Exploring usage, expected benefits and perceived usefulness of social media in travel agencies: an empirical investigation in Italy

Purpose – The purpose of the present study is to investigate travel agencies ’ social media usage and its perceivedeffectivenessby small-andmicro-Italiantravelagencies;the pre-pandemicperiodiscomparedtothe forecasts for the post-Covid-19 period and different characteristics of firms and entrepreneurs are considered. Furthermore, the study analyses the expected benefits in terms of marketing objectives, such as improving brand image and/or personalizing the offer. Design/methodology/approach – The research was developed through a questionnaire administered electronically to travel agents (282 respondents). The resulting data was analyzed by applying the McNemar test, a pairwise t -test and the multivariate analysis of variance. Findings – Theresultsshowthatsocialmediaarestrategicallysignificantfortravelagents,eventhoughtheir adoptionisinfluencedbydifferentagencyaims;theperceivedeffectivenessresultsarediversifiedaccordingto varying agency typologies. Research limitations/implications – The two main limitations of the study are its focus on the Italian context only and the missing consideration of the consumer ’ s point of view. The latter prevents an exhaustive assessment of future trends regarding the use of social media in the client – agency relationship. Originality/value – The study, which focuses on a little debated topic concerning the relationship between social media and SMEs, organically explores various dimensions related to the adoption of social media by small agencies, also considering the impact of the Covid-19 on the perception of travel agents. As a further elementoforiginality,theresearchtakesintoconsiderationthemainsocialplatformsseparatelyratherthanthe setoftoolsasawhole


