How Does Destination Image Evolve? Introducing the Co-creation of the Destination Image Approach

The co-creation concept has been used in different areas of studies like design (Sanders and Stappers, 2008), product development (Hoyer, Chandy, Dorotic, Krafft, & Singh, 2010), innovation (Lee, Olson & Trimi, 2012), brand (Hatch & Schultz, 2010), public sector (Osborne, Radnor, & Strokosch, 2016; Wise, Paton, & Gegenhuber, 2012), engagement in service delivery (Auh, Bell, McLeod, & Shih, 2007), and service recovery (Dong, Evans, & Zou, 2008; Heidenreich, Wittkowski, Handrich, & Falk, 2014). Co-creation has been, perhaps, mostly connected to the concept of value in the literature (e.g., Grönroos & Voima, 2013; Payne, Storbacka, & Frow, 2008; Prahalad & Ramaswamy, 2004; Vargo, Maglio, & Akaka, 2008; Yi & Gong, 2013).

In value co-creation, consumers assume an active role and create value together with the firm (Ranjan & Read, 2016). Edvardsson, Tronvoll, and Gruber (2011) stated that value co-creation should be understood in a social context that is shaped by social forces and reproduced in social structures. Ind and Coates (2013) extend this scope and believe that co-creation is to be recognized by taking into account the perspectives of different disciplines and approaches like management science, innovation, design, and literary theory rather than adopting a narrow view of the concept. They also see co-creation as a way of creating value through co-opting the skills and creativity of individuals. The co-creation process is more related to customer interaction, and thus it moves away from a firm-centric view of customer orientation (Binkhorst & Dekker, 2009; Chathoth, Altinay, Harrington, Okumus, & Chan, 2013). Improvements in information technologies also ease the co-creation process and enable new forms of producer–consumer collaboration (Cabiddu, Lui, & Piccoli, 2013; Füller, Mühlbacher, Matzler, & Jawecki, 2009; Füller, 2010).

In tourism literature, co-creation studies have tremendously increased over the last decade (Campos, Mendes, Valle, & Scott, 2018). Service-Dominant (S-D) logic was mainly used as a base in these studies (Cabiddu et al., 2013; Chathoth et al., 2013; Grissemann & Sauer, 2012). S-D logic suggests that value occurs through customer engagement at every stage of the value creation process (Chathoth et al., 2013; Grönroos, 2006). Empirical evidence on value co-creation, on the other hand, is limited in tourism literature (Binkhorst & Dekker, 2009; Liang, 2017), and the full potential of co-creation has not been fully explored in the tourism industry (Hoyer et al., 2010). The literature review of Campos, Mendes, Valle, & Scott (2018) shows that the scope of this research varies from specific tourism experience contexts to broad industry or destination analyses. These studies support the idea that the co-creation of experiences impacts the satisfaction and loyalty of the tourists positively (Grissemann & Sauer, 2012) and thus results in increased value for consumers and others (Prebensen, Vittersø & Dahl, 2013). As it is seen from the literature, co-creation is based mostly on the interaction of the customer and company, where experience value is discovered and innovatively used in the production process of new goods or services (Campos et al., 2018). However, the interaction does not only occur between customer and company but also occurs between the individuals (Bitner, 1992). Moreover, co-creation should not be considered as if it were only a part of the design, production, or value creation. Although these are highly critical elements of competitiveness in the market, they are only effective if the beliefs and perceptions of the customers about a product or brand are positively distinguished from the others. Beliefs and perceptions are developed through many interactions with other people and organizations, so beliefs are also co-created because humans are social creatures that affect and are affected by other humans’ beliefs, knowledge, and perceptions. If this is the case, it is reasonable to assert that the image of a product or destination can be an output of a co-creation process. As the interaction level gets higher, the co-creation process is felt more intensely. Furthermore, the magnitude and content of the interaction differ under different conditions and contingency factors.

The main idea of the TALC model is to show the development stages of destinations and each stage's characteristics. As the destination passes from one stage to another, it means that the destination is not the one it used to be anymore. For instance, changes occur in product development strategies (e.g., in numbers or variety), pricing strategies (e.g., premium or competitive pricing), promotion strategies (e.g., creating awareness or branding), and distribution strategies (e.g., direct or through intermediaries). Besides structural features of the destination (e.g., political and economic), the development level of infrastructure and technology can affect the visitors’ experiences and images as well (Murphy, Pritchard, & Smith, 2000; Stevens, 1992).

