type of information sources accessed in the purchase process. Their results seem to suggest that smaller organizations tend to rely extensively on external sources of information because they do not possess the resources or personnel to develop and accumulate purchase-related information internally (Moriarty and Spekman 1984, p. 143).
Other research suggests that the well-established and formal bureaucratic structures in place in larger organizations can facilitate access to more and better information. However, this same bureaucratic structure can become cumbersome and inhibit decision making (Rogers and Agarwalla-Rogers 1976; Webster 1980). In such cases, DMT could be slowed considerably.
Along with the buyclass model, DMU composition and size comprise perhaps the most widely researched components of organizational buying behavior (e.g., Robinson, Faris and Wind 1967). Johnston and Bonoma (1981) found that organizational size was not related to any dimensions of the DMU but a measure of organizational formality was positively related to DMU size.
Gronhaug's (1975) work suggests that DMU size is affected by, among other things, the type of buying situation confronted: more routine problems are likely to yield a smaller number of decision makers. Johnston and Bonoma (1981) found that perceived increases in the importance and complexity of the decision led to a larger DMU.
Weigand (1968) reported that the composition of the DMU is affected by the complexity of the buying function: more complex situations engender greater participation by organizational members outside the traditional purchasing department. While one might be led to expect that such changes in DMU participation would result in an increase in the DMU size, caution must be used -- Johnston and Bonoma (1981) found that although novel purchase situations led to participation in the DMU by higher levels of management, the DMU size actually did not change significantly. They also found that service acquisition decisions involved smaller DMUs than did equipment decisions.
Another area of inquiry in organizational buyer behavior has focused on information search activities of the DMU. The classic buyclass model suggests that information search activities will be more extensive in new buying tasks than for rebuy situations. That is, more novel buying situations should lead to accessing larger quantities of information. This could occur through utilizing a larger number of information sources and/or through information acquisition about a larger set of attributes, reflecting the presumed strategy of information acquisition as a method of risk reduction. This expectation is supported by the work of Choffray and Johnston (1979), Moriarty and Spekman (1984), Lee and Rethans (1984) and Anderson, Chu and Weitz (1987). Puto, Patton and King (1985), however, discovered a restricted pattern of search activity when purchase risk was high. Specifically, Cardozo and Cagley (1971) found greater reliance on known "in" suppliers in high-risk situations. As suggested by McMillan's (1972) work, this behavior may occur because "known" suppliers are perceived as less risky information sources than are new sources of information. In the telecommunications industry, for instance, the break up of the Bell System allowed several competitors in the switching and transmission equipment industry access to the telephone market but the vendors who won the telephone company contracts were primarily "established suppliers with good track records" (Business Week 1983, p. 180). Also, it may reflect a desire to build long-term favored relationships with a few respected suppliers.
Complicating research and measurement problems, at least to some degree, are the findings that organizational size (Moriarty and Spekman 1984; Webster 1980), stage in the decision process (McTavish and Guillery 1986), and personal characteristics of DMU numbers (Peters and Venkatesan 1973) may influence the degree of search activity undertaken.
Search activities result in the evoked set -- that set of alternatives seriously considered as a solution to the purchase problem. The magnitude of effort devoted to information search has serious consequences for the size of the evoked set. In one purchase situation, for example, Ameritech assembled a list of 110 equipment vendors, requested proposals from 45 and trimmed the final set to four before it made its equipment acquisition decision (Business Week 1983).
It has already been suggested that in some situations the risk perceived in more novel purchase situations might actually result in restriction of search activities rather than a more intensive seeking of information. Therefore, it is not surprising that the extant literature contains equivocal findings. For example, Ferguson (1979) and Saleh et. al. (1971) suggest that more routine purchase situations lead to less serious consideration of additional suppliers. Lee and Rethans (1984) also found that purchase scripts in modified rebuy situations contained fewer suppliers than did purchase scripts for newbuy situations. In addition, Anderson, Chu and Weitz (1987) found that the more novel the purchase situation the more willing members of the DMU seemed to entertain proposals from "out" suppliers and the less willing to favor "in" suppliers. Not all research, however, has produced results consistent with these findings. Puto, Patton and King (1985) in a simulation study found strong loyalty to existing suppliers in modified rebuy situations, particularly when purchase risk was high.
