A Re-Examination of Consumer Decision Making For a Repeat Purchase Product: Variations in Product Importance and Purchase Frequency

Alain d'Astous, University of Sherbrooke
Idriss Bensouda, University of Sherbrooke
Jean Guindon, University of Sherbrooke
ABSTRACT - This paper reports on a replication of Hoyer's (1984) study on in-store consumer decision making. The study uses a more important, less frequently purchased product category for the choice task. Differences in information search and choice tactics are presented and discussed. The results show that the decision making process of consumers is more complex as hypothesized. However, the extent of deliberative decision making is shown to be rather low.
[ to cite ]:
Alain d'Astous, Idriss Bensouda, and Jean Guindon (1989) ,"A Re-Examination of Consumer Decision Making For a Repeat Purchase Product: Variations in Product Importance and Purchase Frequency", in NA - Advances in Consumer Research Volume 16, eds. Thomas K. Srull, Provo, UT : Association for Consumer Research, Pages: 433-438.

Advances in Consumer Research Volume 16, 1989      Pages 433-438


Alain d'Astous, University of Sherbrooke

Idriss Bensouda, University of Sherbrooke

Jean Guindon, University of Sherbrooke


This paper reports on a replication of Hoyer's (1984) study on in-store consumer decision making. The study uses a more important, less frequently purchased product category for the choice task. Differences in information search and choice tactics are presented and discussed. The results show that the decision making process of consumers is more complex as hypothesized. However, the extent of deliberative decision making is shown to be rather low.


The rational approach to consumer decision making where it is proposed that consumers go through several stages in the purchase process problem recognition, information search, evaluation of alternatives, decision and purchase, post-purchase evaluation (see Wilkie, 1986 for a representative discussion) - has been challenged by numerous marketing authors. Studies have shown that for many products consumers either spend very little time or do not even engage in some of the sequential activities that are hypothesized to form a "rational" purchase process (Olshavsky and Granbois, 1979). Several empirically supported facts are inconsistent with the economic man model. Thus, it has been repeatedly found that many consumers undertake little or no prepurchase external information search (Beales et al., 1981; Claxton, Fry and Portis, 1974; Katona and Mueller, 1955; Newman and Staelin, 1972). Some researchers also argue that there is often limited planning prior to a purchase decision and that consumers construct decision rules at the time of choice (Bettman and Zins, 1977; Biehal and Chakravarti, 1986). In addition, most researchers believe that consumers do not typically apply analytical decision rules to make optimal choices but instead rely on heuristics that lead to "satisficing" solutions (Bettman, 1979; Payne, 1976). The rational view also overlooks the fact that most consumer decisions are repetitive and relatively unimportant and that, consequently, consumers need not go through an elaborate purchase process but may make their choices on the basis of habit or brand loyalty (Kassarjian, 1978).

A study by Hoyer (1984) provides a recent attempt at showing that consumers do not engage in extensive in-store deliberation when making purchase decisions. In that study, 120 consumers were observed while making a laundry detergent purchase. The results showed that a great majority of consumers (more than 70%) did not examine more than one package, did not make comparisons between brands or between different sizes within a single brand and did not examine any shelf tag. The observed consumers spent an average of 13 seconds from entering the aisle until making the purchase decision. Finally, when asked why they had made that particular brand choice, 91% of the consumers gave a single reason, related to price (e.g. its the cheapest brand), affect (e.g. I love it), performance (e.g. it's the best) or social norms (e.g. my wife likes it).

Two factors make the Hoyer (1984) study a particularly interesting one. First, its focus on an uninvolving frequently purchased product contrasts with some of the studies that have examined consumer decision making in the context of more involving purchases. Second, lt uses an unobtrusive methodology to study how consumers actually make their brand choice in the store and complements it with standard interviews with the observed consumers.

However, as Hoyer (1984) himself notes, the generalization power of his research is greatly limited since only one product category was studied. Clearly, in order to better understand consumer decision making, research must be carried out with other products. Consistent with this observation, this paper reports on a replication of the Hoyer (1984) study with a different consumer product. The product was chosen on the basis of importance and purchase frequency. Since one objective was to examine a distinct purchase situation, a more important, less frequently purchased product was selected for study.


According to Hoyer (1984), when a product is purchased repeatedly and is relatively not important, the decision making process of consumers is characterized by low cognitive effort and use of very simple choice tactics. These tactics reflect the fact that consumers have acquired experience with similar product purchases in the past. Therefore, the decision process does not unfold at the moment of choice, but rather evolves over time on the basis of perceived satisfaction with purchased brands. For example, a common choice tactic for a consumer who is satisfied with a particular brand could be "buy my favorite brand as I usually do."

