Competing for attention with in-store promotions

Journal of Retailing and Consumer Services 26 (2015) 141–146 Contents lists available at ScienceDirect Journal of Retailing and Consumer Services journal homepage: www.elsevier.com/locate/jretconser Competing for attention with in-store promotions Megan Phillips a, Andrew G. Parsons a,n, Helene J. Wilkinson a, Paul W. Ballantine b a b Department of Marketing, Advertising, Retailing, and Sales, Auckland University of Technology, Private Bag 92006, Auckland 1142, New Zealand Department of Management, Marketing, and Entrepreneurship, University of Canterbury, Private Bag 4800, Christchurch 8140, New Zealand art ic l e i nf o a b s t r a c t Article history: Received 9 August 2014 Received in revised form 24 May 2015 Accepted 24 May 2015 Supermarkets typically have an in-store demonstration located near the promotional end-of-aisle (or end-cap) area due to space requirements. Using a field experiment, we examine whether the occurrence of these in-store promotions competing for attention and engagement can disrupt each other, using binary logistic regression to analyse shopper behaviour. Results show the best way to attract attention to the end-of-aisle is not to have an in-store demonstration near it, or if required, a complementary product to the end-of-aisle should be used. Inferences based upon shopper characteristics are also given, providing important nuances in the attention to, and engagement with, in-store promotions. & 2015 Published by Elsevier Ltd. Keywords: In-store promotions End-of-aisle Shopper behaviour Attention 1. Introduction A supermarket shopper will encounter a range of in-store promotional activities during a typical shopping trip, including store coupons, manufacturer coupons, product demonstrations, and end-of-aisle (or end-cap) displays. Each of these is designed to make an item stand out from the competition, attract the shopper's attention, and engage them with that item. However, with a plethora of activities competing for attention, does the “surprise” element, which has been shown (Itti and Baldi, 2009) to positively moderate the planned behaviour of a shopper, become obfuscated or even lost, and thus reduce the effectiveness of in-store promotions? This study examines the behaviour of shoppers confronted with in-store promotions competing for their attention and action. Specifically, this study will explore whether the effectiveness of an end-of-aisle display is diluted if there is a product demonstration occurring near the end-of-aisle. The implications of this have significant value for suppliers, who typically pay for both these locations. A supplier will be able to make more effective promotion choices if they know their activity is enhanced or reduced by the accompanying promotional activity. In-store promotions are increasingly favoured over external advertising because a connection can be made between the ability of in-store promotions to actively engage shoppers and increase sales (Abratt and Goodey, 1990; Bava et al., 2009; Wilkinson et al., n Correspondence to: Retailing Faculty of Business and Law, Auckland University of Technology, Level 4, WY Building, 120 Mayoral Drive, Auckland 1010, Private Bag 92006, Auckland 1142, New Zealand. E-mail address: [email protected] (A.G. Parsons). http://dx.doi.org/10.1016/j.jretconser.2015.05.009 0969-6989/& 2015 Published by Elsevier Ltd. 1982). Intensification of competition between brands has led to more vigorous promotions and significant expenditure (Gilbert and Jackaria, 2002). Research to-date has examined the value of in-store promotions and the responses of shoppers. These responses have included generating attention (Bava et al., 2009), stimulating purchase (Wilkinson et al., 1982), enhancing impulse buying likelihood (Abratt and Goodey, 1990), and increasing traffic (Spiekermann et al., 2011). Nordfalt and Lange (2013) in their comprehensive study took the novel approach of considering the timing (day-of-week) of in-store promotions. In the context of our study, Nordfalt and Lange (2013) also looked at whether a demonstration was with or without a display. What has not been examined, however, is whether the occurrence of in-store promotions competing for attention and engagement can disrupt each other, even when they are not in the same product category. 