Further complicating these equivocal findings, Anderson, Chu and Weitz (1987) found that after controlling for product class effects, serious consideration of various supply alternatives related positively to the number of people in the decision making unit.
Our primary focus is to examine causal antecedents of the length of decision making processes in organizational purchases. Though, in and of itself, DMT has been largely neglected in the literature, the above discussion suggests a number of structural and process variables which are likely to influence it. As a guide for this study of decision making time in organizational purchasing, the following propositions were formulated for examination:
P1: The more novel the acquisition situation (buyclass), the greater the number of people involved in the decision (i.e., larger the DMU);
P2: The larger the firm, the more people involved in the purchase decision process (i.e., larger the DMU);
P3:The larger the firm, the more sources of information accessed in the acquisition process;
P4: The greater the number of individuals involved in the DMU, the more sources of information accessed in the acquisition process;
P5: The more novel the acquisition situation (buyclass), the larger the number of alternatives considered (evoked set size);
P6: The greater the number of individuals involved in the DMU, the greater the number of alternatives in the evoked set;
P7: The greater the number of information sources accessed, the more alternatives in the evoked set;
P8: The more novel the acquisitions situation (buyclass), the more time involved to arrive at an acquisition decision;
P9: The more sources of information accessed, the more time required to make an acquisition decision;
P10: The larger the DMU size, the more time required to make an acquisition decision;
P11: The more alternatives in the evoked set, the more time required to make an acquisition decision;
P12: The larger the firm, the more time required to make an acquisition decision.
The setting for this study was organizational decisions regarding acquisition of telecommunications products and services. This setting is desirable because, depending on the experience of the organization, characteristics of organizational needs, and the products/services being considered, purchase situations can involve the full range of buyclass categories.
The target population for this study was defined as small to mid-sized organizations located in the United States that have recently considered adopting or modifying their telecommunications system. Specifically, in terms of size, for-profit and non-profit organizations that use between 6 and 500 phone lines were targeted. In the telecommunications industry, this delineates small to medium sized firms. In addition, these organizations had to be considering the modification of their existing telecom products or services, or have made such an adoption decision in the 18 month period prior to data collection for this study. This information, as well as a variety of other data on qualifying organizations, was obtained from FOCUS Research of Hartford, Connecticut, a nationally recognized firm specializing in database services for the telecommunications industry.
In recent studies of organizational buying behavior, researchers have employed elaborate data collection procedures referred to as "snowballing" (Moriarty and Bateson 1982; Moriarty and Spekman 1984). "Snowballing" entails establishing a contact within an organization and querying that informant about who else participated in the decision process. In turn, this second wave of informants is queried about others in the DMU. The process is repeated until no new informants emerge. Thus, data are collected from a number of informants in each organization and, ostensibly, the researcher has access to information from all members of the DMU.
Other approaches involve gathering data from a single key informant external to the organization making the purchase, e.g., the salesperson (Anderson, Chu and Weitz 1987). This offers the obvious advantages of a less cumbersome, more controllable and executable data collection procedure. In addition, this single key informant has most probably observed and participated in the entire purchase process, and enjoyed a broadbased and objective view of it.
In this study, like Anderson, Chu and Weitz (1987), we collected information from a single key informant with an ongoing, broadbased view of the entire purchase process. However, we depart from the above study in that we selected this key informant from within the organization. Information was collected from a single critical key informant who was identified as knowledgeable and important in this purchase decision. Most often this key informant was in the upper management echelon. The selection of this knowledgeable key informant, the relatively small size of the buying organization, and the nature of the variables under investigation rendered an elaborate "snowball" data collection procedure unnecessary in this case.
