Abstract
In this paper, we propose an experiential approach to tourist segmentation aimed at overcoming the limits of both socio-demographic and context-specific approaches widely adopted in the literature and in practice. In this study, segmentation is carried out based upon the expected experiences of Chinese tourists at the Shanghai World Exposition. The segmentation reveals four tourist clusters with different interests in relation to their experiences in visiting the World Exposition. The clusters showed insignificant differences in the demographics but proved to be powerfully discriminant in determining tourists' satisfaction and loyalty, which affirms the potential of the tourist experience being a segmenting variable. Moreover, thanks to the analysis, an evaluation of the Shanghai World Exposition's success in terms of visitors' satisfaction is provided.
1. Introduction
Segmentation is a key process for developing more efficient and effective tourism marketing plans and hence to maximize the benefits for tourism facility managers, policy makers and communities [1].
Several segmentation approaches have been provided over time and there is no absolutely correct way of segmenting: such aspects as methods and bases for segmentation should be determined, in fact, by a strategic purpose and should reflect the strategic view of the organization segmenting the market [2]. For this reason, many variables have been utilized so far in segmentation. Whereas traditional, product-based segmentations have proven to be only slightly effective, demographics has been one of the most frequently-used segmenting variables because of its simplicity and intuitiveness (e.g., [3, 4]). However, there have been increasing arguments that demographics are not wholly reliable or adequate as segmenting variables (e.g., [5–7]), since often motivations, values and attitudes are better predictors of behaviours [8]. Psychographics - i.e., the variables and attributes relating to the personalities, values, attitudes or lifestyles of a population (e.g., [9]) - as an alternative background variable for segmentation is receiving increasing attention, especially in tourism research.
In time, several different psychographic variables have been considered for segmenting: motivation to travel, satisfaction, lifestyle, benefits sought, and so on [10–13]. Their common trait is the idea of segmenting according to the importance attributed to a series of ad hoc parameters developed according to the specific setting of the study. We believe that this approach has a significant advantage and a potential disadvantage. The advantage lies in the fact that building segmentation on context-specific variables helps in understanding punctual patterns in tourist behaviours and decisions that are extremely relevant for the specific venue where the surveys for segmentation are conducted. The potential disadvantage is the fact that such an ad hoc approach may limit both the theoretical generalizability of the segmentation models developed and the possibility of replicating the same study in a different context.
Hence, finding sound theoretical backgrounds for grounding segmentation analysis may provide tourism managers and researchers with more powerful and generalizable tools for interpreting tourist behaviour. Moving from this principle, over the last few years, tourist behaviour has been increasingly analysed by scholars under an experiential perspective (e.g., [14–16]). However, while the literature has mostly looked at tourist experiences as a variable through which satisfaction and a propensity to re-visit were analysed (e.g., [15,17]), quite surprisingly customer experience has seldom been considered as a segmenting variable.
The objective of this research note is to provide the results of an exploratory study aimed at understanding whether and (if so) how an experience-based segmentation can provide insightful information to support tourism planning and marketing. We will do this through the analysis of the outcomes of a survey conducted on a sample of Chinese visitors at the 2010 Shanghai World Exposition. In this way, we aim to make a twofold contribution: (i) a testing by experience-based variables of the ability to identify clusters of tourists of differing sensitivities to tourism stimuli and, accordingly, differing degrees of satisfaction in relation to the same stimulus; (ii) an evaluative model of the overall visitor satisfaction at the 2010 Shanghai World Exposition, which was the largest event in humankind in terms of the number of visitors (over 84 million), answering to recent calls from tourism management (e.g., [18]).
The rest of the work is articulated as follows: first, we will summarize the theoretical and conceptual background of our study in order to highlight the experiential variables adopted in the study. Next, we will introduce the research method of the study and the results, suggesting interpretations and guidelines for the further refinement of the segmentation method. Finally, we will briefly review the satisfaction data about the World Exposition to provide an overview of its ability to fulfil its visitors' requirements.