2. Theoretical background 2.1. Literature There are three typical in-store promotions for a supermarket. The first – in-store demonstrations – can be the presentation of a product (e.g. cooking demonstration), trial of a product (e.g. free tasting), or a combination of both. While there is little research on in-store demonstrations, and in particular trials (Heilman et al., 2001, 2011), the expectation is that they would be effective in gaining attention due to their physical presence and difference from surrounding aisles, and that they would be effective in creating engagement because the shopper is spoken to and has the 142 M. Phillips et al. / Journal of Retailing and Consumer Services 26 (2015) 141–146 option to trial the product (see Nordfalt and Lange (2013), for an excellent discussion). Furthermore, demonstrations should have a direct impact on sales given sampling is in the store where the product is available for immediate purchase, and is often available for selection at the demonstration point. The second in-store promotion is the end-of-aisle display. These are considered highly visible by manufacturers and supermarkets, and are changed frequently to maintain interest (Suher and Sorenson, 2010). The end-of-aisle is also the fixed in-store promotion space, and has traditionally been regarded as the prime real-estate of a supermarket. The normal expectation from shoppers is that there is some discount associated with the displayed item, though they are used to display new merchandise as well, which may or may not be discounted (Chevalier, 1975). End-ofaisle displays expose shoppers to items in a space free from direct aisle-based competition (Dulsrud and Jacobsen, 2009; Schindler et al., 1987). While the display itself is only passive engagement, the call-to-action is the associated discount price, which is announced with prominent signage, or the “newness” of the item, similarly announced. The third in-store promotion is the one that has received a greater level of research interest – coupons (Barat and Ye, 2012). Coupons are redeemable vouchers used to receive a specified reduction in price. They may be a straight discount (e.g. 10c off regular price), or they may be some value deal (e.g. two for the price of one). They are used by both manufacturers and retailers, and have been found to be more effective at increasing sales than advertised price discounts. They are known to have an impact on sales because they encourage shoppers to purchase more than they normally would of the promoted item, purchase earlier than their normal cycle of purchase, and they encourage switching behaviour from competing brands due to trial risk reduction (e.g. Aggarwal and Vaidyanathan, 2002; Heilman et al., 2002; Kahn and Schmittlein, 1992). In practise, a supermarket is likely to have an in-store demonstration located near or at the end of an aisle. This is simply because in a typical supermarket layout, this is where the most clear space is available to set-up a temporary stand. The location often has no relationship to the item′s normal product category location in an aisle. Therefore, an end-of-aisle display is competing for attention with the demonstration. In-store coupons, on the other hand, are typically located with the item in the normal aisle space. Sometimes they are associated with the end-of-aisle display as well, but we treat this as the same case as the end-of-aisle display. Thus, competition for attention is most likely to arise between product demonstrations and end-of-aisle displays. This potential conflict needs to be addressed because of the pragmatic implications for both manufacturers and retailers. Manufacturers often seek access to end-of-aisle space to gain an awareness advantage over competitors, and can be charged a premium for displays in these locations (Rossiter and Percy, 1987). Similarly, retailers looking to shift stock use this space to gain attention for a product that otherwise may be relatively unnoticed. At the same time, manufacturers also use demonstrations to create awareness. They are often used to introduce new products in an attempt to secure a viable level of sales to warrant allocation by the retailer of limited shelf space. Manufacturers pay retailers for the demonstration space, and of course also incur the demonstration costs (e.g. giving away trial items; the demonstrator wages), while retailers incur an opportunity cost by allowing selling space in their store to be used by the manufacturer. Any reduction in attention or engagement due to competing attractions harms the potential value of the in-store promotion. 2.2. Hypothesis development 2.2.1. Product demonstrations and end-of-aisle displays In-store product demonstrations are largely under-researched (Heilman et al., 2011; Nordfalt and Lange, 2013), with only a handful of studies focused specifically on sampling in the retail environment (e.g. Heilman et al., 2011; Lammers, 1991; Steinberg and Yalch, 1978), or considering sampling as part of the wider study of in-store promotions (e.g. Gedenk and Neslin, 1999; Shi et al., 2005). Trade and academic research shows that in-store product demonstrations have an impact on sales – including those of other products – and trial (Lammers, 1991; Laposky, 2007; Lawson et al., 1990; Major, 2002; Moses, 2005; Troy, 2005; Zwiebach, 2005), and it is believed that a key reason for this is that the demonstrator grabs attention, and the product is there for the shopper to purchase (Heiman et al., 2001). Surprise is also likely to have an impact, given that the in-store product demonstration is where the shopper normally