Introduction
The advent of social media (SM) has enormously impacted the tourism intermediation industry (Leung et al., 2019;Xiang et al., 2015).In particular, Facebookamong the first massively subscribed-to platformshas transformed information searches related to travelling not only through destination pages, but also through the content shared by users who publish photos and videos of their travels on social networks (Asongu and Odhiambo, 2019); this is especially true among Millennials (Carnoy, 2017) who, according to a study by Fotis et al. (2012), might perceive such user-generated content as more reliable than tourist portals, travel agents and mass media advertising.
The importance of SM for travel agencies (TAs) is not limited to the influence they can exert on the choice process, but extends to that of co-creation of value and the related enhancement of the customer's role.According to Yang et al. (2014), the SM interactions between businesses and consumers and among consumers lead to the creation of roles of varying intensity that the consumer can assume, from being a simple co-creator of value of the service that he/she will use to becoming a promoter of the enterprise on its social networks.This influence and this new role of the consumer clearly also has potentially negative repercussions when users generate complaints, even causing processes of co-destruction of value (Dolan et al., 2019), which requires appropriate organizational responses as well as the ability to manage operational tools.
In order to deal with these challenges, all tourism operators need to develop more or less intense/strategic digitization processes to maintain their relationship with the customer and to seize possible new opportunities (Hu et al., 2023).
Several studies, in fact, show that digitization, in its various forms, positively influences the growth, organizational learning, innovative capacity and competitiveness of small businesses (Xue et al., 2022), and that social networking, microblogs, and social commerce services have a positive impact on the relationship with the customer (Park and Oh, 2012), providing added value within the shopping experience and beyond (Brun et al., 2020).The adoption of SM affects the innovative capacity of the service, encourages processes of co-creation of value and improves satisfaction (Hollebeek and Rather, 2019;€ Ozturan et al., 2019) as well as customer loyalty (Viljoen and Roberts-Lombard, 2016).
However, such advantages do not guarantee that SM are used effectively or that agents consider them useful in achieving their strategic objectives.There are also a number of issues related to the full adoption of SM that go beyond the usual obstacles related to the small or micro size of the enterprise.Taiminen and Karjaluoto (2015) highlight that SM usage in SMEs is challenging because, in most cases, they are unable to develop an effective SM publishing policy due to a lack of specific skills (Matikiti et al., 2016).
Despite this being a promising topic (Donthu et al., 2020), the literature about SM adoption by small enterprises appears fragmented (Felix et al., 2017;Chu et al., 2020), often dedicated to cross-sectoral studies (Hu et al., 2023), or mainly concentrated on the consumer (Lee et al., 2015).
By focusing on this research gap, the objective of this study is to explore the use of SM in the tourism intermediation industry, by endeavoring to understand, by means of a quantitative study, the extent of the phenomenon, the objectives sought and the factors that can determine differences among TAs.Data collection took place in the early months of the Covid-19 pandemic and its impact was investigated by having the interviewees assess and compare their pre-pandemic situation with the forecasts for what would happen once the crisis was over.
The purpose of the study is to investigate the following three themes: (1) the use of the most popular social platforms among Italian TAs for marketing purposes before Covid-19 and the post-pandemic forecast (RQ1) TQM 35,9 (2) the benefits that travel agents expect from this usage, based on those already identified in the literature (RQ2) (3) the effectiveness perceived by travel agents regarding the adoption of SM (RQ3) and the differences in this perception among groups of TAs identified on the basis of independent variables (RQ4).
2. Literature review and research objectives 2.1 Social media usage in travel agencies Social media (SM) have had a very strong impact on the tourism intermediation industry, mostly because of how they have transformed the whole decision-making process of potential travellers (Xiang et al., 2015;Mar G alvez-Rodr ıguez et al., 2020;Asongu and Odhiambo, 2019;Fotis et al., 2012).Despite having historically focused their competitive advantage on off-line aspects, like travel agent expertise and handling capacity (Kumar et al., 2021;Aguiar-Quintana et al., 2016;Marinkovi c et al., 2013;Fang et al., 2020;Huang, 2012), tourist intermediaries such as TAs are facing a digital transformation as SM tools have become pervasive, affecting their strategies as a consequence.The importance of SM for TAs goes far beyond the influence they can exert on the choice process, extending to the processes of co-creation of value and the customer's increasingly important role (Yang et al., 2014;Dolan et al., 2019).
Today, there is a great variety of constantly evolving SM available at the global level to users.In particular, according to the data provided by blogmeter.it, in 2022 among the most used SM by Italian users in order of popularity YouTube, Facebook, Instagram, TripAdvisor, Linkedin and Twitter complete the list.Considering the messaging services used by Italians, WhatsApp continues to be the most popular with 97% and is followed by Telegram with 63% which has been growing strongly in recent years.
Several studies on the adoption of SM by agencies have revealed that Facebook is by far the most used SM platform; others worth mentioning are (not in order of importance): YouTube, Twitter, WhatsApp, LinkedIn, Instagram and Telegram (Sengtong et al., 2022;Sharma et al., 2020;Parvez et al., 2018;Fernandes et al., 2016;Oji et al., 2017;Ainin et al., 2015;Aspasia and Ourania, 2014;Fedushko and Bekesh, 2014).The various web platforms are usually adopted in parallel and their usage is not exclusive to any one of them.In their study, Hu et al. (2023) highlight that Facebook and Instagram are used to promote and sell products, while LinkedIn and Twitter are used for stakeholder communication and recruiting.
On the other hand, some studies show that the use of SM is not an obvious choice for many, even though it appears to be low cost and does not require high technological skills (Hu et al., 2023).In fact, in their analysis of a group of Finnish tourism companies, Taiminen and Karjaluoto (2015) reveal that nearly half of them (46%) use digital marketing rarely or very rarely and only 7% judge their digital marketing activity to be very good or excellent.Company investments in digital marketing also vary substantially; just over a third (35%) of respondents stated that their investments in digital marketing absorb less than 5% of their marketing budget, while just over a quarter of respondents (26%) have allocated over 41% of their marketing budget for digital channels.In the same vein, the results of the survey on a group of Greek SMEs (Aspasia and Ourania, 2014) show that fewer than 50% use SM for digital marketing purposes.
To all this was added the Covid-19 pandemic in early 2020, which representedand still representsa watershed moment for TAs all over the world.According to a recent study of over 700 Spanish TAs by Ruiz and Garcia (2022), the current situation is perceived as marking a turning point in the market: changes that must take place range from technological innovation to the marketing of more sustainable products, and potential changes in business Social media in travel agencies models.The pandemic may have represented an exogenous factor capable of changing TAs' propensity to adopt SM and, in general, digital technologies.According to Sigala (2020), in fact, the pandemic could facilitate and accelerate that digital revolution capable of changing the way companies will operate in the coming years.According to Hu et al. (2023), before the pandemic offline activities prevailed over online ones and the advent of Covid-19 has changed SM adoption.The authors identified pushing factors, such as Covid-19 countermeasures, digitalization of the environment, customers' quest for digital communication and slowing factors, such as the absence of digital skills and lack of organizational support.
For this reason, in our analysis it is interesting to not only look at SM adoption before the pandemic but also to predict if and how the situation could change once things revert back to normality.
From the above considerations, the first research question investigates the use of the main SM platforms by Italian TAs before and after the pandemic.
RQ1.What social media platforms have been used by TAs before and which ones will be used after the end of the Covid-19 pandemic?
2.2 Benefits and perceived effectiveness of social media adoption SM can offer several advantages to TAs, especially regarding their relationship with customers (Ali Abbasi et al., 2022;Abou-Shouk et al., 2016;Foltean et al., 2019).
Other studies have highlighted the potential of platforms to build (3) customer loyalty (Hollebeek and Rather, 2019;Abou-Shouk et al., 2016;Viljoen and Roberts-Lombard, 2016;Van Asperen et al., 2017), also through the (4) improvement of customer-agent interactions (Hollebeek and Rather, 2019) and the activation of value co-creation processes (Yang et al., 2014).In a survey of Indonesian tourism businesses, Setiadi (2019) reveals how the use of SM marketing relates to objectives such as retaining customers, actively communicating with them but also acquiring new ones, as well as developing value co-creation processes related to the travel experience (Salvado et al., 2011).Digital platforms also make it possible to offer content and information in real-time, allowing interaction with customers according to a dialogic approach and impacting the quality of the relationship with them (Park and Oh, 2012) and having an effect on (5) customer satisfaction ( € Ozturan et al., 2019;Abou-Shouk et al., 2016;Asongu and Odhiambo, 2019).
SM also contribute to digital marketing strategies by (6) improving brand image and notoriety (Abou-Shouk et al., 2016;Barcoe and Whelan, 2018;€ Ozturan et al., 2019;Asongu and Odhiambo, 2019).In particular, Matikiti et al. (2016), using the Digital Marketing Framework (Kierzkowski et al., 1996) to analyze the SM marketing performance of 150 small South African TAs and tour operators, affirm that the best effectiveness of SM use lies in obtaining new potential customers by attracting them to their social channels.

