Impulse purchases in retailing

Contents

I. Table of Tables

II. Table of Figures

III. Table of Abbreviations

1. Introduction
1.1. Purpose of the thesis
1.2. The framework and related research questions
1.3. Organization of the research

2. Literature Review
2.1. Introduction
2.2. Who makes impulse purchases?
2.3. What is bought on impulse?
2.4. Where are impulse purchases conducted?
2.5. How much is bought on impulse?
2.6. When are impulse purchases made?
2.7. How is impulse purchasing defined?
2.8. How are impulse purchases measured?
2.9. Types of impulse purchases
2.10. Why are products bought on impulse?
2.11. How do people act upon impulses? Resistance strategies
2.12. Factors that influence consumers’ impulse buying behavior
2.13. How to boost impulse purchases? Tactics for retail managers
2.14. Areas for further research

3. Study 1
3.1. Method section
3.1.1. Participant characteristics
3.1.2. Sampling procedures
3.1.3. Measures
3.1.4. Measurement instruments
3.1.5. Data collection, procedures, and analysis
3.2. Result section
3.2.1. Respondents’ definitions of impulse buying
3.2.2. Emotions and Cognition
3.2.3. Impulse Buying Motives
3.2.4. Buying determinants
3.2.5. Involvement, perceived risk, and impulse buying tendency
3.2.6. Shopping company and store visit frequency
3.2.7. Results concerning gender differences
3.2.7.1. Emotions and cognition
3.2.7.2. Impulse buying motives
3.2.7.3. Buying determinants
3.2.7.4. Involvement, perceived risk, and impulse buying tendency
3.2.7.5. Shopping company
3.2.8. Correlations
3.3. Discussion section
3.3.1. Respondents’ definitions of impulse buying
3.3.2. Emotions and Cognition
3.3.3. Impulse Buying Motives
3.3.4. Buying determinants
3.3.5. Involvement, perceived risk, and impulse buying tendency
3.3.6. Shopping company
3.3.7. Gender differences
3.3.7.1. Emotions and Cognition
3.3.7.2. Impulse buying motives
3.3.7.3. Buying determinants
3.3.7.4. Involvement, perceived risk, and impulse buying tendency
3.3.7.5. Shopping company
3.3.8. Correlations
3.3.9. Limitations and recommendations for further research

4. Study 2
4.1. Method section
4.1.1. Setting
4.1.2. Participant characteristics
4.1.3. Sampling procedures
4.1.4. Measurement instruments
4.1.5. Data collection, procedures, and analysis
4.2. Results section
4.2.1. Individual and situational difference variables
4.2.1.1. Significant effects
4.2.1.2. Insignificant effects
4.2.2. Emotions and cognition
4.2.3. Impulse Buying Motives
4.2.4. Buying determinants
4.2.5. Non-buying reasons
4.2.6. Results concerning gender differences
4.2.6.1. Emotions and cognition
4.2.6.2. Impulse buying motives
4.2.6.3. Buying Determinants
4.2.6.4. Involvement, perceived risk, and impulse buying tendency
4.2.6.5. Shopping company
4.2.7. Correlations
4.3. Discussion section
4.3.1. Individual and situational difference variables
4.3.1.1. Significant effects
4.3.1.2. Insignificant effects
4.3.2. Emotions and cognition
4.3.3. Impulse buying motives
4.3.4. Buying determinants
4.3.5. Non-buying reasons
4.3.6. Results concerning gender differences
4.3.6.1. Emotions and cognition
4.3.6.2. Impulse buying motives
4.3.6.3. Buying determinants
4.3.6.4. Involvement, perceived risk, and impulse buying tendency
4.3.6.5. Shopping company
4.3.7. Correlations
4.3.8. Limitations and recommendations for further research
4.3.9. Managerial implications

5. Conclusion

6. List of References

Appendices
Appendix 1: Questionnaire Study 1
Appendix 2: Impulse purchase definitions and episodes
Appendix 3: Factor analysis Study 1
Appendix 4: Correlation tables and graphics
Appendix 5: Photos of inspiration island
Appendix 6: Floor plan of the store
Appendix 7: List of products on the inspiration island
Appendix 8: Questionnaire for buyers Study 2
Appendix 9: Questionnaire for non-buyers Study 2
Appendix 10: Factor analysis Study 2

I. Table of Tables

Table 1: Classification scheme of buyer types (Cobb & Hoyer, 1986, p. 392)

Table 2: Overview of individual and situation difference variables, consumer-generated stimuli, and marketer-controlled stimuli

Table 4: Number of times each dimension of impulse purchase definitions was mentioned

Table 5: Means and standard deviations for emotions and cognition

Table 6: Means and standard deviations for impulse buying motives

Table 7: Means and standard deviations for buying determinants

Table 8: Frequency distribution of shopping companions

Table 9: Frequency distribution of store visit frequency

Table 10: Means and standard deviations for emotions and cognition according to gender

Table 11: Example of Mann-Whitney test statistics

Table 12: Means and standard deviations for impulse buying motives according to gender

Table 13: Means and standard deviations for buying determinants according to gender

Table 14: Means and standard deviations for IBT according to gender

Table 15: Frequency distribution of shopping companions according to gender

Table 16: Significance levels and effect sizes for individual and situational difference variables

Table 17: Means and standard deviations for emotions and cognition

Table 18: Means and standard deviations for impulse buying motives

Table 19: Means and standard deviations for buying determinants

Table 20: Means and standard deviations for non-buying reasons

Table 21: Means and standard deviations for emotions and cognition according to gender ... 52 Table 22: Means and standard deviations for impulse buying motives according to gender ..

Table 23: Means and standard deviations for buying determinants according to gender

Table 24: Frequency distribution of shopping companions according to gender

II. Table of Figures

Figure 1: Conceptual framework

Figure 2: Research questions in the style of Punch (2009, p. 58)

Figure 3: Organization of the research

Figure 4: Summary of findings

III. Table of Abbreviations

illustration not visible in this excerpt

1. Introduction

“ If we went into stores only when we needed to buy something, and if once there we bought only what we needed, the economy would collapse ” (Underhill, 2009, p. 24).

While this statement might be mildly exaggerated, impulse buying is a widespread phenomenon that generates over 4$ billion in annual sales volume in the United States alone (Kacen & Lee, 2002, p. 163). The earliest endeavors into impulse buying date back more than

90 years (Barnard, 1920). Yet, even today, it continues to intrigue economists, psychologists, and consumer behavior researchers alike. For a long time, the dominant paradigm used to describe consumer behavior has been the rational choice model. Consumers decide whether to make a purchase, and which purchase to make, by pondering the costs and benefits. Consumers are seen as unemotional information processors, evaluating alternatives in a rational fashion and effortlessly implementing decisions (Hoch & Loewenstein, 1991, p. 492). There is only one major problem with this assumption of human rationality: it is wrong (Lehrer, 2009, p. xv). Whereas a prototypical homo economicus would indeed purchase products on the basis of an evaluation of costs and benefits, most people are much less rational in their purchase behavior. By now, consumer behaviorists have acknowledged that consumers’ purchase behavior rarely follows the principles of economic theory. Rather, consumers’ purchases often seem to be driven, at least to some degree, by desires, moods, or emotions (Verplanken & Herabadi, 2001, p. 71).

Going to the stores is no longer solely motivated by having to fill the fridge or buying a new pair of jeans to replace ones that are worn-out. Shopping has become a major leisure and lifestyle activity. This may help explain the increase in unplanned, non-necessity purchases (Bayley & Nancarrow, 1998, p. 99). Underhill (2009, p. 25), for instance, even talks of an explosion in the levels of impulse buying. According to him, more and more purchasing decisions are being made on the premises of the store itself - in supermarkets and everywhere else too. Shoppers even make important decisions right there on the selling floor. They have money to spend and open minds, and they are yielding to their impulses (Underhill, 2009, p. 64). For many shoppers, their intent to purchase is far from fixed and retailers can alter it right up to the point of purchase (Phillips & Bradshaw, 1993, p. 52).

As industries move from mass marketing into an era of mass customization, marketers and retailers look for characteristics that can be used to differentiate consumers and streamline segmentation strategies. As a result, there is increased interest in several areas of consumer behavior including impulse buying (Coley & Burgess, 2003, p. 282). Are certain shoppers more likely to make impulse purchases? What kinds of customers are most susceptible to impulse purchasing (Kollat & Willett, 1967, p. 21)? Impulse purchasing has long been considered a significant form of consumer buying action. Yet today, 65 years after the first large-scale studies in this area, researchers continue to have difficulties operationalizing the concept and uncovering useful predictors (Cobb & Hoyer, 1986, p. 384). Thus, a profound lack in understanding this purchase pattern still prevails (Baun & Groeppel-Klein, 2003, p. 290).

For many stores, these add-on impulse sales mean the difference between black ink and red (Underhill, 2009, p. 57). Therefore, marketers need to understand impulsive consumer behavior in order to formulate appropriate marketing strategy, allocate marketing budget below-the-line and design effective marketing tactics (Bayley & Nancarrow, 1998, p. 99). If retail managers do not have a solid and shared understanding of what constitutes an impulse purchase, then there is no way for them to boost them.

