#Wanderlust : measuring the social media factors that influence tourism

(1)

Master Thesis

#Wanderlust: Measuring the Social Media Factors that Influence

Tourism

University of Amsterdam

Faculty of Economics and Business

MSc. In Business Administration – Marketing Track

Under supervision of: dr. Jonne Y. Guyt

Student: Diana Brosiu

Student number: 11083921

(2)

2

Statement of Originality

This document is written by Student Diana Broşiu who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3

Acknowledgement

By finalizing the thesis, I have reached the final step in achieving my master’s degree in Business Administration (Marketing Track) at the University of Amsterdam. The writing process of the thesis was very challenging, but rewarding at the same time. My interest in the topic was sparked by the growing importance social media has on our daily lives (including politics and economics) combined with my love for the travel industry. The specific focus of researching what are the most influential information sources for niche versus popular destinations managed to increase my curiosity for this field and continue my learning process.

I would like to take the opportunity to thank my supervisor dr. Jonne Y. Guyt for his constant support, feedback and social media expertise that guided me through the writing process. Furthermore, I would like to thank my family and friends for their endless encouragements and for having patience and faith in me.

I hope you will all enjoy reading this thesis and become interested in this important research field that has numerous practical implications.

Kind regards,

Diana Broşiu

24th June 2016

(4)

4

Abstract

Social media is becoming increasingly popular and increasingly influential consequently. Its influence spheres include economics, politics, communication patterns, dating information research and even stress levels. Tourism is no exception to this. Given the time people spend on social media, it has become an effective communication channel for marketers for branding, but also crisis management since it enables them to directly communicate with consumers.

The research tried to analyze what are the information sources most influential for travelers and how does their effect vary for niche versus popular destinations. In order to achieve this, a literature review was used to determine the key influence factors. Therefore, National Tourism Organizations and Travel Guides are the factors chosen. Afterwards, a regression analysis was done, that featured time lag effects (they were added due to the booking pattern of travelers). The results suggest that National Tourism Organizations are more effective for popular, while Travel Guides are better for niche destinations. For the analysis, six different countries were studied and the data collected reflected actual visitors number and the social media activity for the 2014 – 2015 period on a monthly basis

The author discusses further research directions and the limitations of the study at the end of the paper, together with managerial implications.

Keywords: tourism marketing, social media, Instagram, destination marketing, country of origin, information sources, expert publications, travel guides, national tourism organizations, time lag effects, regression analysis

(5)

5

Table of Figures and Tables

Figure 1: Conceptual Framework ... 10

Figure 2: Branding strategy for a destination ... 19

Figure 3: Overview of analyzed data – Singapore ... 35

Figure 4: Overview of analyzed data – Canada ... 35

Figure 5: Overview of analyzed data - Germany ... 35

Figure 6: Overview of analyzed data - Ireland ... 35

Figure 7: Overview of analyzed data - Serbia ... 36

Figure 8: Overview of analyzed data - Croatia ... 36

Table 1: Overview of information sources per each country ... 32

Table 2: Overview of Instagram information sources ... 33

Table 3: Regression Analysis - No Lag ... 38

Table 4: Regression Analysis - One Month Lag ... 38

Table 5: Regression Analysis - Two-Month Lag ... 39

Table 6: Regression Analysis - Three-Month Lag ... 39

Table 7: Regression Analysis - Four-Month Lag ... 40

Table 8: Regression Analysis - Five-Month Lag ... 40

Table 9: Regression Analysis - Six-Month Lag ... 40

(8)

8

1. Introduction

In recent years, tourism has recovered from the slowdown caused by the 2008 recession and other macro events (epidemics) and continued its growing trend. In the last 60 years, “tourism has experienced continued expansion and diversification, to become one of the largest and fastest-growing economic sectors in the world” while “many new destinations have emerged in addition to the traditional favorites of Europe and North America”, as shown in a report by United Nations World Tourism Organization (UNWTO, 2015).

It is already known that the new generation – Millennials or Gen Y – has a different spending pattern than their predecessors (Goldman Sachs, 2015). For example, they prefer to have access to goods, rather than to own them, “prioritizing access over ownership”, thus having resources available for other purposes. In terms of expenditures, they prefer to invest in experiences such as traveling, “things that cannot be taken away from them” (Suddath, 2015). At the same time, being a generation that has grown up with access to internet, they are obviously more prone to use the web to research before making a purchasing decision. They rely more on social media (34% vs. 16% of previous generations) (Goldman Sachs, 2015) and blogs (33% rely on blogs vs. 3% on TV) (Schwabel, 2015). Moreover, they also prefer to purchase online as shown by their previous behavior: in 2013 and 2014 over 90% of millennials have made an online purchase (Goldman Sachs, 2015). Thus, it is safe to say that tourism will continue gaining importance.

Currently, the literature regarding the impact of social media on tourism is in incipient phase, therefore there are numerous gaps in the literature regarding this topic (Zeng & Gerritsen,

(9)

9 2014). One of the most important is a classification regarding the power of influence of various social media factors on the purchasing decision of travelers. Usually, papers published are focusing on only one factor, such as destination marketing organization use of social media (Hays, Page, & Buhalis, 2013) or user generated content (Akehurst, 2009). Of course, the relationship between the social media and tourism can be mediated by different factors such as the type of destination (whether it is a popular destination or a more niche one). Therefore, the proposed research question for this paper is:

How is the tourists’ purchase decision influenced by social media factors (content generated by destination marketing organizations, and travel sites) and how do these effects compare between niche and popular destinations?

The present study will make a classification of the most important social media factors based on their power of influence over the travelers’ purchase decision. At the same time, it will offer a better understanding of the role of expert publications (such as travel guides) in today’s interconnected world. In order to achieve this, a social media network research will be conducted. The research will have a quantitative approach. Social media data will be collected for different countries across a two-year period on a monthly basis in order to enable for a longitudinal comparison to be made across different type of destinations (niche versus popular destinations).

The conceptual framework on which for this research can be observed in Figure 1 at the end of this chapter. It is easily visible from it what the goal of the current paper is. The independent variables are represented by the content produced by National Tourism Organizations and Travel Guides. The author examines what is the effect of this factor on actual

(10)

10 visiting behavior moderated by the level of the destination’s popularity. The hypotheses are further explained in the Literature Review chapter.

In conclusion, a deeper understanding of the influence social media has over tourism could help marketers better promote and address the needs of their consumers. This paper aims at increasing the understanding of this topic and generating directions for further research. Most importantly, the classification of social media factors achieved can be tested in subsequent studies to generalize to other industries.