2. Framework
2.1 Theoretical Background: Experiential Marketing
Through experiential approaches to the analysis of customer behaviour have long been long debated in the literature (e.g., [19]), the importance of customer experiences has been increasingly recognized by researchers and practitioners only recently: moving from competitive and macro-economic arguments, scholars have supported the idea of moving from an analysis of purchase decisions based upon specific product-level features to an interpretation of customer behaviour as a response to a comprehensive and holistic sets of rational and irrational variables able to give customers a memorable experience [20–22]. From this perspective, organizations stimulate customers with sensory, cognitive, emotional and behavioural stimuli [23] in an attempt to make customers sense, feel, think and act in a way that is desirable and memorable [24]. Though the experiential industry as a conceptual landscape may be applied to very diverse industries [25], some specific industries have a stronger experiential content than others, since they more deeply impact the hedonic, emotional and cognitive self of the customer. Tourism is certainly one such highly experiential industry (e.g., [14–16,26]), and in fact preliminary studies on the impact of experiential variables on themes such as visitor satisfaction and a propensity to re-visit have been proposed [15,17].
Previous studies on customer experience have all suggested its multi-dimensionality and the centrality of customers' personal perceptions [19–21,27], which is suggestive of how the importance attributed to the different experiential stimuli might actually represent an interesting segmentation variable. This is the main assumption at the basis of our work.
Different conceptual frameworks for classifying customer experiences have been provided over time, but they can generally be pulled back to two main streams that are conceptually entwined. On the one hand, the approach developed and beginning with Pine and Gilmore [20] analyses the experiences on the customer-side, identifying four ‘realms’ of consumer experiences according to the degree of customer involvement (passive vs. active participation) and the desire with which the customer connects or engages with the event performance (absorption vs. immersion). As a result, four types of experiences are generated: entertainment (passive/absorption), educational (active/absorption), escapist (active/immersion) and aesthetic (passive/immersion). This conceptualization has characterized studies in the field of marketing [15] and, interestingly, in the specific setting of tourism [17]. On the other hand, there is the approach moving from Schmitt's conceptualization [23], which assumes a brand- or company-side viewpoint and which defines five main areas of experiential stimulation (sensory, emotion, cognition, behavioural and relational). Further refinements of this approach have led to different scales and conceptualizations, including that of Brakus et al. [22], which neglects the relational dimension in defining a brand personality, and that of Gentile et al. [21], which subdivides behavioural stimuli into two subgroups: pragmatic stimuli, aimed at stimulating the customer to do something, and lifestyle stimuli, aimed at showing the customer a series of cultural traits (e.g., values and beliefs) characterizing a specific way of living somehow considered to be desirable by customers themselves.
Considering the evident overlaps between the two streams, we endorse the former approach and especially that of Gentile et al. [21] in our analysis because, compared with that of Oh et al. [15], it makes it possible to analyse not only the kinds of experiences looked for by customers but also the constituting elements that customers value the most, and this seems a more suited viewpoint in approaching a segmentation. As a consequence, in our study we characterize visitor's experiential expectations according to the importance attributed to four main dimensions:
Sensory experience: related to elements that can be seen, heard, touched, smelled or tasted. Emotional experience: related to the affective self and the feelings of the customer. Cognitive experience: related to the conscious self of the customer, including the acquisition of new knowledge, etc. Pragmatic experience: related to active engagement in the representation/experience.
2.2 Context of the Study
We set our study within the context of a theme park within a mega-event, namely the World Exposition Park during Shanghai World Exposition 2010. Our choice was motivated by a threefold argument. First, theme parks created a ‘fantasy world’ that aimed to bring their visitors an escapist and elevated experience in relation to their daily-lives [28]. Therefore, customer experiences have always been essential to theme parks. Not surprisingly, theme parks play an important role in tourism worldwide. In 2010, despite the wide-spread economic crisis, attendance at theme parks has still generally increased - especially in Asia - by 7.3%, with China showing the strongest growth [29]. Second, the Chinese tourism market is by far the most interesting worldwide and relatively little is known about the perceptions and expectations of Chinese tourists [30–33]. Thirdly, the 2010 World Exposition was the largest event in history in terms of numbers of visitors (over 84 million in 6 months), and the literature has expressed calls for an understanding of whether such an anticipation around the event and attendance during the event has actually generated satisfaction [18].
