A regional survey of current practices on destination marketing organizations ’ Facebook

Constantly changing nature of social network sites creates the need for continuous process of online benchmarking for identifying practices used by other parties. Facebook as the most used SNS still plays an increasingly important role as a marketing channel for destination marketing organizations (DMO). This paper explores basic characteristics of the official DMO Facebook Pages in order to quantify and present those characteristics in a regional context on the case of two travel markets (EU countries and U.S. states). The results show inconsistent practices in the EU and the USA. When comparing those two markets most similarities in practices are present in general usage of Facebook Pages, while indicative differences are recorded in terms of Page popularity, some posts’ characteristics and most evidently in users’ engagement. Understanding the Facebook usage practice under the regional spotlight can help DMOs and other service providers to evaluate their activities and if necessary to harmonize it to region-


Introduction
Facebook is currently the most used and most influential social network on the Internet and represents one of the most popular websites in Europe and worldwide (Azevedo, 2011;Wells, Link, 2014).On December 31st, 2017, Facebook had reached the number of 2.13 billion monthly active users (Facebook, 2017), that is almost half of the world's estimated online population (Internet World Stats, 2018).This arguably makes it one of the biggest media organizations in the history of humankind (Rieder, 2013).For many companies this social network is an indispensable element of the marketing activities (Stankov et al., 2016).Unlike in-dividuals who use Profiles (web pages containing user's information) to present themselves on Facebook, companies mostly use Facebook Pages.User can interact and affiliate as a fan of a company's Page in the same way they interact with other Profiles (Cooper, 2010).
Many destination marketing organizations (for the consistency in the text, term destination marketing organizations -DMO is used to represent an organization on a country/state level which is responsible for tourism marketing) recognized growing popularity of Facebook.The focus in tourism indus-try and academic research has predominately moved from the question of adopting Facebook as a communication channel to the question how to use it, that is, how to find effective strategies for managing existing Facebook Pages.Even though some progress has been made since the beginning of social network sites (SNS) adoption, still a rather small number of DMOs fully understand and effectively use SNS (Hamill & Stevenson, 2012).In many cases, destination managers are not well informed about success strategies to maintain and manage their Facebook Pages (Lalicic & Gindl, 2018).
Having a bad practice of using Page can be more detrimental than not adopting this social network at all (Hays et al., 2013).Constantly changing nature of SNS makes research on this subject outdated in a very short period of time (Zouganeli et al., 2011).Therefore, DMOs should follow constant changes in the social media platforms.Monitoring of the competitor's use of Internet marketing needs to be a continuous process (Chaffey et al., 2006).This kind of benchmarking can help identify the best practices and marketing standards with the goal of improving online promotional strategies (Luna-Nevarez & Hyman, 2012).New SNS features and approaches that are accepted by customers and competitors could soon become standard elements of social media strategies and they can be expected by the new customers as well.To overcome that goal, there is a necessity for low-time consuming and for technically low demanding procedure for DMOs to overview a larger number of Facebook Pages that are part of a regional market context.The selection process of characteristics to review should be flexible and open with the respect to the dynamic nature of Facebook platform and needs of the organizations.
The discussion above justifies the purpose of this study, which is to explore the basic characteristics of the official DMO Facebook Pages on the tested travel markets to quantify those characteristics and to present them in a regional context (Stankov et al., 2017).Following the above mentioned, for the case studies of this research, two separate groups of DMO Facebook Pages were chosen: (1) EU countries and (2) U.S. states.Two regional markets are chosen to strengthen and support the results of the presented methodology by comparison of results in order to determine potentially existing common practices.
The rationale behind choosing DMOs from the EU and the USA was that those two groups are among the leading travel markets in the world.At the same time, their comprising members sometimes represent mutually inner-competing tourism markets for the same niches (Jeffries, 2001;Jordan, 2006) that underline the need for regional benchmarking.From international visitor's viewpoint, each member of the EU and the USA can be seen as integral part of that group and can be expected to follow common standards in their practice, representing recognition, identity, etc.There are more general similarities between these two travel markets than any other markets in the world that also justify comparison, among most obvious are open borders between comprising members, the use of single currency in the U.S. and good parts of the EU; decision making process in most cases is the responsibility of members, etc.Finally, both markets have well organized DMO structures.
The results of this research will provide insights into the regional practice of using Facebook Pages in these two world's leading tourism markets.The paper stresses the importance of determination of common Facebook practice in regional settings by pointing out structural problems rather than focusing on separate country cases.In that sense, knowledge of the common Facebook usage standards can help DMOs and other service providers to evaluate and adjust their practice to collective regional brand efforts.Coordinated and collaborative marketing efforts at regional levels can help in communication of the desired destination image to tourists (Hudson, 2014).For example, efforts of the collective marketing are recognized in the USA (Hudson, 2014), where the decisions of international tourists for visitation are not typically driven just by a single destination, but they also include a wide scope of experiences, products, and services in several destinations (Oxford Economics, 2014).
Unlike other studies that usually examined national DMOs of one or of a limited sample of the countries, this paper, similarly to the study of Zouganeli and colleagues (2011), gives a broad territorial overview of the phenomenon and focuses on Facebook that is a most preferred and favoured platform by the DMOs at a global scale.For example, previous study of Hays and colleagues (2013) examined the usage of Facebook and Twitter among the DMOs of the top 10 most visited countries by international tourists and Roque and Raposo (2016) compared the use of social media applications of 13 key player DMOs across all continents.Other authors analyzed DMOs in one country such as, Yang and Wang (2015) in the case of China or Mariani and colleagues (2016) in the case of Italy.For that reason, insights of this overview could be of interest to the examined destinations and, if deemed appropriate, they can be used to transfer the good practice and experiences to other tourist destinations.
Finally, the findings of this paper could contribute to the existing research of the use of SNSs, specifically Facebook, by DMOs.Our research highlights the need for constant monitoring of SNSs managed by DMOs and adjusting to the good practices of their competitors.

