A case relating the effects of changes to a retail environment for customer satisfaction is described. Unexpectedly, satisfaction was found to decrease after changes were implemented, although the modifications were made in response to customer complaints. Analysis of this case identifies reduced crowding as the most probable rationale for diminished customer satisfaction. Crowding is seen to have advantages in some situations, and can facilitate positive customer-customer and customer-employee interactions. A discussion of the effects of crowding and how to manage crowding as a beneficial environmental attribute is relevant to retailers.
CUSTOMER RESPONSE TO RETAIL ENVIRONMENTS
The level of satisfaction claimed by a firm's customers is related to its profitability and ability to retain customers (LaBarbera & Mazursky, 1983; Anderson & Sullivan, 1993; Anderson, Fornell, & Lehmann, 1994; Rust, Zahorik, & Keiningham, 1995). These relationships provide a rationale for the marketing concept, which councils that satisfying customer needs and wants is a desirable strategy that should guide a firm's actions. Firms exhibiting a marketing orientation, or propensity to apply the marketing concept, have been shown empirically to perform better than those that do not (Kohli & Jaworski, 1990; Narver & Slater, 1990; Kohli, Jaworski, & Kumar, 1993). The benefits of satisfying customer needs and wants are therefore apparent, and managers should aim to achieve this objective.
Unfortunately, it is not always easy to understand what customers need or want. Direct communication with customers is an obvious solution, and is one of the primary benefits of market research. However, customers may not know what they want, or customers may not really want what they think they want. Reading between the lines or attempting to understand customers better than they understand themselves is one of the goals of marketing research that is not always realized.
The business literature is replete with examples of firms or individuals that-as a consequence of market research-make changes to the business or product, only to obtain a poor outcome. Probably the best known examples come from new product development studies, such as Coca-Cola's "New" Coke formula that, despite extensive market research, proved to be unpopular in most markets. Similarly, the Ford Edsel failed dismally despite a tremendous amount of marketing research that predicted success. Examples come from other areas, such as university classrooms. Brunner (1997) recounts her experiences with modifying course syllabi in response to comments on students' evaluation forms and complains that her teaching ratings fell after making the alterations. A further case of the problem of failed reactions to customer input is the focus of this article.
After listening to its customers and making several significant changes, a retail firm was surprised to find that its customers were less satisfied with the firm than before the changes were made. In attempting to understand the rationale for the decreased customer satisfaction, the authors searched the marketing literature for explanations. Specifically, literature related to the retail environment was studied, and it became obvious that no comprehensive theory or model exists that could explain the customers' response to the changes implemented by the firm.
This article presents what is effectively a case study, yet the analysis is interwoven with a degree of theoretical development. The firm introduced above is described both before and after modifications to its physical environment were made, and the measurement of customer satisfaction prior to and subsequent to the changes is also explained. Measurement results were surprising both to the firm and the researchers, and attempts at explanation led to the search for a general model of retail environments. The most valuable model found comes from the services literature. Bitner's (1992) Servicescape model is explored, as are its implications for entrepreneurs and managers who design the physical retailing environment in retail settings. Retail crowding is the environmental factor selected by the authors as being the most useful for explaining the measurement results, and this topic is discussed with respect to ways in which perceptions of crowding can be managed.
THE ORIGINAL PROJECT
This project began when a small retail firm's manager met with the authors to discuss customer satisfaction. The focal organization (hereafter referred to as "Wines Plus") specializes in the sale of wine in a large western Canadian city. The firm has a single retail outlet located in a busy strip mall. The original store was approximately 2400 square feet, which included a small area for employees, a bathroom, a small office, a counter and till, and a raised platform for seating customers during wine tasting promotions. The remainder of the floor space was devoted primarily to the display of wines, with a small amount of space for beers and liquors. Unfortunately, and perhaps because the firm was quite successful, the primary floor area was rather crowded, even to the point that it was difficult to browse if there were more than several customers in the store at one time. Many of Wines Plus' patrons complained about the crowding to the staff and manager. It was clear that the firm would need to make some changes.
In response to the obvious crowding and associated customer complaints, Wines Plus made some important decisions. First, the owners and management chose to move to a larger retail space. The new location is across the parking lot from the former one (roughly 200 feet away), and has approximately double the previous outlet's floor area. Most of the floor space is dedicated to displays and shelves, but room is set aside for a small office, employee rest area, bathroom, and a shipping/receiving zone. There is even space for a small stockroom in the new store. Shelves are now in a conventional grid layout, instead of the boutique layout or casual network of shelves and displays that existed previously. Additional changes include an increase in selection and a more vibrant color scheme. The name of the company was also changed to one that reflected a broadening of product offerings. Although wine remains the primary product sold, there is now a slightly increased emphasis on beer and liquor.
