INTRODUCTION

India's diversified culture has resulted in varied food habits in different parts of the country. Although traditionally Indians preferred home-prepared food, the growing influence of western culture has resulted in a shift in the food consumption patterns among some classes of Indians. In particular, the urban Indian youth have started to eat out-of-home more frequently and are switching to western food eating habits.1, 2 Specifically, fast food has gained increased acceptance among Indians following liberalization and urbanization in the early 1990s and the subsequent entry of numerous global and local fast food companies such as McDonalds, KFC and Mamu. According to Merriam-Webster online dictionary fast food is ‘of related to, or specializing in food that can be prepared and served quickly’. This emergence of fast food as a new option for food consumption has resulted in a significant change in the lifestyle and food habits of Indian consumers.3

Over the past few years, fast food has become one of the fastest growing industries in the world as well as in India. According to RNCOS (a leading marketing research firm), India's fast food industry is witnessing a robust growth of 30–35 per cent a year and anticipates the entry of numerous international players over the next few years.4 It has also been stated that over the past decade the amount that urban Indians spend on eating out and fast food has increased dramatically. In another survey by ACNielsen, India was ranked one of the top 10 countries in terms of frequency of eating fast food with approximately 37 per cent of adult Indians consuming fast food at least 2–3 times a week.5 McDonalds has over 100 outlets across major cities and was found to be the most frequented fast food chain in India, along with Domino's and Pizza Hut. Among the local brands, Nirula, Pizza Corner and Hot Breads were identified as providing tough competition to the multinational companies. It should be noted that while the western world relates fast food restaurants with convenience, the majority of Indians relate (western) fast food restaurants with life style and eclectic taste.

In this context of changing food and eating habits, this study aims to understand the drivers of satisfaction with consumption of fast food among urban, young, Indian consumers. In addition, this study will provide a segmentation of fast food consumers based on the drivers of satisfaction. As satisfaction is directly linked to customer loyalty and a firm's profitability,6, 7 the authors identified segmentation based on satisfaction drivers as worth exploring. Furthermore, past research contends that as expectations differ from customer-to-customer, the factors that influence customer satisfaction may also vary.8, 9, 10 Hence, understanding the drivers of satisfaction on an individual level rather than at aggregate level could provide key insights for managers on product or service improvements. Accordingly, in this article the authors use Kano's theory of attractive quality to derive satisfaction-based segmentation.11

The rest of the article is organized as follows. First, the article provides a brief review of the literature on customer satisfaction and argues the need for segmentation based on the importance of quality attributes in customer satisfaction. Next, an empirical study is presented that examines the differences in the satisfaction drivers between the four clusters. The managerial implications and conclusions are discussed in the final section of the article.

LITERATURE REVIEW

Customer satisfaction

In today's global market, customer satisfaction has become a major issue and an essential pre-requisite for creating and sustaining a competitive advantage. In fact, increasingly, firms are using customer satisfaction as a key performance metric in assessing their current and future prospects.12 Customer satisfaction is considered as an important part of the firm's corporate strategy and a key driver of long-term consumer loyalty and market value. In this context, it is argued that ‘customer satisfaction has come to represent an important cornerstone for customer-oriented business practices across a multitude of companies operating in diverse industries’.13 In recent years this has led to various consulting firms promoting and offering strategies for enhancing customer satisfaction.14, 15 The shift in focus on customer satisfaction is grounded on the assumption that satisfied customers are likely to increase the share of purchases in the category (that is share of wallet), leading to greater customer loyalty. In one study, it was found that customer satisfaction explained about 37 per cent of the differences in the customer loyalty levels given the influence of external factors such as competition.16 Furthermore, customer satisfaction leads to an increased price tolerance level, thereby potentially increasing the consumers’ willingness to pay a price premium for products/services.17 With higher levels of customer satisfaction, firms will be able to achieve higher levels of future cash flow, above-market returns, lower customer transaction costs and increased shareholder value.18, 19 To sum up, the ultimate goal of every business is not to sell, supply or service, but to satisfy the needs that drive customers to buy, therefore maintaining higher satisfaction levels.20

The Kano model

Prior research within marketing literature has often conceptualized customer satisfaction in a one-dimensional view – higher product attribute performance leads to higher customer satisfaction levels.21, 22, 23However, recent studies have found that this relationship is non-linear that is improvement in attribute performance will not always lead to a proportionate increase in customer satisfaction.8, 15 It has been argued that since the attribute importance weights can vary, the performance of some attributes do not necessarily lead to higher levels of customer satisfaction.24 In this context, it is important to determine the different types of product attributes that determine customer satisfaction.