TQM 35,9
Residual benefits also emerge, deriving from activities such as contests and prize draws (Aspasia and Ourania, 2014), also useful for the purpose of (11) collecting customer data (Abou-Shouk et al., 2013) in order to profile customers based on their needs and purchasing patterns.
Several scholars have highlighted the lower costs and greater convenience of using SM by TAs to promote their offers (Sharma et al., 2020;Taiminen and Karjaluoto, 2015).According to Ainin et al. (2015), the benefits of using Facebook would also extend to financial performance and cost reduction (Ali Qalati et al., 2022), allowing investments in SM marketing activities to be profitable.In this sense, Fedushko and Bekesh (2014) focus on sponsorships through Facebook, highlighting the ability to attract new target customers by this means.In other words, SM enhance traditional marketing activities with cost savings (Park and Oh, 2012;El-Gohary, 2012) allowing the (12) improvement of the effectiveness of promotional activities (Rambe, 2017;Ali Abbasi et al., 2022).
The following two research questions emerge from these considerations, namely: RQ2.What benefits do travel agents expect from social media adoption?
RQ3.What is the owner/manager's perceived effectiveness of social media platforms prepandemic and what will it be post-pandemic?
2.3 Factors determining differences in social media adoption Several scholars have investigated factors influencing SM adoption in SMEs through the T-O-E model (Tornatzky et al., 1990), confirming that technological, organizational and environmental dimensions influence SM adoption, albeit with differences (Ali Abbasi et al., 2022;Ali Qalati et al., 2022;Matikiti et al., 2018;Salwani et al., 2009).
As pointed out by Taiminen and Karjaluoto (2015), SMEs are generally in an early stage of adoption while large firms are more likely to have the resources and knowledge needed to successfully adopt new digital channels and tools.In the same vein, He et al. (2017) highlight that large companies generally have ample financial resources and can designate full-time staff to manage SM; (2) firm seniority ( € Ozturan et al., 2019;Olanrewaju et al., 2020); (3) network membership (Abrate et al., 2020;Diaz-Chao et al., 2016); (4) the main activity carried out by the agency.By focusing on TAs, some scholars have highlighted that adopting ICT becomes advantageous when a TA organizes tours as its main activity rather than exclusively selling holiday packages (Abrate et al., 2020;Wu et al., 2016); (5) target of consumers to which the offer is addressed.This may also represent a relevant factor, as there are significant differences in the use of SM by type of audience.According to Xiang et al. (2015) who mention a survey conducted by the US Bureau of Transportation Statistics, the Y generationor Millennials (persons born between 1981 and 1996)demonstrates stronger activism and involvement in the use of digital channels for travel planning, compared to other generations (Kim et al., 2015).Similarly, Darmoyo and Sustaningrum (2022) state that SM activities developed by TAs have a significant and positive effect on Millennials' intention to purchase online holiday packages Consequently, it is conceivable that agencies that

Social media in travel agencies
focus their offering on younger targets, namely the Y and Z generations, will perceive greater effectiveness of their SM usage.
Additional factors, such as: (9) expertise and skill in using new technologies (Taiminen and Karjaluoto, 2015;Oji et al. (2017) and ( 10) daily use of SM for personal activities (Taiminen and Karjaluoto, 2015;McLaughlin et al., 2022;Matikiti et al., 2018).According to research conducted by He et al. (2017), in fact, most of the managers or owners of SMEs who used Facebook in their business for digital marketing purposes also adopted it on a personal level, declaring that keeping in touch with customers was a pleasant activity.
From those considerations the fourth research question is thus formulated: RQ4.What differences among agencies emerge regarding the perceived effectiveness of social media platforms?

