Postconsumption Competition: the Effects of Choice and Non-Choice Alternatives on Satisfaction Formation

Cornelia Dr÷ge, Michigan State University
Robert D. Mackoy, Butler University
ABSTRACT - Current models of satisfaction formation focus on the single product or service chosen; this paper proposes that, in a competitive environment, alternative products or services not chosen can also impact satisfaction formation. Five propositions regarding a revised conceptualization of satisfaction formation processes are offered. The first is a general proposition stating that multiple pre-choice alternatives can remain relevant in the ultimate satisfaction formation process. The second proposition states that multiple paths of satisfaction processing are interrelated. The remaining propositions specify three conditions under which this reconceptualized process of satisfaction formation is likely to hold.
[ to cite ]:
Cornelia Dr÷ge and Robert D. Mackoy (1995) ,"Postconsumption Competition: the Effects of Choice and Non-Choice Alternatives on Satisfaction Formation", in NA - Advances in Consumer Research Volume 22, eds. Frank R. Kardes and Mita Sujan, Provo, UT : Association for Consumer Research, Pages: 532-536.

Advances in Consumer Research Volume 22, 1995      Pages 532-536


Cornelia Dr÷ge, Michigan State University

Robert D. Mackoy, Butler University


Current models of satisfaction formation focus on the single product or service chosen; this paper proposes that, in a competitive environment, alternative products or services not chosen can also impact satisfaction formation. Five propositions regarding a revised conceptualization of satisfaction formation processes are offered. The first is a general proposition stating that multiple pre-choice alternatives can remain relevant in the ultimate satisfaction formation process. The second proposition states that multiple paths of satisfaction processing are interrelated. The remaining propositions specify three conditions under which this reconceptualized process of satisfaction formation is likely to hold.


The vast literature on consumer choice behavior addresses the question of how consumers decide which of multiple competing alternatives to purchase. The notion of competition is an integral element of the choice process. Yet postchoice processes, such as satisfaction formation, are modeled in a manner inconsistent with the competitive reality in which they, too, occur. Current models of satisfaction formation focus exclusively on the single, chosen product while assuming that evaluations of competing products are no longer salient.

We contend that in many situations competing alternatives remain salient in postconsumption satisfaction formation processes. We propose a satisfaction formation model which explicitly considers how the ongoing evaluation of competitive alternatives affects satisfaction with the consumption experience. The traditional model of satisfaction formation is a special case of the proposed model.

In this paper, the current conceptualization of satisfaction formation, referred to as the traditional model, is summarized briefly. Then, five propositions are presented which jointly offer a revised conceptualization of the satisfaction formation processes. The first and second propositions provide an overview of the new model specifying the primary structural components of the new conceptualization. The remaining three propositions indicate three conditions under which this expanded conceptualization is likely to hold. Finally, conceptual and managerial implications are presented.


Study of consumer satisfaction is one topic which resides primarily in the post-decision, consuming realm of consumer behavior. Traditional models of satisfaction formation typically regard consumer choice as a given and specify satisfaction to be a function of antecedents such as expectation, desire, attitude, perceived performance, and disconfirmation relative to the one choice already made. Thus, most satisfaction models have been structured and tested in a manner consistent with Alderson's (1957) distinctions, that is, with consumption as totally separate from the "buying" process.

However, should the fact that it is possible to distinguish between the purchase decision and consumption imply that it is desirable to study each as if they were unrelated? Consider the implications of the current literature as illustrated in Figure 1. The consumer considers purchasing one of three alternative products or services. Numerous models of how the choice is made have been proposed, tested and supported (e.g., Bettman, Johnson and Payne 1991). Often such choice processes include intensive information manipulation on an attribute-by-attribute basis, comparing expected characteristics or performance levels. Such comparisons are made with the consumer's internal ideal standards and/or with the characteristics and performance levels of the other alternatives. The end result of this "choosing" or "buying" process is a selected alternative, in this case, Alternative B (see Figure 1).