In order to achieve these contributions, the research proposal is divided in two important parts. First, the social media and tourism literature will be reviewed. Secondly, the proposed research design will be described together with the research schedule. The paper will of course end by detailing the limitations and possible further research directions and by offering a conclusion to the analysis.

Number of Visitors National Tourism Organization

Travel Guides

Destination’s level of popularity

H3a H3b

H2 H1

(11)

11

Note: There are many other factors (control variables i.e. macro-trends or global events) which also impact travelers’ buying behavior. However, within the scope of our study the author does not capture them in our analysis and solely focus on social media.

2. Literature Review

2.1 Tourism Marketing

The author decided to focus her research on the tourism industry since it is one of the most dynamic and one of the fastest to recover after an economic crisis such as the one the world economy faced in 2008 (Dekimpe, Peers, & van Heerde, 2016). Also, given the purchase behavior of the new generations (millennials) that value experiences over products and spend more time on internet, but especially on social media, being also more easily influenced by it (Goldman Sachs, 2015; Suddath, 2015; Schwabel, 2015), it is safe to say that tourism is a very interesting industry to study in the next couple of years since it will probably have a significant growth overall in the following period. This trend was already visible in 2014 as a report by UNTWO (2015) perfectly illustrates.

Travelers are influenced by a series of factors when they are evaluating a particular destination, such as destination attractiveness as introduced by Ritche & Zins (1978). Marketers’ goal is to persuade tourists to choose their products or services while at the same time assuring a high level of tourist satisfaction (meaning there should not be a discrepancy between the

(12)

12 expectations and the reality) as defined by Pizam, Neumann, and Reichel (1978). Therefore, it is necessary for marketers to better understand the differences between the experiences vistirors have or look for, in order to better satisfy their needs, as emphasized by Taylor (1980).

Given the turbulent times that everyone is living in, where terrorist attacks are becoming increasingly frequent and there is political unrest in numerous countries (Brexit is the perfect illustration of this), marketers must try to address travelers’ concerns and reassure them that they are safe while visiting. This is necessary so that people will feel secure enough to make the holiday booking, since people tend to be risk-averse as demonstrated by Kahneman & Tversky (1979). Egypt was also affected by a political crisis that has decreased the number of visitors. Avraham (2016) has analyzed in depth the strategies employed by the Egyptian officials to address this issue. The approach taken in the paper is of “multi-step model for altering place image” as descriebed by Avraham and Ketter (2008). Thus, the Egyptian marketers have taken a number of measures to change the country’s image and they can be grouped in the following categories: source (cooperating and developing PR or media relations; stoping and/or addresing negative media; finding alternatives for traditional media and replacing it), message (ignoring the crisis; reducing or limiting the scale of the crisis; positive brand associations; hosting special events; geographic dissaciation) and audience (emphasizing common traits with audience; adressing new market niches).

Even from the approach taken by the Egyptian marketers, the importance of social media starts to become visible as it can have a central role in the source category of possible approaches. Therefore, the author finds it useful for marketers to include also social media in the marketing strategies that they employ, especially given the growing importance of the channel (Pew Research Center, 2015). Furthermore, the need for a better understanding of how tourism

(13)

13 marketing being changed by societal changes such as the rise in popularity of social media are emphasized also by Dolnicar and Ring (2014). Thus, they encourage researches to deep dive into the matter. In this affirmation they are not alone, since also Zeng & Gerritsen (2014) and Leung, et al., (2013) point our that further research in this area is needed.

2.2 Social Media

Social Media (SM) is one of the mega-trends that have changed the way people interact in today’s society. Web 2.0 has empowered consumers and has transformed the traditional marketing communication into a dialogue, by facilitating a communication channel between the brand managers and the users of social media networks (one-to-many communication) (Xiang & Gretzel, 2010). Now users have the possibility to self-select the content and advertising shown to them, which further complicates the challenges marketers are facing (Holt, 2016). Formally, SM can be defined as “a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0 and that allow the creation and exchange of User Generated Content” (Kaplan & Haenlein, 2010).

Tourism is defined by the United Nations World Tourism Organization (UNWTO) as the “the activities of persons traveling to and staying in places outside their usual environment for not more than one consecutive year for leisure, business and other purposes” (UNWTO, 1995). As recent years have demonstrated, tourism is easily impacted by macro trends. It has been slowed down by negative events such as the world recession in 2008 (UNWTO, 2013) or the

(14)

14 outbreak of epidemics (H1N1 for example, or more recently the Zika virus) (UNWTO, 2009; Dekimpe, Peers, & van Heerde, 2016).

Social Media impacts almost all aspects of our life and tourism is not an exception. A research found that of the consumers who used social media to research travel plans, only 48% stuck with their original plans and 52% of Facebook users were inspired by their friends’ photos to make their holiday choice and plans (Bennett, 2012). Despite of its importance, the literature for social media in tourism is currently still in its incipient phase, but it is continuingly gaining more attention in the recent period, as it is clearly visible from the literature review done by Leung, et al., (2013).

A simplified model of decision making process that any consumer goes through in order to make a certain purchase includes the following stages: problem recognition, search, purchase and post-purchase (Neslin, et al., 2006). Consumers cannot yet purchase travel packages directly from social media sites, but there are different methods by which social network platforms influence the way customers make the purchase decision for our travels, across all the decision stages, thus influencing the way they research, read and trust pieces of information (Sigala, Christou, & Gretzel, 2012; Xiang & Gretzel, 2010). At the same time, the users are collaboratively producing content regarding various destinations. In this context, social media forces marketers to reevaluate the strategies and tactics by which they address the needs of consumers, by challenging the current view on customer service and marketing/promotional processes. (Zeng & Gerritsen, 2014).

The main reasons consumers choose to use social media for organizing and taking vacations are the perceived benefits (social, functional and psychological and hedonic). This

(15)

15 relationship is moderated by a series of factors that can act as incentives for SM adoption and usage. Those factors include trust on the users’ information exchanged and written online; availability of the technology; the environment; altruism and individual predisposition (Parra-Lopez et al, 2011).