The move and name change were anticipated, but some other changes were not. As the move was being planned and the new location made ready, the store temporarily lost its license to sell wine, beer or liquor for a period of nearly two months (a critical necessity for a firm selling only these products!). For this the previous manager resigned from her position, and a new manager was installed. In some respects the timing for these negative incidents was acceptable because the move required a great deal of extra work by management and employees. Fortunately, the closure did not threaten the survival of the firm.
The proposed changes created an exciting opportunity for the authors, who wanted to evaluate customer satisfaction and expectations before and after the modifications. The hypothesis developed at the time was that customers would originally be very happy with the new store, but over time their satisfaction would regress to the previous levels. How the hypothesis was developed is the focus of another paper, though it is clear that a longitudinal research design was required.
What followed was a very time consuming and expensive data collection process. An instrument capable of measuring customer satisfaction accurately and reliably was developed, and administered several times to the same group of respondents. The sample was provided by Wines Plus' manager, who was given instructions to choose 500 customers at random from the firm's customer files. The representativeness of the final sample cannot be ascertained, but respondents gave a broad range of responses. Thus, it is assumed that the sample was representative of all frequent Wines Plus customers at that time.
At the same time the sample of customers was selected, the questionnaire used to assess customer satisfaction was carefully developed. There exists a large body of literature concerning the measurement of customer satisfaction, and the information contained therein was put to use for the survey of Wines Plus' customers. Tse and Wilton (1988) appears to best represent the research associated with satisfaction measurement, and this article suggests that expectations, perceptions of performance, and disconfirmation can all contribute to the explanation of satisfaction. Multi-attribute measures of these three constructs were included in the questionnaire and covered such facets as product selection, price, store displays and atmosphere, and staff expertise and helpfulness. Experience has shown that summary questions can be useful as concurrent validity checks, so this type of question was employed also. Seven-point scales were employed except for the first two summary questions, which employed nine-point scales.
The questionnaire just described was intended to address the academic issues of interest to the researchers. However, Wines Plus' management and owners had other questions, and a second questionnaire was developed to address these issues. This form asked practical questions such as customers' newspaper reading habits, factors affecting wine buying decisions, and the convenience of store hours. The second questionnaire was distributed just once, but the first was employed in a longitudinal study and distributed four times to the same customers.
The dates that questionnaires were sent out is as follows: the first wave was mailed out at the beginning of March 1995; the second in late April 1995; the third in late May 1996; and the fourth in the middle of October 1996. Questionnaires were initially mailed to a sample of 488 customers in the first wave of data collection. Usable responses were received in a timely fashion from 245 of this group and twenty were undeliverable, so in the first wave 245 questionnaires were returned out of a possible 468 for a 52.4% response rate. This figure is considerably higher than is typically experienced with mail surveys, but several procedures and factors are believed to be responsible for the high response rate. First, the survey method was modeled after Dillman's (1978) Total Design method, which suggests the utilization of a letter of introduction to carefully explain the importance of the research and why it is beneficial to both the academic researchers and Wines Plus. The letter was addressed to individual customers and signed by the store's manager and the lead researcher.
Another factor for the high response rate in the first wave of data collection is that once the return of surveys slowed-a few weeks after its initial mailing-a reminder note was mailed to those individuals that had not returned their surveys (as recommended by Dillman (1978)). This procedure was followed for all four waves of data collection. Surveys were plainly numbered for this reason, and also because potential respondents were offered further incentives to respond to each and every wave of data collection. Coupons with a value of $1 were inserted with questionnaires at every wave as a token of the management and researchers' appreciation. Furthermore, the individuals that responded to all four waves of the study were entered into a drawing for a $250 gift certificate.
Unfortunately, the number of respondents decreased over time, as predicted by Wall and Williams (1970). In the second wave-conducted approximately seven weeks after the first-182 of the original 245 respondents returned questionnaires. The measures thus far indicated a consistent level of satisfaction with the original store across the two data collection periods. After this time, Wines Plus made the significant changes described earlier. Once the store reopened, the research was resumed. The third wave of data collection saw 92 customers respond, but this was approximately one full year after the second wave. The final wave experienced a further drop in response, with 45 complete questionnaires. Some initial respondents moved or stopped purchasing at Wines Plus. Others were presumably lost when the store did not remain open for business after temporarily losing its license. Another portion can be attributed to a loss of interest due to the long period between measurement waves, and fatigue is likely to be a further factor.