Consumer satisfaction, as part of the "consuming process," is modeled as a response or a judgment of the consumption experience related only to Alternative B. Expectations, desires, attitudes, perceived performance, and disconfirmation related to Alternative B are considered sufficient to understand satisfaction with the consumption experience.

Thus when we compare the prechoice decision models with the postchoice satisfaction models, we note three important contrasts. First, all alternatives in the choice set are salient prechoice, but only the chosen alternative remains salient in satisfaction formation postchoice. Second, prechoice models typically offer some explanation as to how the alternatives are compared to one another, while postchoice satisfaction models typically offer no explanation as to how the alternatives not chosen potentially interact or impact overall satisfaction. Third, prechoice models often incorporate moderating factors such as involvement into explanations of alternative processing while postchoice satisfaction models assume that no moderating factor can make alternatives not chosen salient.

In summary, satisfaction is conceptualized as a single process related to a single target (see Figure 1). When one considers the intensive prechoice processing which involved Alternatives A and C in addition to B, it seems unreasonable to assume that thoughts or feelings about alternatives not selected become totally irrelevant to postchoice satisfaction formation processes. Expectations, desires and attitudes regarding the chosen Alternative B evolved in relation to expectations, desires and attitudes regarding the two other alternatives. Likewise, in some usage contexts, the perceived performance of Alternatives A and C may influence the consumer's interpretation of the consumption experience related to Alternative B. Given that many consumer decisions of interest to marketers occur in a competitive context C one in which alternatives exist, information (including trial) is available, processing is encouraged, and risk is involved C it becomes clear that the satisfaction literature has not developed a satisfaction model which explicitly considers the comprehensiveness of the processing of competitive alternatives. Figure 2 depicts a model proposed to address these issues. The basic propositions related to Figure 2 are delineated below. Throughout the subsequent discussion the chosen alternative will be called the "choice" while the non-chosen alternative(s) that was (were) in the consumer's original choice set will be called the "nonchoice(s)."

Proposition 1: Those alternatives salient pre-decision can remain salient post-decision. The evaluation of both the choice and the nonchoices can impact ultimate satisfaction with the consumption experience.



Support for Proposition 1 comes from a number of sources. First, in A Theory of Cognitive Dissonance, Festinger (1957) explicitly recognizes that dissonance can be created by the simple act of choosing one alternative while rejecting others. He states:

"....There will be some cognitive elements corresponding to the positive aspects of the unchosen alternative and some elements corresponding to the negative aspects of the chosen alternative which will be dissonant with the cognition of having chosen one particular alternative...." (p. 36).

Thus, Festinger provides some support for hypothesizing that nonchoices may continue to have an impact on consumers' postchoice experiences, at the very least in the period immediately following the act of choosing.

Second, theoretical support may also be found in economic regret theory. Loomes and Sugden (1982) state that ultimate utility derived from the choice is not

"....independent of the nature and combination of actions simultaneously rejected...." (p. 82).

In their theory, the anticipation of regret (and/or "rejoicing") related to the nonchoices is combined with factors derived from conventional utility theory. The entire constellation of alternatives in the choice set is evaluated in terms of expectations following the decision. For our purposes, the important point is that consumers develop expectations regarding their potential future responses of regret or rejoicing to both choice and nonchoice alternatives, and the degree to which these responses to both choice and nonchoices are realized determines ultimate utility (i.e., satisfaction).

We propose that nonchoices be incorporated in the satisfaction formation process through their direct impact on overall satisfaction; i.e., that "satisfaction" with nonchoices as well as satisfaction with the choice can determine overall satisfaction (see Figure 2). Two issues need to be addressed. First, what is overall satisfaction? Satisfaction scholars appear to be moving toward broadly defined targets of satisfaction (e.g., Bitner and Hubbert 1993). There seems to be a transition from focusing on a single, specific target of satisfaction (e.g., the product chosen) to focusing on an overall target (e.g., the consumption experience). Some studies have included both specific and general targets (e.g., Bitner 1990). Thus we conceptualize satisfaction to be realized at two levels: satisfaction with each of multiple targets (one of which is the choice) and satisfaction overall.