Questionnaire and data collection
The questionnaire consisted of two sections.In the first, relating to the owner/manager, the demographic aspects taken into consideration were: gender, age and level of education.Furthermore, in order to grasp the person's level of expertise with respect to their use of the Internet, the amount of time spent daily online and their self-perceived level of technological experience were considered.
The second section, relating to the TA, took into account the characteristics of the company, and specifically: year of foundation, number of employees, participation in professional networks and turnover.In addition, the main activity carried out by the company (tour organizer and/or travel agency) and the type of audience (in terms of generation) were also taken into account.The final part of the questionnaire analyzed the online presence of the company, focusing on the use of different SM platforms and the underlying aims of the agency; on this point, respondents were asked to refer to the state of the art at the time of completing the questionnaire, as well as to predict the scenario at the end of the pandemic.
Afterwards, in order to ensure the clarity and exhaustiveness of the questionnaire, it was discussed with two managersthe director of marketing and the director of salesof one of the top players in tourism intermediation in Italy.These managers were chosen for their extensive experience in advising and supporting the thousands of TAs belonging to their customer network.
To date, as far as the authors of the present paper know, there is no exhaustive list of TAs operating in Italy.Therefore, a database provided by one of the main tour operators in Italy, which includes 9,000 companies, was used to constitute the reference population.The questionnaire was pre-tested on 20 companies, in order to identify any ambiguities in the formulation of the questions and the items.Subsequently, the companies were contacted by e-mail and the questionnaire was administered electronically.Data collection took place in the last four months of 2020.In total, 282 companies replied.
As regards the relatively low number of replies, it should be noted that the data collection took place in the midst of the Covid-19 pandemic emergency, making it particularly difficult to quantify the number of companies that were actually operational at that time.In fact, 22% TQM 35,9 of Italian TAs never reopened due to the effects of the pandemic (Confcommercio, 2020), which brings the reference population down to around 7,020.Moreover, a further percentage of companies, not easily quantifiable, were only temporarily inactive due to the restrictive measures adopted in Italy to tackle the pandemic.

Measures
Of the variables considered for the analysis, some are categorical: the gender of the owner/manager and the type of clientele in terms of incoming and outgoing tourists, and some are dichotomous: participation in networks and the use of social networks before and after Covid-19.The social networks referenced are those most used in Italy and most studied in the literature, namely WhatsApp, Facebook, Instagram, YouTube, Twitter and Telegram.Ordinal variables were also used, e.g.owner/manager's daily use of social networks and their level of education.The reasons behind the use of SM were measured using a five-point Likert scale.Many of the originally continuous variables used, such as the owner/manager's age, the year the company was founded and the number of employees, were treated as ordinal or dichotomous, thus coming to define new variables, and namely: the generation of the owner/manager, the size of the company, the seniority.As regards the generation to which the owner/manager belongs, the following classification was used as reference: baby boomers (born between 1946(born between and 1964(born between ), generation X (born between 1965(born between and 1980(born between ), and generation Y, or Millennials (born between 1981(born between and 1996)).Other variables, measured on five-point Likert scales, were also treated as ordinals, resulting in different "levels": the self-perceived ICT experience of the owner/manager and the level of focus on younger customers, belonging to generations Y and Z.As regards the main activity, the percentages of turnover attributable to activity as a TA vs those attributable to activity as a tour organizer were taken into consideration; a comparison between these percentages allowed us to define a categorical variable, adding a third modality, for hybrid activity, which defines companies whose turnover is attributable (less than 60%) to the activity as a travel agency and tour organizer.

Data analysis
Data analysis was carried out through the use of several techniques suitable for answering the research questions.To trace any significant differences in the use of social networks before and after COVID (RQ1), the McNemar test, which is considered for paired proportions of dichotomous variables (McNemar, 1947), was used.To address the second research question relating to the benefits sought from the use of SM (RQ2), the average and standard deviation for each of the reasons included in the questionnaire were calculated.A different technique was used to understand whether significant differences are found in TAs' perceived effectiveness of social networks before and after Covid-19 (RQ3); a pairwise t-test (Xu et al., 2017), which can be used if the dependent variable is continuous, was performed.Finally, to examine the perceived effectiveness of the use of different SM in the pre-COVID period, considering multiple characteristics of the owner/manager and of the company (RQ4), a multivariate analysis of variance (MANOVA) was carried out (Huberty and Olejnik, 2006).This choice was made because MANOVA allows several continuous dependent variables to be considered simultaneously (in our case, the perceived effectiveness of the use of the SM considered), identifying significant differences due to an independent variable (the characteristics of the owner/manager and of the company, considered individually).As mentioned in the previous subsection, some independent variables that were originally continuous were treated as ordinals.In particular, starting from the original five-point Likert scale, three levels were identified, thus leading to a configuration of the independent groups.