Wall and Williams (1970) identify several difficulties with longitudinal studies that can be summarized as follows:
1) Response rates will probably decrease with time.
2) Those respondents willing to endure a longitudinal study may not be representative of the population as a whole. Therefore, generalization can be suspect.
3) Continued study of individuals may modify their behavior in some way.
4) Analysis of data collected prior to study completion may cause revisions in study goals.
5) Because longitudinal studies generally require more time to complete, evidence derived from such studies is necessarily delayed. This can affect standing within academic or other circles, creating a preference for studies that produce results quickly.
6) Due to reliability issues, longitudinal studies may focus on what is easily and accurately measurable. For example, it is easier to measure changes in height than changes in IQ, especially if there are different observers administering a survey from one year to the next.
Wall and Williams (1970) believe the first difficulty is the most problematic, and that proved to be the case with this study. Nonetheless, although significant reductions in responses occurred, all signs consistently indicated decreased satisfaction with the new store. It is therefore assumed that the responses are a valid measure of the original sample's opinions over time.
The values in Table 1 and Table 2 represent the mean responses to four of the summary measures for two different samples. The first question asks if Wines Plus offers the best value and service when compared to competitors. The second question asks customers their degree of overall satisfaction with Wines Plus, and the third asks how respondents feel about their overall experiences with the firm. The fourth question simply asks if the respondent would recommend Wines Plus to a friend or acquaintance; the mean figures in Table 1 and Table 2 related to this question represent the proportion of customers that would recommend Wines Plus.
The figures in Table 1 demonstrate that for those individuals who responded to questionnaires in all four periods, changes took place between the second and third waves of data collection. Every one of the differences in means is significantly different across the second and third periods (P < 0.05). Differences between waves 1 and 2, and waves 3 and 4 are not significant. The pattern of responses is remarkably stable. When all respondents from the four periods are considered the results are almost identical. Table 2 repeats the differences seen in Table 1, though differences are slightly more significant because of the larger sample sizes involved (P < 0.01).
It is apparent from Table 1 and Table 2 that the response means hardly varied across the first two data collection periods, but there was a drop in overall satisfaction from the second to third waves. Because figures from the fourth period mirror those of the third, the decrease in satisfaction seems to be enduring. The obvious question is: Why did satisfaction decrease after the move to the new store when the move was undertaken largely in response to customer complaints of crowding in the old location?
The answer to this question is complicated by the many simultaneous changes that occurred. The change in managers and the temporary closing could have turned off some loyal customers. On the other hand, the new store's physical environment should have improved customer satisfaction because it is larger and offers less crowding and better product selection. Memory effects are assumed to be minimal because of the fairly lengthy period between all measurements. Green, Tull, and Albaum (1988) recommend a minimum of two weeks between measurements to avoid memory effects, and there is always more than six weeks between measures in the current study.
Further analysis of the data was performed in an attempt to identify the source of the differences across the second and third periods. One of the questions in the instrument asks respondents to indicate their level of satisfaction with a series of fourteen firm characteristics. Table 3 lists the characteristics and the average difference between associated scores obtained from the second and third waves of data collection. Negative values indicate increased satisfaction across periods. There are just two attributes about which respondents are more satisfied: these are satisfaction with the selection of beer and liquor and satisfaction with the complimentary home delivery. Respondents were, on average, less satisfied with the remaining store features.
Regressing the changes in measures of overall satisfaction on the differences of the fourteen characteristics across the two periods brought mixed results. The stepwise method for variable selection was employed simply to identify the most important variables. This exercise finds that only differences related to the helpful staff attribute are significant in explaining variance in changes to responses to the first summary question (best value and service). Selection of wine and store atmosphere are the only variables significantly related to changes in the degree of overall satisfaction with Wines Plus. Stock availability is added to the latter characteristics as one of the significant explanations of changes in feelings about the overall experience. Logistic regression is employed for the dichotomous recommendto-a-friend variable and finds that only the helpfulness of staff explains a significant portion of the variance in differences for that variable. On the whole there is no single variable identified as being important for explaining the behavior of all four summary measures, which implies that the fourteen characteristics-and perhaps others-combine to be responsible for the reduced satisfaction revealed by Wines Plus customers.
The stage is now set for the analysis of the case addressed by this paper. Clearly, there is a marked reduction in satisfaction with Wines Plus after the changes that were made to it. This observation is supported by decreases in satisfaction scores associated with Wines Plus' characteristics. Why would decreases in satisfaction occur, especially when alterations to the business were initiated in response to customer complaints regarding crowding? It is assumed that something in the retail environment was responsible for the reduced satisfaction. Therefore, the next section reviews literature related to the retail environment.