1.1. Purpose of the thesis

Given the importance of the topic to retailers, the primary purpose of this thesis is to discover how retail managers can increase the level of impulse buying in their stores. How can retailers inspire customers to not only buy what is on their shopping list, but “push them over the edge” and stimulate additional impulse purchases? A second purpose is to unearth suitable segmentation criteria that can be applied to distinguish between impulse and non- impulse shoppers. What shopper characteristics are associated with differential propensities to impulse purchasing? Thirdly, this thesis takes on a customer perspective of impulse buying, offering retailers a glimpse into how shoppers perceive and experience impulse purchases.

1.2. The framework and related research questions

The framework depicted in Figure 1 does not present any cause and effect chains. Rather, it serves to give the reader an overview over the factors that are dealt with in the two empirical studies that were carried out. Individual and situational difference variables are on the left side of the Figure. All of these are presented in detail in the next paragraphs. Gender is shown separately, due to its importance as a segmentation criterion and the special attention it will receive throughout the thesis. Examining these difference variables (specific research questions one and two) is expected to lead to a better understanding of the typical characteristics of the impulse shopper. Before entering the store, and once inside as well, the shopper experiences certain emotions and harbors different impulse buying motives. Additionally, emotions and impulse buying motives interact. Specific research questions three, four, and five incorporate these factors into the process. Once inside the store, the shopper is subject to retailer-related buying determinants, such as the store’s atmosphere or the presentation of the products. Specific research question six examines the influence of these determinants on the shopper.

illustration not visible in this excerpt

Figure 1: Conceptual framework

Following Punch (2009, p. 58) Figure 2 gives on overview of the research area, the research topic, and the general and specific research questions that are examined in this study.

illustration not visible in this excerpt

Figure 2: Research questions in the style of Punch (2009, p. 58)

Fifteen individual and situational difference variables have been identified, serving as segmentation criteria; and data on these were collected in both studies. The results concerning these variables serve to answer the first specific research question, namely, to identify characteristics that distinguish buyers from non-buyers. In the following, I present the rationale for including each of these fifteen variables.

Rook (1987, p. 196) suggests to examine situational factors that influence impulse buying. For example, how does the social environment (group vs. solo buying) affect impulse buying decisions? Bayley and Nancarrow (1998, p. 102) state that those accompanying the shopper, or those who are not present but important in the shopper’s life, impede or encourage impulse buying behavior. Luo (2005, pp. 288-291) puts forward that the presence of peers increases the urge to purchase whereas the presence of family members decreases it. Luo’s central assumption is that peers consider it socially desirable to make impulsive purchases, whereas family members do not. Therefore, it is hypothesized that:

H1: Customers shopping in the presence of others will be more prone to making impulse purchases than customers shopping alone.

If there is support for this hypothesis, retailers should encourage people to shop in company, for instance, by offering “partner specials” or “family specials” when more than two people from a family are shopping together.

A curvilinear relationship (inverted U-shape) concerning age and the propensity to buy on impulse has been found in various studies (e.g., Bellenger et al., 1978; Cobb & Hoyer, 1986). Therefore, it is hypothesized that:

H2: There is a curvilinear relationship between age and impulse purchases. Young (<25 years) as well as older (>60 years) customers will be more prone to buy products on impulse than middle-aged customers.

Whereas Wood (1998, p. 306) found that low educational levels appear to be a primary cause of buying on impulse, Kollat and Willett (1967, p. 24) did not find any significant correlation between education and the level of impulse purchasing. The author hypothesizes that:

H3: There is a negative relationship between education and being an impulse buyer.

Verplanken and Herabadi (2001) developed and applied an impulse buying tendency (IBT) scale which correlated significantly with reported purchase frequencies of typical impulse products and the number of recent impulse purchases. It is therefore hypothesized that:

H4: There is a positive relationship between impulse buying tendency and being an impulse buyer.

Various studies (Abratt & Goodey, 1990; Kollat & Willett, 1967; Prasad, 1975; Rook & Hoch, 1985) have examined the effects of shopping lists on levels of impulse buying. Whereas Prasad (1975, p. 7) did not find levels of impulse buying to vary with the existence of a shopping list, Abratt and Goodey (1990, p. 117) found that customers who do not write shopping lists are more prone to making impulsive purchases. Therefore, the hypothesis is put forward that:

H5: Customers who go shopping without a list are more prone to making impulse purchases.

According to Iyer (1989, p. 45), the quantity of impulse purchases is a function of time pressure - the lower the time pressure, the higher the number of impulse purchases and vice versa. If there is a correlation between perceived time pressure and impulsive purchases, retailers should make every endeavor to make sure that customers do not feel pressed for time when shopping at their stores. It is therefore hypothesized that:

H6: The less perceived time pressure customers feel, the more prone will they be to make an impulse purchase.

The internet teems with tips helping consumers curb their impulse buying¹. One frequently presented advice is for consumers to set a specific budget before heading to the shops and stick to it. It is therefore assumed that:

H7: Customers who have not set a specific budget for their shopping trip will be more likely to make impulse purchases.

D’Antoni and Shenson (1973, p. 75) stated that risk-reducing measures such as money back guarantees, generous return policies, and free samples can be an effective way to induce more impulse purchases in customers. It is therefore hypothesized that:

H8: Impulse buyers will generally rate lower on perceived risk than non-buyers.

Underhill (2009, p. 54) established a direct link between the percentage of shoppers using a cart or basket and the size of the average transaction. Shoppers might be more willing to grab a product on impulse if they are equipped with a basket or cart. Shoppers already juggling several items in their hands will think twice about burdening themselves with yet another one. It is thus hypothesized that:

H9: Customers who use a shopping help are more prone to make an impulsive purchase.

Maybe some impulsive purchases are triggered because customers know that they need to reach a certain threshold (e.g., 30€) in order to get more bonus points as part of a loyalty program. If this was the case, retailers should further intensify their efforts of persuading customers to join their loyalty programs. Hence, it is hypothesized that:

H10: Customers who are part of a loyalty program are more prone to making impulse purchases.

Jones et al. (2003, p. 508) found that higher levels of product involvement are associated with higher levels of product-specific IBT. Individuals who have a high level of involvement were found to be more likely to produce the emotions required for an impulsive purchase. Furthermore, shoppers who are involved with a specific product category are more likely to browse stores stocking such products, making an impulsive purchase more likely given the intensified frequency of browsing or shopping (Jones et al., 2003, p. 508). It is therefore hypothesized that:

H11: Impulse buyers will show higher levels of involvement than non-buyers.

The majority of truly impulsive purchases should be accompanied by a short inspection time or “decision time lapse” as D’Antoni and Shenson (1973, p. 68) dubbed it. That is, the time span between the customer catching sight of the product and putting it into his shopping cart (or walking away in case of a non-buyer) should be a few seconds at the maximum for a buyer. It is assumed that the longer a customer looks at the stand or product and the more he starts to reason (i.e., the longer his decision time lapse becomes), the more likely it will be that he does not end up buying. It is therefore hypothesized that:

H12: The decision time lapse will be shorter for impulse buyers than non-buyers.

Most extant research found a positive relationship between disposable money and impulse purchases. This finding is rather intuitive, because all other things being equal, people with less disposable money have a smaller array of products that they are able to purchase impulsively (D’Antoni & Shenson, 1973, p. 69) and cannot act on as many impulses as people with more money on hand. Additionally, the money available at a specific time tends to affect people’s moods. If customers sense that they have a fair amount of money available they will tend to be in a more positive mood. In turn, being in a positive mood has been shown to positively affect impulse purchases (Beatty & Ferrell, 1998, pp. 173-185). It is therefore hypothesized that:

H13: On average, buyers will have more money disposable than non-buyers.

Iyer (1989, p. 50) found that people who bought something on impulse came to the store less often than people who did not buy something on impulse. Consequently, it is hypothesized that:

H14: The lower the knowledge of the task environment, measured via frequency of visits to this particular store, the higher the probability of being in the buyer group.

The human brain tries to minimize its effort when taking in visual information. Thus, the presentation of products on a shelf can either be arranged “brain-friendly” (which customers prefer) or it can be arranged “brain-unfriendly”. Since clearly arranged product presentations are preferred to confusing ones which overload the brain (Häusel, 2008, p. 217), it is hypothesized that:

H15: Customers who evaluate the presentation of the products on the inspiration island more positively are more prone to make impulse purchases than customers who evaluate it more negatively.

Both studies contain a results and discussion section concerning notable gender differences. After a phase of egalitarianism concerning the sexes, the differences between men and women seem to attract growing attention again, especially in the shopping context (e.g., Brennan, 2011; Kreienkamp, 2007; Underhill, 2011). Changing social roles and growing purchasing power of women have reinforced the focus on gender as a segmentation variable. It is no longer enough to know why shoppers buy; it is essential to know why men and women buy. Especially in a drugstore environment, where many products are either squarely aimed at women or men, it is vital to explore the gender differences in (impulse) shopping behavior, and thereby answer the second specific research question.