Second, what is the meaning of "satisfaction" with nonchoices? As stated by Houston, Sherman and Baker (1991):

"....we can and do encounter alternatives we rejected when visiting another university that we had considered attending or seeing someone driving another car we considered. At such times the focus for feelings of satisfaction or regret may be the rejected alternatives...." (p. 425).

The authors demonstrate, as proposed by Festinger (1957), that continued processing of information about the nonchoices could have an impact on overall satisfaction formation. Partial experiences (visiting the nonchoice university) or vicarious experiences (seeing someone driving the nonchoice car) may provide the necessary points of reference ("target") in this process. Remembered satisfaction with alternatives not chosen at a particular time, but previously chosen, could also impact overall satisfaction. For example, consider someone who has chosen to dine at restaurant B instead of restaurant A on a given night, where both restaurants have been frequented in the past. On this particular evening, a long wait is experienced. After 45 minutes, the consumer is poignantly aware that the restaurant A wait was never longer than half an hour. After an hour, images of restaurant A's dessert exasperate a growing hunger and impatience. Remembered satisfaction with restaurant A has an impact on the growing dissatisfaction with restaurant B.

Proposition 2: The paths of satisfaction formation processing leading to satisfaction with both choice and nonchoice targets are interrelated.



If satisfaction with the choice and satisfaction with the nonchoices both contribute to overall satisfaction (see Figure 2), it is reasonable ask what processes lead to satisfaction with each target. In addition to assuming a single target of satisfaction (i.e., the choice), the traditional model posits that all relevant antecedents of satisfaction also have a single target (though possibly consisting of multiple relevant attributes). For example, common antecedents of satisfaction are expectations and disconfirmation. The traditional satisfaction literature has investigated in great detail the standards of comparison to which the one target is compared, but such standards are assumed to be incorporated in disconfirmation (or possibly expectations). For example, when Cadotte, Woodruff and Jenkins (1987) talk about "best brand norms" or when Tse and Wilton (1988) talk about "ideal standards," they are referring to a set standard against which the person judges the target's performance.

In contrast we propose, as indicated in Figure 2, that it is likely that the processes themselves are interrelated. Expectations, desires, and attitudes formed prior to the act of choice continue to be updated and compared in the processing following choice. For example, actual prechoice expectations can be different from postchoice memory of prechoice expectations. In addition, disconfirmation related to nonchoice targets may directly impact satisfaction with the choice target.


The model that we propose is more comprehensive than the traditional model of satisfaction formation. We do not state that nonchoices are always salient in postchoice processing; indeed the seminal model with one target is a special subset of our model that probably adequately describes many consumption experiences. Thus the question arises of when it becomes more likely that the traditional model's explanatory power falls short. The remaining three propositions focus on factors which may influence the degree to which nonchoices affect satisfaction formation: dissatisfaction with the choice, involvement, and the number of close competitors.

Proposition 3: The higher the level of dissatisfaction with the choice, the more likely that nonchoices are salient postchoice.

Consumers who are dissatisfied may process differently than consumers who are satisfied. Satisfied consumers do not feel the need to rethink or reanalyze the factors which are responsible for their largely positive state, perhaps because there has been closure on the process. On the other hand, consistent with cognitive dissonance theory, dissatisfied consumers do not reach closure as easily. They experience "psychological discomfort" (Festinger 1957) and become motivated to reduce this dissonance. Empirical evidence is consistent with this explanation. For example, Maddox (1981) hypothesized that dissatisfied consumers need to rationalize their negative feelings. Westbrook and Newman (1978) hypothesized that dissatisfied consumers experience an "amplified sensitivity" to the satisfaction formation process. Dr÷ge and Halstead (1991) found strong empirical evidence of differential processing among dissatisfied consumers; they hypothesized that dissatisfied consumers engaged in greater cognitive processing.

