Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1598-7248 (Print)
ISSN : 2234-6473 (Online)
Industrial Engineering & Management Systems Vol.16 No.2 pp.205-214

Is Your Check-Out Counter’s Profile Display Impulsive Enough for Your Impulsive Goods?

Paulina K Ariningsih*, Subagyo
Industrial Engineering Department, Gadjah Mada University, Indonesia
Corresponding Author,
January 29, 2016 February 20, 2017 March 13, 2017


Difficulties of developing strategy for improving impulsive buying (IB), major purchase transaction in supermarkets, had been general case. There is research gap on examining and assessing efficiency of decision variables from supermarkets’ point of view, since most IB studies were still focusing on perception and psychological variables of customer’s point of view. This article is focusing on fundamental exploration of display, one of supermarket’s decision variables, and on development of assessment model to improve IB transaction. IB improvement will enhance supermarket’s competitive advantage.

Mathematical model is delivered by logistic regression, which then becomes an assessment model. The model was developed by interviewing 110 respondents and observing 55 product group displays in Check-Out Counter (COC) of five supermarkets in Indonesia. The model developed indicates that IB occurred because of the proper arrangement of display profile: space width (SW), product mix complexity inside a group product (COM), and height of level from eye level (LVL). IB responses for the same display profile maybe vary depends on the consumer’s queuing time (TM). As managerial implication, this research is suggesting to provide product mix which will expand its space width, to display product closer to eye level, and to maintain proper length of queuing.



    Even though the e-commerce practice had been widely spread, brick and mortar store, such as supermarket, would still dominate grocery sales because the enjoyable and engaging experience inside the store (Nielsen, 2015). Economics Intelligence Unit noted that retail growth in Asia emerging countries are predicted to continue with an increase of 5-6% annually until 2018 (PWC, 2015). This is supported by the shifting buying behavior of urban people which move away from traditional market. In Indonesia, annual modern retail growth had arrived at 15% and its market proportion achieved 30% (ICN, 2011).

    Impulsive buying (IB), spontaneous or unplanned buying behavior, is a major (more than 50%) transaction in modern retail (Rook and Fisher, 1995; Madhavaram and Lavarie, 2004; Stilley et al., 2010). Experimental study of IB in consumer behavior demonstrates the contribution of habits, environment and human cognitive limitation to spontaneous yet irrational decisions (Ariely, 2012; Kahneman, 2011; Solomon, 2011). The irrationality of buying appears through some causal patterns, such as habits, tendency of jumping to conclusions during quick thinking, availability of variance and alternatives, and the inability of identifying risk (Ahmad, 2011; Kahneman, 2011; Ariely, 2012; Solomon, 2011). Supermarket as a sales institution will gain more sales if they can increase consumer’s IB behavior.

    IB is influenced by product’s price and promotion, hedonic motivation, self-esteem fulfillment, store atmos- phere, store layout (Tafesse dan Korneliussen, 2012) and product display (Kacen et al., 2012). Impulsive products are supposed to be placed in the front space of the supermarket or near COC and Check-In Counter (CIC) so consumer recognizes products effortless which lead to higher sales (Amos et al., 2014). Therefore in most retail, COC and CIC are considered as most valuable Place of Purchase (POP), a place where the decision of buying happens. Furthermore, Levy and Weits (2012) noted that COC has the highest consumer traffic. Samson and Lite (1993) shows that 40% of sales come from 25% front area where the COC and CIC is placed on most modern retail grid layout type.

    Most IB researches are more focused on analysis in finding the causal variables of IB from consumer’s cognitive point of view-emotional and perceptions, such as motivational influences or product characteristics (Amos et al., 2014). Meanwhile physical environment has been known as stronger stimulus to impulsive buying decision (Herabadi et al., 2009). Therefore, it is important to deeply investigate the physical environment.