TOPICS IN RETAIL ENVIRONMENT
It has long been recognized that the environment of a retail setting can affect customers (Kotler, 1974). However, a search of the marketing literature indicates that a general model of the retail environment does not exist:
The effect of atmospherics, or physical design and decor elements, on consumers and workers is recognized by managers and mentioned in virtually all marketing, retailing, and organizational behavior texts. Yet, particularly in marketing, there is a surprising lack of empirical research or theoretically based frameworks addressing the role of physical surroundings in consumption settings. Managers continually plan, build, change, and control an organization's physical surroundings, but frequently the impact of a specific design or design change on ultimate users of the facility is not fully understood (Bitner, 1992, p. 57).
In response to this shortcoming, Bitner (1992) develops her Servicescapes model, which conceptually maps the service environment's effects on customers, employees, and customer-employee interactions. Other than Bitner's model, a general framework for studying the effect of the retail environment on customers could not be found. However, studies of specific environmental attributes are not uncommon and include such factors as scents (Spangenberg, Crowley, & Henderson, 1996), music (Milliman, 1982, 1986), and crowding (Harrell, Hurt, & Anderson, 1980; Hui & Bateson, 1991). Such attributes are summarized by Baker, Grewal, and Parasuraman (1994). Sherry (1998) edits an entire volume devoted to various aspects of servicescapes and environments that further demonstrate their importance.
Although the Servicescapes model is directed at service environments, it is employed by this case study to examine potential explanations for the response to changes in the retail environment made by Wines Plus. The Servicescape model is directly relevant because the job of retailing is itself a service (e.g., Berry, 1986). Bitner's model focuses on three dimensions of the service environment, which are: (1) ambient conditions such as temperature, air quality, noise, music, odor, lighting, and colors; (2) spatial layout and functionality, defined as the arrangement and design of the physical environment, including floor space, furniture, and other equipment; and (3) signs, symbols, and artifacts, which communicate impressions of image and quality, plus expectations of behavior. The latter dimension also includes customers' reactions to employee behaviors and appearance.
All of these dimensions and the specific attributes may affect employees, customers, and the interaction between the two groups, and their implementation will largely depend on the type of service and its objectives in meeting customer needs. Bitner (1992) delineates the effects of service environments as customer and employee cognitive, emotional, and physiological responses that prompt customer and employee approach or avoidance behaviors. For example, in some organizations creating increased interaction between service employees and customers is a positive response, but in other firms this response may not be desirable. A further example is the amount of time a customer will spend browsing in a store.
From examining the various dimensions and their specific attributes it appears that the Servicescape model can be extended to all retail settings, although services retail outlets can be viewed as a subset of all retailers. Service retailing involves the sale of an intangible product, like car rentals. Some aspects of retailing are more important for services than for typical goods retailers. One example is that a high degree of customer-employee interaction tends to require highly trained employees, which cannot generally be said for frontline employees in a traditional retail setting. Kelly and George (1982) offer an inventory of aspects that show how services retailers differ from goods retailers. Most of the differences focus on the factors that differentiate services from goods: intangibility, inseparability, variability, and perishability (Zeithaml, Parasuraman, & Berry, 1985). Kelly and George (1982) identify store organization as being different across services and goods retailers, though differences here focus on employees.
Upon returning to Bitner's (1992) Servicescapes model it appears that the critical aspect for how goods retailing environments differ from service environments is simply in the product purchased by customers. A goods retailer's customers largely serve themselves by making choice decisions, selecting products, and transporting them to a checkout counter. This means that the display of products is important, and customers must be able to find and clearly view all products, physically transport them, and interact with one another in doing so. This procedure is true of most goods retailers, including Wines Plus.
APPLYING THE SERVICESCAPES MODEL TO WINES PLUS
What are the implications of Bitner's (1992) Servicescapes model for Wines Plus? None of the factors discussed above offer a straightforward explanation for why Wines Plus' customers' satisfaction decreased after the store's move. Several store characteristics changed during the move and temporary closure, including the physical location, the retail environment, the name of the store, and the manager. Because of their interaction with the store managers, employees, and customers, the researchers were in a unique position to directly observe the effect of these changes. Unfortunately, the alterations to the firm somehow combined to produce a response that-under the circumstances-would be difficult to predict.