From the commercial database pertaining to telecommunications purchases provided by FOCUS Research, approximately 4,000 organizations were identified as meeting the criteria of size and acquisition recency. A mail survey was designed to collect information from these organizations. Surveys were directed to an individual (by name) within each organization identified by the database as having an important role (probably the most important role) in decisions regarding adoption of telecommunications products and services.
Data collection procedures entailed several steps (Dillman 1978). First, contact was made by an advance letter which introduced and described the research project, and alerted respondents to expect the survey instrument. Second, one week after the advance notice, a cover letter, the survey questionnaire and a postage-paid reply envelope were mailed. Third, a reminder postcard was mailed to all participants approximately 10 days after the survey package.
The survey resulted in 1,147 usable responses. This response rate of 29% compares favorably with response rates in other studies involving organizational purchasing processes (e.g., Anderson, Chu and Weitz 1987; Moriarty and Spekman 1984). As a check for response bias, early and late respondents were compared on the basis of respondent characteristics, firm size in terms of employees, length of decision time, and current telecommunications system. Chi-square tests indicated no significant difference between early respondents (those who responded within the first 12 days of the initial mailing) and late respondents (those who responded on or after the 13th day after the initial mailing) on any of these characteristics. To further insure the quality of the data, information on the firm size, respondent, and acquisition of equipment generated by the survey was compared with the information from the database from which the sample was drawn. It did not differ significantly, ensuring that information from the appropriate firms and respondents had been gathered. Table 1 summarizes other descriptive characteristics of the sample.
In general, measures were developed on the bases of academic literature, trade literature, and field interviews in the industry. Initial versions of the research instrument were pretested on a small pool of managers from organizations in the Northeast who had recently acquired a telecommunications system. The general format, content, and specific measures were revised on the basis of the pretest results.
As indicated in the research propositions, six variables were included in this study with the major focus (i.e., the major dependent construct) on Decision Making Time. An open-ended item asking respondents to indicate the number of months from initiation to completion of the purchase process was used to assess DMT.
For Buyclass, respondents indicated which of three categories best described their purchase situation. The three categories were: Routine Decision, Partially New Decision, and Completely New Decision. Consistent with Robinson, Faris and Wind (1967), in the questionnaire, each category was described in terms of the newness of situation, the amount of information needed, and the effort involved in the purchase. Nine categories of employee size assessed Firm Size. Since the study focused on small to midsized firms, the categories ranged from "less than 5 employees" to "1001 or more employees" on the questionnaire. The measure of DMU Size consisted of an open-ended item where respondents indicated the number of other individuals within the firm involved in the decision making process for this purchase. For the Number of Information Sources Used, a list of five external information sources was used which included vendors, consultants, seminars, trade shows, and professional associations or publications. Information sources in other research have been operationalized on the dimensions of personal/impersonal and commercial/noncommercial (Moriarty and Spekman 1984). However, preliminary field interviews strongly suggested that these five adequately covered the range of sources in this research setting. Respondents were asked to check all the sources accessed in the purchase decision. The sum of these sources served as the measure. For Evoked Set Size, respondents indicated how many vendors were asked to submit proposals for this purchase. Table 2 summarizes the operationalization of relevant variables. Table 3 shows the means, standard deviations and correlations among the variables being investigated.
Structural equation modeling with maximum likelihood estimation techniques (Joreskog and Sorbom 1984) was used to test the expected relationships. Research propositions one through twelve (P1 - P12) suggest the LISREL model shown in Figure 1. As the model indicates, the exogenous constructs, Buyclass ( 1) and Firm Size ( 2) are expected to affect the endogenous constructs DMU size ( 1), number of information sources ( 2), evoked set size ( 3), and DMT ( 4). Specific parameter estimates addressing the individual research propositions are given in the and matrices. However, before discussing individual parameter estimates, the overall fit of the LISREL model must be examined.
In LISREL, assessment of model fit involves verifying that the covariance structures derived from the model adequately represent those actually existing in the data. A statistically significant chi-square indicates that the covariance structures specified in the model do not replicate those existing in the data. In this analysis, the statistically nonsignificant chi-square of 1.14 (df=2, p=.57) indicates that covariance structures specified fit existing covariance structures in the data quite well.