    Physical environments as impulsive stimulus are grouped into ambience stimulus (temperature, music background, noise, and aroma), design stimulus (layout, spaces, display, colors and décor), and social factors stimulus or sales person (Tafesse and Korneliussen, 2012). Strong arguments exist on the positive influence of display to IB (Wakefield et al., 2007; Geetha et al., 2009) that the display is strongly influencing IB behavior. Stilley et al. (2010) add the importance of shopping pathway route into display stimulus. There is also an important influence of product appearance display (Karbasivar and Yarahmadi, 2011). However, the study of impulsive stimulus is still in a debate since some study shows that hedonic motivation are more significant than layout (Liu et al., 2007) or display (Kacen et al., 2012).

    1.1.Purpose of the Study

    Most researches of IB are still emphasizing their research in the perception of consumer behavior. Amos et al. (2014) only distinguish a few researches that determine and measure display factor (physical stimulus) that influence IB, such as: Kacen et al. (2012). Chang et al. (2011) studied the three characteristics of retail environment (ambient, design and social) toward positive emotions build inside the store which triggers IB. However, those researches had given broad strategic foundations but not detail tactical implementations of display factors strategy. Chandon et al. (2009) reveal the essential of studies in measuring the effectiveness of POP display factor on sales. Vieira (2013) tries to distinguish the relationship among stimulation-human and human-respond by meta analysis. Because there is none of previous works available as foundation for tactical activity of visual merchandising, it is necessary to develop the evaluation tools of display profile characteristics towards IB. Therefore, the research questions tried to be answered in this research are: (1) What variables are affecting IB behavior on the COC?; (2) What tools can be used to measure the effectiveness of COC in terms of IB?

    This research would completes Vieira (2013) and Chandon et al. (2009) in measuring the effectiveness of display profile and closing the gap of IB from retailer’s point of view as revealed in Amos et al. (2014), thus this article would proposes the basic approach on creating measurement tools for product display with a mathematical model. Using the model, supermarkets would be able to assess their product display in an objective, cheaper, less time consuming and easy manner.

    This article is divided into 5 major sections. In the first section readers could expect the introduction and conceptual models. Second section is presenting the research methods and importance findings. Section 3 reveals discussion and the managerial implication of this research, while in the last section reader will find conclusion and future potential research.


    Framework of this research is made based on consumer’s decision making process from the retailer point of view as depicted in Figure 1. Concept model to be tested in Figure 1 is inspired by Belk’s concept of SOR, Stimulus – Organism – Respond paradigm for modeling consumer behavior (Belk, 1975) which constructs all objects required to make a buying decision making. In SOR model, environment stimulates are required by inner state of organism to make a particular buying response.

    The Stimulus is determined by situation stimulus (environmental) and object stimulus (product characteristics). Wakefield et al. (2007), Geetha et al. (2009), and Chang et al. (2011) found out that design characteristics of supermarket would create impulsive behaviors, thus, a retailer might manipulate store stimuli to create a particular emotional responses (Vieira, 2013) which leads to buying decision. In this research, the Stimuli are focused on two different type of variables: Decision Variable and Noise Variable. Display profiles factors denote as Decision Variables because these factors are supposed to be managed and manipulated by the retailer. They are coming from display parameter which is developed and scaled to ease quantitative measurement: space width (SW), level (LVL), complexity (COM) and direction (DIR) (Chandon, 2009; Tafesse and Korneliussen, 2012; Kacen et al., 2012; Amos et al., 2014). SW is the length of display per product category. LVL is the place or which rack level a particular product group is displayed in (LVL). COM is complexity of product mix in a certain product category in a certain POP. DIR refers to the way the product is placed upon the consumer’s sight facing (DIR).

    Each display parameter is determined to catch actual condition in the simplest way. Each category displayed in the store holds a single display profile regardless the number of SKU and or COC embedded in the particular category. This method will make a better prediction for overall performance of COC. Partitioning product character by its product category had been proven to be better predictor (Jones et al., 2003).

    This article is introducing the queuing time as Noise Variable. Queuing time (TM) represents time needed by consumer to browse the display in COC. Although TM is not possible to be managed and manipulated by supermarket, it is nearly impossible to be controlled at the IB and visual merchandising system. It can only be controlled at a wider operational supermarket system, such as by scheduling operation of workstation or discount setting. As an example, an addition of new operating station might decrease overall stations’ queuing. In other hand, discount would attract consumers to shop thus the queuing might be increasing.