In quality management literature, Kano and his colleagues proposed a classification model that recognizes the diverse relationship between the attribute performance and customer satisfaction.11 The Kano model consists of two dimensions on which customer satisfaction is determined. The first is an objective aspect that involves the product functionality and the second is a subjective component that includes the customers’ perception of satisfaction. On the basis of this, five types of quality attributes were identified that determine customer satisfaction, namely: attractive quality attributes, one-dimensional quality attributes, must-be quality attributes, indifference quality attributes and reverse quality attributes.

Attractive quality attributes involve those attributes that if fulfilled give satisfaction, but do not produce dissatisfaction if unfulfilled. These attributes are not normally expected and often unintentionally surprise and delight customers.25 One-dimensional quality attributes are those that are linearly related to customer satisfaction – that is, the greater the attribute performs the greater the level of customer satisfaction will be. These attributes are often spoken and are the ones with which companies compete.26 An attribute whose absence will result in customer dissatisfaction, but whose presence does not significantly contribute to customer satisfaction, is referred to as must-be attributes. The must-be are viewed as basic attributes and customers expect that the company understand these attributes.27 Indifferent attributes are those that customers perceive as not contributing to their satisfaction; hence their presence or absence has no effect on satisfaction. Reverse quality elements are those attributes, which, if present, would lead to consumer dissatisfaction, and if absent would lead to consumer satisfaction.28

The Kano model has important managerial implications as it addresses the shortcomings of the one-dimensional view by creating a better understanding of how customers evaluate and perceive quality attributes. It enables fresh understanding of customer needs and suggests a new way to prioritize customer requirements. When this deeper understanding of customer requirements is gained the Kano model can be used to identify important attributes in determining customer satisfaction. Furthermore, by focusing on the most important attributes to improve, the model helps managers identify opportunities for differentiation. For example, a study was conducted to investigate the differences across various quality dimensions in determining destination satisfaction.29 It was found that interaction and the physical environment of the design are the important attributes that managers should prioritize so as to create an attractive offering, resulting in higher customer satisfaction. Similarly, another study used a five-level Kano approach to classify the attributes in an e-services context.26 The results for the seven attributes showed that web-navigation possibilities, the treatment of professional information, select seating online and the availability of e-services were classified as one-dimensional, watching trailers online was seen as an indifferent attribute, advance booking was an attractive attribute and reliability of information was a must-be attribute.

The Kano model has gained increased acceptance since its introduction and has been adopted by many researchers to study product development and design, strategic thinking, employee management, service quality management, customer experience management and industrial marketing.30, 31, 32, 33, 34 Surprisingly, however, prior studies have not addressed the issue of market segmentation. This is a major limitation as it is apparent that market segments vary in their expectations and requirements with regard to products or services.31 As a result, market segments differ in the importance they assign to quality attributes in product evaluation. These differences in attribute importance reflect the drivers of satisfaction with regard to various market segments. Hence, Kano-based segmentation seems to be particularly well-suited to assess the drivers of satisfaction in the changing Indian fast food industry.

METHODOLOGY

Pre-test – Elicitation of drivers of satisfaction

In order to elicit the drivers of satisfaction, a focus group setting was utilized. Students who had previous experience in visiting fast food restaurants were invited for a focus group discussion; they were then asked to describe their most satisfying and dissatisfying experiences in fast food restaurants and their minimum expectations from fast food chains. By the end of the fourth focus group, researchers had saturation of information and hence the focus group exercise was called off. The focus group exercise yielded 11 drivers of satisfaction. Further, four participants, differing in their levels of patronage were independently interviewed in depth and were further coded by a different researcher; this resulted in a similar set of drivers of satisfaction. Table 1 presents the drivers of satisfaction with fast food restaurants following the focus groups and interviews.