If dissatisfied consumers process more extensively, it is reasonable to propose that this processing encompasses nonchoices. Some of the additional processing likely involves reexamining expectations and desires with regard to the choice. However, in addition, the "...positive aspects of the unchosen alternative...." described by Festinger (1957) will likely be recalled during this processing, as the dissatisfied consumers experience the actualized regret described by Loomes and Sugden (1982). By contemplating the now relatively more attractive nonchoices, consumers may lay the groundwork to avoid being dissatisfied in the future (Westbrook and Newman 1978), or they may simply be trying to come to terms with their dissatisfaction.

Proposition 4: The higher the level of involvement, the more likely that nonchoices are salient postchoice.

Involvement may be thought of as a person's degree of interest or level of arousal with a product or service due to its personal relevance. Although different kinds of involvement have been identified (e.g., Richins and Bloch 1986), there appears to be agreement that involvement usually results in higher levels of attention to and search for information (Houston and Rothschild 1978; Tigert, Ring and King 1976), in increased processing of information (Petty, Cacioppo and Schumann 1983), and a narrower "latitude of acceptance" (Rothschild and Houston 1977).

The increased level of information search and processing coupled with the more stringent acceptance criteria of highly involved consumers logically implies an intensive comparison process among the members of the choice set. The intensive comparison process results in heightened awareness of desires, expectations and attitudes with regard to each of the choice set members. For high involvement consumers, it is unlikely that nonchoice alternatives will suddenly become irrelevant in postchoice processing. Conversely, for low involvement consumers, wider latitudes of acceptance and reduced information processing make it less likely that nonchoice alternatives will continue to have much salience postchoice. Festinger (1957) stated that favorable attributes of nonchoices and unfavorable attributes of the choice produce postchoice dissonance. The larger the number of salient attributes and the more intensive the processing, the greater the likelihood that postchoice dissonance will result. The greater the dissonance, the greater the motivation to reduce the dissonance, and hence, the more likely it is that nonchoices will remain salient during post-consumption satisfaction formation.

Proposition 5: The more difficult the tradeoffs among competing alternatives, the more likely that nonchoices are salient postchoice.

If products and services are conceived of as bundles of attributes, then closely competing products/services ("close competitors") would be those which possess both similar attributes, and similar amounts of each attribute. Choosing between close competitors may require more extensive processing than does choosing between competitors which are not perceived as close. The additional processing may be required because no matter which type of decision strategy is being used (e.g., compensatory, lexicographic, elimination by aspects, etc.), similar ratings of the attributes of each alternative complicates the decision process. Even if competitors are not "close" in this sense, more extensive processing may be necessary if the tradeoffs between the alternatives are difficult. The increased processing due to similarity between alternatives or due to the difficulty of making tradeoffs has the same consequences as the increased processing due to involvement: it can result in the formation of highly refined expectations and attitudes regarding multiple alternatives and/or in postchoice cognitive dissonance. In short, nonchoices are more likely to remain salient in the postchoice formation of overall satisfaction.


We have presented a comprehensive model of satisfaction formation which builds on the traditional model by including the potential incorporation of competitive factors postchoice. Postchoice competition exists when nonchoices remain salient in the satisfaction process. When this occurs, parallel yet interrelated processing paths (associated with the choice and each of the salient nonchoices) are proposed to lead to satisfaction with each target, and hence to satisfaction overall. When nonchoices are not salient postchoice, the model reduces to a single path model identical to the traditional model. This is the sense in which the traditional model is a special case of the competitive satisfaction model.

The traditional models' focus solely on the choice has been critical to the development of satisfaction formation constructs and to the specification of fundamental relationships. To address critical issues of internal validity, seminal works in developing the current disconfirmation-based models of satisfaction formation have utilized experimental methods in which subjects were exposed to a single product in a non-competitive context (e.g., Churchill and Surprenant 1982; Tse and Wilton 1988; Spreng and Olshavsky 1993). While useful for construct and model development, single product empirical tests artificially remove examination of the satisfaction process from the competitive environment in which it actually occurs. Although some questions remain regarding the conceptualization of and relationship between key constructs (e.g., Yi 1990), we believe that theoretical breakthroughs are unlikely to occur unless the reality of continuing competition is incorporated explicitly into a more generalizable model.