    Organism in the original concept is stated as a person who builds emotional respond as base of the Respond (Belk, 1975). However, this article avoids perceptions and emotional variables in developing the model which is very difficult to predict. For that reason, we made adjustment by making an allowance for the decision making process as black-box process as suggested by Gutterez (2004). That black-box system decides the IB response. Here, the Organism behaves as input of the black-box system which represents supermarket’s different demographic segments and target market. By linking the organism with demographic target market and segments, the supermarket may develop specific strategy and may understand their consumer behavior better. Although, Mihic and Kursan (2010) found in Croatia that IB is not significantly affected by gender, age, income; the similar study shall be investigated for Indonesia market. It is assumed that in order to pursue IB, consumer shall have good buying capacity, therefore the higher the segment market (higher SP, BT, and FR) the higher IB would be.

    The IB as response suits with Chang et al. (2011). The binary scaling approach is used to ease actual condition measurement. A respond or an IB action is fit if determined in binary variable to eliminate perception which may occurs in other nominal scale. By having binary variable, it is clear that if in the latest purchase, a consumer underwent with an impulsive buying, then they were scored 1, or else they were scored 0.

    Hypothesizes developed based on literatures are:

    • H1:  There will be positive impacts of IB from positive arrangement of decision variables and noise variables)

    • H2:  There will be positive impacts of IB from higher market segment of supermarkets

    Scaling system for each variable is following the incumbent condition on Indonesia’s common supermarkets. CA, SP, BT, and FR scaling are based on the target markets’ conditions. TM scaling is based on the supermarket’s target. The scaling of each display variables is based on the common COC’s fixture in Indonesia which is implemented in studied supermarkets. Readers may take a look on Appendix A for detail scaling for each variable.


    Data is taken from 5 supermarkets that belong to a biggest sales and oldest local retail group in Yogyakarta which is targeting low-middle consumers. This retail group has highest visitor per store compared to other supermarkets. Amongst 5 supermarkets, only one supermarket is located in downtown. The other 4 supermarkets are spread over sub-urban area around Yogyakarta. The downtown located supermarket has the biggest sales proportion as well as visitor proportion amongst others. In the other hands, Yogyakarta is chosen because this city has the biggest potential market for modern retail in Indonesia. The increase of market growth for Yogyakarta’s modern retail had improved by 52% in 2007-2009 (compared to 10-15% of Indonesia average) (Indonesia Commercial Newsletter, 2011). This growth is supported by local players as well as by national (Hypermart, Superindo) and international brands (Giants, Carrefour).

    Primary data (person characteristic and respond) was gained from consumer in-depth interview and observation. A total of 55 display profiles of 11 product categories were observed. Interview is done for 110 random consumers of the 5 supermarkets. Sampling methods used is purposive quota sampling without neglecting consumer’s proportion amongst all 5 stores. Purposive quota sampling is done because there were many limitations in consumers’ willingness to fill the data. Hence, there were limitations on generalizing the result but, however, this method ensures all the subgroups are represented satisfactorily and suits for exploratory research (Sekaran and Bougie, 2013).

    The interview had been done in 2 hours each day for 3 months. Due to technical limitation, each day was covering one supermarket only. This interview was not compromising the framework resulted by literature review.

    Right after settling their shopping, respondents were asked about their latest buying behavior, their queuing behavior in COC and how they choose COC counter. The categorizing will be good differentiation to make average prediction of the overall display performances. The product groups are categorized based on Schreiber (2002).

    Primary data is then modeled using a binary logistic regression. This regression is used to develop prediction of a binary dependent variable from non-binary independent variables (Hair et al., 2010). This particular regression may fit to both parametric and non-parametric data.