One of the objectives of this thesis is to more fully model central precursors of impulse buying. Thus, emotions, impulse buying motives, retailer-related factors (buying determinants), and individual and situational difference variables are included in the examination.

Emotions are a crucial part of our decision-making process. When we are detached from our feelings, the most trivial decisions become impossible. A brain that cannot feel cannot make up its mind. Higher levels of emotion mean more impulsivity. Whichever emotion you feel most strongly tends to determine your shopping decisions (Lehrer, 2009). The results concerning this aspect serve to answer the third specific research question.

There are various motives that shoppers can have for buying on impulse. Five of these motives are examined in both studies. The results concerning predominating impulse buying motivates contribute to answering the fourth specific research question.

Given the relevance of interactions in explaining behaviors (Punj & Stewart, 1983), relations within emotions, relations within impulse buying motives, and relations between emotions and impulse buying motives seem worth to uncover. The results of these correlation analyses shed light on the fifth specific research question.

Once inside the store, the shopper is influenced by myriad factors, ranging from a store’s atmosphere to the presentation of the products. Both studies seek to determine which of these factors are most crucial in shoppers’ impulsive buying decisions and thereby help to answer the sixth specific research question.

1.3. Organization of the research

illustration not visible in this excerpt

Figure 3: Organization of the research

2. Literature Review

2.1. Introduction

In the course of compiling the literature review, thirteen areas of interest emerged. Each of the articles that were reviewed was scanned on whether it contained substantial and enlightening information on any of these thirteen areas. Every impulse purchase is made by a person. The first section sheds light on these persons’ characteristics. Most purchases (except for services) involve a physical product. Section two clears up the question of whether “impulse” items exist. This thesis focuses on physical retail settings. Section three lists the types of stores where impulse purchases are conducted. What percentage of all purchases do impulse purchases make up? Section four summarizes the findings concerning this question. Are impulse purchases mostly made at the beginning of a shopping trip? Section five addresses these, and related questions. One of researchers’ favorite pastimes is to come up with elegant definitions for complex concepts. Section six presents some of the most influential concerning impulse buying. In order for science to advance, concepts and behaviors have to be measured. Section seven brings the most common methods to measure impulse buying to the fore. One impulse purchase is not like another. Section eight distinguishes different types of impulse purchases. Nothing happens without a reason. The reasons for why impulse purchases happen are examined in section nine. Not every impulse to purchase is welcomed by the shopper who experiences them. Section ten explores the strategies that shoppers adopt to resist their impulsive urges. A plethora of factors decides on whether an impulse purchase is made or not. Section eleven sheds light on the factors that influence shoppers’ impulse buying behavior. What practical implications and guidance for retailers have emerged so far concerning the stimulation of impulse purchases? Section twelve summarizes the actions retailers can take to increase the level of impulse purchasing. Finally, what promising research opportunities have been identified in the realm of impulse buying? Section thirteen presents the most promising ones.

2.2. Who makes impulse purchases?

Who is the impulsive shopper? The answer to this question is essential to retailers if they plan on targeting this elusive type of person. Almost 90% of people purchase goods on impulse occasionally (Welles, 1986). In essence, this is basically everyone! Out of this mass of people who shop on impulse, what are the characteristics that can at least give retailers a clue as to which group of people is most prone to buy on impulse? This first section gives an overview of the most commonly encountered characteristics derived from research on impulse purchasing.

The two most researched characteristics of impulse shoppers are their age and sex. Surely, one reason for this is the fact that these two characteristics can be observed rather easily and without complication, whereas, for instance, a consumer’s attitude is much harder to determine. Regarding age, two findings deserve to be pointed out: First, there is extensive evidence for a negative relationship between age and impulse buying (Rook & Hoch, 1985, p. 25; McGoldrick, 1982, p. 30; Verplanken & Herabadi, 2001, p. 79; Tuyet Mai et al., 2003, p. 18). Second, the relationship between age and impulse buying seems to be curvilinear (inverted U-shape), i.e., people in the middle-aged group are the ones most prone to buy on impulse, whereas younger and older people tend to buy less on impulse (Cobb & Hoyer, 1986, p. 394; Bellenger et al., 1978, p. 17; Wood, 1998, p. 312). These findings contradict each other to some degree and need further explication. An explanation for the curvilinear relationship may lie in the fact that people tend to “underconsume” during the periods before and after they are employed, i.e., roughly until they reach 25 and after they turn 65. In turn, people tend to “overconsume” during their working years, i.e. between 25 and 65 (Courant et al., 1984). This period of overconsumption will naturally contain a certain number of impulse purchases which could result in the curvilinear relationship that has been observed by the aforementioned studies.

Intuitively, one might be tempted to assume that women buy more on impulse than men. For instance, Coley and Burgess (2003, p. 285-286) state that women’s behavior is commonly thought of as being more emotionally and psychologically ingrained than the behavior of men. Such perceptions give rise to the frequently held assumption that women are more inclined to impulse buying. However, findings are relatively inconclusive in this respect. While a number of studies (Verplanken & Herabadi, 2001, p. 81; Coley & Burgess, 2003, p. 291; Dittmar et al., 1996, p. 204) finds that women tend to buy more on impulse than men, a similar amount of studies (Bellenger et al., 1978, p. 17; Kollat & Willett, 1967, p. 24;

Herabadi et al., 2009, p. 23), finds no significant relationship between gender and impulse buying. It is noteworthy though that only two of the roughly 40 articles reviewed find that men buy more on impulse than women (Tuyet Mai et al., 2003, p. 27; Cobb & Hoyer, 1986, p. 406). Coley and Burgess (2003, p. 292) take a category approach towards impulse buying and find that men are more likely to impulsively purchase products from categories related to technology and entertainment, whereas women impulsively purchase more health and beauty items. They explain this fact by stating that women are more emotional and image-guided and purchase goods that project emotional and appearance aspects of self. In a similar fashion, Dittmar et al. (1995, p. 503) conclude that men seem to have a more functional and leisure- oriented perspective on impulse buying, whereas women pay more attention to appearance and body-related matters in their choices. The importance of taking a category approach is a recurring theme in studies on impulse buying. Furthermore, the motives behind impulse purchases have been found to differ between men and women. Women are more likely to use impulse purchases as a method to manage their mood and reduce stress than men. They were also found to be more likely than men to show feelings of regret or mixed feelings of pleasure and guilt over making an impulse buy (Coley & Burgess, 2003, p. 290-291).

Apart from age and gender, impulse buyers’ income is a characteristic which merits closer inspection. Whereas Kollat and Willett (1967, p. 24) find no significant effect of income on the rate of unplanned purchasing, Tuyet Mai et al. (2003, p. 28) indicate the possibility of a curvilinear relationship (inverted U-shape) with the middle income group buying most on impulse. Hence, whereas a low income might deter many people from acting on their impulses, a certain level of income is conducive to impulse buying.

Another feature of customers which could be related to impulse buying is their intelligence, and by this token, their education. However, evidence on this characteristic is rather thin and inconclusive. Intelligence has been found to correlate positively with the capacity to delay gratification (Rook, 1987, p. 190). This is an important finding, because impulse purchases often act as immediate gratification, thus, customers who have a higher capacity to delay gratification should conduct fewer impulse purchases all other things being equal. However, Rook (1987) does not follow up with any empirical results that could lend support to his hypothesis. Whereas Kollat and Willett (1967, p. 24) find no significant effect of education on the rate of unplanned buying, Wood (1998, p. 315) reports that the highest level of impulse buying was among respondents who had some college experience, but lacked a college degree, suggesting a curvilinear (inverted U-shape) relationship between education and impulse buying where those with a medium level of education are most prone to engage in impulse buying.

To conclude this first approximation of what characteristics might be typical for an impulse buyer, some characterizations have been drawn from the papers reviewed. Kacen and Lee (2002, p. 164) describe impulsive buyers as likely to be unreflective in their thinking, to be emotionally drawn to the object, and to covet instantaneous gratification. In addition, they hardly pay attention to potential negative consequences that may be brought about by their actions. Impulse buyers may also be hungry for arousal in the sense that they enjoy upholding strong and stable emotional arousal levels by pursuing emotionally stimulating daily activities, and choose shopping to satisfy this need (Herabadi et al., 2009, p. 29). Impulsive shoppers engage in very little in-store information processing, but cherish quality almost as much as do planners. They lean towards enjoying shopping more, are less likely to arrange shopping on particular dates and are less likely to compose shopping lists (Rook & Hoch, 1985, p. 407; Cobb & Hoyer, 1986, p. 25). These results portray the impulsive customer as a recreational shopper who shops when the mood strikes, finds satisfaction in shopping activities, and often ends up buying more than planned (Cobb & Hoyer, 1986, p. 25).

2.3. What is bought on impulse?

Although they are often mentioned in popular science books (e.g., Underhill, 2009, p. 87) as well as in scientific articles (e.g., Rook & Hoch, 1985, p. 23), there really is no generally accepted classification of impulse items as opposed to non-impulse items. However, there are certain characteristics that may cause a product to be bought more on impulse than others. Impulse products are regularly characterized as low-cost, repeatedly purchased goods that require little cognitive effort from the customer. These products produce immediate and “mindless” reactive behavior (Rook & Hoch, 1985, p. 23). This is not to suggest that any one product is inevitably bought more on impulse than other products; rather for the population taken together particular goods (e.g., magazines) tend to be bought more on impulse than other goods (e.g., sailing boats) (D’Antoni & Shenson, 1973, p. 73).