Instagram is one at of the latest social media platforms to appear. It was launched on the 6th of October 2010 and it quickly grew in popularity reaching 400 million users as of September 2015 (Instagram, 2015). It focuses on enabling their users to edit and share photos and videos publicly or private. One important feature of the platform is that it allows users to search for different (public) posts that contain a certain hashtag (for example: #wanderlust). Because of its focus on sharing visual content, it can be a powerful tool that can help create destination brand as it is argued by Fatanti and Syadnya (2015). Although their study lacks some methodological rigor, it does point out the fact that Instagram has an increasing influential power and that can encourage, leverage and systemize user-generated content (UGC) with the help of different strategies such as: use of hashtags (e.g. #VisitBali); sharing tourist photos in their official account; geo-tagging the photos shared; requesting for a response from the users; organizing promotions and so on.

2.3 Destination Marketing Organizations use of Social Media

Social media can be used by destination marketing organizations (DMOs) as a powerful tool to persuade travelers to choose a particular destination or hotel that they may have not previously considered. A study by Hays, Page, & Buhalis (2013) analyzed the social media

(16)

16 usage of the top 10 most visited countries by international travelers. Their results show that SM usage among DMOs is mostly experimental since they are still at the initial stages of utilizing SM to promote their destinations, thus confirming Gretzel et al.’s (2006) proposition which affirms that DMOs still need a greater understanding of how to better utilize social media in their promotional campaigns. At the same time, the strategies employed by DMOs varied extensively, most of them being rudimentary, from the analyzed national tourism organizations analyzed, only two were an exception from this. Those two National Tourism Organizations (NTOs) do provide a series of best practices, but there is still room and need for further research that could help NTOs be more creative and use social network sites (STSs) to their maximum capacity. The quality high variation in the content produced by NTOs was observed also by the author in the case of the countries analyzed in the paper.

There are numerous studies done on how to better manage a brand on social media in order to get a better response from consumers (Kaplan & Haenlein, 2010; Hollebeek, Glynn, & Brodie, 2014; de Vriesa, Genslera, & Leeflang, 2012) that can help marketers in NTOs create a brand for the desired destination. One of the important findings relate to enhance the popularity of a brand post. The drivers found by de Vriesa, Genslera, & Leeflang (2012) are: interactivity, vividness and position of branded post.

Destination Branding

One method that marketers consistently use to promote a particular destination is branding. By adopting branding, marketers can achieve different goals, such as: brands can

(17)

17 simplify consumers’ choice, guarantee a certain quality level, reduce perceived risk and/or increase consumer’s trust (Keller & Lehmann, 2006). Keller & Lehmann (2003) consider brands to be “the most valuable assets for any firm” for their ability to generate long-term income for the company. Given that the hospitality industry is service based, it translates into a higher perceived risk by the customers because of its intangible and experiential nature (Murray & Schlacter, 1990). Therefore, brands are very important in the tourism and hospitality industry since they have the role of reducing the perceived risk of consumers when they decide to book a holiday (Kahneman & Tversky, 1979).

However, since the relationship between marketers has changed drastically after the internet boom, practitioners need to be more aware of the network dynamics that govern social media platforms. Most importantly, people with similar interests can gather in groups and discuss about topics of interest to them, including traveling (Holt, 2016), thus transforming the once-isolated communities around the globe into influential online crowdcultures.

Due to the industry’s characteristics, there are several limitations that marketers need to take into consideration when developing the marketing strategy for a particular location. A few of them are:

 Tourism is dependent of macro-environmental factors (politics, disease outbreaks, terrorism, currency fluctuations, climate conditions) (Dekimpe, Peers, & van Heerde, 2016; UNWTO, 2009)

 Inherited names and history (heritage), culture and country of origin (COO) perception cannot be easily change (Balakrishnan, 2009)

(18)

18 These are aspects that marketers need to address with the campaigns promoting a specific location: they need to highlight the positive aspects from their cultural background and adapt their marketing efforts to the fluctuation of various macro-environmental factors. Christou (2015) has proved that a strong travel social media contributes to travelers’s brand loyalty. Therefore, it is necessary for marketing professionals to invest their efforts and budgets in building a trustworthy social media brand for their destination designed especially for the desired target group in order to obtain a maximum return on investment.

Destination branding has become a tool vital for marketers trying to differentiate a city or location since the global competition between places is growing continuously (Hultman, Yeboah-Banin, & Formaniuk, 2016). As Michelson & Paadam (2016) conclude, it is necessary to develop in paralel the destination brand and the construction of the symbolic capital to leverage the interconnectedness between the two. There are a series of methods that can be utilized in order to achieve this goal. The main types of engagements that Michelson & Paadam (2016) propose are: engamenents with images prima facie, engagements with local offers of products and services; engagements with spatial opportunities. Of course, the first type of engagements (engagements with images prima facie) can be shared also in an online environment, taking advance of today’s technology advancements.

Of course, in order to successfully develop a strong social media brand or a strong destination brand on social media it is important to first have a very clear idea about what are the necessary steps that need to be taken. A very helpful and clear strategic framework for destination branding was proposed by Balakrishnan (2009), after an in-depth literature review that can serve as guidelines for marketers starting to develop the branding strategy for a particular destination. It is reproduced in Figure 2. The most important thing to emphasize is that

(19)

19 as Balakrishnan (2009) argues Vision should be starting point of the strategy development proces as without defining it properly, the rest of the strategic parts (customer targeting, positioning brand components and so on) will probably not be correctly defined, nor implemeneted correctly. Therefore, it would be improbable to obtain the maximum level of impact.

Figure 2: Branding strategy for a destination Source: (Balakrishnan, 2009)

As it was mentioned before, one of the techniques that marketers use to engage with consumers and to influence them to purchase is by generating branded content, both in the offline and online environment (Keller & Lehmann, 2006; Gensler, et al., 2013). By doing so, they can educate the consumer regarding several target attractions/activities, but also build a brand destination brand that will help bring more tourist in the country on long term. Therefore,

(20)

20 the quantity of marketer-generated content should have a positive effective on tourists purchase behavior. However, because in the niche condition people are more likely to not be informed regarding the destination than in the popular condition, it is more likely that for niche destination marketer-generated content will be more effective, consumers being more interested in receiving information about this type of countries. Based on this, the author formulates the first hypothesis of the paper:

H1: Marketer generated content by National Tourism Organizations will have a positive effect on the number of visitors.

2.4 Travel Guides

A travel guide is defined as “a book about a city, country, or area” (Macmillan Dictionary, n.d.). There are a number of popular travel guides such as Lonely Planet, Rough Guides, Frommers, Michelin Guide and In Your Pocket City Guides. Traditionally, these travel guides represented the main source of information used both before trip and during it (Hyde & Lawson, 2013). At the same time, printed travel guides have declined in sales and they were forced to offer a big quantity of content online on their sites due to the change in consumer behavior (Mesquita, 2012). Little research is known on the current role of travel guides in the age of social media, where the consumers rely on their smartphones to research, plan and enhance the traveling experience (Wang, Park, & Fesenmaier, 2012).