All of the changes listed have the potential to create customer dissatisfaction, but in this particular situation none of them stands out as being responsible. As noted earlier, the physical location hardly changed. Parking space was at least as easy to obtain as before the move. Despite not being as well known to customers, the new manager appeared to the researchers to be more organized and better equipped to manage a larger store. Moreover, there were no complaints about the new manager in space provided for comments on the questionnaire.
The store's original name did not appear to be any better or worse than the new label; both were quite innocuous yet indicated the company's purpose and offerings. Again, no complaints about the name change were noted. The last change relates to the store's environment. Recall that store atmosphere was found to be a significant predictor of two of the summary questions listed in Table 1 and Table 2. Of all the various dimensions and specific attributes identified by Bitner (1992), the one that stands out most is crowding because it is the characteristic about which customers originally complained. Crowding led to the decision to make the significant changes to Wines Plus' retail space and is likely the only variable that-in this case-had the potential to create the sort of strong affective response that caused satisfaction to decrease so markedly.
There is limited research regarding the effects of crowding in a retail setting. Stokols (1972) distinguishes between the terms crowding and density. In a retail context, density can be measured as the number of shoppers per area of open floor space. Crowding, on the other hand, exists when the density is perceived by retail customers as excessive. Hui and Bateson (1991) determine that perceived control can mediate the relationship between density and perceived crowding. Reductions in perceived control are found to diminish pleasure with a service. However, increased customer density in a bank scenario is found to reduce perceived control but strengthens perceived control in a bar scenario. Harrell, Hutt, and Anderson (1980) suggest that shoppers will tend to employ adaptive strategies when crowded conditions are perceived. Such strategies may include reduced shopping time and conformity to traffic patterns. Harrell, Hutt, and Anderson find that strategies for adapting to perceived crowding mediate the relationship between crowding and attitudes toward the retail outlet. Eroglu and Harrell (1986) develop a model of retail crowding that leads to several propositions regarding the effects of perceived crowding under various conditions. In a subsequent study, Eroglu and Machleit (1990) empirically examine perceived risk and time pressure as antecedents of retail crowding and shopping satisfaction as a consequence. However, these researchers recognize the exploratory nature of their investigation and suggest that retail crowding remains an under-researched topic.
A search of the literature for more recent research in this area was not fruitful, and little is known about the effects of perceived crowding on retail customers under varying conditions. The terms crowding and crowded have negative connotations and might be predicted to create unfavorable feelings toward the retail outlet, but it may be that under certain conditions crowding will be viewed as a positive characteristic of retail environments. Eroglu and Harrell (1986) note that in some situations crowding can be a desirable feature, such as with sporting events and bars. This is precisely what is found by Hui and Bateson (1991).
As noted, Wines Plus originally had a rather crowded retail environment, which led to the decision to move to a larger retail space. Although the new store was very nice, it lacked much of the ambience afforded by the former setting. Crowding may have had much to do with this location's atmosphere. It created a cozy environment that appeared to make customers feel as if they were part of a community; this sensation is likely a positive attribute of sporting events and bars as well. Customers in crowded retail spaces typically must interact with one another and with employees to a much greater degree, and these interactions may be sought by individuals under particular circumstances. For example, Wines Plus' customers are able to discuss wines with one another or with store employees; such exchanges are facilitated by their closer proximity in a smaller store. Although coziness and crowding are two terms for perceptions of high shopper density within a retail setting, which one of the two is brought to the minds of customers is likely dependent on most or all of the factors that are already known to affect impressions of retail environments. Important factors include, among others, the type of product (good or service) sold, lighting, and customer and employee actions and communications. All of these features will be critical for turning a crowded environment into one that is comfortable and inviting.
The effects of crowding have been the focus of some research, but the circumstances under which crowding may be beneficial to a retail firm have not been established. This section attempts to identify the characteristics that define when crowded retail conditions may or may not be appropriate. Although the concepts presented here are not tested in this article, they can still provide business people and managers with ideas that can be considered in their own firm's context. The concepts described also suggest a series of tests that may provide the basis for future research.
Positive characteristics of crowding offered by the discussion thus far are the feeling of camaraderie, warmth, and coziness created by some bars, and the energy of a crowd at a sporting or other such venue. Crowding may also facilitate customercustomer and customer-employee interactions that can be helpful to all parties. Customers can obtain advice and recommendations from one another and from employees. Employees and managers can learn from customers and come to understand their preferences and complaints (Alicke et al., 1992). Personalized service from employees can significantly affect customer satisfaction (Surprenant & Solomon 1987; Mittal & Lassar 1996).