expects nothing to be. Similarly, research on end-of-aisle displays has tended to examine them in the context of wider in-store promotions (e.g. Curhan, 1974; Fader and Lodish, 1990; Haans and Gijsbrechts, 2011; Lemon and Nowlis, 2002; Wilkinson et al., 1982). Underhill (1999) suggests that, like product demonstrations, they act as a prompt for immediate purchase. They are seen as very effective attention generators (Dulsrud and Jacobsen, 2009; Schindler et al., 1987), even when goods are not discounted (Chevalier, 1975). Bezawada et al. (2009) examined the placement in aisles and endof-aisle displays in the context of cross-category placement (i.e. complementary goods), using aggregate sales data to measure placement effectiveness. Together, in-store product demonstrations and end-of-aisle displays appear to have similar characteristics – they locate in similar areas, they focus on gaining attention from their physical presence rather than price, and they promote immediacy of purchase by having the product available for the shopper to select easily. This suggests that competition for attention is likely to occur when both are present in neighbouring spaces. As we are unable to remove the end-of-aisle (by definition) we therefore couch our hypothesis in terms of the end-of-aisle being the basepoint. H1. The presence of an in-store demonstration near an end-ofaisle will affect shoppers' attention paid to the end-of-aisle. ‘Near’ is defined as having to pass between the in-store demonstration and the end-of-aisle to navigate past the end-of-aisle. Our expectation from the literature is that an in-store demonstration in proximity to the end-of-aisle will grab attention away through surprise and immediacy. 2.2.2. Surprise and shopper profile Itti and Baldi (2009) state that the strongest attractors of attention are stimuli that pop-out from their neighbours in space or time, and that surprise is the best known attractor of human attention. The same authors (Baldi and Itti, 2010) note that novel or salient events attract attention. In-store promotions fall into this description, and the study by Heilman et al. (2002) considers the impact of unexpected in-store coupons as a pleasant surprise, showing that shoppers encountering this surprise are more likely to make unplanned purchases. This notion of planning (i.e. the Theory of Planned Behaviour, Ajzen, 1991) is important in our context because the disruption of the planned behaviour – which is prevalent in shopping (Iyer, 1989), because of the multiple buying goals characterising supermarket shopping (Park et al., 1989) – is likely to be modified by whether the overall shopping trip goal is very precise and concrete (e.g. to take advantage of a specific promotion) through to relatively abstract (e.g. to fill up on weekly needs; Bell et al., 2011). Lee and Ariely (2006) suggest that M. Phillips et al. / Journal of Retailing and Consumer Services 26 (2015) 141–146 the success of marketing actions, such as promotions, depends on the goals shoppers have when they are exposed to such promotions. In another context we see this occur with the use/prominence of shelf facings when it comes to brand attention (Chandon et al., 2009). We expect these goals to manifest themselves through the shopper profile. The profile of the shopper will impact upon attention, and the ability to engage. Shopper orientation – hedonic or task oriented – is known to impact attention paid to environmental cues, such as in-store displays/demonstrations (Breugelmans and Campo, 2011), and we also know that the gender of the shopper is a moderator of shopper orientation (Mortimer, 2012; Mortimer and Clarke, 2011; Tifferet and Herstein, 2012). The type of shopping trip also has an effect (Mazumdar and Papatla, 1995). “Fill-in” and “major-trip” shoppers have different responses to displays than “intermediatetrip” shoppers. The shopping party (e.g. size, relationships) can interfere with the shopping environment and influence engagement (Borges et al., 2010), and even the shopping device – anything from a cart to carrying items in one's hands – can influence attention and the ability to engage (Cochoy, 2008). Given the preceding discussion, the focal research objective of this study, addressed in H1, is to examine whether having an instore product demonstration situated near an end-of-aisle affects shoppers' attention to the end-of-aisle display. Addressing this objective will help to understand where the most effective place to locate a product demonstration is, and will help to determine whether two promotional activities next to each other are beneficial or not. Moreover, this objective will help address whether the attention paid to the in-store promotion (the end-of-aisle and/ or product demonstration) will be affected by whether the in-store promotions are competing with each other, and whether that competition involves the same or different product categories. To this end we add a second hypothesis to the study: H2. The presence of an in-store demonstration near an end-ofaisle with the same (different) products will affect shoppers' attention paid to the end-of-aisle. We also explore how the characteristics of the shopper, including gender, the shopping device being used, and the direction and movement of their travel, impact on the attention given to these two in-store promotion types. In essence we examine whether differences in any of the shopper profile characteristics affect attention, establishing the following pair of generalised hypotheses: H3. Differences in gender/shopping device/direction of movement/travel will affect shoppers' attention paid to the end-of-aisle. H4.Differences in gender/shopping device/direction of movement/travel will affect shoppers' attention paid to the in-store demonstration. 3. Method 3.1. Overview A field experiment following an after-only with control group approach was undertaken in a large supermarket situated in a suburban area of a mid-sized city. The supermarket cooperated by sourcing manufacturer product demonstrators and coordinating them with product change-outs in the end-of-aisle displays, allowing combinations of demonstrations and displays to be deployed. Juice and cookies were selected as the two product lines for the promotional activities. Each was easy to display at the endof-aisle, and the juice was simple to demonstrate (offer to taste). They were not in the same (or even similar) categories, but were likely to appeal to a similar market. Nordfalt and Lange (2013) cleverly noted that there might be a relationship between day-of- 143 Table 1 Daily demonstration and display combinations; indicator variables. Group Product demonstration End-of-aisle display A (Day 1) B (Day 2) C (Day 3) Juice Juice n/a Juice Cookies Juice and cookies the-week and product, when it came to promotion effectiveness in-store. The two products selected do not appear to have any discernible cyclical pattern, so this was not considered an issue. Video-recorded observations were made of shopper behaviour over a three-day period. The three daily combinations that were observed are shown in Table 1. The two product demonstration days (A and B) were located adjacent to the end-of-aisle, allowing approximately seven feet of space for shopper movement, which was the same space as in the aisles. The end-of-aisle only day (C) used the two adjacent end-ofaisles that the product demonstration (A and B) had been located next to. Collection of data was for the duration of each product demonstration, and for one hour on the non-demonstration (end-ofaisle only) day, with 1807 observations being collected overall. Also observed was the gender of the shopper, the shopping device being used, and the direction of movement/travel (i.e. from which direction did the shopper approach the location and where they then went). The observations were made via security cameras so that the shoppers were unaware that their behaviour was being observed, which should result in natural behaviour, and reliable observation. A detailed map of the observation area was used for the manual recording and coding of shopper movement patterns in order to evaluate their activity and behaviour. In addition to the variables discussed in Table 1, there were a number of sample variables that were used. These are expanded upon in the next sub-section, and are shown in Table 2. All the variables utilised were categorical and the coding scheme was indicator-variable coding (i.e. a 1 for present; 0 for not present). Reference groups for each of the independent variables were Table 2 Summary of sample data. Group A (Day 1) B (Day 2) C (Day 3) Total sample Gender Femalen Male 178 (65.9) 92 (34.1) 194 (71.9) 76 (28.1) 176 (65.2) 94 (34.8) 548 (67.7) 262 (32.3) Shopping device Nothing Trolley Basketn 93 (34.4) 109 (40.4) 68 (25.2) 97 (35.9) 90 (33.3) 83 (30.7) 88 (32.6) 91 (33.7) 91 (33.7) 278 (34.3) 290 (35.8) 242 (29.9) Movement Anticlockwisen Clockwise Both 218 (80.7) 208 (77.0) 208 (77.0) 634 (78.3) 44 (16.3) 8 (3.0) 54 (20.0) 8 (3.0) 49 (18.1) 13 (4.8) 147 (18.1) 29 (3.6) Travel Perimetern Aisle Both 97 (35.9) 20 (7.4) 153 (56.7) 135 (50.0) 11 (4.1) 124 (45.9) 75 (27.8) 30 (11.1) 165 (61.1) 307 (37.9) 61 (7.5) 442 (54.6) Paid attention to End-of-aisle Demonstration 78 (28.9) 88 (32.6) 85 (31.5) 116 (43.0) 176 (65.2) n/a 339 (41.6) 204 (37.8) Frequencies and percentages (in parentheses). n Reference category. 