Theoretical Framework
Organizational Aspects of DMOs in the EU and the USA By the way of the establishment, organizational structure and its roles, official DMOs in the EU and the USA have some differences.The first obvious reason is that official DMOs, or national tourism organizations (NTOs), in the EU represent different individual countries that all are part of the EU, a union of states, compared to the USA where DMOs represent states that are part of a federal political system.For that reason, the internal organization of DMOs in the EU differs from internal organization of DMOs in the USA.DMOs in the EU are ranging from French DMO established in 1910(Siguax, 1966, as cited in Pike 2004, p. 21), all the way to DMOs of the new EU members.The activities of DMOs are usually not limited just to tourism promotion but also include overall tourism administration in the countries.The EU policies and programs actively support tourism promotion, although there is no official EU DMO.DMOs (except France, Netherlands, UK, and Sweden) are full members of European Travel Commission (ETC), a nonprofit organization responsible for supporting its members and promotion of Europe as a tourist destination in the third markets.However, ETC is not an institution of the EU.ETC also includes DMOs from European countries that are not members of the EU (European Travel Commission, 2014).ETC operates a website www.visiteurope.com in cooperation with European Commission in order to increase the attractiveness of Europe as a collection of sustainable and high-quality tourist destinations (European Commission, 2010).
Tourism administration in the USA is quite decentralized as policy decision making and funding is in a jurisdiction of state governments and rests largely on the private sector (Timothy, 2006).State DMOs or state tourism offices (STO) are organizations with overall responsibility for marketing states as tourist destinations.Many DMOs were established during the 1940s, but not until the 70s that most states had STOs (Pike, 2004).Until 2010, USA did not have a national-level public tourism organization, when the public-private marketing entity Brand USA started operating with the purpose of acting as the destination marketing organization for the USA.Brand USA is nation's first cooperative destination marketing organization with the focus on delivering programs and platforms to promote the USA worldwide (Brand USA, 2015).The official website of Brand USA tourism marketing program is www.visittheusa.com(previously www.DiscoverAmerica.com).Social me-dia plays an important role in Brand USA's marketing strategy.The USA is among the first travel destinations that launched country-specific social media pages (Hudson, 2014).