    Inside the medium sized supermarkets, more than one workstations or COC are available. Each supermarket has unique product display characteristic. Inside each of the supermarket, minimum 3-5 COC are operated low hours. A consumer couldn’t browse all products through all COCs, because she/he could only be served by one COC station at a shopping time. Therefore all COC in a supermarket is considered as an aggregate therefore it gives the idea or assumptions of equal chances of all COC and its display product in a supermarket to be chosen by consumer.

    A total of 11 product categories were observed in this article. Each supermarket has different product mix for each category even though not all product categories were displayed in all supermarkets. All observed product category were: vitamin, candy, chocolates, battery, razor, cigarettes, dry snack, RTD, drugs, fresh bread, and ice cream. Ninety two percent of the products displayed in COC were recognized as TOP Brand based on Top Brand Awards, 2014, only 6% of the products were new products developed, and the other 2% are neither. Top Brand product scattered evenly in each product category.

    Each category product displayed in COC was having different respond of impulsiveness. Table 1 shows the proportion of consumer impulsive buying to particular product category from a supermarket’s COC. The highest respond are happen to be vitamin (12.5%). Meanwhile razor and cigarette shared the lowest respond (1.3%). However, 45% of people who passed COC were buying something during their check out time. Indulgence product- such as candies, chocolate, dried snack, RTD, bread ice cream, and cigarettes; has higher proportion than health and body care-such as vitamin, razor, drugs. Thus fulfilling hedonic need still overcomes other motivations of impulsive buying in consumer goods retail (Beatty and Ferrel, 1998).

    As introduced in Figure 1, there are 4 decision variables (DIR, COM, LVL, SW), 1 noise variable (TM) and 5 uncontrollable input variables (CA, SP, BT, FR). There are 3 scenario proposed in this article each to find fittest yet simplest model to represent reality. Scenario 1 introduces a focus on decision variables. Thus, in scenario 1, we only test H1. By using scenario 1, supermarket managerial staffs could directly use the model neglecting the characteristics of consumer or supermarket’s target market as supported by Mihic and Kursan (2010). Scenario 2 involves all variables and tests H1 and H2. In the scenario 2, we propose to involve consumer characteristics that could be quantified easily by the supermarket into the model. Scenario 3 checks the multicollinearity amongst variables to simplify application of testing H1 and H2. The multicollinearity of each variable is checked to understand how a variable strongly correlated to another. If a variable’s correlation to another variable is above 0.30, we could consider both variables are strongly correlated, thus one chosen variable is enough to represent both variables in a model (Hair et al., 2008). As seen in Table 2, CA, SP, BT are strongly correlated therefore it can be represented by CA in scenario 3.

    Concordant value shows the level or proportion of ability a regression to fit their prediction, dis-concordant shows the proportion of model making wrong prediction, and ties shows the proportion of model making a blur prediction (Hosmer and Lemeshow, 2000). As shown in Table 3, scenario 3 has the maximum likelihood value, maximum P-value Pearson Goodness of Fit and 72.8% concordant. Meanwhile scenario 1 has 70.7% and scenario 2 has 72.7% concordant value. Then, all scenarios were validated to investigate the performance on predicting new data as seen in Table 4. The validation is performed by comparing the result of Hosmer and Lemeshow Goodness of fit test and R-square as suggested by Allison (2016).

    In Table 4, all scenarios showed high R-square score (above 0.97), where scenario 3 has the highest Goodness of fit test. Moreover, scenario 3 has the highest value of concordant (72.8%) and least log likelihood value (-149.89). Therefore scenario 3 is chosen to be used as a predictive model to assess the efficiency of the display profile.

    The concept model thus revised under equations:

    Logit ( π I B ) = 0.67 C A + 1.19 F R + 1.45 T M + 0.07 C O M + 0.46 L V L + 1.12 S W 0.13

    π I B = e ( 0.67 C A + 1.19 F R + 1.45 T M + 0.07 C O M + 0.46 L V L + 1.12 S W 0.13 ) 1 + e ( 0.67 C A + 1.19 F R + 1.45 T M + 0.07 C O M + 0.46 L V L + 1.12 S W 0.13 )


    • πIB =  IB decision (0, 1)

    • Logit(πIB ) =  probability of IB

    Odd ratio in logistic model measures the likelihood increment of respond variable over increment a level of a dependent variable (Hosmer and Lemeshow, 2000). In equation (1), TM in scenario 3 has the highest impact to IB (odd ratio = 1.45), while LVL has lowest impact (odd ratio = 0.46). It means that difference of TM: from not waiting to waiting will likely increase the probability of impulsive buying 1.45 times. Increasing percentage of SW, LVL, DIR and COM will increase impulsive buying 1.12, 0.46, 0.77, and 1.1 times. The impact of display profile apparently is not as high as the noise variables.