(21)

21 More recently, online travel guides have appeared. The most popular is TripAdvisor reaching 350 million unique monthly visitors (TripAdvisor, 2015). TripAdvisor is a platform where users share content (reviews, pictures), thus influencing each other. At the same time, on TripAdvisor also the marketers have the possibility to share pieces of information, but it is clearly signaled on the site which is marketer-generated and which is user-generated content (Miguens, Baggio, & Costa, 2008). The site’s popularity does not come as a surprise since it takes advance of people’s tendency to create online crowdcultures that influence each and become viral on social media (Holt, 2016).

By the nature of their business, travel guides offer advice and important information to tourists as to when to go to a particular destination, what to visit there, where to eat and so on. Therefore, they act as experts in their field, representing a reliable source of information and communicating with their customers accordingly. As Cialdini (2006) argues, experts (or authority) are among the universal factors identified by his research that can be used to easily influence other people. The other factors identified by him are: reciprocity, scarcity, consistency, liking and consensus.

Nevertheless, internet enabled users to connect and form specialized communities and has made expert information easily available. Consequently, one can wonder about what is the role of traditional travel guides (such as Lonely Planet, Rough Guides and Frommer’s) in today’s interconnected world. Do they still have a place on the market or will continue losing users to sites such as TripAdvisor that rely on user-generated content.

Since travel sites advertise all destinations and strive to provide useful and high-quality content, they are seen as impartial and as experts in field. Thus, if they officially recommend a

(22)

22 particular destination as the place that must be visited that year, then travelers are more likely to be influenced, especially if Cialdini’s (2006) theory about influence. Therefore, the following hypothesis has been proposed:

H2: Content generated by Travel Guides will have a positive effect on the number of visitors.

Of course, it is expected for differences between the niche and popular destinations to be observed. Given that in the niche condition, people are more likely to be uninformed regarding the destination than in the popular condition, it is more likely that for niche destination marketer-generated content will be more effective. This is probable because the information regarding niche destinations represents new and interesting information which people would like to receive but also to share it with their social group since traveling can be considered also self-relevant for the tourist (Berger & Schwartz, 2011; Chung & Darke, 2006). In the case of travel sites, people will have the tendency to react more positive given their lack of information regarding the touristic possibilities in a particular niche destination. Overall, due to the need for information, the tourists will be impacted more strongly by the content about niche destinations rather than the content about popular destinations. Thus, the author proposes the last two of the paper’s hypothesis:

H3a: The relationship between content generated by national tourism boards and the number of visitors will be influenced by the popularity of the destination.

H3b: The relationship between travel guides the number of visitors will be influenced by the popularity of the destination.

(23)

23

2.5 User-Generated Content

User generated content (UGC) is a powerful influence tool, 70% of global consumers say online reviews are the second most trusted form of advertising (Bennett, 2012). Also, the most trusted form of advertising is earned media such as e-WOM and recommendations from friends and family.

UGC can have also a negative impact on a destination brand. The content released by NTOs follows clear branding rules in order to accomplish previously stated internal goals, whereas UGC does not follow these rules and can either praise the location or critique it if the tourists were not satisfied with their stay (Lim, Chung, & Weaver, 2012). Because of this, consumers have different perceptions of a destination brand created by consumers than the one created by DMOs. This might end up being even more powerful than the brand image built by marketers.

Moreover, there are differences also in the profile of users that so share content online. Lo et al. (2011) concluded that users, who share pictures online from their travels tend to be better educated, earn a higher income and are younger than those who do not post. At the same time, they usually use different platforms to distribute their photographs. Even if currently, only a minority of people post pictures online, this percentage is expected to increase. Thus, this could have an even greater impact on other potential tourists in the planning phase.

(24)

24 There are various reasons why consumers choose to distribute content online regarding their traveling experiences. Motivational factors for instance differ depending on type of content and type of social media (Munar & Jacobsen, 2014). For example, altruistic and community-related motivations are most relevant for information sharing. Moreover, users feel the need to control their content, thus social networks allowing audience control are most popular for online sharing. There is low ‘real-time’ use of social media for holiday content sharing, probably due to situational factors. Interestingly, visual content is preferred for sharing by travelers rather than narrative or textual content.

A specific type of UGC is represented by blogs (or web logs) that do also have a high percentage of textual information in their content, not only visual. Nevertheless, the quality of the text written on the site can vary greatly among blogs, as well as the relevance for readers (Akehurst, 2009). In the case of blogs, it is easy to see how the “long tail” effect described above is applied. They do have an important role in distributing information relevant for travelers, but also for tourism and marketing managers. Other roles of blogs found in the research by Akehurst (2009) are educating tourists and facilitating tourism transactions.

Despite the disadvantages of UCG, Goh, Heng, & Lin, (2013) have compared marketer-generated content (MGC) with UCG for the apparel industry and have found MCG to be less effective (-22%) in persuading consumers. The study looked on what is the difference in impact of UGC vs. MGC on consumers’ purchase expenditures by looking at a Facebook page community. The differences recorded clearly showed that UGC exhibits a stronger influence power on consumer purchase behavior than MGC. It would be interesting to research whether this effect holds also for touristic services, not only apparel and other social media platforms.

(25)

25 In conclusion, UGC is a very important factor for social media that impacts consumers’ purchasing decision. Initially, it was part of the proposed conceptual framework and research question along with the other independent variables (content-generated by National Tourism Organizations and content-generated by travel guides). Unfortunately, due to the limitations in collecting the data, it was not possible to include also this factor in the analysis. A short literature review for this topic was done because of its importance for the social media literature.

2.6 Long Tail Effect

There are numerous social network sites (SNSs) and research (Xiang & Gretzel, 2010) has established that there is a core and a long tail in the distribution. The long tail is represented in this case is represented by a substantial number of less popular websites, while the core is made out a handful of “big players”. The “tail” manages to obtain profit by providing a greater variety than the core or the “short head” (Anderson, 2006). This principal can be easily compared to the Pareto rule: the top 20% make 80% of the profit, whereas the bottom 80% makes 20% of the profit.