144 M. Phillips et al. / Journal of Retailing and Consumer Services 26 (2015) 141–146 determined and are shown in Table 2 with a * indicating selection, so that when the variables were recoded, the coefficients for the new variables represented the effect of each category compared to the reference category (i.e. similar to the use of dummy variables in linear regression). As both the dependent and independent variables were categorical, binary logistic regression was used, with all variables being entered into the analysis simultaneously. Each of the reference groups was chosen based upon characteristics found within the sample, and these are discussed next. 3.2. Sample Of the 1807 shopper observations, 1186 (65.6%) were female, 494 (27.3%) were male, and 127 (7.0%) were couples. The proportions found in the sample are similar to recent studies, and as female shoppers are the largest group, they were chosen as the reference category for the variable gender. The majority of observations were also of individual shoppers (n ¼1463), with fifteen other shopping party configurations being observed. In order to more clearly focus on individual, as opposed to group level behaviour, only those observations that included a single shopper were retained for later analysis. Of the 1807 observations, there were six types of shopping devices used, with shopping carts (n ¼ 633; 35.0%), nothing (n ¼615; 34.0%), and baskets (n ¼522; 28.9%) being the most common. In this case, baskets were chosen as the reference category. Although they were only the third largest group, speaking with the management of the supermarket revealed that a basketcarrying customer was their primary target. Direction of movement and direction of travel around the supermarket was considered. In terms of movement, and of the 1807 observations, 1457 (80.6%) moved anti-clockwise, 269 (14.9%) moved clockwise, and 81 (4.5%) moved both ways. Anti-clockwise was the natural movement pattern given the store layout (and was the largest group), so was chosen as the reference category. Finally, shoppers had three ways they could travel about the store – around the perimeter, through the aisles, or in both the perimeter and aisles. Of the 1807 original observations, 990 (54.8%) travelled both, 662 (36.6%) travelled the perimeter only, and 155 (8.6%) travelled only the aisles. Perimeter travel was selected as the reference category, as basket shoppers (the target group for the supermarket) were targeted with perimeter products. Due to the differing durations of the two product demonstration days, there was a discrepancy in the number of observations recorded. Specifically, 1102 observations were recorded on Day One, while 347 were recorded on Day Two (358 were recorded on the day when there was no product demonstration). To ensure an equal number of observations were analysed for each of the three days, in addition to the omission of those observations which did not include a single shopper only, it was also decided to only include the three main types of shopping device in the analysis (shopping carts, nothing, or baskets). Given these criteria, and the desire for an equal sample size for each of the three days, observations were randomly deleted to achieve this goal, resulting in an analysis sample of 810 observations, or 270 observations for each of the three days. A summary of the sample used for the statistical analysis presented in the next section (including frequencies and percentages) is provided in Table 2. 4. Results Attention was operationalized as the visual connection the observed shopper made with the in-store demonstration and/or the end-of-aisle. If the shopper slowed down or stopped, and looked at the in-store demonstration, or went further and engaged with the demonstrator and/or examined the product then they were considered to have visually connected with the in-store demonstration and thus gave it attention. Similarly the shopper who slowed down or stopped at the end-of-aisle display, and looked at it, or went further and examined the product, was considered to have given their attention to the display. We can see from the last two rows of Table 2 that there is a clear difference in attention when there is no competing promotion. A simple Chi-square test reveals that there is a significant difference (χ2 ¼90.98, p o.01) and further investigation of the direction shows that attention towards the end-of-aisle improves without competing promotions. Therefore we find that H1 is supported, and that the presence of an in-store demonstration near the end-of-aisle does affect shoppers’ attention paid to the end-of-the-aisle. Binary logistic regression analysis was used to examine the impact of the four independent variables (gender, shopping device, movement, and travel) on the two dependent variables of interest (attention to the in-store demonstration and attention to the end-of-aisle display). This section presents the results for both main and two-way interactions. The significance of each model was indicated by the chi-square statistic, while the Wald statistic was used to evaluate the statistical significance of each independent variable. Finally, the percentage of the number of cases correctly predicted by each model was also determined. 