SNS as Platform for Social Media Marketing of DMOs
The role and functions of SNSs for tourism operations have been widely discussed in tourism literature (Leung et al., 2013).Being one of the major trends attracting global interest of marketers (Baethge et al., 2016), SNSs offer many new resources and opportunities for improving and reengineering operations of travel and tourism organizations (Hvass & Munar, 2012;Minazzi, 2015;Zeng & Gerritsen, 2014).
In order to improve marketing effectiveness, successful DMOs need to develop marketing strategies adjusted accordingly to reflect the new realities represented in overflow of new SNSs (Shao et al., 2012;Yang & Wang, 2015).The role of SNSs to DMOs is especially vital for inbound marketing and content marketing strategies (UNWTO, ETC, 2014).Furthermore, SNS can be used by DMO to dissimilate word of mouth electronically.In that context, Tham and colleagues (2013) suggest that DMOs with the use of SNSs could extend opportunities for communicating travellers' experiences and also they could engage them more and involve industry partners to build desirable destination images.
In June and July 2009, Stankov and colleagues (2010) identified that about half of DMOs, members of European Travel Commission (ETC) did not have an official presence on Facebook.Among those who had, almost one-third had Facebook Pages, and almost a fifth of them had a Facebook Group.One DMO even used Personal Profile as official Facebook presentation.Nowadays, the situation has changed and all DMOs in the EU have official Facebook Page.The study of Yoo and Kim (2013) found that all of 50 state tourism websites in the USA integrated at least one type of social media and all of them provide official Facebook page (Milwood et al., 2013).Recent studies found out that social media adopting capacities of DMOs differ significantly (Roque & Raposo, 2016;Shao et al., 2012;Yang & Wang, 2015).Different social media strategy approaches were also found in the international analysis of most visited countries by Hays et al. (2013).The research of Zeng and Gerritsen (2014) also confirms that there are differences between countries in social media usage.The study of Milwood et al. (2013) found that the USA widely embraces most popular social media whereas Swiss DMOs are quite cautious about social media adoption.Similar practices will probably continue in the future, as social media approaches are not only driven by organizational structure and overall DMO competencies but also by dynamic and innovative nature of social media itself.

Facebook page as a marketing platform and post characteristics
Every Facebook Page has a unique structure to start from, in fact that is "a blank paper" that needs to be filled out by the given company.Based on the review of the contemporary studies on the business usage of Facebook (e.g. Cvijikj & Michahelles, 2013;Mariani et al., 2016;Hays et al., 2013;Hsu, 2012;Kwok & Yu, 2013;Munar & Jacobsen, 2014;Roque & Raposo, 2016;Sabate et al., 2014;Zouganeli et al., 2011) and advances in Facebook usability, in the following text, we will explain some basic characteristics of Pages and posts that DMO should consider when analyzing competitors practices.These characteristics are not conclusive, but are offered to spark consideration and to add to the constantly open debate which is necessary concerning the dynamic nature of this social medium.
When visiting a Page, if the user has not landed from an official destination website, there is always a question of Page authenticity.Simple Google search or search using Facebook internal search engine will often results in various unofficial country Pages (Dwivedi et al., 2011).Therefore, complete authenticity can be guaranteed by Facebook when Page is manually verified with assignation of special verification sign next to Page's title (Facebook Developers, 2015b).
Facebook Pages can be enriched by using Page Tabs.Facebook Page is actually a separate web page containing different content.Besides standard Facebook Tabs, such as ("About", "Photos, "Videos", etc) businesses can make custom Tabs containing different types of apps, welcoming messages, polls, showcase videos, reservation forms, etc.Generally, custom Facebook Tabs can create a much richer user experience and add value to the standard Facebook Page (Pitre, 2015).If business pays attention merely to the main Page News Feed (Page Wall) that can be considered as neglecting and limiting the full potential of this SNS for spreading of the information and collaboration with the users (Hsu, 2012;Zouganeli et al., 2011).
Facebook, as a democratic SNS medium, provides options for user-generated content (UGC), posting and expressing opinions on the official Pages.Even though organizations should promote consumer participation (Belanche et al., 2010), an open expression of customer dissatisfaction can be an aspect of concern (Sarkar et al., 2014).For some type of Page categories, user ratings and reviews section can be enabled.Looking for other consumers' reviews is often practiced travel related activity of Internet users (Gretzel et al., 2007).A Page's star rating is the average of all public star ratings (star ratings that are shared publicly) that the Page has received (Facebook, 2016).
In digital marketing, "call to action" words are often used as a motivation to take a desirable action when visiting websites (Eisenberg & Eisenberg, 2006)."Call-to-action" feature on Facebook Page is a button at the top of the page that links to any destination on or off Facebook and can help Pages to drive business objectives.Currently there are seven call to action options available: book now, contact us, use app, play game, shop now, sign up and watch video (Facebook for Business, 2014).
There is no general agreement on posting frequency of travel related content.Frequent posting provides new content, keeps members engaged and allows greater interaction with the fans (Zarrella & Zarrella, 2010).However, high posting frequency does not necessary guarantee high engagement rates.For example, Australian Tourism Commission advices travel industry to be cautious with the number of posts, putting attention to well-planning of posts, not on quantity (South Australian Tourism Commission, 2015).New study of Mariani and colleagues (2016) for the regional DMOs in Italy found that high post frequency has a negative impact on user's engagement.
Facebook users' personal news feed (i.e.Walls) are constantly filled with content coming from multiple sources (other users' profiles, Facebook Pages liked by users, sponsored content).On average, only about 17% of a business Page's post shows up on fans' walls (Hubspot, 2015).Facebook uses complex ranking algorithm based on machine learning to select and rank the content that shows up in the user's news feed.
Before every content post to Facebook Page, a DMO usually has to choose: what it will post (link, photo, video, status or event), the time of day when content should be posted and the day of week to post (Linnell, 2012).One of the determinants of internet advertising effectiveness can be the length of the message (Baltas, 2003).According to Baltas (2003), lengthy messages that involves paying close attention, can reduce direct response to it.