We contend that satisfaction research may advance more rapidly in the context of the comprehensive, competitive model. There are three reasons for this. First, explicitly incorporating competitive elements in the satisfaction formation model will make it more consistent with current choice models. The traditional model of satisfaction formation lacks theoretical links to the rich choice literature; the competitive model offers the framework for potentially becoming theoretically integrated with this literature. Second, the competitive model offers the potential to explain exactly how satisfaction relates to intentions and future behaviors. For example, the traditional model inadequately explains why many consumers do not repurchase even when they are highly satisfied with their choice. The competitive model provides a more reasonable context for addressing this issue. Third, the generalizability of the competitive model should contribute to its managerial relevance, and the model fits well with the strategic objective of developing long-term competitive advantage.

The proposed competitive model needs to be developed empirically and theoretically. Empirically, the next step is to determine whether there is initial support for the competitive model by showing cases in which nonchoices are salient postchoice. A logical context in which to test the model would be one in which the conditions presented in Propositions 3, 4, and/or 5 applied; i.e., a situation in which there is expected to be widespread dissatisfaction, and/or high involvement, and/or product similarity. Once support for the model is demonstrated, other specific conditions under which the model applies can be explored, and the paths tentatively identified here can be empirically tested.

Opportunities for theoretical development occur on several fronts. First, work needs to be done to determine other conditions under which one can expect a priori that the nonchoices remain salient postchoice. In this paper, one product-related factor (product similarity) and two consumer-related factors (level of dissatisfaction and involvement) are proposed to be criteria affecting nonchoice salience. Second, what specific elements of choice models apply to postchoice satisfaction formation? For example, does the choice heuristic utilized affect whether nonchoices remain salient postchoice? Third, exactly how does the proposed model relate to the consumer's consideration set? Might the continuing salience of nonchoices in postchoice processes provide empirical evidence of which specific elements comprised the consideration set? Fourth, if nonchoices remain salient, what effect do they have on future choices? For example, are such effects mediated by satisfaction or are there also direct effects?

Two major managerial implications should also result from development of the propositions. First, if the model is supported, it will provide clear evidence that competition continues postchoice. Most marketing effort is directed towards attracting customers by focusing on prechoice competition. The model implies that one's competitors can affect the satisfaction formation of one's own customers, and vice versa. This opens explicit consideration of a new competitive arena requiring new strategies and tactics. Defensively, managers may have to alter their strategies for increasing/maintaining consumer satisfaction and preventing dissatisfaction. Offensively, managers can develop methods for affecting the satisfaction formation of their competitors' customers. Second, managers may have to alter the way they monitor customer satisfaction. Not only should process variables relevant to one's own product/service be monitored, but the monitoring system should also be expanded to include relevant process variables pertaining to the products/services of one's major competitors. By doing so, managers can gain insights into potentially exploitable causes of dissatisfaction with competing products, while also monitoring the effects of competitors' actions on one's own customers. Specifically, for one's own customers, managers can determine a) what specific factors affect satisfaction and dissatisfaction, and b) if the satisfaction processes of one's own customers are affected postchoice by one's competitors. For the customers of one's competitors, managers can determine a) what specific factors affect their satisfaction and dissatisfaction, and b) if one's own products remain salient to competitors' customers postchoice. This information can be used to develop strategies to exploit advantages and minimize the consequences of the firm's weaknesses.

In summary, we propose that it is essential for satisfaction researchers to include competitive alternatives in the analysis of postchoice satisfaction not because these nonchoices are always relevant but because a determination of when and how they remain salient has major theoretical and managerial implications.


Alderson, Wroe (1957), Marketing Behavior and Executive Action, Homewood, IL: Richard D. Irwin.

Bettman, James R., Eric J. Johnson, and John W. Payne (1991), "Consumer Decision Making," in Handbook of Consumer Behavior, Thomas S. Robertson and Harold Kassarjian, eds. Englewood Cliffs, NJ: Prentice-Hall, 50-84.