Kozak and Martin (2012) mentioned the differences that happen in the stages of the life cycle. According to the authors, in the introduction stage, product categories represent recent innovations, prices are high, distribution is limited, and word-of-mouth communication exposes the destination to the next tourist wave. The growth stage is where destination awareness increases and more attention is given to the variety of products (i.e., food or accommodation). In this stage, happy tourists mean more positive word-of-mouth and repeat visitors. As the tourism destinations reach the maturity stage, they are no longer differentiated from destinations that offer similar experiences. In the decline stage, destinations have fewer visitors and revenue. Therefore, they need to create unique niches to maintain their existence (Kozak & Martin, 2012).

There are also differences in the perceptions of destinations that are in different stages of their life cycle (Dolnicar & Grabler, 2004; Yun & Zhang, 2017). Such differences can also be observed in studies that search for the personality and image of destinations. For instance, a mature destination like London was characterized as being open-minded, unorthodox, vibrant, and creative (Hall, 2004), and Paris was characterized as romantic (Morgan & Pritchard, 2002). These personality traits need evolutionary periods to develop. It is not possible for a destination at its early stage of development to have such a sound perceived personality.

Thus, it can be inferred that beliefs, perceptions, and opinions about a destination change as the destination matures. The main reasons for this change can be explained by looking at the different paths of the co-creation process that destinations follow throughout their life cycles. Cole (2012) mentioned the missing ingredient of the TALC equation in his paper, which was, as Butler (2005) stated, written in an unpublished paper of Brougham and Butler (1972). In this equation, dV/dt = kV(M-V), the tourism development process is calculated with the help of the ‘k’ parameter, which is called the ‘telling rate, or the spread of knowledge of resort’. This means that there will be few people receiving the message for the first time, and so the increase slows down as V (number of visitors) approaches M (maximum number of visitors) and market penetration of a destination is complete.

There might also be differences for destinations offering similar tourism products or for destinations at the same maturity level because of their unique characteristics (i.e., geographical or cultural). Besides, contingency factors related to destinations or visitors can be influential in this process. Therefore, studying the destination image formation process and contingency factors requires the consideration of a destination's development level and unique characteristics.

The contingency factors can be handled with two dimensions: scope and time of visit. The scope is divided into two groups that are originated from the person and the destination. The time of visit is divided into three phases: before, during, and after visiting the destination (Table 1). Person-related contingency factors such as familiarity, age, education level, motivation, and origin of country mainly affect the DI before visiting the destination (or for non-visitors). Perceived risk and safety, trust, physical and cultural proximity, usage of emotional triggers, and destination offerings are among the destination-related contingency factors of this period. During the visit, perceived quality and satisfaction are the major personal contingency factors; destination governance and creating memorable experiences are among the destination-related contingency factors. The past experience could be the strongest personal contingency factor after the visit; this is called the ‘primary image’ by some authors (Kislali, Kavaratzis, & Saren, 2016). This experience, once shared with other people, affects other people's thoughts about the destination. Destination image differs between those who have never visited a destination and those who have visited one or more times. Studies show that even if people have an image of a place before they visit the destination, this image may change after the visit (Yilmaz, Yilmaz, Icigen, Ekin, & Utku 2009). This change is the result of the personal experience with the destination through the interaction he or she has with other people or institutions in the destination. With the increase in the number of visits, the image may still change, but this change would be not significant (Santana & Gosling, 2018). Organizations, like DMOs, try to use Customer Relationships Management (CRM) effectively to create a favourable image during this period.

Impacts of contingency factors may differ for the various stages of the destination's life cycle as well. For instance, motivation or interaction value is quite different for explorers in the early stage of the destination's development than for the visitors of a mature destination. Likewise, destinations act in quite different ways throughout their life cycle, e.g., for product development or communication strategies. Therefore, the DI formation process can be said to be unique for all destinations due to the different combined effect of the contingency factors for destinations, as well as the different paths the destinations go through in their life cycle.

Destination image has been studied by many researchers for over fifty years which makes it one of the most popular areas of study in tourism literature. The cognitive and affective dimension of destination image, referring to beliefs and knowledge about a travel destination's attributes and emotions or feelings attached to the destination, respectively, have been well formalized in these studies (e.g., Baloglu & McCleary, 1999; Beerli & Martin, 2004; Echtner & Ritchie, 1991; Gallarza, Saura & Garcia, 2002; Gartner, 1994; Santana & Gosling, 2018; Tasci, Gartner & Cavusgil, 2007; Wang, Li & Lai, 2018).