Table 1 Drivers of consumer satisfaction elicited from focus groups and depth interviews
Full size table

Measures and sample

Overall satisfaction with fast food restaurants was measured using a five-point semantic differential scale (‘delighted’ – ‘terrible’35). Responses on the drivers of satisfaction were collected using the functional–dysfunctional form of the question. The first question concerns the respondent's reaction if a particular driver of satisfaction is evident within a fast food restaurant (functional form of the question); the second question concerns if the fast food restaurant does include that factor (dysfunctional form of the question). Both the functional and dysfunctional forms were measured on a five-point scale anchored by ‘I like it that way’ (+2) to ‘I dislike it’ (−2). Behavioral loyalty was measured using a single item seven-point Likert scale ‘how often do you visit fast food restaurants?’, anchored by ‘seldom’ (1) to ‘very often’ (5). Demographics and behavioral information such as ‘fast food restaurants commonly visited’ and ‘fast foods often ordered’ were sought towards the end of the questionnaire.

A survey using a structured questionnaire was administered to around 200 young (less than 31 years of age) urban customers visiting major fast food restaurants in one of the metropolitan cities of India. One hundred and fifty-eight usable questionnaires were retrieved.

Data analysis

Data analysis was performed in three steps. First, based on the functional and dysfunctional question responses, the quality attribute of each customer was classified into one of the five drivers of satisfaction based upon the Kano evaluation table.15 Next, categorical regression was carried out, with the overall satisfaction as dependent variable and the drivers of satisfaction as independent variables. Finally, two-step cluster analysis was employed to identify various segments in urban Indian fast food customers.

Identification of Kano categories

By combining the functional and dysfunctional forms of the questions,15 the product criterion or the drivers of satisfaction of fast food restaurants for each consumer upon each quality attribute was classified as either an attractive (A), one-dimensional (O), must-be (M), indifferent (I) or reverse quality (R) quality attribute for each respondent. The resultant data were nominally coded and were used in categorical regression to identify the key drivers of satisfaction.

Identifying significant drivers of satisfaction

The categorical regression procedure of SPSS 17.0 (CATREG) was carried out to identify significant drivers of satisfaction among fast food customers. The overall satisfaction was used as a dependent variable and the drivers of satisfaction were used as independent variables for the analysis. CATREG rescales categorical data into ratio data by assigning numerical values to the categories, and then uses categorical canonical analysis with optimal scaling procedure (also known as OVERALS) to arrive at partial regression coefficients. The variables (drivers of satisfaction) whose corresponding partial regression coefficients were significant at levels P<0.01 were only considered for further analysis of two-step cluster analysis. Following this analysis, six variables (gourmet taste, ample servings per order, rotating menu, thematic menus, young crowd and quick service time) were found to be significant. These drivers/factors were used for the segmentation using cluster analysis (represented in bold in Table 2). This was done as selection of irrelevant or insignificant variables in the cluster analysis might have the tendency to distort the cluster solution.36

Table 2 Results of the CATREG regression analysis
Full size table

Two-step cluster analysis of relevant Kano variables

In this study, two-step cluster analysis was employed to identify the various segments of the fast food industry. The categorically coded drivers of satisfaction identified in the previous step as significant and important were used in the cluster analysis. The log-likelihood distance measure was used with subjects categorized under the cluster associated with the largest log-likelihood. No prearranged number of clusters was suggested. The Bayesian information criterion (BIC) was used to judge the adequacy of the final cluster solution. On the basis of the BIC and log-likelihood distance, four clusters were found to be appropriate. For four cluster solutions, the BIC ratio of change was 0.582; the corresponding ratio of distance obtained was 1.526 – the maximum among all feasible cluster solutions.