    The odd ratio of consumer characteristic variables: CA, SP, BT, and FR in Eq. (1) notifies us that FR (odd ratio: 1.19) is the most influencing consumer characteristic. Consumer that is more likely to come into the supermarket would increase the probability of impulsive buying. The more consumers trust the supermarket, the more consumers would come; and the more consumers would buy impulsively. Apparently, in-store marketing which is experienced by consumer in the past could create brand anchoring and the excitement which leads to trust and buying decision (Ariely, 2012; Solomon, 2011). “I kind of comfortable shopping in modern supermarket rather than traditional one, I put a trust on them and their goods”, is a sample of consumer comment that communicates consumer encouragement on the supermarket’s reputation which leads to IB behavior. Increment a level of FR will add 1.26 times of probability impulsive buying.

    Somehow, DIR is not significantly influenced IB behavior because in a small space of COC, consumer will turn their body to face cashier during payment. It means that the product will always be placed in most of consumer facing. In the incumbent object, most of the product category is placed based on consumer facing. Therefore, it isn’t any arguments to support significance of DIR on influencing IB.


    In Table 1, all 11 products category displayed in the COC were bought impulsively although each of the products has different probability. It is emphasizing the terms of impulsive as spontaneous decision (Beatty and Ferrel, 1998) which might happen to every product combination put in COC to every consumer passed COC. It is indicating the occurrence of irrational economic decision. The irrationality decision happens in an extremely short time when consumer has no proper time for analysis. The consumers can grab any product as long as easy to recognize a product and the products are able to fulfill their urge and sudden needs. “Well, I am kind of hungry. So I just grab and bought this bakery,” and “This seems so right to wipe my thirst with this drink,” are popular comments from consumer that argue their IB decision for fill their sudden needs.

    For a particular product display, as described in Eq. 1, display height (LVL) and product mix category (COM) will significantly influence IB behavior. Apparently the concept of impulsive buying has shifted from only spontaneous decision (Solomon, 2011) into more considering alternative selection inside product mix as impulsive decision. Amos et al. (2014) is pointing on the influence of product mix category (COM) to IB, which is related to psychological nature of variety seeking. The more variety or substitution appears in a product group, the easier a consumer makes a decision. In COC, while the urge to fulfill a desire tends to motivate the buying, there is still alternative selection of IB happens in a short period of time during waiting, where any options in certain product display will expedite buying decision making process (Ahmad, 2011; Ariely, 2012; Kahneman, 2011). In the case of IB, similar phenomenon appears not only in supermarket, but also other type of retails (Sharma et al., 2010). Thus, it is recommended to mix TOP brand with less priced substitution product or non-TOP Brand inside a display of a category product to accelerate the sales. Mixing TOP brand products with its substitution would create wider variety in terms of price.

    In fact, based on our observations, it appeared that only 5% product displayed in COC that was discounted or bundled product. Indeed, products placed in COC are usually targeted to middle-high segment market which is characterized as more expensive, aired regularly in national TVC, and exclusive packaging. Because of its strategic location, space charging in COC area is normally higher than other area. Therefore, only product with higher profit would be able to rent the space. High percentage of the IB respond, as in Table 1, indicates that consumers are willing to pay more to spoil their hedonic needs as long as they can recognize and hardhearted on the product and brand they bought (Ariely, 2012). Even more, based on our interview, we conclude that more than 90% consumers admit the satisfaction of buying impulsive goods because of product usefulness and urgent needs fulfillment.