Nonetheless, I tend to agree with Bellenger et al’s (1978, p. 15) view that “it is virtually impossible to isolate and label certain products as impulse products since individual purchasers’ motivations, product information levels, environmental variables, monetary and time constraints combine to produce an impenetrable combination of variables that vary from one individual to another”. Therefore, both researchers and retailers should shift their focus from the scarcely manipulable product to the highly influenceable customer who wanders the stores and ultimately buys these products.

2.4. Where are impulse purchases conducted?

Impulse purchasing is neither restricted to one single continent nor to any particular retail channel. Studies on impulse buying have been conducted in the United States, Europe, Africa and Asia. They have been done in supermarkets, department stores, drugstores, pharmacies, and a host of other physical places as well as in direct buying contexts like online-shops, television shopping channels, telemarketing, and catalogues (Bayley & Nancarrow, 1998, p. 99). Although most studies have focused on a stationary retail context, studies into direct buying contexts might yield interesting results, because it is assumed that less public shopping and the associated anonymity may incite more impulse purchasing (Bayley & Nancarrow, 1998, p. 102).

2.5. How much is bought on impulse?

The reported percentages of how much is bought on impulse vary greatly. This is mainly due to a plethora of study designs, large discrepancies in definitions of what constitutes an impulse purchase, and the variety of product categories observed. For a table comprising 42 studies from 1950 to 1978 and their reported rates of unplanned purchasing, see Phillips and Bradshaw (1993, p. 54). To give the reader an idea of how vastly the rates of unplanned purchasing vary between these studies, the lowest and the highest rate reported in this table are zero and 70.1% respectively.

Taking a historical perspective, impulse purchasing seems to have gained in significance during the last century. A study examining supermarket behavior from 1945 to 1965 found that the proportion of impulse purchases rose from 38% to 50% of total consumer purchases over this twenty-year period (Cobb & Hoyer, 1986, p. 385). There is some evidence (Guirl & Stiansen, 1985) that the rate of unplanned purchasing may have waned to some degree as a result of recessionary conditions in the early 1980s. Nonetheless, few would quarrel over the long-term upward trend of consumers making decisions inside stores and on an impulsive basis. If one takes the average of impulse buying rates across different studies, sectors, and product categories, one can conclude that the impulse buying rate hovers at around one third of total purchases (Häusel, 2008, p. 207). It seems that impulse buying rates have stabilized somewhat at this level. Therefore, this proportion of one third of all purchases being impulsive ones, can serve as a first orientation to retail managers when taking on the subject of tracking and/or boosting impulse purchasing.

2.6. When are impulse purchases made?

Kollat and Willett (1967, p. 26) suggest that impulse purchasing is more likely to happen on a major grocery trip and in a large store rather than on a fill-in trip. The exposure hypothesis² justifies this finding by claiming that during top-up trips, the customer’s needs are more clearly identified making him less receptive to in-store stimuli. During major trips, however, as the customer’s needs are less well defined, the customer is more open to in-store suggestion. Cobb and Hoyer (1986, pp. 397-404) find that the typical impulse purchaser that they identified in their study shops earlier in the week than other shoppers, thereby increasing the number of impulse purchases that are made during the first days of the week. Likewise, their group of impulsive shoppers was most likely to shop late in the day, thereby increasing the number of impulse purchases that are made in the evening or just before the store closes. An additional observation that the authors make is that impulsivity was more apparent toward the latter half of the shopping trip. This is surprising insofar as supermarket shoppers tend to “speed up” with the end of their shopping trip nearing completion, giving them less time and attention to make impulse purchases. In fact, this phenomenon is so pronounced and regular that it is referred to as the “checkout magnet” (Sorensen, 2011, p. 60).

In contrary to the exposure to in-store stimuli hypothesis mentioned earlier, the buying impulse does not always depend on direct visual stimulation. Some respondents in Rook’s (1987, p. 193) study described sitting at home and experiencing the urge to go out and buy something out of the blue, with no perceptible visual encouragement. Apparently, the buying impulse and the actual purchase which follows this impulse do not have to happen in the same location. Whether these types of purchases constitute “true” impulse purchases is another matter of discussion.

2.7. How is impulse purchasing defined?

One would assume that a subject such as impulse buying which has drawn quite considerable attention from researchers is clearly defined. Unfortunately, this is not the case. It is not exaggerated to say that almost every researcher who conducts research in this area seems to come up with a new definition of impulse buying. This definitional hodgepodge is one of the main reasons why studies dealing with impulse buying are hard to compare. Moreover, the adopted definition is often at odds with the operationalization even in the very same study. Thus, cautiousness is required whenever comparing studies or generalizing from their results.

A great deal of earlier definitions suffered from “definitional myopia” and simply equated impulse to unplanned purchasing without further ado (Piron, 1993, p. 341). When applying this definition, any purchase not written on the shopper’s shopping list that he ended up buying would qualify as an unplanned and therefore an impulsive purchase. One can easily figure that this definition grossly overestimates the number of real impulse purchases. Obviously, customers use store layout as an external memory aid. When a shopper strolls past the bread shelf and is reminded that he needs to buy bread, but forgot to put it on his shopping list, he will put it into his basket. According to the definition above, this constitutes an unplanned and therefore impulsive purchase. However, closer scrutiny of this shopping act would reveal that there was nothing impulsive whatsoever about it. Thus, the fact that a purchase is unplanned cannot be a sufficient condition for construal as an impulse purchase.

In the following, some worthwhile approaches of categorizing impulse purchases are presented. Cobb and Hoyer (1986, p. 392) came up with a 2x2 matrix which is depicted in Table 1.

illustration not visible in this excerpt

Table 1: Classification scheme of buyer types (Cobb & Hoyer, 1986, p. 392)

By asking customers whether they had intended to purchase the product category and whether or not they had intended to purchase the specific brand prior to entering the store, the authors could identify three groups of customers. Those who intended to purchase both the category and the brand (planners), those who intended to purchase the category, but not the brand (partial planners), and those who intended to purchase neither the category nor the brand (impulse purchasers).

Stern (1962) already takes the line that equating impulsive with unplanned purchases does not make a lot of sense. Rather, he delimits four types of impulse buying. He defines pure impulse 13

buying as a “novelty or escape purchase which breaks the normal buying pattern” (Stern, 1962, p. 59). In addition, he distinguishes reminder impulse buying, which occurs when one sees a product in a store and is reminded that one wanted to buy it anyhow. Suggestion impulse buying describes a purchase where the customer sees the product for the first time and feels a need for it. In contrast to pure impulse purchases, suggestion impulse buying can also be a purely rational act according to Stern. Lastly, Stern mentions planned impulse buying which may seem jarring at first, but according to the author happens when a customer goes into a store with some particular purchases in mind, but with the expectation and objective to make other purchases that depend on price specials and bargains.

When making a pure impulse purchase, the customer breaks with the principle of rational buying and undertakes an emotional act of buying. This makes Stern (1962) one of the first to explicitly demarcate the heavily emotionally influenced act of unplanned purchasing from unplanned purchasing happening for other reasons. By defining the pure impulse purchase as a “novelty or escape purchase which breaks a normal buying pattern” (Stern, 1962, p. 59), he points out what all later research into impulse buying is about: The search for an explanation for a buying pattern that does not fit into any of the traditional buying behavior patterns respectively cannot be explained in a conventional manner (Baun, 2002, p. 14-15).

Weinberg and Gottwald (1982), too, became convinced that measuring impulse buying as unplanned purchases is no longer adequate. They define impulsive buying decisions on the basis of affective, cognitive, and reactive determinants of consumer behavior. Impulsive buying decisions are defined as being accompanied by strong emotional activation, little cognitive control, and highly reactive, mostly automatic, behavior, with a special stimulus situation acting as the trigger (Weinberg & Gottwald, 1982, p. 44).

Piron (1991) puts forward a definition that contains four elements. An impulsive purchase according to his definition is unplanned, the result of an exposure to a stimulus, decided on- the-spot, and can be classified into experiential and non-experiential³ impulse purchases. According to Piron, this definition is both characterized by flexibility insofar as it is possible to categorize impulsive purchases according to the existence of accompanying emotional and cognitive processes, as well as discernment because impulsive purchases which have not been decided upon first contact with the stimulus at the Point of Sale (POS) cannot qualify as impulse purchases (on-the-spot feature).

2.8. How are impulse purchases measured?

The methods used in a certain study depend in part on the definition of impulse purchases that this study adopted. As has been shown in the previous chapter, a great number of studies, especially in the 1950s, 60s, and 70s equated unplanned with impulse purchases. I have already shown that this definition is seriously flawed; nonetheless, it is quite easy to operationalize by recording the items that are on a shopper’s list and comparing them with the ones that appear on his receipt. Any item that is on the receipt but not on the shopping list is recorded as an impulse purchase, because according to this rationale, the customer had obviously not planned to buy it prior to entering the store.