Harris, Baron, and Ratcliffe (1995) recommend that firms should actively manage interactions between customers. Pranter and Martin (1991) describe some roles that employees can use to encourage or discourage customer interaction. Interactions between customers can often increase satisfaction with a service experience (Grove & Fisk, 1997), but the reverse is also true (e.g., Alicke et al., 1992). For example, Grove and Fisk (1997, p.79) find ".. .a propensity to experience dissatisfying incidents when other customers were different in some identifiable way." Thus, when specialized goods and services are retailed it is important for a company to ensure that marketing tools create a consistent image for positioning and targeting purposes. This practice should minimize dissatisfaction arising from inappropriate customer expectations or interactions with other customers.
Crowding may be seen as a means of facilitating customer interaction because it is easier for customers to communicate when they are in close proximity, particularly when their presence in a specialty store indicates a common interest. If encouraging customer interaction can be of benefit to the business, crowding is one method of inducing such contact. This is possibly a strong component of the satisfaction revealed by Wines Plus' customers, because customer-customer and employee-customer interactions were facilitated by the crowding in the former location. The new location is much more spacious, and its grid layout does not encourage communication to the same degree.
Other constructive aspects of crowding in a retail environment exist. Crowds often signal success, popularity, and good values. If a retail store is crowded, it may be assumed that the store is successful, and/or its customers believe that the store offers good value. Crowds may also indicate how the firm is positioned in the market. With many retailers crowding can be indicative of a successful low-end store (e.g., Wal-Mart). A high-end store would probably not want the same customer density as Wal-Mart, and high-end customers are likely to be unhappy in such surroundings. However, the perception of retail crowding can be maintained through environment and layout management. The trick is to develop the sense of belonging and success that comes with a degree of crowding yet at the same time be able to properly serve all customers.
To summarize, based on the case presented here and hints derived from the existing literature, it appears that there is potential for retail crowding to be utilized as a tool to enhance the retail environment. It is not likely that Wines Plus' customers would ask for crowded conditions, but it is possible that-without knowing it-they actually prefer some degree of crowding. Crowding can signal retailer success and an ability to offer customers an affiliation that goes beyond the customary retailerclient relationship, and managers and entrepreneurs should be aware that there is likely an optimal degree of crowding for every retail space. The degree of crowding perceived by customers can be managed by considering the amount of floor space and the organization of aisles, shelves, displays, and other retail accoutrements. It is predicted that further changes to Wines Plus' current physical layout in order to make use of these suggestions would improve their customers' satisfaction. However, crowding must be carefully managed because it is also capable of creating negative feelings between customers, which can produce dissatisfaction with the store and even reduce spending intentions.
DISCUSSION AND CONCLUSIONS
In an effort to explain an unexpected response to a longitudinal survey of a firm's customers, this article has traversed several topics. It began with a description of a study designed to measure customers' response to significant changes a firm made to its retail environment. The firm-Wines Plus-moved to a larger location largely because customers complained of crowding. Surprisingly, long-term customers indicated decreased satisfaction with the store after the move, despite the new location's increased floor space and product selection. The focus of the article became an attempt to understand this response by customers, when it appeared that there was little reason for their decreased satisfaction. Thus, the article is essentially a practical case analysis, although it offers a contribution to marketing theory by exploring the potentially beneficial effects of retail crowding.
A comprehensive framework that might be utilized to predict and explain the effects of changes to a retail environment could not be found. Bitner's (1992) Servicescapes model should perform well when applied to retailing. However, understanding the environmental dimensions of the servicescape (or retailscape) and the potential responses and moderators does not necessarily provide an explanation of the events associated with Wines Plus' move to a larger store. Customer response to changes in these dimensions is a complex process that cannot always be predicted; and that is certainly true in this situation. Ultimately, the researchers determined that retail crowding is the best explanation for the reduced satisfaction exhibited by Wines Plus' customers. It appears that marketing does not have a full understanding of the causes and effects of perceived crowding, and especially the circumstances under which it may be attractive to consumers.
The discussion thus far has much to offer retail managers, entrepreneurs, and marketers. One of the more obvious conclusions is that making changes to a retail store-in response to customer requests-may not always produce a favorable outcome. If customers' reactions to a change in a retail setting cannot be predicted with reasonable accuracy it may not be advantageous to go ahead with such alterations. Market testing will likely be of considerable benefit to firms evaluating modifications to their retail environment, but even this can produce erroneous conclusions, e.g., the Coca-Cola example mentioned earlier.