These arguments depict our context as an ideal research locus for investigating tourists' expectations and satisfaction from an experiential point of view.
2.3 Methodology
Our study involved 510 Chinese visitors of the 2010 World Exposition, who attended the event for the first time over the course of the week between September 20 and 24, 2010. They were randomly selected at the entrance and asked to fill out an initial questionnaire aimed at measuring the importance that they attributed to a set of experiential stimuli and requesting them to pass by the same gate at the exit, where they were required to fill out a satisfaction questionnaire about the same stimuli. This allowed us to avoid any importance attribution bias, which leads respondents to attribute higher importance to those factors which have particularly satisfied/dissatisfied them. Each customer was identified through a numeric ID in order to preserve anonymity. The contact with the visitors was managed by trained native Chinese speakers. The questionnaires were self-compiled by the visitors, who were provided with small gifts as an incentive for participation. Seven questionnaires were unusable, leading us to obtain 503 valid responses.
2.4 Measures
The identification of a comprehensive, though parsimonious, set of experiential stimuli followed the procedure suggested by Curchill [34] for developing solid marketing constructs. We essentially moved from the experiential stimuli (the framework by [21]) so as to define a set of items specifically thought of for the specific context of theme parks; to do this, we reviewed the previous literature on theme park segmentation in order to identify a broad set of stimuli: this led us to identify 26 possible items. We conducted an initial study aimed at reducing the size of the scale. First, we conducted interviews with six marketing and tourism management professors, as well as two expert consumer behaviour researchers in China and in Italy, to identify overlapped and superfluous items. 10 items were dropped after this scoping study, leading to 16 items (5 for sensory experience, 4 for emotional, 5 for cognitive and 2 for action experience).
Next, a scoping study of 228 Chinese students at the undergraduate and graduate level was conducted to assess the reliability of the importance scale. Confirmatory Factor Analysis (CFA) was conducted in order assess the goodness of the conceptual framework. The results of the 5-factor solutions were widely satisfactory (adj. χ2df=116=175.478, p<.001; CFI=.966; TLI=.960; RMSEA=.047, p=.60; SRMR=.043). All the factor loadings were above 0.6, and Cronbach's alpha for the factors exceeded 0.7 for all of them except for ‘action experience’, which was slightly below (.678); nonetheless, with the construct being constituted by two items only, Cronbach's alpha underestimates reliability [35], so we tested whether the Average Variance Explained (AVE) by each factor exceeded the 0.5 threshold [36] (sensory=.524; emotional=.513; cognitive=.534; action=.528), thereby ensuring reliability.
The survey was complemented with a section investigating demographic variables in order to explore whether the experience-based segmentation may provide proactive, practical, and precise suggestions to the management of the World Exposition regarding positioning, targeting and even operational tactics. It also intended to provide a general alternative view of segmenting a market.
3. Results
Following previous studies aimed at segmenting tourists according to psychographics, the responses were segmented through a K-means cluster analysis [10] according to the importance attributed to the different experiential factors (treated as summated scales). Five cluster analyses were carried out and generated solutions of 3 to 7 clusters. The solution with 4 clusters was the best in terms of the Calinski-Harbasz Pseudo-F index, as indicated in Table 1. The standardized importance attributed to the different experiential dimensions is described in table 2.