A Chi-square test of importance was used to identify satisfaction drivers important for each cluster. Variables were considered important within the particular cluster if the chi-square importance measure exceeded the critical value. These variables are indicated as ‘I’ in Table 3. For example, in Cluster 1, gourmet taste, ample serving per order and young crowd were found to be important at level P<0.05.

Table 3 Importance of variables in each cluster
Full size table

Reliability and validity of Kano segments

Before making an attempt to describe the cluster solution, reliability and validity of the cluster solution was carried out. To assess the reliability of Kano segments, about 50 per cent of cases (n=75) were randomly selected and cluster analyzed again using the same procedure as discussed above. A similar four-cluster solution evolved. A Chi-square test was performed to check whether significant association existed between the new cluster membership and original cluster membership. Pearson Chi-square value was found to be statistically significant (χ2=68.594, DF=9, P<0.01), therefore the reliability of Kano segments is established. Kano segments are derived on the assumption that drivers of satisfaction differ across different segments, hence it is desirable to expect mean overall product satisfaction to differ across different segments; a significant difference would establish concurrent validity. To test this assumption, one way ANOVA was utilized with the independent variable as KANO segment membership and the dependent variable as overall satisfaction. A significant F statistic [F(3,154)=3.102, P<0.05] was evident, thus proving concurrent validity of proposed segmentation. Mean overall satisfaction for different segments is reported in Table 4.

Table 4 Means satisfaction and loyalty scores for different KANO segments
Full size table

Prior studies suggest that satisfied consumers are more loyal,37 hence differences in measure of loyalty are expected to vary between different KANO segments. Significant differences in loyalty, if found, would establish predictive validity. To test this assumption, one way ANOVA was utilized with the independent variable as KANO segment membership and the dependent variable as frequency of visit (measure of behavioral loyalty). A significant F statistic [F(3,154)=3.305, P<0.05] was found, thus proving the predictive validity of proposed segmentation. Mean loyalty for different segments is reported in Table 4.

Interpretation of clusters

Following the test of reliability and validity, the four clusters obtained were interpreted. Tables 5 and 6 present the composition of the four clusters along with the drivers of satisfaction.

Table 5 Cluster profiling of respondents
Full size table
Table 6 Composition of important variables in cluster solution
Full size table

Important drivers of satisfaction of the first cluster are taste, ample serving per order and young crowd; these consumers viewed all of these three drivers of satisfaction as one dimensional. This segment has the least number of consumers (n=28, 18 per cent of the respondents). These consumers are generally least satisfied with fast food restaurants and generally visit restaurants less often than other individuals. As features related to taste and quantity primarily guide their visit to fast food restaurants, this group is named as ‘value for money seekers’.

Important drivers of satisfaction of the second cluster are taste, rotating menus, thematic interiors, young crowd and quick service time. This cluster considers taste, rotating menus and thematic interiors as attractive quality attributes, and young crowd and quick service time as one-dimensional quality attributes. This segment has the highest number of consumers (n=49, 31 per cent of respondents). These consumers are moderately satisfied with fast food restaurants and rank second in terms of frequency of visit. As these consumers can be most satisfied by features relating to food and ambience, this group is named as ‘novelty/experience seekers’.

Important drivers of satisfaction of the third cluster are taste and young crowd. These consumers view taste as a one-dimensional quality attribute and young crowd as an indifferent requirement. As indifferent drivers do not contribute to satisfaction, the only driver that is considered as important by this segment is taste. These consumers are less satisfied compared to other clusters but frequently visit fast food restaurants. As these consumers can be satisfied only by features relating to taste alone, this group is named as ‘taste seekers’.

In the case of Cluster 4 the important drivers of satisfaction were taste, ample serving, rotating menus, young crowd and quick service time. Most of these consumers consider ample serving as an indifferent attribute quality; hence this was not considered a driver of satisfaction for these consumers. This segment ranks highest in terms of overall satisfaction, but lowest in terms of frequency of visit. Taste and quick service time are identified as the important drivers of satisfaction in this segment; also satisfaction among these consumers can be improved by the absence of young crowd and standardized menus. As the expectations of the respondents belonging to this segment conform to the expectations of traditional Indian food joints, consumers in this segment are named ‘traditionalists’.