    The other significant decision variable of IB in COC is height or level of display (LVL). LVL is influencing the ease of consumer’s sensory contact to identify certain product category. When consumer can identify the product, consumer most likely can touch it (make a physical contact). The physical contact can be direct stimulus for buying decision (Solomon, 2011). The closer the display to eye level, the higher the identification possibility that leads to the higher the IB probability would be (Amos et al., 2014). However, this is quite in contrast with the way certain group product display is seen as more predominant, which is reflected by SW variables. During the research, most product category had been placed as closer as possible to eye level; some categories are arranged vertically which seems effective in in-store marketing point of view (Chandon et al., 2009).

    The slower a product category is being sold, the closer the product shall be placed towards consumer’s eye level to ensure the visibility, so that the buying probability is enlarged for slow moving products, such as: razor and battery.

    COC denotes as a dilemmatic multifunction fixture inside supermarket. As a POP fixture and a main part of grid supermarket layout, COC is created as interesting as possible to attract consumer to browse. The browsing activity will create purchase transactions (Solomon, 2011). Meanwhile, as main work station, COC must serve the consumer in shortest processing time and minimize queue waiting time. As depicted in Figure 2, almost half of the respondents are not interested to browse products displayed in COC. Even more, 60% of the respondents choose COC from its short queue, meaning the shortest browsing time. Some supporting comments to rationalize that decision are such as, “Oh, if I need a particular product (displayed in COC), I would just take it from the shelf and would still lining to the shortest one,” and, “I have already had all my needs from normal rack, and would like to finish the transaction as fast as possible.” This phenomenon shows that spontaneous of thinking that is triggered by intuition has dominated decision making process during IB (Sharma et al., 2010; Kahneman, 2011). This phenomenon might also trigger IB since the time pressure given during waiting (Amos et al., 2014). Consumer is pushed to make a short decision about what they have browsed before cashier is finishing his/her transaction which will leads to the irrational decisions (Ariely, 2012; Kahneman, 2011).

    Since most of the displayed products in COC were TOP Brand product (Top Brand Index, 2015) as mentioned earlier, there is an intuition that resembles in consumer minds about its performance and quality. The anchoring of trust will empower irrationality and trigger spontaneous buying. Such comments of, “I have no doubt about this product (quality) though haven’t yet consume it, (I have) seen the advertising a lot in TV,” appears most time when a respondent is being asked about their surety of the quality that is impulsively bought. In the meantime, the product displayed in COC may be classified as daily needs commodity. Grocery TOP Brand product characteristics may cause a high tendency of overconfidence buying decision making (Kahneman, 2011) which was becoming a stronger motivation to the respondents to impulsively buy. Consumers have high confident to their decision that the impulsive product bought will be useful and deliver satisfaction someday if not used at a certain time. “Yes, this is my favorite brand. Although I still have some little of this (in my house), I think it is OK to buy it now, while I am here.”

    Since impulsive behavior in modern retail less depends on the internal consumer condition but more influenced by stimulus display condition as well as browsing time, IB had been re-conceptualized into not only unplanned and spontaneous buying but also fulfillment of hedonic motivation, desire satisfaction, self-esteem fulfillment, based on past experience (Rook, 1987; Hausman, 2000) and closely related to alternative and variance selection (Sharma et al., 2010).

    Based on Eq. 1, retail management is recommended to add product alternatives inside a product category’s display profile. That will add the score of COM and SW at once. Primary product and or slow moving product should be placed closer to eye level, of course. Retail management also needs to allocate more space in COC for any product that has instant ability to fulfill consumer urge needs. Ready to use product; such as: canned drink, chocolate, ice cream, candies and bread, shall be best to fit this requirement. While trying to improve the goods exposure time by setting the queuing in COC, retail management should be able to maintain its best consumer satisfaction in queuing. Opening only a few COC which displayed particular targeted product category during low time will be able to maintain product exposure and consumer satisfaction as long as they don’t wait too long.