Cluster analysis stopping rules
Importance attributed to experiential dimensions by clusters
Cluster analysis was performed on the standardized scores of the importance attributed to each experiential dimension. The original score ranged from 1 being ‘Irrelevant’ to 5 being ‘Fundamental’
Analysing the data, it is possible to depict the key characteristics of each factor:
Cluster 1: Involved cluster. This cluster attributed relatively high importance to all the experiential dimensions, which suggested their high involvement in the experience of visiting the Shanghai World Exposition. Cluster 2: Indifferent cluster. On the opposite side, this cluster attributed relatively low importance to all the experiential dimensions, which suggested that they were not particularly interested in the experience of visiting the Shanghai World Exposition from the very beginning. Cluster 3: Introverted cluster. By comparison, this cluster attributed higher importance to the experiential dimensions of emotion and cognition, which suggested that they were more concerned with their own feelings and thoughts during the experience. Cluster 4: Extroverted cluster. This cluster attributed relatively high importance to the experiential dimensions of sensory and action, which suggested that they were more concerned with the physical and social environment during the experience
A discriminant analysis was carried out to test the existence of possible demographic elements antecedent to belonging to a specific cluster, as shown in table 3. Interestingly, none of the demographics investigated in the study resulted in a predictor of cluster attribution, testifying to how the importance attributed to the different dimensions of the experiential vector is generally not dependent upon demographics.
Demographic characteristics of clusters
In order to verify that the clustering proposed is actually a powerful discriminant of some tourists' behaviour, discriminant analysis through ANOVA was also performed in relation to satisfaction with respect to the different experiential dimensions, intention to revisit, category attitudinal loyalty and intention to recommend peers to visit the Exposition afterwards. As shown in Table 4 and table 5, different targets demonstrated significantly different levels of satisfaction with respect to the experiential performance of the World Exposition: Involved visitors showed the highest levels of satisfaction on all the dimensions, while introverted visitors were satisfied by the action and emotional experiences more than extroverted visitors, who in turn were more satisfied by sensory and cognitive experiences. ANOVA has also shown how involved visitors - the most satisfied - actually also showed higher attitudinal loyalty, both for the Exposition and the theme park category, supporting the overall reasonableness of the predictive power of this segmentation when analysing the relationship between satisfaction and loyalty.
ANOVA of Satisfaction on experiential dimensions by cluster
The scale of satisfaction ranged from 1 being ‘Much worse than I expected’ to 5 being ‘Much better than I expected’
ANOVA of visitors' loyalty by cluster
The scale of loyalty ranged from 1 being ‘Strongly disagree’ 5 being ‘Strongly agree’
4. Discussion
4.1 Testing of the Segmentation and Implications for Further Research
The results of this study support the idea that an experiential segmentation may provide interesting insights into tourists' behaviour, especially by matching importance and satisfaction in predicting attitudinal loyalty. It is very remarkable that the segmentation provided detects clusters that are substantially indifferent in terms of demographics, testifying once more to how demographics-based segmentation is increasingly less powerful in anticipating tourist behaviour [37]. This outcome, remarkably, calls into question some established knowledge about Chinese tourists' behaviour, which attributes a predictive role to demographics in identifying satisfaction drivers and levels of expectations when travelling (e.g., [38]).
Furthermore, the study suggests that the experiential stimuli are not to be considered as a trade-off, but rather that different tourists may attribute higher or lower levels of importance to all of the experiential dimensions, suggesting that a link exists between the experiential importance attributed and the overall level of involvement in the experience.
These results suggest that in fact experiences may represent valuable theoretical viewpoints through which segmentation in tourism could be approached. Moreover, as the experiential approach that we took was non-context-specific, it provides a potentially replicable tool, at least for further segmentation analysis in theme parks. Of course, the main limitation of the study lies in its being based on a single case: we warmly encourage further research adopting the scale in different contexts so as to observe its statistical reliability and usefulness.