The long tail principal can apply to destinations because there are some locations that continuously attract a larger number of tourists, even in the off-season period, whereas there are some other places that have a smaller number of visitors, but they can make profit by offering a higher-value selection of products/goods/services. Of course, a small village in the mountains cannot be compared to a major tourist destination such as Paris or Rome, but it does have unique

(26)

26 advantages, such as the scenery and less polluted and hectic environment. These can be perceived as unique attributes by some travelers who would prefer to go to a peaceful place rather than to an overcrowded touristic metropolis. In this example, the small village can be considered niche destination due to its characteristics.

Social media can be leveraged to promote niche destination marketing given its targeting capabilities (Lew, 2008). Moreover, it can help create a community of target consumers in order to build a sense of trust and a strong identity that will increase tourism in the desired regions over time. By considering the long tail approach, it can help better understand and explain the possible differences between niche and popular destinations.

2.7 Literature gap and Research Question

Because, as previously stated, the literature regarding the applications of social media in tourism is still in its incipient phase the gaps that are signaled by researchers (Leung, Law et al., 2013, Zeng & Gerritsen, 2014) are numerous. However, they accentuate the fact that quantitative research in general is needed in order to generalize the findings of previous studies. At the same time, it is important to point out that usually studies analyze only one influence factor for the tourism industry, not realizing across multiple factors.

However, Fotis, Buhalis, & Rossides (2012) have compared the influence power of various types of information (photo & video sharing websites, blogs, microblogs, social networking websites, wikis, and travel review websites) by surveying travelers from Russia and

(27)

27 Former Soviet States about their last holiday trip. The main differences versus the present study are:

(1) The researchers used a non-random survey to collect data, which means that the answers can influenced by cognitive biases and not represent the truthful image (2) The respondents are from the same region which means the results cannot be

generalized

(3) The research does not make a comparison between niche and popular destinations

For these reasons, there was an identified gap in the literature regarding social media in tourism about measuring the influence power and comparing the most important factors across multiple social media platforms for both niche and mainstream tourism. Thus, the proposed research question for this paper is:

How is the tourists’ purchase decision influenced by social media factors (content generated by destination marketing organizations, and travel sites) and how do these effects compare between niche and popular destinations?

The proposed main objectives of the study are the following:

1. Analyzing two of the most important factors based on previous research

In order for the results of the research to have a maximum impact on the industry, the most important factors will be considered. With the purpose of respecting time and data collection constraints for finalizing this research, only two factors were chosen for this analysis.

(28)

28 2. Measuring and comparing their power of influence across on one of the main social

media platforms

Once again, one of the main social media sites was chosen so that the influence level could be at a maximum level, while at the same time taking into consideration practical constraints like the availability of the necessary data.

3. Compare data across niche destinations and popular destinations

It is assumed by the author that travelers have a different behavior when they are considering visiting a niche country versus a popular country. Therefore, it is likely that the sources used for gathering information and the sensibility that they have to a particular information source vary as well depending on the level of the country’s popularity.

At this moment, the literature on the topic of social media in tourism is still in its incipient phase, but this topic is getting more and more attention from researchers (Zeng & Gerritsen, 2014; Leung, et al., 2013). There are a number of papers analyzing various singular factors of influence but none that realizes a comparison on one of the main social media factors. Due to easiness of data collection, Instagram was chosen for the present research.

2.8 Contributions

This research will contribute to existing literature by realizing a classification of these factors and will help establish different new directions for further research. Moreover, it will help create a deeper understanding of the differences between niche destinations tourism marketing

(29)

29 and mainstream destinations marketing, especially in terms of influence factors. Also, it will help fill the literature gap regarding quantitative researches on social media (especially, regarding the papers focused on the tourism and hospitality industry), thus having a higher validity by measuring actual behaviors, not reported behaviors (as in the case of surveys). By doing so, the author manages to captures all effects that impact the purchase behavior, even the ones that go unnoticed by consumers. Furthermore, the paper will offer a better understanding on the role and the influence power of traditional expert publications (such as travel guides) in the online environment.

One of the most important insights the travel industry will gain from the present research is the evaluation of the importance of travel for influencing purchasing decisions and how they compare with the content generated with other types of information source. Travel guides act as unbiased experts for the hospitality industry. Therefore, they can be more influential than the content or campaigns created by the national tourism boards.

At the same time, the results of the research will be valuable for managers as they will have better understanding of the importance of these factors and will be able to allocate funds to obtain the maximum impact on consumers. Therefore, the managers will generate a better return on marketing investment. For example, a hotel will know if partnering with the national tourism board or with a travel guide is more advantageous for its business. Both of the factors analyzed in this paper are easily influenced by marketers, thus can be leveraged effortlessly by the marketers to achieve long-term results.

For a better understanding and visualization of the goals of the current research, the author proposes the following conceptual framework (see Figure 1). It has been developed to test

(30)

30 the hypotheses, which were mentioned and explained. Moreover, the framework has also been developed to analyze the effect of content generated by National Tourism Organizations Travel Guides on the number of actual visitors. At the same time, the framework tries to test whether this effect would differ for destinations with different levels of popularity. The method used for collecting the data is explained in the next chapter.

3. Research Design and Methodology

3.1 Sample

The data collected will focus on 6 different countries – 3 popular and 3 niche – out of which 4 – 2 popular and 2 niche – have been featured in articles and been presented as top destinations for 2015 from 2 of the leading travel guides, namely Lonely Planet (Fildes, 2007) and Rough Guides. The following countries were chosen:

 Popular destinations:

o Canada – recommended by Rough Guides (2015) o Singapore – recommended by Lonely Planet (2014) o Germany

 Niche destinations:

o Serbia – recommended by Rough Guides (2015) and Lonely Planet (2014) o Ireland – recommended by Lonely Planet (2014)

(31)

31 These particular countries were chosen in order to ensure diversity, but also test the influence of different travel guides. It was necessary to be able to obtain monthly information for the number of tourists that visited each country analyzed during 2014 and 2015 since this data is used as input for the dependent variable (Purchase decision of tourists). At the same time, it was necessary that the National Tourism Organization (NTO) for each country should have an active Instagram account, because this data is used as input for one of the independent variable (National Tourism Organization Content). The division between popular and niche was realized based on the classification published by World Bank (2015) for 2014 regarding the annual number of tourists. For the Tourist Guides variable, the data was collected from the official Instagram accounts of two different international travel guides, namely: Lonely Planet (Lonely Planet - Instagram, 2016) and Rough Guides (Rough Guides - Instagram, 2016) for the period 2014 and 2015.