    To use the propose equation in tactical activity of supermarket, management shall apply it after drafting the product display of COC in their planogram. Thus they can assess the planogram performances toward IB and make adjustment based on the assessment result. That assessment activity can be repeated until management satisfied on the result. Whenever it is necessary, the assessment can be start by evaluating current planogram.

    Besides of its dilemmatic traits, COC is the only place where the buying consumers are able to exit supermarket. To create a better purchase performance of a planogram, supermarket may prioritize to offer the rental space to TOP brand and medium-high brand.


    As a store, supermarket can increase it sales by doing in-store marketing which targeting the IB behavior. The in-store marketing can be done through properly arranged product display. This article had proposed to examine the effectiveness of a product display towards IB in COC fixtures through mathematical model (Eq. 1 and 2). The equations using three decision display variables in namely: level of display (LVL), complexity of product mix (COM) and space width of display (SW). Those variables are observable and can be qualitatively measured in a unit of product category, so managers can collect the input data easily. Based on the equation, the magnitude of SW, which provides highest impacts, is closely related to COM. Furthermore, retail managers need to arrange waiting or queuing time (TM) which allows consumer to browse products in selected fixture. Proposed equations had shown it sufficiency to argue on the consumer behavior and motivational aspect of IB.

    On the product characteristics, ready to use (RTU) products and TOP Brand product which is indulgencing, or health and beauty caring will give stronger stimulus to IB behavior because of the irrationality during IB decision making. The irrationality is triggered by alternative selections and the inability to recognize of negative risk in IB.

    Given the noise variable TM and SW as presentation of total width of display aggregate, it is possible to perform simulation and variable optimization in near future for obtaining maximum impulsive purchases for each product category. More detail research about product characteristic and pricing that contributes to IB may also be performed in the near future.

    As conclusion, this research had proposed a new approach on assessing a display profile for consumer’s impulsive buying behavior. Its practical application on business activities might support retail’s managerial level to develop its strategic decision on product display. Meanwhile, its theoretical application may close the research gap on finding the assessment tools on impulsiveness of a visual display by measurable parameter.


    Corresponding author would like to thank to Dr. Carles Sitompul.



    Research conceptual model.


    Dilemmatic diagram of COC.