Sample
In this research, we have focused our attention to official international DMO Facebook Pages of countries, that is, states in the EU and the USA.According to Hays and colleagues (2013), there are different approaches in managing Facebook Pages by DMOsfrom those managed by the main office, in one language or multilingual, to those managed by the regional offices in different languages.For the purpose of this research, Pages from the EU that are international/English version of official country DMOs' websites were analyzed.Most of the Facebook Pages are managed by the main DMO's office in the country of origin, primarily in English or using bilingual posts, with few exceptions managed by the UK or the USA regional offices.In the case of the USA, all Facebook Pages are found at the official state DMOs' websites, managed by main office in English.Total data collection included 27 Facebook pages from the EU and 50 from the USA.

Data collection
For the purpose of the data collection for this research, we focused on a manual and automated approach.The data for Pages were collected manually and using the customized requests, based on the Graph API Explorer, a low-level HTTP-based API for reading and writing the Facebook's Social Graph (Facebook Developers, 2015a).Facebook's Social Graph is a graph data structure that represents social interaction and consists of nodes and connections between the nodes (Russell, 2013).The authors used Graph API Explorer v2.4 to query information, such as the number of users who like the Page, Page fans' countries, whether the page is verified, etc.For those information when automated approach was restricted by Facebook privacy policy (such as number of tabs and applications, review values, etc.), authors used manual data gathering.
The data for Page posts was gathered automatically using page data module of Netvizz v1.25 tool.This Facebook tool extracts data from different sections of the Facebook Groups and Pages (Rieder, 2013).Extracted data include information such as: Facebook's post classification, text of the post, picture URL (if a picture is attached to the post), publishing date and time, number of likes, comments, shares, etc.Only the content posted by DMO was collected.
The authors gathered information on all 3401 published posts (976 from the EU and 2452 from the USA) over one month, from April 1 to April 30, 2015.The actual time of data gathering was from June 3 to 29.This was necessary, in order to see how fans interacted with the post.This time span between actual posting time and time of data gathering is believed to be long enough for the purpose of this study.According to Sabate et al. ( 2014), a content post on the net for more than a month is not likely to receive more significant interaction, especially in the case of Facebook that is extremely dynamic SNS.In addition, with the use of Facebook Graph API aggregated location data about the people who like Page are obtained and sorted by top 45 countries.