4.1. Main effects Both models (attention to the in-store demonstration and attention to the end-of-aisle display) were significant when the instore demonstration and end-of-aisle display were complementary to each other. Both movement and travel were significant in terms of their impact on attention being paid to the instore demonstration, while only movement was found to be significant in terms of explaining attention being paid to the end-ofaisle display (see Table 3). On the day when the in-store demonstration and end-of-aisle display were not complementary to each other, neither model was significant (see Table 4). Finally, on the day when there was no in-store demonstration (see Table 5), it was found that shopping device and movement were significant in terms of their ability to help explain the attention being paid to an end-of-aisle display. The results indicate that gender was not a significant predictor in any of the models. Movement was found to be a significant predictor in all three models which were statistically significant, while the impact of the shopping device used and travel pattern were inconsistent. 4.2. Two-way interactions When the in-store demonstration and end-of-aisle display were complementary to each other, it was found that only the model for attention being paid to the end-of-aisle was significant Table 3 Complementary in-store demonstration and end-of-aisle display. Attention to in-store demonstration Attention to end-of-aisle display Wald Sig Wald Sig Gender .10 Shopping device 2.39 Movement 13.17 Travel 7.83 Model chi-square 31.39 Percentage 68.50 .76 .30 .00 .02 .00 .21 5.22 14.54 4.58 37.49 73.30 .65 .07 .00 .10 .00 M. Phillips et al. / Journal of Retailing and Consumer Services 26 (2015) 141–146 Table 4 Competing in-store demonstration and end-of-aisle display. 145 Table 8 End-of-aisle display only. Attention to in-store demonstration Attention to end-of-aisle display Wald Sig Wald Sig Gender .30 Shopping device 2.20 Movement .43 Travel 3.70 Model chi-square 6.42 Percentage 60.40 .59 .33 .81 .16 .49 .25 5.14 5.34 1.41 11.41 69.60 .62 .08 .07 .49 .12 Attention to end-of-aisle display Gender  Device Gender  Movement Gender  Travel Device  Movement Device  Travel Movement  Travel Model chi-square Percentage Wald Sig 2.09 9.21 1.30 19.87 3.07 1.05 57.96 73.30 .35 .01 .52 .00 .55 .90 .00 Table 5 End-of-aisle display only. Looking at our remaining hypotheses, we find that in general, H2 is supported, and that if the product in the in-store demonstration is different (competing for attention) this distracts from the end-of-aisle. Considering our third and fourth hypotheses, we find that in some cases shopper profile characteristics do have an effect on attention paid to the in-store demonstration and/or the end-of-aisle promotion. We discuss these and the broader findings in the next section. Attention to end-of-aisle display Gender Shopping device Movement Travel Model chi-square Percentage Wald Sig 1.29 19.10 16.54 .05 48.62 69.60 .26 .00 .00 .97 .00 5. Discussion and conclusions Table 6 Complementary in-store demonstration and end-of-aisle display. Attention to in-store demonstration Attention to end-ofaisle display Wald Sig Wald Sig Gender  Device .36 Gender  Movement 1.50 Gender  Travel 2.38 Device  Movement 3.14 Device  Travel 3.51 Movement  Travel .30 Model chi-square 27.03 Percentage 69.30 .84 .47 .31 .54 .48 .99 .08 1.58 1.18 .37 12.17 12.31 5.68 42.54 72.60 .45 .55 .83 .02 .02 .23 .00 Table 7 Competing in-store demonstration and end-of-aisle display. Attention to in-store demonstration Attention to end-ofaisle display Wald Sig Wald Sig Gender  Device .67 Gender  Movement .73 Gender  Travel 1.07 Device  Movement 6.05 Device  Travel 3.38 Movement  Travel 4.68 Model chi-square 23.23 Percentage 64.10 .72 .69 .59 .20 .50 .32 .18 2.03 1.33 2.64 2.98 1.58 .53 18.28 70.40 .36 .52 .27 .56 .81 .97 .44 (see Table 6), with two significant two-way interactions being found between shopping device and movement, and shopping device and travel. When the in-store demonstration and end-ofaisle display were not complementary to each other, neither model was significant (see Table 7); consistent with when only the main effects were examined. Finally, on the day when there was no in-store demonstration (see Table 8), two significant interactions were found in their ability to predict attention being paid to an end-of-aisle display: gender and movement, and shopping device and movement. The purpose of this study was to establish whether having an in-store product demonstration situated near an end-of-aisle affects shoppers' attention to the end-of-aisle display. The results shown in Table 2 clearly show (in the final two rows “end-of-aisle” and “demonstration”) that the best way to attract attention to the end-of-aisle is not to have an in-store demonstration anywhere near it. This has clear implications for suppliers being charged for these spaces yet clearly having concerns about erosion of investment and ROI. At present suppliers are investing in these spaces with reported returns on their promotions, yet until now there has been no research into the interaction between end-of-aisle and demonstration. Our results suggest that supermarkets may have incomplete information