In Hoyer's (1984) view, purchase frequency and product importance are basic dimensions, as far as consumer decision making is concerned (see Assael, 1987 for a similar analysis). One would think that a different scenario would characterize the purchase decisions of more important, less frequently purchased products. When a product purchase is made on an infrequent basis, brand satisfaction may be more difficult to evaluate since consumers may not be able to remember precisely how the brand performed. Less frequent purchase decisions may lead consumers to reconsider their decision criteria at the time of choice. Also, the more important a product is to the consumer, the more complex the decision making process. Products with higher social, personal or financial risks are likely to motivate consumers to engage in external information search and more deliberative decision processes. This is not to say that complex decision making always characterize the purchase of risky products. Brand loyalty and following others' advice are effective strategies to reduce risk (Roselius, 1971). However, we argue that even when these strategies are used, consumers will devote more cognitive effort to the purchase of an important product.

Based on the above discussion, the following research propositions are therefore put forward:

P1: In the case of a more important, less frequently purchased product, consumers will be more likely to seek in-store information and display deliberative choice processes.

P2: In the-case of a more important, less frequently purchased product, consumers will use different choice tactics. In particular, normative choice tactics should be used more.


Product Category

Analgesics were chosen as choice stimuli since that product category seemed to possess interesting characteristics given the objectives of the study. Thus, contrary to laundry detergent, analgesics are less frequently purchased. Also, since their use is associated with health care, they should be more involving to consumers. Finally, there are many brands of analgesics available with different components and they come in a variety of sizes and forms. This last characteristic implies that some sort of in-store decision making process must occur.


The study took place in a drug store considered as one of the largest in a mid-size (100,000 inhabitants) Canadian city. A total of 98 subjects were observed while shopping for analgesics and interviewed once their final brand choice was made. Given the nature of the product (i.e. low purchase frequency), the data were collected at times of high customer traffic (i.e. Thursday and Friday in the evening, Saturday and Sunday all day). Table 1 presents some descriptive information about the sample.

Observation Phase

The first phase of the data collection process consisted in observing consumers while they were shopping for analgesics. An observer stood near the analgesics display to take various measurements. There was no apparent consumer awareness of being watched. In order to facilitate the data collection, the observer used a chart-like reproduction of the display. A coding system similar to that used by Hoyer (1984) was employed. Each time a consumer stopped in front of the display, the observer circled all brands that were picked up and placed an "x" on the chosen brand. In addition, the observer recorded the total time of the purchase episode, from the moment the consumer stopped at the display to the final choice, as well as the time spent on each brand that was picked up.

Interview phase

Immediately after having completed their purchase, consumers were approached by a different experimenter and offered a 50-cent coupon redeemable for the purchased brand to participate in the study. Only 11 of the 109 consumers that were approached refused to participate. The primary purpose of the interview was to assess consumers' choice tactics. This was accomplished with the help of an open question asking why the chosen brand was purchased. The rest of the interview was structured and involved questions about the extent of purchase planning, purchase experience, brand knowledge, brand loyalty, purchase frequency, preference for the chosen brand. influence of in-store and out-of-store advertising, degree of attention to package information, and standard demographic information.


Search Indices

Table 2 summarizes the results associated with the observation phase. We should note first that the observation set-up did not permit an evaluation of the number of packages and the number of shelf tags examined by consumers. The analgesics display was fairly small and such measures would have necessitated the use of an eye-monitoring device.

As shown in Table 2, the results are generally supportive-of the first research proposition. Compared with Hoyer's (1984) findings, the mean number of packages picked up by consumers in this study is higher. A similar pattern is obtained with the mean number of within-brand comparisons (e.g. between different sizes), the average time spent on examining a brand and the average total time of the purchase episode. The only search index which is lower in the present study is the mean number of across-brand comparisons.

In order to test if the differences between the mean indices observed in this study and those of Hoyer (1984) are statistically significant, information on the variances of the variables is needed. Hoyer (1984) did not provide this information, but he supplied the frequency distribution of all variables (see Table 2). Such data allow the use of chi-square goodness-of-fit statistics to test if the frequency distributions obtained in this study are statistically different. The statistical results are presented in Table 3. They confirm the conclusions made earlier. The only variable which does not show higher information search is the number of across-brand comparisons. Apparently, consumers shopping for analgesics do not compare a greater number of brands but make more within-brand comparisons on attributes such as size, ingredients, strength and so on.









Choice Tactics

The responses given to the open question on choice tactics were coded as either related to price, affect, performance or social norms. Table 4 presents the distribution of consumers among the choice tactics. The results seem to partially support the second research proposition since the percentage of consumers having mentioned a normative choice tactic (e.g. my doctor has recommended that I buy this brand) is higher than in Hoyer's (1984) study. However, the major difference clearly occurs at the level of price tactics where the percentage of consumers having come forward with this type of reason is much smaller. It seems that consumers do not think that price is an appropriate attribute for selecting a brand of analgesics. The observed distribution of choice tactics among consumers in this study is statistically different from that reported by Hoyer (1984) (Chi-square = 12.684, df = 4,p = 0.013).

Additional Analyses

Several analyses of variance (ANOVA) were conducted to investigate the relationships between choice tactics and the other study variables. Table 5 presents the statistically significant relationships that were uncovered. It should be noted that the exploratory nature of this analysis makes it likely to obtain some statistically significant results by chance. However, three observations are inconsistent with this interpretation: (1) the relationships are fairly strong (the R2's are all > 0.12), (2) the results are highly significant (all but two p-values < 0.01), and (3) the results seem to make sense.