Findings
Descriptive statistics for all the variables considered are shown in Table 1.

Social media usage (RQ1)
The first research question aims to investigate the use of SM platforms among the TAs in the sample before and after the COVID-19 pandemic.The McNemar test revealed significant differences among the agencies with reference to their use of most of the SM considered (see Table 2).The most evident increase concerns the use of YouTube (χ 2 5 45.833; p < 0.000), with 56 agencies that began using it with the advent of the pandemic.Twitter also shows a significant increase ðχ 2 5 37.123; p < 0.000), with 47 agencies adopting it.The same happens for Telegram ðχ 2 5 34.588; p < 0.000), with an increase of 43 agencies among its users.Finally, significant differences emerge also with reference to Instagram ðχ 2 5 19.314; p < 0.000), undertaken by 27 agencies.It should be noted that, despite the growth in the use of YouTube, Twitter, Telegram and Instagram, they continue to be less widespread than WhatsApp and Facebook. Moreover, between  Telegram and Twitter was more limited (54 and 63% of the agencies, respectively) while YouTube and Instagram, in particular, accounted for a large part of the sample's SM use (77 and 88%, respectively).

Benefits of social media adoption (RQ2)
With reference to the second research question, the mean and the standard deviation were calculated for each of the potential benefits related to the use of SM.As shown in Table 3, almost all the variables have a rather high average value, considering that they are measured on a scale from 1 to 5. The main motivation for SM use is related to strengthening of the corporate image (x 5 4.401), whose standard deviation (SD 5 0.884) denotes a lower variability around the mean, compared to the other variables.The other reasons with higher average scores are improvement of the effectiveness of promotional activities (x 5 4.160), whose dispersion around the mean is rather low (SD 5 0.950), opportunity to provide assistance to customers during the holiday (x 5 4.106; SD 5 1.088) and creation of a community with customers (x 5 4.082; SD 5 1.032).The reasons that appear to be relatively less important for agencies include sale of online packages (x 5 3.284; SD 5 1.309) and customer loyalty ðx 5 3.560; SD 5 1.159).

The perceived effectiveness of social networks (RQ3)
The third research question required the employment of a pairwise t-test.The objective was to trace (any) significant differences that emerged with reference to the perceived effectiveness of the different SM considered, between the pre-pandemic and post-pandemic periods.The results are shown in Table 4.As shown in Table 5, the pairwise t-test reveals significant differences (p < 0.000) in the perceived effectiveness of all the SM considered with reference to the two periods.In all the cases shown, the TAs' perceived effectiveness of SM increases significantly.The greatest rise concerns YouTube, which passes from a fairly low perceived effectiveness in the pre-COVID period (x 1 5 2.007) to a much higher perceived effectiveness in the subsequent period (x 2 5 2.901).There is also a significant increase in the perceived efficacy of Instagram (x 1 5 3.018 and x 2 5 3.762).A smaller but still substantial upsurge is noticed with Facebook (x 1 5 3.638 and x 2 5 4.007); it should be noted that in this case, as also happened with WhatsApp, the increase is significant despite the fact that the perceived effectiveness of SM in the pre-Covid-19 period was already very high.Likewise, Twitter and Telegram, which started from rather low average levels of perceived effectiveness (respectively x 1 5 1.574 and x 1 5 1.525), registered a sizeable uptick, although they are still perceived as the least effective SM platforms (x 2 5 2.103 and x 2 5 2.004).

4.4
The differences among agencies regarding the perceived effectiveness of social media (RQ4) An assessment of TAs' perceived effectiveness with reference to firm size highlights significant differences (F 5 2.063; p 5 0.018).As shown in evident for YouTube (F 5 4.184; p 5 0.016) and Twitter (F 5 3.932; p 5 0.021).For YouTube, as the size of the agency increases, the perceived effectiveness increases, as well; the gap appears particularly stark between small enterprises (x m 5 3.000) and the other groups (1.915 for sole proprietorships and 1.964 for micro-enterprises).Smaller, but still significant, is the difference among the groups in their perceived effectiveness of Twitter.In this case, however, micro-enterprises (x m 5 1.511) report a slightly lower level of effectiveness than individual enterprises (x i 5 1.660), while both these groups are very far from small enterprises (x s 5 2.286).
TAs' perceived effectiveness of SM was also examined with reference to firm seniority.As shown in Table 6, there are no significant differences in the perceived effectiveness of nearly all the SM platforms considered between companies operating for less than ten years and those in business for more than ten years.Only in the case of Instagram are there significant differences between the groups (F 5 7.516; p 5 0.007): younger companies attribute to SM an average effectiveness (x y 5 3.440) that is considerably higher than that attributed by the more senior ones (x s 5 2.865).
As regards network membership (Table 7) and the main activity carried out by travel companies (Table 8), no significant differences were found in perceived effectiveness for each of the SM platforms under consideration.