when reporting success of previous instore promotions. However, assuming the store wants to deploy in-store demonstrations, and given the end-of-aisle space is the easiest place for one, it is worth discussing some nuances from the results. First, it is also clear that both in-store promotions have attention paid to them when they are not competing with each other, but do not have attention paid when they are competing. Thus, it would seem that if a store is going to have a product demonstration, the management should ensure they have a complementary end-ofaisle display. Management can add significant value and benefit to their stakeholders by strategically planning these types of in-store promotions. A supermarket which is aware of and actively engaged in such planning activities will create a mutually beneficial relationship beyond the single in-store promotion. Second, it is clear that when a store does have both (and they are complementary), movement that is against the natural movement pattern (clockwise in this case) is an important factor. This makes sense if we think about the shopper going with the store layout flow, easily moving around the end-of-aisles and looking ahead to what is expected next; not noticing as much the in-store promotions, whereas the shopper moving against the flow is navigating around points not designed with them in mind. This resonates with the Itti and Baldi (2009) and Baldi and Itti (2010) suggestions around standing out and surprise. We see that attention to both the end-of-aisle and the in-store demonstration promotions are affected by movement against the flow, and in- 146 M. Phillips et al. / Journal of Retailing and Consumer Services 26 (2015) 141–146 store demonstrations are also noticed when the travel is through the aisle rather than around the perimeter. When there is no instore demonstration – so end-of-aisle promotion only – pushing a shopping cart increases attention paid to the end-of-aisle promotion. Furthermore, when looking at the interactions, we see that pushing a shopping cart up and down the aisles, and pushing a shopping cart against the natural flow, increases attention paid to the end-of-aisle when the two in-store promotions are complementary. Again, this notion of the shopping device being important (Cochoy, 2008) makes sense when we think of the shopper with a shopping cart needing to negotiate the obstructions more carefully, but also perhaps having more time for the overall trip so more willing to explore promotions. Thus the supermarket and their suppliers need to consider the profile of the shopper in terms of how they move about the store and the shopping device used, when employing in-store promotions. We have shown that attention to in-store promotions is best when “competing” promotions are actually for the same product– so complementary, and not actually competing. Or, they can work when the promotion is restricted to the end-of-aisle only. This is important for the marketer working with the retailer, as spend on marketing efforts that are disrupted by promotions competing for attention is largely ineffective. The marketer is better to ensure agreement that the end-of-aisle is maintained as a clear way for promotional purposes, or to invest in complementary in-store demonstration promotional activity when their product is on the end-of-aisle display. Our results clearly show that the effectiveness of the end-ofaisle display is diluted if there is a product demonstration occurring nearby. This has meaningful implications, as previously discussed, for suppliers, shoppers, and supermarkets. To maximise the return on these in-store promotional activities, suppliers and supermarkets must work together. References Abratt, R., Goodey, S., 1990. Unplanned buying and in-store stimuli in supermarkets. Manag. Decis. Econ. 11, 111–121. Aggarwal, P., Vaidyanathan, R., 2002. Use it or lose it: purchase acceleration effects of time-limited promotions. J. Consum. Behav. 2, 393–403. Ajzen, I., 1991. The theory of planned behavior. Org. Behav. Hum. Decis. Process. 50, 179–211. Baldi, P., Itti, L., 2010. Of bits and wows: a Bayesian theory of surprise with applications to attention. Neural Netw. 23, 649–666. Barat, S., Ye, L., 2012. Effects of coupons on consumer purchase behavior: a metaanalysis. J. Mark. Dev. Compet. 6, 131–144. Bava, C., Jaeger, S., Dawson, J., 2009. In-store influences on consumers' grocery purchasing decisions: a qualitative investigation. J. Cust. Behav. 8, 221–236. Bell, D., Corsten, D., Knox, G., 2011. From point of purchase to path to purchase: how preshopping factors drive unplanned buying. J. Mark. 75, 31–45. Bezawada, R., Balachander, S., Kannan, P., Shankar, V., 2009. Cross-category effects of aisle and display placements: a spatial modeling approach and insights. J. Mark. 73, 99–117. Borges, A., Chebat, J., Babin, B., 2010. Does a companion always enhance the shopping experience? J. Retail. Consum. Serv. 17, 294–299. Breugelmans, E., Campo, K., 2011. Effectiveness of in-store displays in a virtual store environment. J. Retail. 