Despite the vast amount of research on destination image, the process of destination image formation attracted less attention. The main contribution to our body of knowledge to the destination image formation process may be rooted in the study of Gartner (1994), who categorized the information sources as over induced, covert induced, autonomous, unsolicited organic, solicited organic, and organic information sources. After-wards, the impact of information sources on destination image has been investigated widely (Yilmaz & Yilmaz, 2020). Impact of induced sources (e.g., Beerli & Martin, 2004; Lee, Busser, & Yang, 2015; Santana & Gosling, 2018; Yacout & Hefny, 2015), autonomous sources (e.g., Hudson, Wang, & Gil, 2011; Kim & Kerstetter, 2016; Lee et al., 2015), and organic sources (e.g., Baloglu & McCleary, 1999; Lee et al., 2015; Santana & Gosling, 2018) on destination image formation have been researched. Moreover, the number of information sources that the people used may be influential in the destination image formation process. There are pieces of evidence that the amount of information sources influences cognitive image positively (Baloglu & McCleary, 1999; Santana & Gosling, 2018). However, no evidence was found for its effect on affective or unique images of destinations (Santana & Gosling, 2018). On the other hand, the credibility of sources is defined as the believability that the destination management is willing and capable of delivering on its promises related to a specific destination (Veasna, Wu, & Huang, 2013), which can contribute to destination image formation process (Spry, Pappu & Cornwell, 2011).

Although information sources are one significant determinant of destination image formation, impacts of many other constructs have been researched as well, such as the effect of pictural elements (MacKay & Fesenmaier, 1997), destination personality (Kim, Malek, Kim, & Kim, 2018), country image (De Nisco, Mainolfi, Marino, & Napolitano, 2015; Palau-Saumell, Forgas-Coll, Amaya-Molinar, & Sánchez-García, 2016), cultural value (Yacout & Heftny, 2015), place identity (Foroudi, Akarsu, Ageeva, & Foroudi, 2018), familiarity (Tan & Wu, 2016), and event quality (Jin, Lee & Lee, 2013; Moon, Forgas-Coll, Amaya-Molinar, & Sánchez-García, 2011). It is evident in the literature that the destination image formation process is a phenomenon affected by various intrinsic and extrinsic factors that need to be investigated from a different point of view. In this study, the destination image formation process is treated from individual and destination perspectives. In the following sections, firstly, the co-creation process of destination image will be introduced, and then impacts of this process in the development stages of destinations will be discussed through twelve co-created destination images (CoDIs).

Each person has his or her own networks and these networks behave as guides through life (Binkhorst & Dekker, 2009), which are also effective in shaping people's beliefs, including about places. The nature of interactions is critical in this regard. Interactions take place between people to people, an organisation to people, and other stimuli to people (like movies or books). Each interaction makes more or less a contribution to one's belief in an object positively or negatively, and it creates an interaction value that is determined by the power and impact of the interaction.

If we could calculate the interaction value, then formula (1) can be used for this purpose. Formula (1) indicates that interaction value is the result of the power of interaction and the impact of the interaction. Power of interaction refers to the importance given to the interacted person or organisation (source), whereas the impact of interaction refers to the magnitude and direction of the interaction. The power of the interaction is related to the trustworthiness of the source. Impact, on the other hand, is related to the amount (much or less) and content (positive or negative) of information in the interaction process. Interaction value occurs as the result of this process. (1) Xi=aij·pijj=1,..........,k \matrix{{{{\rm{X}}_{\rm{i}}} = {{\rm{a}}_{{\rm{ij}}}} \cdot {{\rm{p}}_{{\rm{ij}}}}} \hfill & {{\rm{j}} = 1,} \hfill \cr } ..........,{\rm{k}} where X is interaction value; i is the person who is interacting; a is the power of interaction; p is the impact of interaction; j is who is interacted with, and k is the number of the interaction of i person.

The impact of interaction (p) could be positive or negative. Therefore, it leads to interaction value to be either positive or negative. As the magnitude of information amount passed in the interaction process increases, the interaction value is affected highly in positive or negative directions. Power of interaction (a), regarded as the importance of interaction, intensifies the total interaction value, which is affected by the trustworthiness of the interacted source. People tend to give higher importance to an interaction process when they feel trust in a person or organisation. Hence, believability and trustworthiness of the source determine the importance and, consequently, the power of interaction.

Adolphs (2003) states that according to the social information processing system, people make distinctions between different kinds of information, categorize and generalize them, and lastly, incorporate them with their past experience. Reappraisal and self-regulation are specific feedback modulation modes in which social stimuli evaluation and emotional response can be voluntarily affected.

A person has many interactions throughout his or her life, and the combined effect of these interactions leads to having a total belief about an object (i.e., product, company, or destination), which is called a self-regulated image (SI). If it is about the image of a place, a self-regulated image occurs through the composite value of all interactions, which are directly or indirectly related to that place. It then plays an active role in the co-creation process of the destination image. SI evolves along with the interactions and values assigned to these interactions.