The Goodness of Fit Indices (GFI and AGFI, adjusted for degrees of freedom) in LISREL reflect the relative amount of variances and covariances accounted for by the model (Joreskog and Sorbom, 1984, p. 41). Despite their limitations, researchers continue to recommend heavy reliance on the GFIs in model fit evaluation (Mulaik, et al., 1989). The GFI of .996 and the AGFI of .990 for the model estimated here fall well within the acceptable range of past research (e.g. Anderson and Naurus 1990; McQuiston 1989) and suggest that the model explains a more than adequate level of observed-measure covariation.
The Root Mean Square Residual (RMR) which assesses the average magnitude of the residuals suffers from the same weakness as the GFIs. However, based on past research (e.g., Anderson and Narus 1990; McQuiston 1989), the RMR of .007 for the model in Figure 1 falls well within acceptable ranges. These four measures for overall model fit assessment lead us to judge the model in Figure 1 as representing the data quite well. An examination of the individual parameter estimates for tests of the research propositions follows in the next section.
Proposition 1 states that more novel the acquisition situation (Buyclass), the greater the number of people who would participate in the decision making process. A parameter estimate of .133 for 11 (p>.05) indicates support for this proposition (see Table 4). In P2, organization size was also expected to relate positively to DMU size. The data suggest support for this relationship ( 12=.215, p>.05).
P3 and P4 suggest that the larger the Firm Size and the larger the DMU Size, the more the number of different sources of information that are likely to be accessed in the acquisition process. The results in Table 3 support expectations regarding the effects of Firm Size ( 22=.291, p>.05), and the effects of DMU Size ( 21=.202, p>.05) on the number of information sources accessed.
Buyclass, that is, the perceived novelty of the acquisition situation (P5), DMU Size (P6), and the number of information sources (P7) were expected to affect the number of alternatives appearing in the evoked set positively. Parameter estimates of .105 and .190 (p>.05) for 31 and 31, respectively, suggest support for P5 and P6. However, the statistically nonsignificant parameter estimate for 32 indicates no relationship between the number of information sources accessed and the size of the evoked set.
In P8, we expected that the more novel the acquisition situation, the longer it would take to complete the decision process. With a statistically nonsignificant parameter estimate for 41, this proposition received no support in the data. Number of information sources accessed (P9), DMU size (P10), and number of alternatives in the evoked set (P11) were expected to lead to longer decision making time. Parameter estimates of .120 for 42, .126 for 41, and .121 for 43 (for all, p>.05) indicate support for P9, P10 and P11. Proposition 12 states that the larger the size of the organization, the longer it would take to make the purchase decision. The statistically nonsignificant parameter estimate for 42 indicates lack of support for P12.
In general, the results of this exploratory study conformed to our expectations for the relationships among the variables examined. Consistent with past research, purchase situations involving new tasks lead to more individuals participating in the purchase process. In addition, new task purchases result in more alternatives being seriously considered. This result is inconsistent with the idea of reducing risk by limiting alternatives and staying with known suppliers. Firms in this study facing new buying tasks were apparently willing to put in more effort at generating alternatives to consider.
Surprisingly, buyclass did not affect decision making time directly. This is logically inconsistent given the theoretical underpinnings of the buyclass model. The only plausible explanation is that the measure of buyclass in this study consisted of a simple three group classification. Perhaps a more sensitive measure involving a continuum anchored at the two extremes of new task and straight rebuy would have captured the more subtle variances in purchase situations. Another approach could be to develop multiple item indicators of the three dimensions of the buyclass typology such as those used by Anderson, Chu and Weitz (1987).
The results also suggest that in larger firms, more individuals get involved in procurement processes. In addition, larger firms engage in more extensive information search since we found use of a larger number of sources and a larger number of vendor proposals. However, firm size did not affect decision making time directly. We had expected larger firms to be more cumbersome and slow in their decision processes. However this lack of findings could be due to our sample which included only small to midsized firms. Perhaps none of the firms in our sample were large enough to have cumbersome established bureaucracies which could lengthen the acquisition process.