Apart from the issue that not every unplanned purchase also qualifies as an impulsive one, there are at least two more grave concerns regarding this method. First, the customer may be unwilling or unable to fully articulate his pre-purchase intentions resulting in an overstatement of impulse purchases (Kollat & Willett, 1969, p. 82). Second, both pretest questioning and recitation of shopping plans as well as the awareness of later inventory tallying at the check-out counter may bias responses. This guinea pig effect may be especially pronounced for research on such delicate subjects as impulsive purchasing because impulsive behavior can be influenced by everything that is going on at a given time (Pollay, 1968, p. 323). Moreover, socially desirable responding seems to be a recurrent problem in impulse buying research. It is not uncommon to encounter some slight reluctance on the part of the respondents to admit to lower levels of planning. Impulsive behavior is often seen to be negative and can therefore be classified as a “dark side variable” (McGoldrick, 1982, p. 28; Bayley & Nancarrow, 1998, p. 101; Cobb & Hoyer, 1986, p. 408).

In the meantime, more diverse research methods have been applied to the subject of impulsive buying and more often than not, a combination of different methods is used in any given study. For example, Cobb and Hoyer (1986, p. 391) make use of direct observation of shoppers, a personal interview directly following the brand selection, and a self-administered mail questionnaire for more detailed inquiries. While many studies have been positivist or quantitative in format, atomistic, and have disregarded the problem of socially desirable responding (SDR), Bayley and Nancarrow (1998, p. 104) present a qualitative approach with a distinctive combination of features. The authors leave the definition of impulse purchasing to the interviewee, include direct marketing, and focus on benefits, costs, motivations and emotions associated with impulse buying behavior. By using friendship pairs who interview each other and self-scripts which minimize self-censorship, they strive to neutralize SDR and foster full disclosure.

Nevertheless, impulse buying research still suffers from the lack of a suitable theoretical framework to guide empirical work and extant definitions are unsuccessful at adequately capturing impulse buying by centering on one element of the phenomenon (i.e., the purchase itself) to the expense of another (i.e., the customer).

2.9. Types of impulse purchases

Stern’s (1962) typology of impulse purchases has already been presented in the chapter on definitions which are in use for impulse purchases. One type of impulse purchasing which he mentions - reminder impulse purchasing - is frequently picked up by other researchers. Rook and Hoch (1985, p. 23) state that customers use store layout as an external memory aid, for instance, seeing the milk provides a handy cognitive reminder and culminates in a reminder impulse purchase. In the same notion, Stern (1962, p. 60) remarked that customers increasingly use the store as a substitute for a shopping list and treat it as if it were a gigantic catalog from which to make the selections. Like Stern (1962), McGoldrick (1982, pp. 27-28) also defines four forms of unplanned buying, one of which is labeled reminder purchase. The three other categories from McGoldrick’s (1982) study which are “specifically planned purchase”, “generally planned purchase”, and “entirely unplanned purchase” seem to have been lifted from the Popai/Du Pont Consumer Buying Habits Study (1977). In contrast to a reminder purchase, this study introduced the category of a substitute purchase where the customer changes from a specifically planned item to another.

Bayley and Nancarrow (1998, pp. 110-111) differentiate two types and four styles of impulse shopping. These are all original to their article and quite original. The two types of impulses are “self-willed impulse” and “captivated impulse”. While in the former one, the customer semi-consciously directs the impulse towards the purchase, in the latter one the customer succumbs to the passion of the impulse and relishes in a feeling of being totally out of control. The four styles of impulse purchases differ in terms of the purchase experience and in terms of benefits and rewards desired. These are 1) accelerator impulses, prompted by a sudden desire to stock-up for a future need, 2) compensatory impulses, serving as a reward or an emotional lift, 3) breakthrough impulses, which unfold in an unforeseen and unrepeatable way, and 4) blind impulses, often dysfunctional, where the customer has a sensation of being overwhelmed by the product.

A rather peculiar observation is that some people seem to plan on being impulsive. Take someone for instance who withdraws 200 dollars from a cash dispenser and goes to the mall. He does not know what to spend it on, he only knows that he wants to spend the money on something. By indulging in this behavior, people can enjoy the feeling of being overpowered by spontaneous desires and at the same time keep up some form of impulse control imposed by a budget (Rook & Hoch, 1985, p. 25).

Wood (1998, p. 302) differentiates impulse buys into akratic and compulsive purchases. While the latter ones are compelled actions and motivated by an addiction to buy, the former ones are free and intentional actions contrary to the buyer’s better judgment. According to Wood, there are no sensible impulse purchases because they all go against the buyer’s better judgment.

2.10. Why are products bought on impulse?

There is general consensus that impulse purchasing is the result of a combination of situational variables and individual shopper characteristics. It is the interaction of a person with its environment that leads people to impulsively purchase or not (Herabadi et al., 2009, pp. 21 & 30). The standard economic explanation of impulse buying falls short of giving a plausible explanation for why it occurs. The discounting model assumes that people who buy on impulse discount the future at too rapid a rate. Thus, the benefits of the coveted object at the point of purchase overshadow the future problem of paying the bill. Yet, these preferences switch later, when the buyer comes to pay the bill and regrets the acquisition (Dittmar et al., 1996, p. 189). The fact alone that many people who impulsively purchase an object do not regret this purchase later discredits the standard economic explanation for why impulse purchases occur.

Although it may seem mundane, price is probably one of the most important factors for why impulse purchases occur (Stern, 1962, p. 61). A reduced price or special offer is a clear indication to customers that they might be in for a bargain. A red price tag, explicitly stating the rebate, can exert a very direct effect on customers and convince them to buy the product within a split second.

Many laypeople share the misconception that most impulse purchases are irrational and go against the usual practice of evaluating alternatives and making an informed choice when buying products. Whereas some impulse purchases surely can be considered irrational acts, most of them are not. In fact, buying something on impulse can be more rational than going the conventional way. In a world of endless choice and alternatives, impulse buying can help reduce cognitive effort, help prevent information processing overload and be a rational alternative to more time-consuming search behaviors (Hausman, 2000, p. 410). Stern (1962, p. 62) denotes impulse buying “an efficient and sensible way to buy goods”. Hence, impulse buying behavior can be seen as a rational answer to complexities in a shopping environment that gets ever more abundant and confusing.

One of the major reasons for why impulsive purchasing occurs is its ability to manage and alter the mood state of the person who takes part in it. In a study conducted by Gardner and Rook (1988, p. 128), 75% of the respondents reported they felt better after making an impulse purchase. Impulse buying seems to be an effective technique for breaking out of unwelcome mood states. Especially when feeling depressed, bored, or frustrated, impulsive buying seems to be the method of choice for leaving undesirable mood states behind (Gardner & Rook, 1988, p. 129). Dittmar et al. (1996, p. 203) even go so far as to say that for some shoppers, mood may be the most important psychological consideration leading to impulse buying. Whereas Herabadi et al. (2009, p. 29) state that the result of an impulse purchase is a pleasure experience, at least at the time of making the purchase, impulse purchases can also emotionally backfire. While over 90% of the sample in Gardner and Rook (1988, p. 128) fell into the somewhat to extremely happy range after making an impulse purchase, 38% fell into the somewhat to extremely guilty range, showing the double-edged nature of impulse buying and its consequences for one’s emotional state.

Hedonic motives can play a weighty part in shoppers’ decisions to buy impulsively. Needs for novelty, surprise, fun, and social interaction can all contribute to impulsive buying behavior (Hausman, 2000, p. 404). For some shoppers, the act of shopping itself seems to satisfy certain needs (Hausman, 2000, p. 405). Not every impulse purchase has to serve selfish needs though. Rook and Fisher (1995, p. 305) draw attention to the fact that some impulse purchases may be motivated by generosity, e.g., when somebody decides to buy a gift for an ill friend. However, customers do instrumentalize impulse purchasing as a means of identity construction and to satisfy their esteem. The socio-psychological value can become a leading driver for an impulse purchase in the shopping arena (Bayley & Nancarrow, 1998, pp. 102 &109). Dittmar et al. (1996, pp. 188 & 204) put forward that people buy impulsively to acquire material symbols of personal and social identity. Their social constructionist approach assumes that people are especially likely to buy on impulse if they sense aspects of their self to be discrepant from their ideal.

2.11. How do people act upon impulses? Resistance strategies

People develop myriad strategies to avoid purchasing on impulse. Following Hoch and Loewenstein’s (1991) categorization, these strategies can be classified into desire reduction and willpower strategies. Desire reduction strategies often include the manipulation of a person’s reference points and will be presented in the following.

Avoidance strategies can be an effective means to prevent impulse purchases. People may choose not to go near stores that have a history of provoking impulsive purchases. By applying this strategy, people impede possible impulse purchases from the outset. Once inside the store, people can still use avoidance strategies even after having spotted an item that they would like to buy on the spot. Customers can choose to literally walk away from products in order to bring physical distance between them, and to wait for the impulse to subside.