    Impulsive Proportion for each COC

    Correlation score

    Logistic regression result

    Validity of logistic regression

    Detail Variables


    1. Ahmad T (2011) The impulse buying behavior of consumer for the FMCG products in jodhpur , Australian Journal of Basic & Applied Sciences, Vol.5 (11) ; pp.1704-1710
    2. Allison P D (2016) Measures of Fit for Logistics Regression, Available from:
    3. Amos C , Holmes G R , Keneson W C (2014) Ameta analysis of consumer impulse buying , Journal of Retailing and Consumer Services, Vol.21 (2) ; pp.86-97
    4. Ariely D (2012) Predictably Irrational: The Hidden Forces That Shape Our Decision, Harper Colins,
    5. Beatty S M , Ferrel M E (1998) Impulse buying: Modeling its precursors , Journal of Retailing, Vol.74 (2) ; pp.169-191
    6. Belk R W (1975) Situational variables and consumer behavior , Journal of Consumer Research, Vol.2 (3) ; pp.157-164
    7. Chandon P , Hutchinson J W , Bradlow E T , Young S H (2009) Does in store marketing work?Effects of the number and position of shelf facings and brand attention and evaluation at the point of purchase , Journal of Marketing, Vol.73 (6) ; pp.1-17
    8. Chang H J , Eckman M , Yan R N (2011) Application of the stimulus-organism-response model to the retail environment: The role of hedonic motivation in impulsive buying behavior , The International Review of Retail, Distribution and Consumer Research, Vol.21 (3) ; pp.233-249
    9. Geetha M , Shivakumaran B , Sharama P (2009) Role store image in consumer impulse buying behavior , Advance in Consumer Research, Vol.7 ; pp.193-194
    10. Gutterez B P B (2004) Determinants of planned and impulse buying: The case of the Philippines , AsiaPacific Management Review, Vol.9 (6) ; pp.1061-1078
    11. Hair J F Jr, Black W C , Babin B J , Anderson R E (2010) Multivariate Data Analysis, Pearson Prentice Hall,
    12. Hausman A (2000) A multi-method investigation of consumer motivations in impulse buying behavior , Journal of Consumer Marketing, Vol.17 (5) ; pp.403-419
    13. Herabadi A G , Verplanken B , Van Knippenberg A (2009) Consumption experience of impulse buying in Indonesia: Emotional arousal and hedonic considerations , Asian Journal of Social Psychology, Vol.12 (1) ; pp.20-31
    14. Hosmer D W , Lemeshow S (2000) Applied Logistic Regression, John Wiley & Son,
    15. (2011) Available from:
    16. Jones M A , Reynolds K E , Weun S , Beatty S E (2003) The product-specific nature of impulsive buying tendency , Journal of Business Research, Vol.56 (7) ; pp.505-517
    17. Kacen J J , Hess J D , Walker D (2012) Spontaneous selection: The influence of product and retailing factors on consumer impulse purchases , Journal of Retailing and Consumer Services, Vol.19 (6) ; pp.578-588
    18. Kahneman D (2011) Thinking, Fast and Slow, Gramedia Pustaka,
    19. Karbasivar A , Yarahmadi H (2011) Evaluating effective factors on consumer impulse buying behavior , Asian Journal of Business Management Studies, Vol.2 (4) ; pp.174-181
    20. Levy M , Weitz B A (2012) Retailing Management, Mc. Graw-Hill,
    21. Liu S , Melara R , Arangarasan R (2007) The effect of store layout on consumer buying behavioural parameter with visual technology , Journal of Shopping Centre Research, Vol.14 (2) ; pp.63-72
    22. Madhavaram S R , Laverie D A (2004) Exploring impulse purchasing on internet , Advances in Consumer Research, Vol.31 ; pp.59-66
    23. Mihic M , Kursan I (2010) Assessing the situational factors and impulsive buying behavior: Market segmentation approach , Management, Vol.15 (2) ; pp.47-66
    24. Nielsen (2015) The Future of Grocery: E-commerce, Digital Technology, and Changing Shopping Preferences Around the World , Available from:,d.dGo
    25. (2015) 2015-16 Outlook for The Retail and Consumer Products Sector in Asia , Available from:
    26. Rook D W (1987) The buying impulse , The Journal of Consumer Research, Vol.14 (2) ; pp.189-199
    27. Rook D W , Fisher R J (1995) Normative influences on impulsive buying behavior , The Journal of Consumer Research, Vol.22 (3) ; pp.305-313
    28. Samson H E , Lite W G (1993) Retail Merchandising: Consumer Goods and Services, South-Western Publishing, Co,
    29. Schreiber N (2002) Planning for impulse , 16th Annual PBAA Convention,
    30. Sekaran U , Bougie R (2013) Research Methods for Business: A Skill-Building Approach 6th edition, John Wiley & Sons Ltd,
    31. Sharma P , Sivakumaran B , Marshal R (2010) Impulse buying and variety seeking: A trait-correlates perspective , Journal of Business Research, Vol.63 (3) ; pp.276-283
    32. Solomon M R (2011) Consumer Behavior: Buying, Having and Being 9th editions, Pearson Education Inc,
    33. Stilley K M , Inman J J , Wakefield K L (2010) Planning to make unplanned purchases? The role of in-store slack in budget deviation , Journal of Consumer Research Inc, Vol.37 (2) ; pp.264-278
    34. Tafesse W , Korneliussen T (2012) Identifying factors affecting consumers purchase incidence at retail trade shows , Journal of Retailing and Consumer Services, Vol.19 (4) ; pp.438-444
    35. (2015) Available from:
    36. Vieira V A (2013) Stimuli-organism-response framework: A meta-analytic review in the store environment , Journal of Business Research, Vol.66 (9) ; pp.1420-1426
    37. Wakefield M , Germain D , Henriksen L (2007) The effect of retail cigarette pack display on impulse purchase , Society for Study of Addiction, Vol.103 (2) ; pp.322-328