Social media in travel agencies
Turning to agencies' focus on younger targets, as reported in Tables 9 and 10, no significant differences, overall, were found, either with reference to generation Y (F 5 1.928; p 5 0.076) or to generation Z (F 5 1.194; p 5 0.310).As for the perceived effectiveness of individual SM, considering the Generation Y target, there are significant differences in perception only for WhatsApp (F 5 6.857; p 5 0.009), as illustrated in Table 9.In this case, the perceived effectiveness is higher for companies that are strongly focused on consumers belonging to this generation, rather than for those in other groups.
With regard to TAs that focus on Generation Z, however, only the perceived effectiveness of Telegram is significantly different among groups (F 5 4.053; p 5 0.041).In particular, the perception is higher for companies that are moderately focused on consumers belonging to this generation, as shown in Table 10.
When factoring in the characteristics of the TA owner/manager, overall, there no significant gender-based differences emerged between the groups.As reported in Table 11, among the different SM platforms, only in the case of WhatsApp was the level of perceived effectiveness significantly different for male and female owners/managers (F 5 6,623; p 5 0.011): it was higher for the men (x m 5 4.074) than for the women (x f 5 3.681).
As regards the education level of the owner/manager (see Table 12), significant differences in perceived effectiveness between the groups were found (F 5 6.558; p < 0.000).This was true for nearly all the SM considered.The most relevant differences were found for Facebook (F 5 35.150; p < 0.000) and Instagram (F 5 22,262; p < 0.000), which are considered significantly more effective by owners/managers with a medium-high level of education; YouTube, Telegram and Twitter showed similar results, albeit with less marked differences.The outlier, WhatsApp, is the only SM platform whose effectiveness is perceived similarly among owners/managers with a medium-high and medium-low level of education; this is Considering the generation to which the owner/manager belongs, significant differences emerged overall (F 5 5.225; p < 0.000) and, in particular, for Instagram (F 5 15.621; p < 0.000), Facebook (F 5 11.610; p < 0.000), WhatsApp (F 5 10.758; p < 0.000) and Telegram (F 5 3.094; p 5 0.047).As shown in Table 13, in all these cases, the perceived effectiveness is higher for owners/managers belonging to the Generation Y, the youngest among the respondents.And, generally speaking, owners/managers that are Gen Xers have a higher perception of efficacy than do Baby Boomers.
Another factor worth considering that could affect TAs' perceived effectiveness of SM is the owner/manager's level of experience with new technologies (see Table 14).In general, the results highlight significant differences among owners/managers according to whether they believe they have a high, medium or low level of experience (F 5 4,218; p < 0.000), for most or all of the SM platforms, except for WhatsApp.The differences in perceived effectiveness are particularly evident as regards Instagram (F 5 22.947; p < 0.000), Facebook (F 5 14.047; p < 0.000) and YouTube (F 5 9.002; p 5 0.002).In general, a higher level of experience corresponds to a higher perceived level of effectiveness; however, Twitter and Telegram are exceptions to this rule.Finally, the amount of time spent daily online by the owner/manager for personal activities significantly affects their perceived effectiveness of SM.As shown in Table 15, the overall difference among the groups is significant (F 5 7,699; p < 0.000), which the highest differences for Facebook (F 5 39.168; p < 0.000), Instagram (F 5 23.431; p < 0.000) and Telegram (F 5 8.819; p 5 0.003).In all these cases, however, as the TA owner/manager's time spent online increases, their perception of the effectiveness of the various SM platforms grows.