Customer perceptions of crowding might be expected to have a negative connotation for retailers, but the notion that, under the right circumstances, consumers can perceive crowding as a positive attribute deserves further consideration. Creating a club-like atmosphere could have benefits for many traditional retailers, and not just bars and sporting arenas. People may seek out that kind of atmosphere, perhaps in order to gain a sense of belonging or camaraderie. The retail environment would need to be carefully designed to fashion impressions of warmth, coziness, and intimacy. A certain degree of crowding will probably be necessary to establish the desired atmosphere. This kind of environment might be especially appropriate with-but not restricted to-products that have a degree of luxury or exclusivity. Exceptional employee knowledge and an emphasis on service are likely required in this type of environment, and all other environmental attributes would have to signal and support this strategy.
A final point that should be emphasized is that it is critically important that the firm consider the interactions between customers, employees, and other customers in the retail environment (Eiglier & Langeard, 1987; Pranter & Martin, 1991; Bitner, 1992; Harris, Baron, & Ratcliffe, 1995). If the design of a crowded retail space does not manage these interactions appropriately, there is considerable opportunity for some form of conflict between different groups. For example, an expensive wine store that must employ the "cozy strategy" because of space restrictions will have to ensure that employees can easily move around to serve customers adequately. To effectively serve customers, it may even be necessary to have more employees than with a less dense floor plan. Signs will be important to minimize search times and will reduce the likelihood that a customer will bump into employees or other customers. Yet, at the same time, it is important to recognize that employee-customer and customer-customer communications can be of substantial benefit (Pranter & Martin, 1991; Harris, Baron, & Ratcliffe, 1995; Grove & Fisk, 1997).
Given the complicated design of the study described earlier in this article, it would be surprising if difficulties were not encountered. An obvious weakness of the data obtained is that there was a decreased response rate over time. While such a decrease was expected (Wall & Williams, 1970) it was, nevertheless, somewhat disappointing. The decreased response rate reduces confidence in the results and interpretation. However, despite the decreasing response rate there are no signs that the quality of the data was compromised, and it is assumed that the measures are accurate.
A second limitation of the study is that there are many confounding factors related to the observed changes in customer satisfaction. Although Wines Plus' move and name change were planned, the temporary closing and loss of the original manager were not. Other confounds may be possible. For example, diminished satisfaction with price (Table 3) could be driving decreases in overall satisfaction. The effect of all changes has been carefully examined, but there is no way of knowing exactly what caused the decreased satisfaction. Such confounds are endemic with longitudinal designs, and the effects of changes can usually only be inferred. Future experiments can more strongly establish any relationships of interest.
Finally, the current study would be improved had data regarding new customers been collected. It is not known whether or not the sample reflects the opinions of customers familiar only with the new store. However, the original study was initially concerned with measuring changes in the opinions and satisfaction of existing customers and is assumed to have accomplished that task accurately.
The authors would like to acknowledge the editors' helpful comments regarding an earlier version of this article. James B. Wiley and Paul D. Larson provided valuable advice during instrument development and data collection, which was funded by the Canadian Social Sciences and Humanities Research Council.
Alicke, M. D., Braun, J. C, Glor, J. E., Klotz, M. L., Magee, J., Sederholm, H., & Siegel, R. (1992). Complaining behavior in social interaction. Personality and Social Psychology Bulletin, 18, 286-295.
Anderson, E. W., & Sullivan, M. W. (1993). The antecedents and consequences of customer satisfaction for firms. Marketing Science, 12, 125-143.
Anderson, E. W., Fornell, C, & Lehmann, D. R. (1994). Customer satisfaction, market share, and profitability: Findings from Sweden. Journal of Marketing, 58, 53-66.
Baker, J., Grewal, D., & Parasuraman, A. (1994). The influence of store environment on quality inferences and store image. Journal of the Academy of Marketing Science, 22, 328-339.
Berry, L. L. (1986). Retail businesses are service businesses. Journal of Retailing, 62, Editorial.
Bitner, M. J. (1992). Servicescapes: The impact of physical surroundings on customers and employees. Journal of Marketing, 56, 57-71.
Brunner, C. J. (1997). Learning from student evaluations. Marketing Educator, 16, 10.
Dillman, D. A. (1978). Mail and telephone surveys: The total design method. New York, NY: John Wiley & Sons.
Eiglier, P., & Langeard, E. (1987). Servuction. Paris, France: McGraw-Hill.
Eroglu, S. A., & Harrell, G. D. (1986). Retail crowding: Theoretical and strategic implications. Journal of Retailing, 62, 346-363.