Consistent with our expectations and with past research, DMU size has ubiquitous effects throughout the acquisition process. These results indicate that when more individuals are involved in the decision process, information search becomes more extensive and the sources of information more varied. Additionally, a greater number of alternatives are considered as possible solutions to the purchase problem probably because more and different organizational constituencies with more varied concerns are present in larger DMUs.
This study indicates that DMU size both directly and indirectly increases decision making time. Directly, the larger the DMU, the more time it takes to get all the individuals involved to agree on a decision. Indirectly, because larger DMUs result in more information and more alternatives, the decision process takes longer.
Although the number of information sources consulted increased the decision making time, the most unexpected finding in this study was that the use of a greater number of different information sources did not result in any greater number of alternatives being considered. Given the complex nature of the solutions being considered, it is likely that different sources are used to obtain information on a larger and varied number of attributes, rather than elimination or addition to the pool of alternatives considered. Our operationalization of the sources of information -- which included five different types of sources -- may have contributed to the observed result. Different information sources may be accessed because of knowledge and experience on product attributes as well as to corroborate available information. Although this was not directly tested in this study, it may explain the effect of the number of sources on decision making time but not on the number of alternatives considered.
For managers, the results of this study suggest that longer decision making times could be critical for both "in" and "out" suppliers. Longer decision making time is more hazardous for "in" suppliers. It could change the operating and competitive environment with more alternative vendors and solutions being considered by the purchaser. Even those suppliers enjoying source loyalty may find their positions eroded when decision making time stretches out. In the fast-changing and complex environment of high-technology sectors such as the telecommunications industry studied here, established or "in" suppliers have to develop specific strategies to cope with long decision times. The complexity of the decision tends to increase DMU size, number of information sources, and the size of the evoked set all of which lead to elongated decision time. To avoid unduly long decision time, "in" suppliers could segment the market and focus on smaller firms. In smaller companies, the DMU is smaller as well as the number of information sources consulted is smaller, both of which would curtail decision making time. This would minimize opportunities for "out" suppliers to get into the vendor consideration process. When dealing with larger firms, "in" suppliers have little choice but to deal with long decision times. To remain a preferred supplier in such cases, the established supplier has to ensure that consistent, positive information about that supplier is available through all the varied information sources the DMU of the larger firm is apt to consider.
In high-technology settings, the novelty of the purchase decision is not only unavoidable but even desirable -- after all, the appeal of high technology derives from its newness. This would in general lead to larger evoked sets and more extensive information search. There are, however, technologically complex situations in which suppliers are able to position themselves as known, reliable, low-risk providers of solutions and systems (e.g., Cardozo and Cagley 1971). As a popular saying in the computer industry goes: "Nobody ever got fired for buying an IBM computer." While some of the factors that lead to such "preferred" positioning in technologically complex settings are known (McKenna 1986), a lot more needs to be learned about such positioning (Loomis 1991). For "out" suppliers, lengthy decision times offer opportunities to influence DMU participants favorably. As the decision process lengthens, even source loyal buyers may be convinced to consider proposals from other alternatives.
Strategies appropriate for "out" suppliers are a mirror image of those appropriate for "in" suppliers. Suppliers who are not established ("out") can benefit from focusing on larger firms, emphasizing the novelty of the decision, conveying strong comparative images about themselves through varied information sources, and in general stressing the need for extensive deliberation before a decision is reached.
In a sophisticated market where both "in" and "out" suppliers pit against each other using well-formulated strategies of the type discussed here, the quality of the strategies and the marketing efforts may prove decisive in the vendor selection process.