Discussion
The findings are in line with others on SMEs (Ali Abbasi et al., 2022;Abou-Shouk et al., 2016), confirming that social network usage represents a competitive factor for the TA industry, even for very small firms with a lack of financial, technical and human resources.In fact, the present study has found that travel agents extensively use SM, making their companies multi-platform organizations that try to reach customers through a variety of touch points, in an omnichannel perspective common to many retail sectors (Ruiz and Garcia, 2022;Sharma et al., 2020).
Regarding specific SM tools, the sample TAs have shown a wide adoption of WhatsApp (96%), Facebook (95%) and Instagram (78%), but with very different levels of perceived effectiveness, however.While WhatsApp (3.9) and Facebook (3.6) were considered to be the most effective tools in the pre-pandemic period, Instagram (3.0) achieved only a just-above-average evaluation; other social networks (YouTube, Twitter, Telegram) were perceived as having low effectiveness.In line with Oji et al. (2017), WhatsApp appears to be the tool considered by micro and small agencies as the most effective for managing direct relationships with customers, probably due to its affordability and ease of use.
In line with other studies (Hu et al., 2023;Sigala, 2020), additional evidence concerns the intensification of SM use following the Covid-19 pandemic.This acceleration could be interpreted as increased use of the main social tools (WhatsApp and Facebook) by those agencies which, until that moment, had maintained a traditional profile; it could also be the result of the need to expand the number of social networks used in order for TAs to take advantage of the individual peculiarities of each one (YouTube is the social network that scored highest on the McNemar test).
The increased adoption of SM by TAs is closely linked to the significant awareness that emerges from the study about their prospective or predicted effectiveness in the The motivations that lead TAs to increase their SM usage (RQ2) seem to be in line with other studies, namely those on corporate brand promotion (Barcoe and Whelan, 2018;Asongu and Odhiambo, 2019), on relational purposes (Ali Abbasi et al., 2022;Foltean et al., 2019) such as activation of value co-creation processes (Yang et al., 2014), improvement of information quality (Albattat, 2020;Sharma et al., 2020;€ Ozturan et al., 2019), technical assistance during the holiday (Fernandes et al., 2016;Asongu and Odhiambo, 2019) and creating communities with customers (Roth-Cohen and Lahav, 2019).The significant values attributed to relational motivations emphasize the need to maintain contact with customers, which remains a critical success factor for traditional TAs (Dini et al., 2022), as well.Unlike other studies (S ¸engel et al., 2021;Oji et al., 2017) the purely economic/income motivations, although present, do not seem to play a decisive role in the choice of using social networks (Bocconcelli et al., 2017).
Table 16 summarizes the findings relating to the differences among the groups regarding the perceived effectiveness of social platforms (RQ4).
The results show that within the sample of TAs analyzed, the personal characteristics of the entrepreneur/manager play a critical role in influencing their attitude toward SM adoption (Taiminen and Karjaluoto, 2015); these features include innovation capability (Sahoo, 2019), level of education, age, personal time spent on social networks and (perceived) level of experience.
The emphasis on the personal dimension of the TA owner/manager appears to be particularly true in the context of micro-enterpriseswhich represent 95% of our sample where the approach to SM is strongly correlated to the private habits of entrepreneurs and employees (McLaughlin et al., 2022;He et al., 2017) entrepreneur is younger, more qualified and better skilled in online communication, the use of SM turns out to be more intense and is perceived as a driver of business success (Matikiti et al., 2018).
Comparing the platforms, the following main differences emerge: (1) WhatsApp is the most used tool, with no relevant differences among TA groups; (2) Facebook, Instagram, YouTube and Telegram show more differences among the TA groups relative to owner/manager characteristics than to firm-specific factors; (3) Twitter is perceived to be of greater use by companies on the larger end of the SME scale and by more experienced owner/managers; (4) YouTube is considered more effective by larger companies (having 10 to 49 employees), probably due to the skills and financial resources that video-production requires, which much smaller companies often lack.

Conclusions
Traditional TAs have been able to assert their role in the tourism industry, despite the digital revolution taking place that could be jeopardizing their survival.However, notwithstanding the importance of these players in the tourism system, academic research shows several gaps, regarding digital marketing, in particular, and specifically with respect to the use of SM to improve the customer experience and the agency's competitive performance, in general.The present study fits in this gap, also highlighting the future impact that Covid-19 could have on these digital processes (Rahman et al., 2022).
The main findings indicate that although small TAs are experiencing a phase of digital transition, as emerged from previous studies (Sharma et al., 2020), there are substantial differences among them regarding their propensity to adopt technology.Moreover, the study shows that SM are considered important by travel agents, who perceive them as tools that involve the opportunities for increased notoriety-reputation, aspects of marketing information (to profile customers), strategic marketing (to strengthen positioning), operational marketing (to create, communicate and deliver value), and relationship marketing (to create engagement, retaining, satisfying customers), all along the customer journey (Pop et al., 2022).
Our findings point to Facebook, WhatsApp and Instagram as the most-used SM networks, but the pandemic is pushing their expansion in terms of variety and intensity, in order to synergistically exploit the different characteristics they possess (Park and Oh, 2012).