4.2 An Analysis of Visitors' Satisfaction at the Shanghai World Exposition 2010
The results show quite clearly that the Shanghai World Exposition was not an extremely satisfactory event for visitors: this can be observed by analysing, for instance, how the average satisfaction exceeded the threshold value of 4 on 5 only for the involved segment (hence, the most, enthusiastic visitors), while it only episodically exceeds 3.5 for the other clusters. A general trait in the satisfaction expressed lies in the generally greater appreciation of the visceral and behavioural stimuli provided by the Exposition (hence, sensory, emotional and pragmatic experiences), while the purely cognitive stimuli were generally less appreciated by visitors. Looking at the specific elements of satisfaction, pragmatic and sensory stimuli were appreciated the most by visitors, and particularly by the involved and extroverted segments for sensory and by the involved and introverted segments for pragmatic experiences, as shown in table 6. This outcome suggests a twofold idea. First, observing that involved visitors were the most satisfied and the indifferent the least satisfied, a general perception of an attitude by Chinese customers to associate the concept of ‘importance‘ to the concept of ‘likeability of being positively impressed’ or ‘sensitivity to stimuli‘ seems to emerge. This is particularly remarkable, as it provides an important cue for interpreting further studies on attributed importance and declared satisfaction with reference to Chinese tourists. Second, the World Exposition seems to have reached appropriate levels of performance with reference to aesthetics and the offering of moments providing visitors with an active engagement in the event, while it has fallen short in terms of cognition and emotional involvement; this element depicts a stronger focus for the organizers in relation to the objective of providing an impressive showcase rather than a more intimate and brain-intensive representation, which can be associated to the overall idea of the Chinese Government as placing pressure on the Shanghai World Exposition to show to the world the astonishing advances of the country itself and of Shanghai in particular [18].
Visitors' satisfaction on the experiential dimensions
The scale of satisfaction ranged from 1 being ‘Much worse than I expected’ to 5 being ‘Much better than I expected’
5. Managerial Implications
The first, paramount managerial implication of this study lies in the need for tourism operators, when approaching the segmentation of Chinese tourists, to go beyond demographics in their segmentation. The results, in fact, testify to how deeper motivations and preferences lie at the basis of tourists' satisfaction. These elements, actually, are barely understandable by a third party; hence, tourist operators - in order to fully accomplish the explanatory potential of experiential approaches to segmentation - must develop a genuine awareness of customer needs, which means a market orientation [40–42].
The second important implication in this study lies in the actual heterogeneity observed in the sample: the analysis performed testifies to how different customers and visitors display different needs and, hence, require different experiences. This should imply the need for tourism operators to be able to “transform” their destination to meet the diverse expectations of visitors. This outcome is consistent with previous studies (e.g., [43]). This means that the managers of theme parks in China should reflect thoroughly about how to provide different visitors with different experiences: almost one quarter of our sample were introverts and almost another quarter were extroverts; this means that the development of different experiences to different targets may represent a valuable lever of development for the increasingly demanding Chinese targets.
6. Limitations and Conclusions
This study is subject to two main limitations that suggest the next steps in developing this research stream. First of all, the World Exposition represents a mega-event, and these kinds of events depict several specificities influencing the level and the nature of the expectations of visitors (e.g., [39]). Hence, the study and the measure should be replicated in different contexts in order to test their adaptability to different tourism offers. A second limit lies in the testing of the segmentation model only on Chinese tourists; despite this choice, it provides a significant insight considering how Chinese tourists are becoming one of the most important tourism populations in the world, though the experiential mix sought for could be affected by cultural parameters to be tested in further cross-national samples in order to validate the segmentation proposed.
Despite these limitations, we believe that the study provides further evidence of the diminished effectiveness of demographics in approaching tourist segmentation and that it suggests once more that psychographics are better predictors of visitors' satisfaction. In particular, we propose an experiential approach that, thanks to its general conceptual scope, seems ideally replicable in different contexts and catches the emerging experiential view of marketing (e.g., [20]). Such a view provides a tool that is not only able to interpret visitors' satisfaction in a more comprehensive way but also, through a learning process and a thorough analysis of the experience provided by the venue, a useful tool for tourism planning and marketing, at an event, a venue and a territorial level.