Eroglu, S. A., & Machleit, K. A. (1990). An empirical study of retail crowding: Antecedents and consequences. Journal of Retailing, 66, 201-221.
Green, P. E., Tull, D. S., & Albaum, G. (1988). Research for marketing decisions. Englewood Cliffs, NJ: Prentice-Hall, Inc.
Grove, S. J., & Fisk, R. P. (1997). The impact of other customers on service experiences: A critical incident examination of "getting along". Journal of Retailing, 73, 63-85.
Harrell, G. D., Hutt, M. D., & Anderson, J. C. (1980). Path analysis of buyer behavior under conditions of crowding. Journal of Marketing Research, 17, 45-51.
Harris K, Baron, S., & Ratcliffe, J. (1995). Customers as oral participants in a service setting. Journal of Services Marketing, 9, 64-76.
Hui, M. K. M., & Bateson, J. E. G. (1991). Perceived control and the effects of crowding and consumer choice on the service experience. Journal of Consumer Research, 18, 174-184.
Kelly, J. P., & George, W. R. (1982). Strategic management issues for the retailing of services. Journal of Retailing, 58, 40-42.
Kohli, A. K., & Jaworski, B. J. (1990). Market orientation: The construct, research propositions, and managerial implications. Journal of Marketing, 54,1-18.
Kohli, A. K, Jaworski, B. J., & Kumar, A. (1993). MARKOR: A measure of market orientation. Journal of Marketing Research, 30, 467-477.
Kotler, P. (1974). Atmospherics as a marketing tool. Journal of Retailing, 41,48-64.
LaBarbera, P. A., & Mazursky, D. (1983). A longitudinal assessment of consumer satisfaction/dissatisfaction: The dynamic aspect of the cognitive process. Journal of Marketing Research, 20, 393-404.
Milliman, R. E. (1982). Using background music to affect the behavior of supermarket shoppers. Journal of Marketing, 46, 86-91.
Milliman, R. E. (1986). The influence of background music on the behavior of restaurant patrons. Journal of Consumer Research, 13, 286-289.
Mittal, B., & Lassar, W. M. (1996). The role of personalization in service encounters. Journal of Retailing, 72,95-109.
Narver, J., & Slater, S. (1990). The effect of market orientation on business profitability. Journal of Marketing, 54, 20-35.
Pranter, C. A., & Martin, C. (1991). Compatibility management: Roles in service performers. Journal of Services Marketing, 5, 43-53.
Rust, R. T., Zahorik, A. J., & Keiningham, T. L. (1995). Return on quality (ROQ): Making service quality financially accountable. Journal of Marketing, 59, 58-70.
Sherry, J. F., Jr. (Ed.). (1998). ServiceScapes: The concept of place in contemporary markets. Chicago, IL: NTC Business Books.
Spangenberg, E. R., Crowley, A. E., & Henderson, P. W. (1996). Improving the store environment: Do olfactory cues affect evaluations and behaviors? Journal of Marketing, 60, 67-80.
Stokols, D. (1972). On the distinction between density and crowding: Some implications for future research. Psychological Review, 79, 275-278.
Surprenant, C. F., & Solomon, M. R. (1987). Predictability and personalization in the service encounter. Journal of Marketing, 51, 86-96.
Tse, D. K., & Wilton, P. C. (1988). Models of consumer satisfaction formation: An extension. Journal of Marketing Research, 25, 204-212.
Wall, W. D., & Williams, H. L. (1970). Longitudinal studies and the social sciences. London, England: Heinemann Educational Books Ltd.
Zeithaml, V. A., Parasuraman, A., & Berry, L. L. (1985). Problems and strategies in services marketing. Journal of Marketing, 49, 33-46.
New Mexico State University
Shaun McQuitty is an Assistant Professor of Marketing at New Mexico State University. He received his Ph.D. from the University of Alberta. His research interests include consumer satisfaction, service quality, consumer behavior, structural equation modeling, and measurement issues. His work has appeared in the Journal of Macromarketing, the International Journal of Research in Marketing, and Structural Equation Modeling.
Kevin Shanahan is a Ph.D. student in the Department of Marketing at New Mexico State University. He received his MBA from Thunderbird - The American Graduate School of International Management. His research interests include services, advertising, and ethics. His work has been presented at the Western Marketing Educators Conference.
Eric Pratt is an Associate Professor of Marketing at New Mexico State University. He received his Ph.D. from the University of Mississippi. His research interests include marketing education and consumer behavior. His work has appeared in academic and trade journals such as the Marketing Education Review, New Accountant, and the Small Business Institute Review.…