Variables
Based on the literature review above and capabilities of Facebook API the information gathered for each Facebook Page is provided in Table 1.Information about Pages is grouped into general information about usage with main page elements used and page popularity performance.Authors arranged Facebook post metrics into two main groups that can be statistically compared.First group included basic metrics of users' engagement: "Likes" (indicating interest in an existing post), "Comments" about the post content, "Shares" of the content on personal Profile or other Pages or Groups and posting content on the Page's wall (depending on the communication policy set by the Page owner) (Linnell, 2012;Sabate et al., 2014;Treadaway & Smith, 2012).For the purpose of this research, we included additional variables -comment replies (number of replies to the existing comments to posts) and comment likes (number of likes on comments).Second group is content characteristics, including length of post, time of posting, day in the week when content is posted and type of post (link, photo, video, status or event).Average number of all tabs used in the EU is 8.37.In the EU 55.55% of the Pages had a tab containing app of some other popular social network site.Among those apps, four SNSs are among the most frequent -Pinterest (24%), Instagram, (24%), Twitter (17.78%) and You-Tube (26.67%).Other social networks (Google plus, Foursquare, Flickr, Tumblr, etc.) are less frequently used as tab apps (EU 6.67%).Tabs with other type of apps (links to polls, contact forms, games, travel planners, brochures, galleries, etc.) are present on 74.07% of European Pages.

General
Average number of all tabs used in the USA is 9.16.For this travel market, 76% Pages had a tab containing app of some popular social network site, which is higher when compared to EU.Among those apps, as in the case of EU, the same four SNSs are among the most frequent -Pinterest (29.70%),Instagram, (28.71%),Twitter (24.75%) and YouTube (9.9%).Other social networks are less used as tab apps (6.94%).Tabs with other type of apps are present on 80% of American Pages.
Only 22.22% in the EU and 26% in the USA had a review option enabled on Page.Interestingly, in the USA, every other Page with less than 100.000fans has this option enabled, while this trend is not recorded in the EU.
In most cases, users can post comments to Page's news feed.That option is allowed on 85.2% of European and on 95.6% of American Pages.In both markets, Wall posts often have to be approved by Page administrator in order to be visible to Page fans.
"Call-to-action" button is used on 48.15% of the Pages in the EU and on 58% in the USA.Table 4 shows proportion of available "call-to-action" buttons used in the both travel markets."Sign up" is the most frequent "call-to-action", showing the intention of DMOs to continue the interaction with the Fans.Average number of monthly posts per country in the EU was 36, that is on average 1.20 times a day.DMO posts in the USA are more frequent as they have 49 posts per state, on average 1.63 posts per day.

Page popularity
In the case of EU, the number of Page fans varies from a very small number (below a 1000 fans) to large fan communities (over 1 million fans) (see Figure 1).
In the USA none of the DMOs has less than 10 thousand fans, two Pages exceed 1 million fans, and one Page has the highest rank with 2 million fan base (see Figure 2).Total number of official DMO Page fans reached 11.9 million in the EU and 13.1 million in the USA.However, average number of fans per country in Europe is much higher than in the USA,442,214.59 and 261,937.42 respectively.Distribution of Page fans by their origin continent is plotted on the map in Figure 3.In general, EU Pages have good geographical distribution of people liking their pages when compared to the USA.While in the USA, North America is the dominant continent of origin of Page fans, in the EU, besides Europe, many fans are from North America, Asia and South America.Furthermore, among all fans recorded for the USA, domestic fans make 90.30%, and just 1.3% of fans are from other parts of North America.On the other hand, in the EU, from total number of fans, 44.53% are domestic fans (from EU countries) and 2.83% are fans from other countries in Europe.
Figure 4 shows a detailed insight into ratios of Page fans by their origin for EU countries.Interestingly, in most of the new member states, dominant group of fans are fans from the European continent.On other hand, Germany (91%), Netherlands (84%) and United Kingdom (78%) have the highest percentage of fans from other continents.
Figure 5 shows a detailed insight into ratios of Page fans by their origin for USA.It is clearly evident that, significant portion of fans from other continents are present in only four states -New York (81%), Nevada (54%), South Dakota (43%) and Vermont (30%).In all other states there are less than 10% of fans from the other continents.
As mentioned above, about every fifth Page in the EU and every fourth in the USA has review option enabled.Average review scores are: in the EU 4.