The major limitation of this study involves the buyclass measure. As noted above, future research could benefit from a more sensitive and sophisticated treatment of the buyclass construct. Another limitation could be the sampling technique employed in this research. However, it is highly questionable whether, in this particular case, any further or more valid information could have been gained from a "snowballing" technique. Comparison of selected data provided by our targeted informant with data available in the database from other informants in the firm suggests that we can accept the survey information with a great deal of confidence. Perhaps some limited gains on information sources could have been realized by a different operationalization. In the future, researchers should consider this. Future research could extend the ideas presented in this exploratory study by investigating whether or not differences in the length of decision making time changes or impacts the acquisition decision. Some factors which could play a role in the acquisition decision as it stretches over time are changes in the firms' external environment, particularly the technological environment, but also in other components such as product safety and quality standards, and employee safety regulations, to name just a few. Internally, the firm's needs may also change, for example, because of changing personnel and/or management.
In summary, this study investigated the causal antecedents of decision making time in organizational purchase processes. DMU size, the number of different information sources, and the size of the evoked set were found to lengthen decision time. Contrary to expectations, buyclass and firm size did not directly lengthen decision time. Buyclass and firm size indirectly lengthened decision time through their positive influence on the size of the DMU. Future research should address the implications of longer decision making times in terms of changes in the purchase decision. In addition, the effects of changes in the internal and external firm environment need to be addressed in extreme situations when firms equivocate and take inordinately long over acquisition decisions.
Anderson, E., W. Chu and B. Weitz (1987), "Industrial Purchasing: An Empirical Exploration of the Buyclass Framework," Journal of Marketing, 51 (July), 71-86.
Anderson, J. C. and J. A. Narus (1990), "A Model of Distributor Firm and Manufacturer Firm Working Partnerships," Journal of Marketing, 54, 1 (January), 42-58.
Bellizzi, J., (1978), "Buyclass vs. Product Type as Predictive of Industrial Behavior," in Proceedings, Southern Marketing Association, Robert Franz, Robert Hopkins, and Al Toma, eds., Atlanta: Southern Marketing Association.
Blau, P. M. (1970) "A Formal Theory of Differentiation in Organizations," American Sociological Review, 35, 201-218. Business Week (1983) "A Scramble to Supply the New Phone Companies," Business Week (December 5), 179-180.
Cardozo, R. (1980), "Situational Segmentation of Industrial Markets," European Journal of Marketing, 14 (June), 264-76.
__________ and J. Cagley (1971), "Experimental Study of Industrial Buyer Behavior," Journal of Marketing Research, 8 (August), 329-334.
Choffray, J. and P. Johnson (1979), "Measuring Perceived Pre-Purchase Risk for a New Industrial Product," Industrial Marketing Management, 8, 333-340.
__________ and G. Lilien (1978), "A New Approach to Industrial Market Segmentation," Sloan Management Review, 19 (Spring), 17-30.
Cullen, J. B., Anderson, K. S. and D. D. Baker (1986), "Blau's Theory of Structural Differentiation Revisited: A Theory of Structural Change or Scale?" Academy of Management Journal, 29 (June), 203-29.
Dillman, D. (1978), Mail and Telephone Surveys, New York: John Wiley & Sons. Ferguson, W. (1979), "An Evaluation of the BUYGRID Analytic Framework," Industrial Marketing Management, 8, 40-44.
Gronhaug, K. (1975), "Autonomous vs. Joint Decisions in Organizational Buying," Industrial Marketing Management, 4 (October), 265-271.
Jackson, D.W. Jr., J. Keith and R. Burdick (1984), "Purchasing Agents' Perceptions of Industrial Buying Center Influence: A Situational Approach," Journal of Marketing 48 (Fall), 75-83.
Johnston, W. (1981), "Industrial Buying Behavior: A State for the Art Review," in Review of Marketing 1981, Ben M. Enis and Kenneth J. Roering, eds. Chicago: American Marketing Association 75-88.
__________ and T. Bonoma (1981), "The Buying Center: Structure and Interaction Patterns," Journal of Marketing 45 (Summer), 143-156.
Joreskog, K. G. and D. Sorbom (1984), LISREL VI: Analysis of Linear Structural Relationships by the Method of Maximum Likelihood 3rd ed., Mooresville, in: Scientific Software.