3.2 Research design

At the same time, the measurement of visiting behavior will be done using estimates provided by the tourism boards or by the national institutes of tourism as it is detailed in Table 1. This method was chosen because it provided the necessary data in a timely manner in order to achieve the main research objectives mentioned above while still maintaining a quantitative approach. The data collected in this manner was used to measure the effect of the paper’s only dependent variable (actual visiting behavior).

(32)

32

Table 1: Overview of information sources per each country

Country Popularity level National Statistics Institute Type of data collected

Singapore Popular (Singapore Tourism Board, 2015; Singapore Tourism Board, 2016)

International Visitors Arrival

Canada Popular (Destination Canada, 2016) International Visitors Arrival

Germany Popular (De Statis, 2016) Visitor Number

Ireland Niche (Central Statistics Office - Ireland, 2016)

International visitors – all means of transportation Serbia Niche (Република Србија, 2016) Number of Tourist Arrival Croatia Niche (Croatian Bureau of Statistics,

2016)

Number of Tourist Arrival

For the social media content, the data will be collected from Instagram since it is one of the leading social media platforms and for practical reasons (it can provide data from previous years and it will be accessed easily using specialized software by accessing the sites’ application program interface). The sources from which the Instagram data was collected are detailed in Table 2.

(33)

33

Table 2: Overview of Instagram information sources

Account Name Account Type Instagram citation Type of data collected

Visit_Singapore Popular national tourism board

(Visit_Singapore - Instagram , 2016)

Monthly number of posts and average number of likes per post Explore_Canada Popular national

tourism board

(Explore Canada - Instagram, 2016)

Monthly number of posts and average number of likes per post Germanytourism Popular national

tourism board

(Germany Tourism - Instagram, 2016)

Monthly number of posts and average number of likes per post Tourismireland Niche national

tourism board

(Tourism Ireland - Instagram , 2016)

Monthly number of posts and average number of likes per post Serbiatourism Niche national

tourism board

(Serbia Tourism - Instagram, 2016)

Monthly number of posts and average number of likes per post Croatiafulloflife Niche national

tourism board

(Croatia Full of Life - Instagram, 2016)

Monthly number of posts and average number of likes per post Lonelyplanet Travel guide (Lonely Planet -

Instagram, 2016)

Monthly number of posts mentioning the analyzed areas Roughguides Travel guide (Rough Guides -

Instagram, 2016)

Monthly number of posts mentioning the analyzed areas

With the intention of measuring the dependent variables (national tourism boards and travel guides), only one indicator was used, namely the actual number of monthly visitors

(34)

34 reported by the national statistics institutes. In order to measure the independent variables (national tourism boards and travel guides), the following indicators were used:

 The number of monthly posts created by the national tourism board on the chosen social media platform

By using this indicator, the author aims to measure the quantity of the marketing effort done by the organization

 The number of the monthly average of likes per posts created by the national tourism board on the chosen social media platform

By using this indicator, the author aims to measure the quality of the marketing effort done by the organization through the posts’ engagement

 The number of social media mentions by the travel guides

With the help of this indicator, the author tries to measure the amount of advocating done for a specific destination done by the travel guides only through the usage of small constant reminders

 The yearly recommendation of countries as the best place to travel in the following year

With the help of this indicator, the author tries to measure the amount of advocating done for a specific destination done by the travel guides only through the usage of a yearly recommendation

For a better understanding of the data that will be analyzed, the author has provided a series of graphs illustrated in Figures 3-8 highlighting the relationship between social media postings and the number of visitors. A general strategy cannot be understood from the graphics presented. Some countries (for example: Canada, Germany, Ireland, Croatia) have the tendency

(35)

35 of trying a social media push by increasing their effort a few months before the peak season or even exactly at the peak season for tourism. As Gretzen, et al., (2006) has also discovered, the strategies used by the national tourism boards seem to be lacking in terms of consistency for post number. In this case, it becomes probable that a lower return on investment will be the result of the marketers’ work.

Figure 3: Overview of analyzed data – Singapore Figure 4: Overview of analyzed data – Canada

Figure 5: Overview of analyzed data - Germany Figure 6: Overview of analyzed data - Ireland

0 200 400 600 800 100 0 120 0 140 0 160 0 0 10 20 30 40 50 60 70 Jan -14 Ma r-14 Ma y-1 4 Ju l-14 Se p -14 N o v-14 Jan -15 Ma r-15 Ma y-1 5 Ju l-15 Se p -15 N o v-15

Overview of analyzed data -

country level

Singapore - posts Singapore - visitors

0 500 100 0 150 0 200 0 250 0 300 0 350 0 400 0 450 0 0.0 0 10. 00 20. 00 30. 00 40. 00 50. 00 60. 00 70. 00 Jan -14 Ma r-14 Ma y-1 4 Ju l-14 Se p -14 N o v-14 Jan -15 Ma r-15 Ma y-1 5 Ju l-15 Se p -15 N o v-15

Overview of analyzed data -

country level

Canada - posts Canada - visitors

0 200 0 400 0 600 0 800 0 100 00 120 00 140 00 160 00 180 00 200 00 0 10 20 30 40 50 60 70 80 Jan -14 Ma r-14 Ma y-1 4 Ju l-14 Se p -14 N o v-14 Jan -15 Ma r-15 Ma y-1 5 Ju l-15 Se p -15 N o v-15

Overview of analyzed data -

country level

Germany - posts Germany - visitors

0 200 400 600 800 100 0 120 0 140 0 160 0 180 0 0 5 10 15 20 25 30 35 40 45 Jan -14 Ma r-14 Ma y-1 4 Ju l-14 Se p -14 N o v-14 Jan -15 Ma r-15 Ma y-1 5 Ju l-15 Se p -15 N o v-15

Overview of analyzed data -

country level

(36)

36

Figure 7: Overview of analyzed data - Serbia Figure 8: Overview of analyzed data - Croatia

3.3 Results and Analysis

In order to test the hypothesis, the author has run a regression analysis between the monthly data of travelers and the number of social media posts published by the National Tourism Organizations, the engagement of these posts (measured by the number of likes per post), whether or not that country was part of the yearly recommendation of any travel guide and the number of mentions a particular country has received on the travel guides Instagram account. All the data analyzed for the regression consisted only of monthly data for the 2014–2015 period.

Due to the purchasing behavior of tourists (usually, they do not purchase a vacation for tomorrow, but they buy in advance), time lag effects were integrated in the analysis. Therefore, the regression was re-run with different time lags – from no lag to a six-month lag. This range was chosen because the majority of tourists (82% in 2014 and 75% in 2015) decide to book a holiday in this time frame (Statista, 2016). The results of this analysis are summarized and reproduced below in the Tables 3-9.