Postponement strategies instruct the customer to put off a choice until some point in the future which can be an effective strategy for controlling transient desires (Hoch & Loewenstein, 1991, p. 500). These strategies often take the form of simple rules like “always consult your spouse before making a major purchase”. Distraction strategies represent one of the most fruitful means of postponing. People might start to focus on a different task in order to get their mind to stray off the would-be impulse purchase.

Substitution strategies often offer a rather cheap alternative to the actual object of desire. Imagine a woman who strolls past a store window and sees a stunning piece of jewelry. However, instead of buying the expensive piece she goes for a much more affordable one. Thus, the rationale of this tactic is to provide adequate instantaneous satisfaction to be able to put up with the deprivation associated with waiting, maybe at the same time diverting oneself from the original desired object (Hoch & Loewenstein, 1991, p. 500). Willpower refers to the various ploys that people use to surmount, rather than to reduce, their own impatience.

Precommitment includes any method through which customers levy constraints on, or modify incentives for, future behavior (Hoch & Loewenstein, 1991, p. 501). For someone who looks to curb his impulsive shopping habit this could mean leaving his credit cards and checkbook at home and only taking as much cash as he knows to need for planned purchases. However, precommitment strategies are often costly or problematic to introduce and are of restricted effectiveness (Hoch & Loewenstein, 1991, p. 501).

Cost-benefit analyses seem to be among the most widely used self-control devices according to consumers themselves (Rook & Hoch, 1985, p. 26). Yet, self-control based on economic cost analyses is not exactly easy. Whereas desire is near and vivid, economic consequences are usually far away and defy definition. In fact, people may make plentiful rationalizations that corrupt their attempts at rational cost assessment (Hoch & Loewenstein, 1991, p. 502).

When using a tactic known as “time binding” people focus on the positive benefits of the delay in addition to thinking about the negative consequences of immediate consumption (Hoch & Loewenstein, 1991, p. 502). By refusing instantaneous gratification and thinking about better ways to spend the money, people overcome the impulse to buy.

By bundling costs, consumers increase the salience of purchase costs. A nice cup of coffee bought at Starbuck’s does not seem to break the budget, but calculating what it costs to drink one cup every day of the year might make some people rethink their willingness to give in to the impulse. Marketers have long used this strategy in the opposite direction, often referred to as “pennies a day”. It is not surprising to see consumers using the same strategy in an attempt to curb their impulsive spending habits.

As seen, customers do have a wide repertoire of resisting impending impulse purchases. However, in these conflicts of pleasure and anxiousness, excitement and guilt, the “heart” often wins the conflict against the “mind”.

2.12. Factors that influence consumers’ impulse buying behavior

There are two broad classification schemes for grouping the factors which can influence consumers’ impulse buying: The first one contrasts individual difference variables and situational variables (Beatty & Ferrell, 1998) and the second one distinguishes between selfgenerated stimuli and marketer-controlled stimuli (Piron, 1991, p. 509). Figure X provides an illustration of the different groups.

illustration not visible in this excerpt

Table 2: Overview of individual and situation difference variables, consumer-generated stimuli, and marketercontrolled stimuli

Individual difference variables, as the name suggests, differ from person to person. Trait buying impulsiveness, or IBT, can be defined as “both the tendencies to experience spontaneous and sudden urges to make on-the-spot purchases and to act on these felt urges with little deliberation or evaluation of consequences” (Beatty & Ferrell, 1998, p. 174). IBT was gathered of respondents in both Study 1 and Study 2 of this thesis using the IBT-scale (Rook & Fisher, 1995). It is assumed that the higher a person rates on the IBT-scale, the more prone he will be to have impulses to purchase and to act on them.

Dittmar et al. (1995) emphasize the role of perceived social image and the manifestation of self-identity in the purchase decision. They theorize that impulse purchases were more likely to be items that represent the preferred or ideal self and as such should be influenced by social categories such as gender (Kacen & Lee, 2002, p. 164). Kacen and Lee (2002) examine the role that culture plays in moderating impulsive buying behavior. They find that collectivist consumers engage in less impulsive buying than individualist consumers. Additionally, Beatty and Ferrell (1998, p. 182) find a positive relationship of in-store browsing on positive affect and felt urges to buy impulsively. Hence, customers who like to browse will be more prone to buy impulsively.

The existence of a shopping list counts as a situational variable. Abratt and Goodey (1990, p. 117) find that customers without a shopping list show higher rates of impulsive purchasing than customers who go shopping with a shopping list. Kollat and Willett (1967, p. 27) found the same effect in their study, but state that a shopping list influenced unplanned purchasing only when more than fifteen articles were purchased.

The effects of shopping companions on the propensity to buy impulsively are uncertain. While Rook and Fisher (1995, p. 312) theorize that when customers shop alone, they may feel less socially visible, lowering their inhibitions about acting on impulse, they concede that the effects of shopping with others are probably more diverse and dependent on others’ assigned or enacted social roles. In fact, Luo (2005, p. 288) finds that that the presence of peers amplifies the urge to purchase, whereas the presence of family members decreases it. Yet, this difference is greater when the group (peers or family) is cohesive and when participants are liable to social influence⁴.

The effects of available time and money on impulsive purchasing behavior are rather intuitive. Before entering a shopping environment, customers are usually aware of the time they have at their disposal. The more time someone has on their hands while in a shopping environment, the more time they will spend browsing. The longer a customer browses through a store the higher the likelihood of impulse purchases, simply because the customer has more exposure to the store’s products. The money available at a specific time tends to affect people’s moods. If customers sense that they have a fair amount of money available they will tend to be in a more positive mood. Being in a positive mood has been shown to positively affect impulse purchases (Beatty & Ferrell, 1998, pp. 173-185).

Emotions and mood states play a crucial role in impulse purchasing⁵. Research so far clearly indicates that positive emotions have a positive effect on impulse buying for a variety of reasons. For negative affect, however, the relationship is less clear-cut. Whereas a negative mood decreases approach behavior, impulse purchases may be used as a means to relieve depression, cheer oneself up, or simply be an act of defiance (Beatty & Ferrell, 1998, p. 174). Cognitive processes - such as thinking, understanding, and interpreting - also affect impulsive buying decisions. However, even though cognitive deliberation plays a part in the impulse buyer’s decision, its influence is smaller as compared to the influence of the affective state (Weinberg & Gottwald, 1982, p. 44).⁶

Impulse buying may occur as a result of marketers’ environmental manipulations through atmospherics like scents and music, shelf and end-aisle displays, shelf and product positioning, tie-ins, the product itself, and the product’s price (Piron, 1991, p. 513). The following section will shed light on how retail managers can apply these stimuli to full effect.

2.13. How to boost impulse purchases? Tactics for retail managers

illustration not visible in this excerpt

Table 3: Physical arrangements, marketing innovations, and other tactics to boost impulse purchases

Various authors stress the importance of creating a positive and attractive shopping environment in order to encourage impulse purchases (Coley & Burgess, 2003, p. 293; Tuyet Mai et al., 2003, p. 30). The atmosphere that a store creates is paramount to attracting and keeping the customer excited about the shopping experience. Attractive decoration, creative merchandising, apposite lighting, music and scents augment mood, which may elicit affective tendencies leading to impulsive buying behavior (Coley & Burgess, 2003, p. 293). Levy and Weitz (2004, p. 614) corroborate this finding, stating that scent, in conjunction with music, has a positive impact on impulse buying behavior.

POS displays⁷ of any kind have received lots of attention in research on impulse buying. McGoldrick (1982, p. 32) mentions “the enormous significance of display and other in-store influences in relation to the customer’s decision to purchase”. In Abratt and Goodey’s (1990, p. 118) study, signs on the shelf, price, and special displays had the greatest influence on unplanned purchases.

Choosing the right products for promotions is essential in a retail manager’s goal to increase impulsive purchasing. Some product groups are far more likely to incite reminder or entirely unplanned purchases and as such possibly deserve special attention concerning displays and other in-store activity (McGoldrick, 1982, p. 28). Bellenger et al. (1978, p. 16) state that items with a high rate of impulsive purchasing usually need to be promoted more heavily in-store. Placing products with companion items that complement each other such as placing chips next to soft drinks has also been found to increase the number of impulse purchases (Rook & Hoch, 1985, p. 23). Dittmar et al. (1996, p. 204) suggest putting those goods up for promotion which relate to self-image concerns (e.g., make-up, magazines) rather than simple utility items, because the former are more likely to be bought on impulse according to their theory. Stern (1962, p. 62) proposes to construct impulse buying centers to complement the heavily trafficked checkout stand.

Marketing innovations can also be used to encourage customers to buy more on impulse. In-store instant coupons⁸ for example, lead to elevated mood states as well as psychological income effects, thereby boosting impulse purchases (Heilman et al., 2002, p. 251).

Measures that reduce perceived risk such as money back guarantees, generous return policies, and free samples can be an effective way to induce more impulse purchases in customers (D’Antoni & Shenson, 1973, p. 75). However, Bayley and Nancarrow (1998, p. 99) put forward the possibility that the ease with which goods can be returned might remove some of the excitement of a purchase and therefore diminish impulsive purchasing⁹.