Users' engagement and post characteristics
The results for the set of engagement variables are shown in Table 5.In both markets the most prevalent way of engagement is "like"."Likes" are followed by "shares" and "comments".
The results of independent sample t-test, as shown in Table 5 indicate that there are significant mean differences for all five variables of engagement in the EU and the USA.There are about 5 times more likes, comments on posts, comments replies in the USA than in the EU, and more than 4 times more comment likes and shares.
To specify the differences and similarities in posts characteristics, t-test was used for interval variableslength of posts and hour of posting and Chi-square was used for other nominal variables -type of post and day of posting.
In the EU, the average length of post is 204.22 characters while in the U.S. the average post length is 155.92.The results of independent-samples t-test for the length of post showed significant difference in scores for the EU posts (M = 204.22,SD = 163.72)and the USA posts (M = 155.92,SD = 101.60;t (1288.14)= 8.58, p = 0.00, two-tailed).The magnitude of the difference in means (mean difference = 48.30,95%, CI: 37.252 to 59.349) was moderate (eta squared = 0.02).
The relationship between post length and users' engagement (as measured by total number of likes, comments and shares) was investigated using Pearson product-moment correlation coefficient.In the EU, there was a positive correlation between the two variables (see Table 6) while in the USA there is no significant correlation, except in case of weak correlation between post lengths and number of comments.Spe-cifically, EU page users seem to prefer to read longer posts unlike USA page users.This finding is certainly interesting and might be a product of cultural differences or some confounding factor such as the content or the quality of the given post.Nevertheless this requires further research in order to determine possible causes for this specificity of two researched destinations.
During working days, posting is relatively equally distributed, with the peaks on Wednesday and Thursday (see Figure 6).Wednesday is the peak for the EU when the most posting is done and for the USA that is on Thursday.Lowest percentage of posts for both regions is recorded on Sunday.A Chi-square test for independence indicated no significant association between day of posting and travel markets, χ2 (6, n = 3401) = 10.20,p = .12.
Table 7 shows average weekly users' engagement.Posts receive most likes on Saturdays in both markets.Lowest number of likes is associated with highest number of posts, Wednesdays for the EU, and Thursdays for the USA.For comments and shares that require higher levels of engagement different patterns are recorded, as shown in Table 7.As shown in Table 8 photos are dominant type of posts as more than half of all posts are of this type.About one third of posts contain links.Lower then 10% are videos, and lower than 1% are events and statuses.Chi-square test indicates a significant association between type of post and region of posting, χ2 (4, n = 3401) = 28.67,p = .00.However, effect size is small as Cramer's V = 0.9.EU DMO Pages use slightly more photo and video posts, compared to the USA.However, Pages in the USA use more link posts than in EU.
In terms of users' engagement, photos on average receive most likes, comments and shares in both markets.In general in the EU videos are more engaging then links.Photos and links receive almost the same number of likes, but number of comments and shares are higher for videos.In the USA links are more engaging, as they have more likes and shares.Statuses and events are the least engaging (see Table 9).