0 50 100 150 200 250 300 350 0 2 4 6 8 10 12 14 16 18 20 Jan -14 Ma r-14 Ma y-1 4 Ju l-14 Se p -14 N o v-14 Jan -15 Ma r-15 Ma y-1 5 Ju l-15 Se p -15 N o v-15

Overview of analyzed data -

country level

Sebia - posts Sebia - visitors

0 500 100 0 150 0 200 0 250 0 300 0 350 0 400 0 0 50 100 150 200 250 Jan -14 Ma r-14 Ma y-1 4 Ju l-14 Se p -14 N o v-14 Jan -15 Ma r-15 Ma y-1 5 Ju l-15 Se p -15 N o v-15

Overview of analyzed data -

country level

(37)

37 There are a few general trends that emerge based on the analysis. These trends will be discussed in detail in the Discussion chapter. For the regression analysis with no lag, without the interaction between the independent variables and the moderator, almost all independent variables are significant. However, this changes when the interaction variables with the level of popularity (the moderator) is added. Only the influence manifested by social media posts and the travel guides’ yearly recommendation are still significant across all steps of the analysis. Curiously, the influence manifested by the travel guides is actually negative, while the one displayed by social media posts by the travel guides is positive when moderated with the popularity variable. This situation is replicated for two-month lag and three-month lag, while for the time lag between four months and six months, the situation was very similar with the only difference that social media posts by national tourism boards were no longer significant in step 3.

If the lag effect for one month is added, the results start changing slightly. Once again, without the effect of the moderator all the independent variables are significant. Although, when the interaction factor is added, this once again changes, the significance for travel guides’ mentions disappearing. It is important to point out that engagement on the national tourism boards Instagram account is significant even after adding the effect produced by the popularity and the relationship between engagement and number is moderated by the popularity. The two trends illustrated above are very interesting since they do add another step in the visiting decision process: shortly before the actual trip, the traveler researches online to find the key attractions that should be visited while traveling. Therefore, this is very important moment for marketers to target the tourist since the willingness to receive information about the country is probably higher than usual.

(38)

38

Table 4: Regression Analysis - One Month Lag

Note: Statistical significance: *p<.05; **p<.01; ***p<.001

One Month Lag R R2 β

Step 1 0.78 0.61 SM_posts -0.65*** Recommended -1.00*** SM_likes 0.19*** TG_Mentions 0.05** Step 2 0.84 0.71 SM_posts -0.53*** Recommended -0.79*** SM_likes 0.01*** TG_Mentions 0.08 Pop*SM_posts 0.43 Pop*SM_likes 0.12*** Pop*TG_Mentions -0.08 Pop*Recommended -0.22 Step 3 0.94 0.88 SM_posts 0.08 Recommended 0.07 SM_likes 0.14 TG_Mentions 0.04 Pop*SM_posts -0.2 Pop*SM_likes 0.01** Pop*TG_Mentions -0.06 Pop*Recommended -1.28 Popularity 1.54*** No Lag R R2 β Step 1 0.79 0.63 SM_posts -0.66*** Recommended -1.02*** SM_likes 0.16** TG_Mentions 0.09 Step 2 0.87 0.76 SM_posts -0.51*** Recommended -0.76*** SM_likes -0.02 TG_Mentions 0.07 Pop*SM_posts 0.50*** Pop*SM_likes 0.09 Pop*TG_Mentions -0.03 Pop*Recommended -0.25** Step 3 0.96 0.92 SM_posts 0.08 Recommended 0.06 SM_likes 0.11 TG_Mentions 0.03 Pop*SM_posts -0.11* Pop*SM_likes -0.02 Pop*TG_Mentions -0.01 Pop*Recommended -1.27*** Popularity 1.48***

(39)

39

Table 5: Regression Analysis - Two-Month Lag Table 6: Regression Analysis - Three-Month Lag

Two Month Lag R R2 β

Step 1 0.77 0.59 SM_posts -0.64*** Recommended -0.99*** SM_likes 0.21*** TG_Mentions 0.02 Step 2 0.82 0.67 SM_posts -0.53*** Recommended -0.79*** SM_likes 0.06 TG_Mentions 0.06 Pop*SM_posts 0.40*** Pop*SM_likes 0.1 Pop*TG_Mentions -0.1 Pop*Recommended -0.20* Step 3 0.91 0.83 SM_posts 0.04 Recommended 0.03 SM_likes 0.19 TG_Mentions 0.02 Pop*SM_posts -0.20** Pop*SM_likes -0.01 Pop*TG_Mentions -0.08 Pop*Recommended -1.22*** Popularity 1.47***

Three Month Lag R R2 β

Step 1 0.75 0.56 SM_posts -0.63*** Recommended -0.97*** SM_likes 0.23*** TG_Mentions 0.01 Step 2 0.80 0.64 SM_posts -0.53*** Recommended -0.78*** SM_likes 0.11 TG_Mentions 0.03 Pop*SM_posts 0.38*** Pop*SM_likes 0.07 Pop*TG_Mentions -0.07 Pop*Recommended -0.19* Step 3 0.89 0.78 SM_posts 0.02 Recommended 0.00 SM_likes 0.22 TG_Mentions 0.00 Pop*SM_posts -0.2* Pop*SM_likes -0.03 Pop*TG_Mentions -0.06 Pop*Recommended -1.16*** Popularity 1.40***

(40)

40

Table 7: Regression Analysis - Four-Month Lag Table 8: Regression Analysis - Five-Month Lag

Four Month Lag R R2 β

Step 1 0.73 0.53 SM_posts -0.620*** Recommended -0.940*** SM_likes 0.230*** TG_Mentions 0 Step 2 0.78 0.60 SM_posts -0.520*** Recommended -0.750*** SM_likes 0.14 TG_Mentions 0.02 Pop*SM_posts 0.370*** Pop*SM_likes 0.04 Pop*TG_Mentions -0.07 Pop*Recommended -0.19 Step 3 0.86 0.73 SM_posts 0 Recommended -0.02 SM_likes 0.25 TG_Mentions -0.01 Pop*SM_posts -0.16 Pop*SM_likes -0.06 Pop*TG_Mentions -0.05 Pop*Recommended -1.10*** Popularity 1.320***