87, 75–89. Chandon, P., Hutchinson, J., Bradlow, E., Young, S., 2009. Does in-store marketing work? effects of the number and position of shelf facings on brand attention and evaluation at the point of purchase. J. Mark. 73, 1–17. Chevalier, M., 1975. Increase in sales due to in-store display. Jo. Mark. Res. 12, 426–431. Cochoy, F., 2008. Calculation, qualculation, calqulation: shopping cart arithmetic, equipped cognition and the clustered consumer. Mark. Theory 8, 15–44. Curhan, R., 1974. The effects of merchandising and temporary promotional activities on the sales of fresh fruit and vegetables in supermarkets. J. Mark. Res. 11, 286–294. Dulsrud, A., Jacobsen, E., 2009. In-store marketing as a mode of discipline. J. Consum. Policy 32, 203–218. Fader, P., Lodish, L., 1990. A cross-category analysis of category structure and promotional activity for grocery products. J. Mark. 54, 52–65. Gedenk, K., Neslin, S., 1999. The role of retail promotion in determining future brand loyalty: its effect on purchase event feedback. J. Retail. 75, 433–459. Gilbert, D., Jackaria, N., 2002. The efficacy of sales promotions in UK supermarkets: a consumer view. Int. J. Retail Distrib. Manag. 30, 315–322. Haans, H., Gijsbrechts, E., 2011. . “One-deal-fits-all?”: on category sales promotion effectiveness in smaller versus larger supermarkets.. J. Retail. 87, 427–443. Heilman, C., Lakishyk, K., Radas, S., 2011. An empirical investigation of in-store sampling promotions. Br. Food J. 113, 1252–1266. Heilman, C., Nakamoto, K., Rao, A., 2002. Pleasant surprises: consumer response to unexpected in-store coupons. J. Mark. Res. 39, 242–252. Heiman, A., McWilliams, B., Shen, Z., Zilberman, D., 2001. Learning and forgetting: modeling optimal product sampling over time. Manag. Sci. 47, 532–546. Itti, L., Baldi, P., 2009. Bayesian surprise attracts human attention. Vis. Res. 49, 1295–1306. Iyer, E., 1989. Unplanned purchasing: knowledge of shopping environment and time pressure. J. Retail. 65, 40–57. Kahn, B., Schmittlein, D., 1992. The relationship between purchases made on promotion and shopping trip behavior. J. Retail. 68, 294–315. Lammers, B., 1991. The effect of free samples on immediate consumer purchase. J. Consum. Market. 8, 31–37. Laposky, J., 2007. In-store demos drive audio sales. Twice 22, 14. Lawson, M., McGuinness, D., Esslemont, D., 1990. The effect of in-store sampling on the sale of food products. Mark. Bull. 1, 1–6. Lee, L., Ariely, D., 2006. Shopping goals, goal concreteness, and conditional promotions. J. Consum. Res. 33, 60–70. Lemon, K., Nowlis, S., 2002. Developing synergies between promotions and brands in different price – quality tiers. J. Mark. Res. 39, 171–185. Major, M., 2002. Show and tell – and sell. Progress. Grocer 81, 68–72. Mazumdar, T., Papatla, P., 1995. Effects of shopping basket size on price and promotion responses. Pricing Strategy Pract. 3, 16–27. Mortimer, G., 2012. Toward a shopping typology of primary male grocery shoppers. Int. J. Retail Distri. Manag. 40, 790–810. Mortimer, G., Clarke, P., 2011. Supermarket consumers and gender differences relating to their perceived importance levels of store characteristics. J. Retail. Consum. Serv. 18, 575–585. Moses, L., 2005. Food city expands sampling program. Supermark. News 53, 43. Nordfalt, J., Lange, F., 2013. In-store demonstrations as a promotion tool. J. Retail. Consum. Serv. 20, 20–25. Park, C., Iyer, E., Smith, D., 1989. The effects of situational factors on in-store grocery shopping behavior: the role of store environment and time available for shopping. J. Consum. Res. 15, 422–433. Rossiter, J., Percy, L., 1987. Advertising and Promotion Management. McGraw-Hill, Singapore. Schindler, R., Berbaum, M., Weinzimer, D., 1987. How an attention-getting device can affect quick choice among similar alternatives. Adv. Consum. Res. 14, 505–509. Shi, Y., Cheung, K., Prendergast, G., 2005. Behavioural response to sales promotion tools: a Hong Kong study. Int. J. Advert. 24, 467–486. Spiekermann, S., Rothensee, M., Klafft, M., 2011. Street marketing: how proximity and context drive coupon redemption. J. Consum. Mark. 28, 280–289. Steinberg, S., Yalch, R., 1978. When eating begets buying: the effects of food samples on obese and nonobese shoppers. J. Consum. Res. 4, 243–246. Suher, J., Sorensen, H., 2010. The power of atlas: why in-store shopping behavior matters. J. Adver. Res. 50, 21–29. Tifferet, S., Herstein, R., 2012. Gender differences in brand commitment, impulse buying and hedonic consumption. J. Prod. Brand Manag. 21, 176–182. Troy, M., 2005. The lure of in-store freebies. DSN Retail. Today 44, 54. Underhill, P., 1999. Why We Buy. Simon and Schuster, New York. Wilkinson, J., Mason, J., Paksoy, C., 1982. Assessing the impact of short-term supermarket strategy variables. J. Mark. Res. 19, 72–86. Zwiebach, E., 2005. Marsh increase in-store sampling. Supermark. News 53, 24.