Discussion and Conclusions
The results of this overview study present current practices and highlight regional similarities and differences in DMO Facebook Page usage in the EU and the U.S. markets.
First and foremost, this study offers a theoretical contribution, as it outlines a regional approach to the analysis of Facebook Pages in the context of travel markets with same geographical and organizational framework of DMOs.
The approach is based on meaningful and open selection of variables for the comparison, with technically low-demanding procedures of data gathering using manual and automatic data mining options freely provided by Facebook.As noted by Mich and Baggio (2015), the implementation of tools for the automatic extraction of values of quantitative variables of a Facebook Page is needed in case of large number of assets to compare.This paper also tries to facilitate the adoption of Facebook metrics by other tourism researchers and practitioner in order to find applicable insights from this SNS.
Practical implications of this overview study are diverse, as examined DMOs can use presented ideas to locate their competitive edge against the current practice of using Facebook Pages within the same geographical travel market and additional regional market for comparison.The results of the study show inconsistent practices in the EU and the USA.By adopting current regional standards, DMOs and other service providers that are lagging behind can add value to their Facebook marketing programs that leverage regional brands.
In case of the comparison of the EU and the U.S. markets this survey shows some interesting findings that can reveal some characteristics of regional markets and indicative differences.
First, there are more similarities than differences in term of general usage of Pages, such as Tab usage, presence of reviews, posting availability and Page verifications.Comparing to some preliminary studies on Facebook Page adoption by EU DMOs that showed slow and partial adoption of full potential of this SNS (Stankov et al., 2010;Zouganeli et al., 2011), the results of this survey shows improvement of DMOs Facebook characteristics.
Second, DMOs in the USA were more agile in adopting Facebook Pages and they are also presently more active, adding more daily posts than EU DMOs.Still, higher number of daily posts does not have to be a priority of the national DMOs (Mariani et al., 2017).For example, this study found out that United Kingdom has the record fan base in the EU but have lowest levels of post frequency.Frequency of posting on Facebook can be important for some company's brands, where too large or too small number of posts can drive away fans.Gretzel and Dinhopl (2014) study found that this is not the case for travel destination.Relationship of travellers on SNS with destinations is at a deeper level of emotional attachment and social media activities therefore have a lesser effect (Gretzel & Dinhopl, 2014;Lalicic et al., 2018).This research also confirms the finding of the Hays et al. (2013) study that the date of joining to Facebook does not necessarily correspond with the number of Facebook fans.Averagely older U.S. Pages attracted slightly more of total fans for all 50 states than 27 Pages of EU countries.Sill, number of fans does not necessarily mean higher engagement, that is, users may like Page of one country and become the fan, but they might never engage with Page's content.
Third, although European countries on average have a larger number of fans per country, differences between numbers of fans in the USA are lower.That makes a more balanced spatial distribution of fans in U.S. market.On the other hand, this study revealed interesting demographic difference in fans' characteristics.Fans of European DMO Pages are much better spatially distributed by continents of their origin.Most of the fans of U.S. DMO Pages are actually the citizens of the USA.This corresponds to the fact that domestic tourism significantly dominates the overall tourism market in the USA (Travel and Tourism Intelligence Center, 2014).In that context, the reason for New York and Nevada to be outliers by the number of international fans can be due to the global recognition of New York City and Las Vegas.Similarly, low international recognition of the new EU member states could lead to the dominance of European fans.As suggested by Luna-Nevarez and Hyman (2012) these interesting results imply the need for an evaluation of demographic and psychographic information provided by SNS in order to better target visitors.
Finally, DMOs have to pay attention to the characteristics of posts and management of posting.For Brand USA one of the key indicators to measure social media success, besides of total number of fans is the percentage of engagement (Hudson, 2014) as tourist involvement has a positive impact on overall destination image (Molinillo et al., 2018).This study showed that user's engagement on posts is different and statistically significant for these two tourism markets.On average, posts by U.S. DMOs attract more user's likes, comments and shares which could be because U.S. DMOs fans are generally from the USA and are therefore more prone to participate and comment (being that English is their mother tongue).Average weekly distribution of posts is not significantly different between examined two tourism markets and corresponds to global post frequency by days of the week, that is, posts are more frequent during workdays than weekends, with small peaks on Wednesdays and Thursdays (Lee, 2014).However, this practice does not result in higher level of average likes.On the other hand, on Saturdays posts receive most likes indicating that less posts and more free time of fans can boost engagement.The sample of DMOs examined in this study shows that statistically significant differences between tourism markets are present in the length of posts and type of posts.EU DMOs have longer posts than U.S. DMOs.For the EU longer posts positively correlate with number of likes, but that is not the case in the USA.It would be interesting to examine in future research the reason for this by analyzing the linguistic characteristics (such as semantics and syntax) of these posts.Similarly, recent study of Italian DMOs suggests that moderately long posts (around 200 characters) have a statistically-significant positive impact on users' engagement (Mariani et al., 2016).Pictures and videos are predominantly used as means of communication with fans for both destinations since they are the easiest to process and are the most evident and concrete marketing tools.But, even though differences are rather small, EU DMOs post more pictures, videos and events when compared to U.S. DMOs, while they post more links and statuses than EU DMOs.Interestingly, links are more engaging in the USA than in the EU.
There are some important limitations of this study.Like in any other similar endeavours, this study only measured post characteristics for the one month period.Therefore, the results could be biased, affected by busy or low seasons or other vacation periods.Particularly, the overview period of this study included the Easter Sunday.As Facebook Graph API provides large amount of data, the analytical power of big data analysis could be used for this kind of regional, crossnational data analyses.The approach of this study did not include content analysis of the post messages that with the use of different analytical tools can very useful in revealing valuable insights to motives of users' engagement and perception towards brands (Cervellon & Galipienzo, 2015).For example in case of users' engagement, Kwok and Yu's (2015) content analysis of Facebook messages posted by hospitality companies reveals that conversational messages receive more users' "Likes" than sales/marketing messages.More precisely, the study of Zouganeli and colleagues (2011) showed that multimedia posts receive mostly "likes" and conversational posts receive more "Comments".The study of Tilly and colleagues (2015) showed that tourism-related social media can be used as a source of representative spatiotemporal macro-level tourism information.However, as it is clearly evident from the results of this study, analysis of separate country cases would require different approaches that are out of the scope of this study and many of which are out of the scope of social media sphere.
The results raise some questions for the future scientific and business research.For example, domestic fans are dominant in most of the countries and it would be interesting to find whether motivation for liking is purely travel-related or are there some patriotic or other reasons involved?(Bodroža & Jovanović, 2016;Ben-Shaul & Reichel, 2017).Could the determined Facebook Page practice be effective in other markets, having in mind different market characteristics and organizational structures of DMOs?Further, user's engagement could be under the influence of different factors in terms of characteristics of the content communicated, as discussed by other authors (Cvijikj & Michahelles, 2013;Sabate et al., 2014, Božić & Jovanović, 2017).It would be interesting to find if there are some seasonal differences in user's activity towards tourism content or in the activities of DMOs in different markets.