Five Month Lag R R2 β

Step 1 0.70 0.49 SM_posts -0.59*** Recommended -0.91*** SM_likes 0.23*** TG_Mentions -0.02 Step 2 0.76 0.58 SM_posts -0.48*** Recommended -0.69*** SM_likes 0.16 TG_Mentions 0.03 Pop*SM_posts 0.43*** Pop*SM_likes 0.03 Pop*TG_Mentions -0.11 Pop*Recommended -0.23* Step 3 0.83 0.69 SM_posts -0.02 Recommended -0.04 SM_likes 0.26 TG_Mentions 0.00 Pop*SM_posts -0.05 Pop*SM_likes -0.06 Pop*TG_Mentions -0.09 Pop*Recommended -1.04*** Popularity 1.18***

(41)

41

Table 9: Regression Analysis - Six-Month Lag

In conclusion, there were two cases that emerged. The general trend was the first one where social media posts and the yearly recommendation by travel guides were the independent variable that had an interaction with popularity and it moderated the relationship the independent variables had with the number of visitors arriving in the target countries. Nevertheless, it is very interesting to point out that in the condition with one month lag the social media likes had an

Six Month Lag R R2 β

Step 1 0.67 0.44 SM_posts -0.55*** Recommended -0.86*** SM_likes 0.22*** TG_Mentions -0.01 Step 2 0.75 0.56 SM_posts -0.43*** Recommended -0.61*** SM_likes 0.16 TG_Mentions 0.03 Pop*SM_posts 0.5*** Pop*SM_likes 0.01 Pop*TG_Mentions -0.11 Pop*Recommended -0.28** Step 3 0.80 0.64 SM_posts -0.02 Recommended -0.03 SM_likes 0.25 TG_Mentions 0.01 Pop*SM_posts 0.07 Pop*SM_likes -0.07 Pop*TG_Mentions -0.1 Pop*Recommended -1.00*** Popularity 1.04***

(42)

42 important influence on the dependent variable and the relationship was moderated by popularity. Therefore, the author argues that this is proof for a change in the behavior of visitors, since in the case of niche destinations they look and engage with the national tourism board Instagram as a source of inspiration for their travels.

4. Discussion

As it was mentioned in the previous section, there were two trends that were observed in the reported results. Time lags where integrated in the analysis in order to capture the distinct behavior of travelers, namely they usually book holidays in advance. The majority of them book their holidays with maximum 6 months in advance (Statista, 2016). This was important since it allowed the author to verify if the behavior of tourists varies in time.

Since across the time lags, almost all the independent variables selected had a significant impact on the number of visitors in step 1 of the analysis (without the moderator). The only exception was the travel guides mentions of the location in social media. Nevertheless, because the yearly recommendation of travel guides was significant with a negative interaction, it signals that H2 is partly supported and that travel guides has a negative influence on the number of tourists. For national tourism boards’ indicators, they were all significant meaning that they do have an impact on the number of tourists. However, because the sign for social media posts’ indicator was minus, it signal that the relationship between the number of posts and the number of visitors is reversed. One possible explanation for this effect is that marketers may have the tendency to overshare online and this makes users to no longer appreciate the content shared by

(43)

43 them. In conclusion, H1 is partly supported because of the negative sign for the social media posts’ indicator.

As part of the general trend of the results, the interactions between the number of social media posts and the travel guides’ yearly recommendation were moderated by the popularity of the destination. When moderated with the popularity, the social media posts’ indicator changed its sign and became positive; therefore, signaling that content generated by national tourism boards is more influential for popular destinations. In conclusion, hypothesis H3a is supported. In the case of travel guides, yearly recommendations are indeed moderated by the level of popularity as it is demonstrated by the data, being more efficient for niche destinations as shown by the negative sign of the variable. As a result, H3b is also supported.

The second trend revealed that social media likes are also moderated by the level of popularity, being more efficient for popular destinations. One possible explanation for this effect is that use Instagram to research different activities or sights in order to decide on a schedule shortly before the actual trip. It is very useful for marketers to take this into consideration when they are developing their content so that they can promote different events taking place or attractions that can visited in the next month. At the same time, this effect further supports H3a.

To sum up, for tourism social media definitely has an impact and this is visible in both the content being produced by national tourism boards, but also in the yearly recommendations by travel guides. However, the influence effect was not illustrated also by the travel guides social media mentions, probably because they were not frequent enough.

(44)

44

5. Limitations and further research

In the initial plan of the research, the author included also user-generated content as a third independent variable due to its demonstrated importance in social media and more importantly for the influencing power social media has on its users. Unfortunately, the complexity of the data collection process forced the author to eliminate this variable. The complexity was driven by the need to differentiate between posts written by travelers and by locals, but also the need to extract only the posts written in the 2014 – 2015 period.

Aside from adding user-generated content as a third independent variable, there are of course other methods through which the insights offered by the present paper could have been further enhanced. One of the ways of increasing the ability to generalize the results is to include more countries in the analysis from different continents and extend the period. By doing so, the researchers can testify that the effects identified by the author are consistent across time and continents. Also, it would be recommended to add more travel guides to be certain that this independent variable is represented correctly. Furthermore, it would be also interesting to test if there are any significant differences between traditional travel guides (like the ones analyzed in this research) and new types of travel guides that depend on user-generated content (such as TripAdvior).

At the same time, the results of the study can be further enhanced if in parallel with a social media analysis, a survey is conducted in order to find out more about their purchase and research behavior for travel services. It would be recommended to include as many nationalities from different geographic regions or continent as possible to have a representative sample across the globe. Another goal for the survey is to clarify the travelers’ perceptions of travel guides as

(45)

45 an expert in the field and of national tourism boards as a source of information and as an influencer. Of course, in order to identify the ideal hypotheses for the proposed survey, it is necessary to first carry out a qualitative study or a pilot study.

In conclusion, the author recommends recreating the current study but improving it by extending the period, adding more countries and more travel guides, but also adding a survey to have a better understanding of how social media impacts the travelers’ behavior across different stages of the shopping process (identification of need, search, purchase and post-purchase). The present research started to illustrate the different influence factors for niche versus mainstream destinations. The author offers an overview of the hypotheses and their results in Table 10 for an easier understanding of the results.

Table 10: Overview of hypotheses

# Hypothesis Status

H1 Marketer generated content by National Tourism Organizations will

have a positive effect on the number of visitors.

Partly supported

H2 Content generated by Travel Guides will have a positive effect on the

number of visitors.

Partly supported

H3a The relationship between content generated by national tourism boards

and the number of visitors will be influenced by the popularity of the destination.

Supported

H3b The relationship between travel guides the number of visitors will be

influenced by the popularity of the destination.

#Wanderlust : measuring the social media factors that influence tourism

Master Thesis