MANAGERIAL IMPLICATIONS

This article explores a segmentation approach based upon the Kano model, contributing to consumer satisfaction. A four-segment solution was found for young urban Indian consumers. Validation of four segments shows the efficiency of Kano-based segmentation to identify variations in consumer profiles that have little variation in demographical, and socio economic features a-priori. It is anticipated that the segmentation procedure developed here will help firms understand what their targeted consumers seek from the offering. For example, if a firm discovers that most of the consumers visiting its restaurant are taste seekers, it might want to withdraw investing in the ambience of the restaurant and instead look at increasing the assortments in the menu to secure continuing patronage of its existing customers.

Firms who endeavor to understand their target segment versus a competitor's target segment, can adopt a similar survey coupled with variables that relate to preference for various fast food brands. After deriving the segments, each consumer is then identified by one brand that he or she prefers most and the segment to which he or she belongs to. A joint space map of multiple correspondence analyses can reveal information regarding the segments that the firm is targeting vis-à-vis segments that competitors are targeting. An endeavor should be made to identify whether the target segment identified on the joint space map coincides with the firm's desired targeted segment. If coincidence is not found, then two options are possible; the product mix and firm's marketing communication may be revisited to tune in with the desired segment characteristics and emphasis may be laid on promoting those elements that the desired targeted segment finds as attractive. Though coincidence is not achieved, but currently visiting segment(s) are viewed as significantly contributing to profits; depending upon the firm's corporate brand positioning and its desired corporate image, if feasible, a status quo call may be taken. In this scenario, those elements that the current footfall sees as attractive may be reinforced. If coincidence is found, no major modifications in the product mix or communication strategy is required; however, care should be taken regarding fulfilling those elements that consumers designate as must be requirements during service delivery; also firms should also constantly explore and innovate with respect to those elements that consumers see as attractive requirements in order to continue satisfying these segment(s).

CONCLUSION AND DISCUSSION

The Kano model of consumer satisfaction has been conventionally developed as a tool to aid the product development process. This article explores a segmentation opportunity using this model. Usage of this model was meant to be in a product context, whereas the current study makes an attempt to integrate the Kano methodology into the service setup. This article thus extends the utility of this pragmatic model by finding new avenues where it can be deployed. Segmentation based upon satisfaction is again a relatively new concept to marketing literature, but the authors opine that this novel approach is actually necessary because it satisfied customers who drive businesses through their loyalty and positive word of mouth.37

Apart from strong managerial relevance, there is also an underpinning theoretical rationale as to why segmentation based on satisfaction may be deemed necessary. Theoretically, a firm can undertake two types of strategies to improve market share – namely offensive and defensive strategies. Offensive strategies aim to attract new consumers whereas defensive strategies aim to retain existing consumers.38 Most popular segmentation approaches (viz. benefit-based, lifestyle-based, demographic-based, psychographic-based segmentation approaches) can be discussed under the repertoire of offensive strategies, as the basic objective of these segmentation approaches is to identify profiles of possible groups of consumers currently existing in the market place and then depending upon desirability and feasibility criteria, the marketer fashions product mix and communication mix elements in a way that attracts the targeted group(s). However, segmentation approaches from defensive strategies have not been adequately explored. The authors are of the opinion that the Kano-based segmentation technique that is introduced in this article will make a contribution to segmentation literature from a defensive strategy perspective. In this article, the authors have investigated the experiences of consumers in fast food restaurants (post hoc evaluation of product) and then have devised a segmentation procedure that gives fresh insights to managers on refashioning their product/communication mix in a manner that maximizes satisfaction within each segment. While an offensive strategy perspective of segmentation gives insights into attracting relevant segment(s), Kano-based segmentation gives insights into how to retain the relevant segment(s). Hence, for those firms that make the majority of their profits from loyal consumers, Kano-based segmentation may be recommended.