Bitner, Mary Jo (1990), "Evaluating Service Encounters: The Effect of Physical Surroundings and Employee Responses," Journal of Marketing, 54 (April): 69-82.

Bitner , Mary Jo, and Amy R. Hubbert (1993), "Encounter Satisfaction Versus Overall Satisfaction Versus Quality," in Service Quality: New Directions in Theory and Practice, Roland T. Rust and Richard L. Oliver, eds. Thousand Oaks, CA: Sage Publications, Inc., 72-94.

Cadotte, Ernest R., Robert B. Woodruff, Roger L. Jenkins (1987), "Expectations and Norms in Models of Consumer Satisfaction," Journal of Marketing Research, 24:3, 305-314.

Churchill, Gilbert A., Jr., and Carol Surprenant (1982), "An Investigation into the Determinants of Consumer Satisfaction," Journal of Marketing Research, 19:4 (November), 491-504.

Dr÷ge, Cornelia, and Diane Halstead (1991), "Postpurchase Hierarchies of Effects: The Antecedents and Consequences of Satisfaction for Complainers versus Non-complainers," International Journal of Research in Marketing, 8:4, 315-328.

Festinger, Leon (1957), A Theory of Cognitive Dissonance, Stanford, CA: Stanford University Press.

Houston, Michael J. and Michael L. Rothschild (1978), " Conceptual and Methodological Perspectives on Involvement," in 1978 AMA Educators' Proceedings, Subhash C. Jain, ed. Chicago, IL: American Marketing Association, 184-187.

Houston, David A., Steven J. Sherman, and Sara M. Baker (1991), "Feature Matching, Unique Features, and the Dynamics of the Choice Process: Predecision Conflict and Postdecision Satisfaction," Journal of Experimental Social Psychology, 27:5, 411-430.

Loomes, Graham, and Robert Sugden (1982), "Regret Theory: An Alternative Theory of Rational Choice Under Uncertainty," The Economic Journal, 92 (December), 805-824 .

Maddox, R. Neil (1981), "Two-Factor Theory and Consumer Satisfaction: Replication and Extension," Journal of Consumer Research, 8:1, 97-102.

Petty, Richard E., John T. Cacioppo, and David Schumann (1983), "Central and Peripheral Routes to Advertising Effectiveness: The Moderating Role of Involvement," Journal of Consumer Research, 10:2, 135-144.

Richins, Marsha L. and Peter H. Bloch (1991), "Post-Purchase Product Satisfaction: Incorporating the Effects of Involvement and Time," Journal of Business Research, 23:2, 145-158.

Rothschild, Michael L., and Michael J. Houston (1977), "The Consumer Involvement Matrix: Some Preliminary Findings," in Proceedings: AMA Educators' Conference, Barnett A. Greenberg and Danny N. Ballinger, eds. Chicago, IL: American Marketing Association.

Spreng, Richard A., and Richard W. Olshavsky (1993), "A Desires Congruency Model of Consumer Satisfaction," Journal of the Academy of Marketing Science, 21:3, 169-178.

Tigert, Douglas J., Lawrence J. Ring, and Charles W. King (1976), " Fashion Involvement and Buying Behavior," in Advances in Consumer Research, Vol. 3, Beverlee B. Anderson, ed. Ann Arbor, MI: Association for Consumer Research, 46-52.

Tse, David K. and Peter C. Wilton (1988), "Models of Consumer Satisfaction Formation: An Extension," Journal of Marketing Research, 25:2 (May), 204-212.

Westbrook, Robert A. and Joseph W. Newman (1978), "An Analysis of Shopper Dissatisfaction for Major Household Appliances," Journal of Marketing Research, 15:3, 456-466.

Yi, Youjae (1990), "A Critical Review of Consumer Satisfaction," in Review of Marketing 1990, Valarie A. Zeithaml, ed. Chicago, IL: American Marketing Association.