Lee, T. and A. Rethans (1984), "A Script-Theoretic Analysis of Industrial Purchasing Behavior," Journal of Marketing, 48 (Fall), 22-32.
Lilien, G. L. and M. A. Wong (1984), "An Exploratory Investigation of the Structure of the Buying Center in the Multiworking Industry," Journal of Marketing Research, 21 (February), 1-11.
Loomis, C.J. (1991), "Can John Akers Save IBM?", Fortune, 124 (July), 40-56. McCabe, D. L. (1987), "Buying Group Structure: Constriction at the Top," Journal of Marketing, 51, 4 (October), 89-98.
McKenna, R. (1986), The Regis Touch: Marketing Strategies for Uncertain Times, Reading, MA: Addison-Wesley.
McMillan, J. (1972), "The Role of Perceived Risk in Industrial Marketing Decisions," in Proceedings of the American Marketing Association, Series 34, Chicago: American Marketing Association, 412-417.
McQuiston, D., (1989), "Novelty, Complexity and Importance as Causal Determinants of Industrial Buyer Behavior," Journal of Marketing, 53 (April), 66-79.
McTavish, R. and P. Guillery (1986), "The Information Search Behavior of Small Companies in the Purchase of Microcomputers: A Buying Process Framework," Industrial Marketing & Purchasing, 2, 30-42.
Moriarty, R. and J. Bateson (1982), "Exploring Complex Decision Units: A New Approach," Journal of Marketing Research, 19 (May), 182-91.
__________ and R. Spekman (1984), "An Empirical Investigation of the Information Sources Used During the Industrial Buying Process," Journal of Marketing Research, 21 (November), 137-147.
Mulaik, Stanley A., Larry R. James, Judith Van Alstine, Nathan Bennett, Sherri Lind, and C. Dean Stilwell (1989), "Evaluation of Goodness-of-Fit Indices for Structural Equation Models," Psychological Bulletin, 105, (3), 430-445.
Patchen, M. (1974), "The Locus and Basis of Influence In Organizational Decisions," Organizational Behavior and Human Performance, 11 (April), 195-211.
Peters, M. and M. Venkatesan (1973), "Exploration of Variables Inherent in Adopting an Industrial Product, "Journal of Marketing Research, 10, (August), 312-315.
Puto, C., W. Patton and R. King (1985), "Risk Handling Strategies in Industrial Vendor Selection Decisions," Journal of Marketing, 49 (Winter), 89-98.
Robinson, P.J., C.W. Farris, and Y. Wind (1967), Industrial Buying and Creative Marketing, Boston, MA: Allyn and Bacon.
Rogers, E., and R. Agarwalla-Rogers (1976), Communication in Organizations, New York: The Free Press.
Ryan, M. J. and M. B. Holbrook (1982), "Decision-Specific Conflict in Organizational Buyer Behavior," Journal of Marketing, 46, 3 (Summer), 62-68.
Saleh, F., B. Lalonde, J. Riley and J. Grabner (1971), "Modeling Industrial Buyer Behavior: The Purchase of Motor Carrier Services," in Proceedings of the American Marketing Association, Series 33, Chicago: American Marketing Association, 402-407.
Spekman, R. E. (1978), "A Macro-Sociological Examination of the Industrial Buying Center: Promise or Problems?," in Research Frontiers in Marketing: Dialogues and Directions, Subhash C. Jain, ed., Chicago: American Marketing Association, 111-115.
_____________. and K. Gronhaug (1986), " Conceptual and Methodological Issues in Buying Center Research," European Journal of Marketing, 20, 50-63. Webster, F. (1980), Industrial Marketing Strategy, New York: Ronald Press.
Weigand, R. E. (1968), "Why Studying the Purchasing Agent Is Not Enough," Journal of Marketing, 32, (January), 41-45.
This research was supported by
The Research
Institute for Telecommunications and Information Marketing (RITIM)
College of Business Administration
The University of
Rhode Island
Kingston, Rhode Island 02881-0802
Telephone
01-792-5065
e-mail: RITIM@URIACC.URI.EDU