Other suggestions include 24 hour retailing, instant credit, discounts tied to opening a charge card, and tying sales events to paydays or tax return periods. However, many of these proposals (e.g., concerning store opening times and credit guidelines) are subject to legislation which varies by country, thus, retail managers thinking about implementing such measures should check with the legal department first.

Wide aisles and effective display information help making shopping more effective and positively influence the perceived time that shoppers have at their disposal (Iyer, 1989, p. 55). This will lead to more and longer in-store browsing and consequently an increase in purchases made on impulse. Beatty and Ferrell (1998, p. 188) recommend identifying the profiles of highly impulsive or recreational shoppers and target promotions and events directly at these groups of customers. In a similar notion, Jones et al. (2003, p. 510) suggest identifying and targeting customers highly involved in a certain product category because they are more likely to be impulse purchasers. Additionally, building relationships and tracking these customers in a database enables the retailer to target specially designed promotions at these customers.

The recruitment and training of employees often go unnoticed as a facet of good retail marketing. Yet, it is vital that salespeople provide shoppers with immediate, anticipated gratification through their respect, courtesy, charm, and helpfulness (Tuyet Mai et al., 2003, p. 30; Hausman, 2000, p. 419).

2.14. Areas for further research

Although an extensive body of research on impulsive purchasing exists, there is still room for further research into impulse purchasing. Almost all of the extant research has been conducted in the United States while Europe and other parts of the globe have been mostly neglected. Consequently, both Kacen and Lee (2002, p. 173) and Tuyet Mai et al. (2003, p. 30) demand that more research be conducted in cultures outside the US.

As early as 1969, Kollat and Willett (p. 83) have called for the concept of impulse purchasing to be precisely defined. While efforts have been undertaken to clarify the concept and find a suitable definition, there is still none in place on which everyone agrees.

Rook (1987, p. 196) suggests to examine situational factors that influence impulse buying. How does the social environment (group vs. solo buying) affect decisions? What role do physical environment variables (retail atmospherics) play? Does it make a difference if the impulsively bought good is a gift or not?

Beatty and Ferrell (1998, p. 187) raise a series of questions which have to be resolved with respect to affect. For example, in what ways can positive affect most easily be stimulated in customers? Are highly impulsive people more likely to make impulse purchases, irrespective of mood, than others? What are the effects of impulse buying on one’s post-shopping affect levels? When and how do guilt or negative reactions to the buying experience affect respondents’ assessments of the experience, as well as their satisfaction and forthcoming purchasing and shopping strategies?

The customers’ inner processes, affective as well as cognitive, which take place just before, during, and right after the impulse purchase offer great research potential to this day. Baun and Groeppel-Klein (2003, p. 298) state that the impact of joy and surprise on other affective processes and the complex interaction of affective processes still present an extensive field for further studies. Herabadi et al. (2009, p. 29) suggest further investigating the role that cognitive facets have on impulsive buying. It may be appealing to comprehend whether customers acknowledge cognitive reasoning as a legitimate reason to purchase a product, maybe more than the more emotional urges.

It is people and not goods who experience impulses, thus, D’Antoni and Shenson (1973, p. 76) rightfully demand to shift the focus on consumer dynamics and the extension of the personality to impulsive behavior rather than on the physical or locational attributes of the goods themselves. Rook (1987, p. 196) proposes extensive post-purchase interviews with customers while they classify their acquisitions into categories ranging from the usual and planned to the uncontrollable and urgent that should pave the path towards a more finely grained continuum of impulsive buying.

3. Study 1

3.1. Method section

3.1.1. Participant characteristics

The sample consisted of 72 participants who completed every question of the questionnaire. Forty (55.6%) participants were male, and 32 (44.6%) female. The sample included respondents between 22 and 33 years of age. The respondents’ median age was 25 (M = 25.4, SD = 2). Basically everyone (97.2%) in the sample held a University degree. Nearly half of the sample (47.2%) was single, whereas the other half (48.6%) lived in a long- term relationship. Only three respondents were married. Nearly half (48.6%) of the respondents claimed to have a disposable income between 0 and 499 euros a month. The median is the income bracket between 500 and 999 euros. Only four respondents declared to have disposable incomes of 2.000 euros or more each month. Germans (41.7%) and Swiss (41.7%) constituted the majority of the sample. Other nationalities constituted 16.7% of the sample.

3.1.2. Sampling procedures

The link to the online questionnaire was spread via the e-mail distribution list of the Marketing, Services & Communication Master program at the University of St. Gallen. Additionally, the link was published on the author’s Facebook page accompanied with a plea to take part in the study. Consequently, roughly 400 people had the opportunity to take part in the study and constituted the population. In total, 189 people clicked on the link leading to the questionnaire. Of these, 72 completed every question of the questionnaire. This corresponds to a completion rate of 38.1%. No compensation was given to the respondents or held out in prospect for participating in the study.

3.1.3. Measures

The study was created and hosted using the EFS survey software by Unipark¹⁰. The complete questionnaire which was filled out by the participants can be examined in Appendix 1. The questionnaire was entirely in English. This may have deterred some people from filling it out, but given the population, the number of people who decided not to take part on language grounds is probably rather small. However, participants were given the choice to answer the two open-ended questions in either German, English, Spanish, or French. Apart from two open-ended questions, the questionnaire contained 14 question blocks in closed format.

3.1.4. Measurement instruments

An online questionnaire was applied to record respondents’ answers. The first two questions were in open format. The first question asked respondents to give a definition of an impulse purchase without consulting books or other resources. The second question presented the respondent with a definition of impulse purchases compiled by the author. Respondents were asked to think about a recent impulse purchase they made and give a detailed account of it. Every answer that was given in a language other than English was translated as accurately as possible by the author. For a complete list of respondents’ definitions of impulse buying and respondents’ impulse buying episodes see Appendix 2.

All following questions were in closed format. First, respondents were asked about the emotions and the level of cognition they experienced during the impulse purchase they described verbally in question two. There were six items in this block; a sample item is “I felt joyful and happy when seeing the product.” One of the items was directly taken from Baun (2002, p. 163) and translated into English, the others were inspired by various studies. Respondents were asked to indicate in how far they agree with each statement on a five-point Likert scale anchored at absolutely disagree and absolutely agree. This scheme was used for every question that had to be answered on Likert scales.

The next block consisting of five items delved into respondents’ impulse buying motives using the Likert scale presented above. A sample item from this block is: “That day I felt like treating myself to something nice. This article just seemed like the right thing for that.” The five items were taken from Baun (2002, p. 110) and translated into English.

Analogously, the following block consisted of seven items that evaluated which buying determinants were most conducive to the respondent’s impulse purchase. A sample item from this block is: “The way the product was presented in the store was crucial in my decision to buy it.” These items were determined via a brainstorming between the author and a colleague.

After indicating their gender, respondents were asked to reveal their nationality. This question was in open format. Next, respondents had to indicate their highest educational achievement, ranging from “no degree” to “University degree” with two intermediate levels.

The next block consisted of three items and was intended to determine the respondents’ level of involvement with the product category from which the impulse purchase was made. Two of the items were taken from the “Involvement with the product” scale and one from the “Involvement with the product category” scale¹¹. This new scale was judged by the author to adequately capture respondents’ involvement and reached a Cronbach’s alpha of 0.89. The corresponding factor analysis can be found in Appendix 3.

Analogously, the next block consisting of two items was intended to capture respondents’ level of perceived risk when buying products from the category from which the impulse purchase was made. One of the items was taken from the “Risk (general) scale”¹² developed by Laroche et al. (2005). The other item is original to the author. The scale reached an acceptable Cronbach’s alpha of 0.75. The corresponding factor analysis can be found in Appendix 3.

After indicating their marital status, respondents had to indicate whether, and if yes, who accompanied them on the day that the impulse purchase was made. Next, respondents were asked to reveal how much money they had at their free disposal each month. Money being a sensitive issue, five income brackets were predefined ranging from “0 to 499” euros to “2.000 euros and more”. Next, respondents were asked to indicate the frequency with which they visit the store from which the impulse purchase was made. Five options ranging from “less than once a month” to “more than once a week” were offered. The last block of items constituted the IBT-scale¹³ (Rook & Fisher, 1995). Nine, five-point Likert-type items were used to measure “a consumer’s tendency to buy spontaneously, unreflectively, immediately, and kinetically” (Rook & Fisher, 1995, p. 306).

3.1.5. Data collection, procedures, and analysis

The online questionnaire was live between September 21st and October 14th, 2011.

September 22nd was the day with the highest participation with almost 30% of the questionnaires filled out on that day. The mean time for completing the questionnaire was around 13 minutes with a Median of 11 minutes and 38 seconds.

Respondents were asked to give a personal definition of an impulse purchase. In total, 91 definitions were collected from the respondents. The goal was to identify the most prominent characteristics of an impulse purchase according to the respondents. As an orientation, Piron’s (1991, p. 511) list of definitional elements was used, slightly abridged, and enhanced by the author. In the end, the following eight categories were established:

(1) Unplanned / spontaneous, (2) Response/exposure to a stimulus, (3) Decision made on the spur of the moment/on-the-spot, (4) Emotional reaction, (5) No research or comparison,

(6) Not in response to a previously recognized problem/need; no intention, (7) Reduction of cognitive evaluation / no rationality or thinking, and (8) Low involvement.