A self-regulated image about a destination is the individual's total beliefs and opinions, which are the results of positive and negative interaction values of all interactions. Hence, it is the composite value of all interactions that a person has about a destination. Formula (2) indicates the self-regulated image of a person about a destination. (2) SIi=∑ι˙kXi {\rm{S}}{{\rm{I}}_{\rm{i}}} = \sum\nolimits_{\dot \iota }^k {{X_{\rm{i}}}} where SI is self-regulated image, X_i is interaction value of i. person, and k is the number of interaction of i person.

Destination image is the aggregate of all self-regulated images attributed to the destination's cognitive and affective character and occurs through a co-creation process. Co-creation is a multidirectional and iterative process in which interaction values determine the self-regulated image, and then the aggregate of self-regulated images determines the destination image. It is also a dynamic process. It evolves as new interactions and self-regulated images are included in this process. The direction of the image change depends on the impact (i.e., positive or negative) and the power (i.e., importance) of interactions. Once the self-regulated images are created and interactively shared in networks (for example, through WOM, e-WOM, etc.), the destination image co-creation process begins, and the co-created destination image (CoDI) is built. Affective CoDI evolves when emotional-based interactions are dominant in self-regulated images and co-creation processes. Similarly, cognitive CoDI evolves when physical elements of the destination are the main subjects in the co-creation process. Formula (3) indicates CoDI is formed through the co-creation process, which is comprised of the total self-regulated images. (3) CoDI=∑ι˙nSIi {\rm{CoDI}} = \sum\nolimits_{\dot \iota }^n {S{I_{\rm{i}}}} where CoDI is co-created Destination Image; n is the total number of people; organisations, and other sources (e.g., movies and books) in the co-creation process, and Si is the self-regulated image of i person.

There is limited interaction between visitors and non-visitors of a new destination. Interaction from visitors to non-visitors determines the self-regulated image creation and co-creation of the destination image. This is the first stage of DI creation, so the image of the destination has not been shaped yet. It is highly affected by the first visitors’ opinions, so first impressions of first visitors reinforce the images of followers, and DI is getting shaped. The level of consistency of thoughts determines the solidity of the image for the further stages of the destination life cycle. At this stage, there is very little or no interaction between people and organizations or other image formation agents (e.g., books, news), so only limited interaction between people is influential in the formation of DI. If there is a low co-creation process, there will be no image at all at the beginning, and there will be tentative positive or tentative negative images following the first visitors of the destinations that depend on whether the co-creation process is positive or negative.

The interaction level between people gets higher at this stage. Besides, there is a moderate level of interaction between people and organizations because organizations (private and public) started to emerge and understand the need for communication with current and prospective customers. They try to use both induced and autonomous sources at this interaction. Image of destination in the minds of people becomes stronger, and the co-creation process of destination image accelerates at this stage. Other image formation agents also have a moderate impact on DI. If the low interaction goes up, self-regulated images cannot be formed easily, and CoDI will be Indecisive. On the other hand, in the case of a high positive co-creation process, the power and impact of interactions will be high and positive, which leads positive self-regulated image. CoDI is called a Promising Image because the aggregate of self-regulated images is positive. On the contrary, a high negative co-creation process, which is the result of high negative interactions between people and the non-existence of organizational efforts to develop a positive image, leading CoDI to be an Imminent Image.

The popularity of destination could be low due to the low interaction level; thus, fragile self-regulated images, which are more affected by contingency factors, occur. Aggregate SIs influenced by the emergent strategies of destination lead CoDI to be an Evasive Image. When a destination becomes a popular mature destination because of the high positive co-creation process, which includes high positive bi-directional interactions (people to people, organization to people, and others), this leads to high positive SIs. CoDI is at its most desired type a Premium Image. If the co-creation process is highly negative, the popularity of the destination decreases as the self-regulated images are affected by an increasing number of negative interactions; then, aggregate SIs turn to have a Dull Image as CoDI.

In the decline stage, CoDI could be Naught Image due to the low and less powerful interaction because of the low popularity of destination leaded self-regulated image, as if the destination never existed. There might be still medium to high positive interactions which could produce positive but not powerful SIs that barely change the behaviour intentions (e.g., visit/revisit the destination). Especially, nostalgic content of aggregate SIs leads to Evocative Image as CoDI. If the co-creation is negative, then it reinforces SIs to be negative that results in Distorted CoDI. Interactions, self-regulated images, and CoDIs in the stages of the destination life cycle are summarized in Table 2.