•
Number and type of page Tabs • Presence of Facebook review options • Availability of posting for page fans • Presence of "call to action" button and type of action • Frequency of DMO posting Page popularity • Number of Page likes/fans • Page fans continent • Average score and estimated number of reviewers obtained by Facebook review option Page usage Average DMO Page in the USA is 5.88 years old, almost one year older than European ones.First Pages were created in 2008 (13 states in the US, but only two in the EU).In 2009, 84% of U.S. DMOs have created their Facebook pages while only 43% were created by EU DMOs.At the moment of analysis only Latvia did not have English version of Facebook Page managed by official DMO.Pages in the EU are verified by Facebook in 40.74% cases.The verification process depends on Facebook and the number of fans clearly plays an important role, as among 10 Pages with the highest number of fans, 7 are verified.None of the 10 lowest ranking Pages, by the number of fans, were verified by Facebook.In the USA 30% of Pages were verified: 6 of the top 10 Pages by the number of fans, and 2 from the 10 lowest ranking Pages.

Figure 3 .Figure 4 .Figure 5 .
Figure 3. Distribution of Page fans by the continents of their origin (in percentages)

Figure 6 .
Figure 6.Average weekly distribution of Page posts by DMOs in the EU and the USA (in percentages)

Table 1 .
Gathered information for Page characteristics

Table 3 .
Number of newly created Pages by year and cumulative percentage of total countries/states in the EU and the USA

Table 2 .
The information (variables) gathered for post's characteristics and users' engagement

Table 4 .
Presence and type of "call-to-action" buttons used 63 (average approximate number of reviewers is 3943.8)and the USA 4.08 (average approximate number of reviewers is 1439.23).

Table 5 .
A comparison of engagement factors on EU and U.S. DMOs Page posts (N-Number of posts; M-Mean; SD -Standard deviation)

Table 8 .
Facebook page posts by type posted by DMOs

Table 7 .
Average weekly users' engagement factors

Table 9 .
Average users' engagement factors by the type of content