Two results essentially replicate what Hoyer (1984) found. Thus, price tactic consumers are less brand loyal and have less purchase experience with analgesics. On the other hand, affect and performance tactic consumers are more brand loyal and have more purchase experience. Consistent with these results, price tactic consumers are less likely to have planned the purchase of the chosen brand, have weaker brand preferences and have purchased the chosen brand less frequently than affect and performance tactic consumers.



Finally, it is interesting to note that the results obtained with some search indices confirm that price tactic consumers expend more effort during the purchase episode. On average, they pick up a greater number of packages, make a greater number of across-brand comparisons and spend more time to make the purchase. Clearly, this is what one should expect from consumers having less purchase experience with a somewhat involving product.


Overall, the results of this study show that the in-store decision making process of consumers is more complex when a product is more important and less frequently purchased. Almost all search indices computed in this study are significantly higher than those reported by Hoyer (1984) for a less important, frequently purchased product. Also, product importance and purchase frequency appear to influence consumers' use of choice tactics. In this study, consumers were more likely to mention affect, performance and normative reasons to explain their brand choice and less price-related reasons than in the Hoyer (1984) study.

However, the above conclusions must be tempered by three observations. First, comparisons between studies conducted at different times, in different places, with different populations and different researchers are hazardous. Such comparisons are not made within an experimental context and, as such, do not permit to draw clear conclusions about true explanatory variables. For instance, we should note that drug stores and grocery stores are different shopping environments. There is generally more consumer traffic in grocery stores and more pressure to speed up the choice process. This fact could well explain why consumers in this study took more time to shop and displayed more search.

Second, even though variations in purchase frequency and product importance distinguish analgesics from laundry detergent, there are obviously other features that make them different as well. Any of these features is a potential candidate for explaining differences in consumer decision making. For instance, analgesics are health products and consumers should be somewhat careful in purchasing them. Within this perspective, it seems understandable that consumers would not confess that they have used a price-related choice tactic.

Finally, although the data presented in this paper support the proposition that product importance and purchase frequency have a significant impact on consumer decision processes, it is worth noting that the extent of in-store deliberative decision making observed in this study is not very great. Thus, almost 70% of the consumers picked only one package of analgesics, near 80% did not make any across-brand comparison and almost 85% did not make any within-brand comparison. In light of these results, we are compelled to conclude, along with other researchers (Olshavsky and Granbois, 1979), that even for a relatively important, less frequently purchased product category such as analgesics, consumers do not engage in complex decision making processes or, perhaps, in any decision making at all.


Assael, Henry (1987), Consumer Behavior and Marketing Action, Third Edition, Boston, MA: Kent Publishing Company, 1987.

Beales, Howard, Michael B. Mazis, Steven C. Salop, and Richard Staelin (1981), "Consumer Search and Public Policy," Journal of Consumer Research, 8, (June), 11 -22.

Bettman, James R. (1979), An Information Processing Theory of Consumer Choice, Reading, MA: Addison-Wesley.

Bettman, James R. and Michel A. Zins, M., "Constructive Processes in Consumer Choice," Journal of Consumer Research, 4 (September), 75-85.

Biehal, Gabriel, and Dipankar Chakravarti (1986), "Consumers' Use of Memory and External Information in Choice: Macro and Micro Perspectives," Journal of Consumer Research, 12 (March), 382-405.

Claxton, John D., Joseph N. Fry, and Bernard Portis (1974), "A Taxonomy of Prepurchase Information Gathering Patterns,-" Journal of Consumer Research, 1 (December), 35-42.

Hoyer, Wayne D. (1984), "An Examination of Consumer Decision Making for a Common Repeat Purchase Product," Journal of Consumer Research, 11 (December), 822-829.

Kassarjian, Harold H.(1978), "Presidential Address 1977: Anthropomorphism and Parsimony," in Advances in Consumer Research, Vol. 5, (ed) H.K. Hunt, Ann Arbor, MI: Association for Consumer Research, xiii - xiv.

Katona, George, and Eve Mueller (1955), "A Study of Purchase Decisions on Consumer Behavior," in Consumer Behavior: The Dynamics of Consumer Reaction, Vol. 1, ed Lincoln H. Clark, New York: New York University Press, 1955, pp. 30-87.

Newman, Joseph W., and Richard Staelin (1972), "Prepurchase Information Seeking for New Cars and Major Household Appliances," Journal of Marketing Research, 9 (August), 242-257.

Olshavsky, Richard W., and Donald H. Granbois (1979), "Consumer Decision Making-Fact or Fiction?", Journal of Consumer Research, 6 (September), 93-100.

Payne, John W. (1976), "Heuristic Search Processes in Decision Making," in Advances in Consumer Research, Vol. 3, ed. Beverlee B. Anderson, Chicago: Association for Consumer Research, 1976, pp. 321-327.

Roselius, Ted (1971), "Consumer Rankings of Risk Reduction Methods," Journal of Marketing, 35 (January), 56-61.

Wilkie, William L. (1986), Consumer Behavior, New York: John Wiley & Sons.