Following this, each definition given by a respondent was analyzed as to which of the eight predefined dimensional elements it contained. The remaining results stem mostly from mean comparisons. All data were analyzed using IBM SPSS Statistics 20. All significance testing for mean differences was done via the non-parametric Mann-Whitney test. Kendall’s tau was applied for correlation analysis, because it is non-parametric and better suited for relatively small sample sizes. Apart from that, one can draw more accurate generalizations from Kendall’s statistic than from Spearman’s (Field, 2009, p. 181).

3.2. Result section

3.2.1. Respondents’ definitions of impulse buying

illustration not visible in this excerpt

Table 4: Number of times each dimension of impulse purchase definitions was mentioned

Note: Totals do not add up to 91 (i.e., number of respondents). Some respondents used two or more of the dimensions in their definitions.

The “unplanned/spontaneous” dimension was by far the most prevalent to appear in respondents’ definitions. More than 60% included this dimension into their definitions.

“ For me, an impulsive purchase is one which is carried out without planning. You see something at the POS and spontaneously decide to buy it. ” (Swiss, Female, 22) “ A spontaneous purchase that was neither planned nor intended. ” (Swiss, Male, 25)

The dimension “response/exposure to a stimulus” was the second most mentioned dimension. Almost 31% of the respondents incorporated it in their definitions.

“ Somebody buys something as a result of coming into contact with a stimulus (such as advertisement, peer pressure or just having a look at it) at the POS. ” (German, Male, 25) “ To me, an impulse purchase is a spontaneous purchase which is unplanned and which is triggered by situational stimuli (e.g., packaging, atmosphere, etc.) ” (Swiss, Female, 24)

An emotional reaction was included by 22% of respondents in their definitions.

“ For me, an impulse purchase is an unplanned purchase that is initiated by a spontaneously generated emotion. Alternatively, it might just be a purchase caused by a good opportunity that leads to a purchase (nice product placement, visual merchandising, or optimal location of a product). ” (German, Female, 25)

“ When you ’ re in the store in front of the shelf and suddenly feel like buying the product. A rather emotional purchase. ” (Swiss, Male, 27)

Three of the dimensions were mentioned by exactly the same number of respondents. These were “decision made on the spur of the moment/on-the-spot”, “not in response to a previously recognized problem/need; no intention”, and “reduction of cognitive evaluation/no rationality or thinking”. Each were mentioned by 16.5% of respondents.

“ Buying actions that just "happen". Grabbing something and buying immediately without long decision making or information search. ” (German, Male, 26)

“ An impulsive purchase is to buy something right in the moment just because you want it without thinking about the price, your needs or even if you like it or if it is useful for you. At the same time, it can be something you have always wanted but you had never decided to buy. ” (Spanish, Male, 25)

“ An impulse purchase to me is to buy something without prior intention, planning, or rational behavior. I am seduced by the effect that the product or the environment has on me. ” (Swiss, Male, 25)

“ To me, an impulse purchase is a purchase in which there is no rational thinking involved. It is activated by an impulse, e.g. cheap price tag or an emotion that is connected to the item etc. Important is the notion that it is not a planned purchasing behavior. ” (Swiss, Male, 27)

A few respondents (5.5%) included the dimension of not making comparisons or no thorough research before a purchase in their definitions.

“ If you buy something in a shop which you initially hadn't intended to buy. You didn't go to that shop to buy that specific item, but when you were there and saw it you decide to buy it without further investigation/price comparison/etc. ” (unknown)

“ An impulse purchase is if I buy without having planned to buy the respective object beforehand. In other words: I see something, want it and buy it without having researched the product, the price, etc. I did not consider all the benefits and costs, I just buy it. ” (German, Female, 25)

Two respondents included the dimension “low involvement” in their definitions.

“ A spontaneous purchase with low involvement and not much cognitive processing, e.g. low-price products at the checkout. ” (German, Female, 30)

These results leave us with two possibilities of coming up with a consumer-driven definition of impulse purchases. First, one could include the three dimensions that were mentioned most frequently (unplanned, exposure to stimulus, and emotional reaction) and discard the other five. Another possibility could include the three aforementioned dimensions plus the three dimensions that were mentioned by 16.5% of respondents each (decision made on-the-spot, not in response to a recognized need, and reduction of cognition/rationality). Two possible definitions derived from the respondents’ statements are presented:

1) An impulse purchase is a spontaneous, unplanned purchase which is made as a response or exposure to a stimulus and which is accompanied by an emotional reaction.
2) An impulse purchase is a spontaneous, unplanned purchase, decided and made on-the- spot, for which there was no previously recognized need, as a response or exposure to a stimulus, and which is accompanied by an emotional reaction and limited cognitive processes.

3.2.2. Emotions and Cognition

illustration not visible in this excerpt

Table 5: Means and standard deviations for emotions and cognition

Note: The values for cognition are reverse-coded, this means that a high value stands for low cognition and vice versa.

Among the emotions joy, surprise, excitement, and guilt, respondents showed the highest levels for joy (M = 3.81, SD = .99). Almost 70% of the respondents agreed or completely agreed with the statement “I felt joyful and happy when seeing the product”. Excitement had the second highest level (M = 3.57, SD = .93). More than half (54.2%) of the respondents agreed or completely agreed with the statement “I felt excited when seeing the product”. The lowest levels in this block of items were recorded for guilt felt after the impulse purchase (M = 2.42, SD = 1.35). Almost 60% of respondents disagreed or disagreed completely with the suggestion that they had felt any guilt. Surprise showed slightly higher levels than guilt, (M = 2.71, SD = 1.25). However, 47.2% disagreed or completely disagreed with the suggestion that they had experienced surprise upon discovering the product. Levels reported for the inner urge respondents felt when buying the product were M = 3.32, SD = 1.09. Less than half (45.8%) agreed or completely agreed with the suggestion that they had felt an inner urge to buy the product. More than one third (36.1%) were undecided concerning this matter and ticked neutral, the highest score for neutral in this battery of six items. Consequently, only 18.1% disagreed or disagreed completely having felt an inner urge to buy the product. Lastly, the levels of cognition were M = 3.17, SD = 1.27. Almost half (48.6%) of respondents agreed or completely agreed with the statement “When I discovered the article, I did not give it much thought and just bought it”.

3.2.3. Impulse Buying Motives

Abbildung in dieser Leseprobe nicht enthalten

Table 6: Means and standard deviations for impulse buying motives

An instant and spontaneous liking of the product, i.e., a pure impulse purchase, was most widely reported as a buying motive (M = 3.72, SD = 1.20). Two thirds of the respondents agreed or completely agreed with the suggestion that they had liked the product instantly and bought it on the spot. Rewarding oneself via impulsive purchasing had the second highest levels (M = 3.10, SD = 1.32). A little more than 40% agreed or completely agreed that their impulse purchase had been motivated by the wish to reward themselves. The perception of the article as being a special discount had a mean score of 2.74 and a standard deviation of 1.32. Reminder impulse purchases as a buying motive had scores of (M = 2.53, SD = 1.43). Buying the article as a substitute was reported at levels of (M = 1.90, SD = 1.10). Almost 78% disagreed or disagreed completely with the suggestion that the product they had bought was indeed a substitute for another product they had actually intended to buy.

[...]

---

¹ See http://www.onemoneydesign.com/20-tips-to-stop-impulse-buying-get-spending-under-control/ for an example.

² The exposure to in-store stimuli hypothesis proclaims that unplanned purchasing is a result of the customer being exposed to in-store stimuli. In-store stimuli apparently create new needs or remind the shopper of temporarily forgotten ones (Kollat & Willett, 1967, pp. 28-29).

³ According to Piron (1991, p. 512) an experiential impulse purchase differs from a non-experiential impulse purchase as only the former is accompanied by emotional and/or cognitive reactions. As a brief example, the customer who buys an expensive designer leather jacket on impulse may experience a mixture of guilt and excitement (experiential impulse purchase), but may not experience such reactions when buying a can of vegetables on impulse (non-experiential impulse purchase).

⁴ See also Brüggen et al. (2011, p. 78)

⁵ For a short and concise summary of the effects of positive and negative affect on impulse purchasing, see: Beatty and Ferrell (1998, pp. 173-174).

⁶ For an illustrative table on affective and cognitive psychological processes and related components to impulse buying, see: Coley and Burgess (2003, p. 283).

⁷ Phillips and Bradshaw (1993) explain in detail how to create effective displays in stores.

⁸ Unanticipated coupons encountered while in the store intended for use on that shopping trip. (e.g., electronic shelf coupons, peel-off coupons on product packaging).

⁹ See Suwelack et al. (2011) for a study on money-back guarantees and how they affect cognitive, affective, and behavioral outcomes.

¹⁰ www.unipark.de

¹¹ For more information on the original scales’ origin, reliability, and validity, see Bruner (2009, pp. 554- 556)

¹² For more information on the original scale’s origin, reliability, and validity, see Bruner (2009, p. 765)

¹³ For more information on the original scale’s origin, reliability, and validity, see Bruner (2009, pp. 486- 487)