Theoretical implications
From a theoretical point of view, the study touches upon and contributes to deepening various aspects of the literature on the relationship between TAs and SM.Six different study perspectives can be identified: (1) usage of SM by SMEs (Matarazzo et al., 2021), in particular with respect to micro-enterprises (Vatanasakdakul et al., 2020: Jones et al., 2015); (2) SM adoption in the tourism industry (Magno and Cassia, 2018;Leung et al., 2019) with reference to TAs, focusing on the importance of social tools for the purpose of achieving marketing objectives; (3) omnichannel strategies in the retail sector (Capriello and Riboldazzi, 2020) implemented by tourism intermediaries in order to ensure a relationship with the client throughout the entire customer journey and achieved by setting up a multiplicity of contact moments, considering the different levels of the digital orientation of businesses and consumers; (4) rule of the pandemic on technology use among tourist enterprises Social media in travel agencies (Sigala, 2020), also in line with recent studies (Hu et al., 2023), where the pandemic seems to have accelerated the digitization process of those who had already undertaken it and, at the same time, induced the most reticent and traditional ones to start it; (5) factors that influence the SM adoption by SMEs and TAs (Ali Abbasi et al., 2022;Ali Qalati et al., 2022;Dahnil et al., 2014).The empirical analysis confirms that in small businesses the personal factors of the owner/manager are more discriminating than firm-specific ones, because the micro/small dimensions of TAs are characterized by a typically entrepreneurial approach to marketing (Morris et al., 2002) particularly suited to an informal organization, structured with few or no hierarchical levels, and strongly projected toward customer needs; and (6) factors that influence the adoption of specific social networks (Facebook, WhatsApp, etc.), expanding research perspectives also considering the evolution of new SM platforms.

Managerial implications
SM have become even more central for TAs, contributing to brand enhancement, improving their promotional activities and facilitating the whole customer journey management (Pop et al., 2022).
However, the central role that digital platforms are taking on in the market must not replace the agency's human resources in any way, but rather should be used to emphasize the human relationship between customer and agent, while also improving the buying experience, thus ensuring the sustainability of the business model of traditional TAs (Law et al., 2004).The prospects of technological development on the one hand (Buhalis et al., 2022;Bolici et al., 2020) and the hopefully imminent end to the pandemic, on the other, will push the digital revolution even further.For this reason, the next few years will be characterized by the need to know how to integrate the physical world with the digital one, valuing all new technologies while remaining strongly oriented toward people and humanity (Pencarelli et al., 2021;Kartajaya et al., 2021).
In this perspective, important considerations arise for micro-and small-TAs which, due to their structural and resource limitations, risk not being able to exploit the potential of digital infrastructures.In fact, the empirical research highlights how the successful use of SM in micro-TAs depends to a significant degree on the personal characteristics of the agency employees (including the owner/manager).Moreover, this is why it is necessary to implement careful recruitment policies and foster employee engagement (Nienaber and Martins, 2020), as well as invest in training (Hu et al., 2023), also through participation in networks and trade associations (Abrate et al., 2020).Furthermore, the importance of increasing travel agents' digital skills through participation in training courses on digital and SM marketing, as well as in communities of practice to share experiences and innovative solutions has become evident.From this point of view, public policies and university institutions can provide valid support to ensure more in-depth digital skills, especially for non-digital natives who risk becoming mired in a dangerous delay and finding themselves victims of a digital divide with younger and more prepared competitors.
Travel agents, even without a marketing background and without having specialized skills, often prove to be naturally market and customer oriented; for this reason the most popular platform is WhatsApp, which plays an important role in "problem solving" and "customer centered" marketing philosophy (Presti et al., 2020;Munikrishnan and Al Mamun, 2021).However, the need to expand the range of social platforms adopted and the increase in the complexity of multimedia communication requires a major effort, technologically speaking, by TAs, especially the smaller ones, which could appear to be inadequately equipped for the purpose ( € Ozturan et al., 2019).In this sense, a further managerial implication concerns collaboration with tour operators, who can represent strategic partners for the production of multimedia content capable of overcoming some of the limitations that microenterprises face.TQM 35,9

Limitations and future research directions
The study has two main limitations, a focus on the Italian context only and the missing lack of the consumer's point of view.The latter, especially, prevents an exhaustive evaluation of future trends regarding the use of SM in the client-agency relationship.
To have a complete overview of the phenomenon, a comparative and international investigation on TAs' strategies could be pursued.Furthermore, a possible line of future research should include studies on the demand side, with the aim of assessing the interest in SM and how these platforms can contribute to a better shopping experience and build loyalty to the agency.In future research studies, it may be possible to evaluate to what extent the substitution processes can be implemented by them.Furthermore, as Del Chiappa and Fotiadis (2019) observe "future research should be developed to examine the different types of innovation that can occur in SME tourism businesses and how these are distributed via the process of knowledge transfer" and how "ICTs, Internet, SM give rise to new business species (both in terms of service providers and intermediaries), thus giving rise to a fierce competition among "old" and "new" actors (e.g.Airbnb, Uber, etc.) that call for a dynamic realignment and redefining of the relationships that underpin the economy (Gretzel et al., 2015).
. In those agencies where the