Keywords

1 Introduction

Researchers are trying to understand which factors influence customer engagement, by investigating both consumers’ behaviors and willingness to engage with the brand, on one hand and by trying to find some metrics to measure this engagement, on the other hand. The majority of studies indicate that certain post characteristics like post richness, post content, post interactivity, timing and frequency of posts should be taken into account when creating and organizing posts. Most of researchers agree that user’s engagement is reflected in the form of liking or disliking a post, commenting to a post or sharing a post to his wall [1] and use the number of likes, number of comments and number of shares as different measures of engagement. Although there is a growing interest from the researchers, there are some gaps: Most empirical works try to investigate engagement in case of multinational brands. For smaller businesses, however, managing customer engagement in social media efficiently is even more important since they see social media as convenient and appropriate and need to make the most of their comparatively limited social media marketing resources [2]. Furthermore, only a limited number of works investigate how various firm-determined characteristics of branded social media content affect different types of engagement in specific sectors. However, it is crucial to understand industry-specific engagement [3] because various engagement behaviors have been connected to various businesses, including tourism [4], airlines [5], and wine [6, 7]. Wineries are increasingly adopting social media, but many still don't have the knowledge to use it effectively in their marketing plans [8]. Despite social media's substantial benefits on firms, digital marketing for small and medium enterprises (SMEs) wineries is still challenging.

The goal of this research is to ascertain the various company created content on consumer engagement on Greek wine brand's postings in order to fill the aforementioned gaps. This study satisfies the requirement to understand how SMEs, which make up the majority of European wine companies, should produce their social media content in order to maximize their limited spending [9].

2 Literature Review

The Uses and Gratification approach has been used widely by researchers for studies on social media as the level of involvement, interactivity, bonding of participants and social capital is higher [10]. Researchers try to investigate which types of content influence if and how consumer’s engagement with brands on social media platforms. Most studies categorize social media content in informative [11], rewarding [6], entertaining [12], or relational content [6, 13]. Some others classify content according to its emotional characteristics or in terms of Persuasive and non-persuasive, Transactional, Promotional, Interactive, Trendy and Customized, Inspirational, Call to action and Shared or original content [14].

The motivation for obtaining information can be categorized as seeking knowledge, guidance, or opinions exchanging information, and receiving real-time information [15]. However, there is limited empirical evidence [4] that supports a significant positive effect of informational content on customer engagement, while many other studies [12, 16] did not find such an effect. Moreover, in works where engagement is measured both in terms of likes, shares and comments, the results are ambiguous. Rewards like financial incentives, vouchers, and special offers are connected to rewarding content and are meant to draw interest and boost engagement [6]. People are expected to be engaged positively with rewarding posts, however, there is also empirical evidence that support a negative effect [17]. According to [18], entertainment posts are entertaining, amusing, and engaging posts, comical videos, tales, wordplay, and generally creative and inventive content that piqued consumers’ attention and prompted them to share it with their peers. Most of previous studies have proved that entertainment is the main driver and result in higher engagement [19]. However, there are several studies that support an insignificant effect [16]. Relational postings are linked to information that promotes conversation by asking and answering questions, offering wishes, celebrating, and mentioning international holidays. Empirical evidence on customer engagement support both positive [16, 20] and negative or no effect [12] of relational content on engagement.

Research suggests that messages with images are richer than text, and videos are more attractive [17, 21]. However, as research has shown that images are even more effective sources of engagement than videos, these two types of information have frequently been included together in one “media richness” category [4]. Another set of variables used by scholars, refer to posts’ characteristics such as length of message, time of posting and day of posting. Researchers place a strong emphasis on identifying the most opportune time to post material on social media to enhance customers’ desire to engage. The availability of working memory, according to [22], rises in the morning when people are more alert, attentive, and looking for information in electronic surroundings.

Most of researchers agree that user’s engagement is reflected by liking or disliking a post, commenting to a post, or sharing a post to his wall and use the number of likes, comments and shares as different measures of engagement [23, 24]. Scholars also use the sum of likes, shares, and comments, either weighted by brand’s followers or not. Dolan [6] include dormancy, separation, and destructing in addition to creating, contributing, and consuming. A weighted engagement index is proposed by [25]. A limited number of works investigate how various firm-determined characteristics of branded social media content affect different types of engagement in specific sectors such as financial [11], hospitality [4, 26], education [23], fashion [27], automotive [28]. There is a limited number of works [7, 29] that investigate how various firm-determined characteristics of branded social media content affect different types of engagement in specific sectors and specially in wine sector.

The following hypotheses are formulated:

H1: Informational posts positively influences consumer engagement

H2: Rewarding posts positively influences consumer engagement

H3: Entertaining posts positively influences consumer engagement

H4: Relational posts positively influences consumer engagement

H5: Network size positively influences consumer engagement

H6: Media Richness positively influences consumer engagement

H7: Posts created on workdays result in higher level of consumer engagement

H8: Posts created in working hours result in higher level of consumer engagement.

3 Methodology and Results

Greek wineries’ Facebook brand pages were chosen for the purpose of conducting the analysis of this research. According to Szolnoki et al. [30] across wineries in multiple nations, Facebook is by far the most used social media platform. Facebook’s market share of social media in Greece is 83.97% [31] and is the most frequented social media adopted by wine marketers [32].

Facebook pages of Greek wineries were analyzed using Next Analytics software. Posts were retrieved daily from January to December of 2019 in order to assure data accuracy and independence from any changing Facebook policies. Number of fans, likes, posts, shares, reactions, links, images, shared stories, videos, comments, post day, and post hour are the major statistics collected for each wine brand page.

We use four dependent variables to measure online engagement on Facebook brand pages: liking, commenting, sharing, and the sum of those three. A post with a lot of likes and shares could mean that its content is interesting, boosting the likelihood that someone will like it and spreading the brand message to more potential customers via the Facebook algorithm. In addition, a post's success or influence can be gauged by the number of comments it receives, as this indicates that readers took the time to voice their thoughts [33]. We selected posts with more than 150 engagements (sum of reaction, shares and comments). A total of 473 posts of 32 wine firms were analyzed. In terms of media elements, according to Nextanalytics data [34], four formats are available: text, photos, videos and shared stories. In our case of 473 posts with the higher engagement, 85.6% are posts with photos, while posts with video are 5.1% and shared stories are 6.76%. According to our data, mean shares per post is 15.3 while mean reaction per post is 295.

In order to assign the content type categories to the posts created by page moderators we performed a qualitative content analysis. Both authors involved in this study reviewed the social media literature to identify pre-existing coding categories for wine brand posts. Keywords, themes, and ideas that seemed to characterize the brand posts were highlighted and discussed. The 473 posts were categorized in four major categories: informative, entertaining, rewarding, and relational. As Informative posts we selected those that were given in form of general information about firm (history, owners, local geographic location, website, open hours, contact information etc.), about specific products (description of products, varieties, their terms of storage and also about harvesting (conditions, characteristics etc.) and cellars conditions. In the category Entertainment we included those posts which were written in a form of word play or slogan, or humorous images and statements. The aim of entertainment posts is to offer some relaxing and interesting content to enjoy their followers, to make them interact [19]. In Rewarding category, we included posts in a form of competitions, contests or prizes to users. Relational posts are associated with content encouraging user participation and interaction, like wishes, celebrations, international days, name days, local events, family and employees’ personal interests, hobbies, opinions, achievements, and so on.

From the total 473 posts with more than 150 engagements, informational posts are the most frequently met 56.5%, while relational ranked in the second place with contribution of 20.9%. These were followed by entertaining posts (15.9%), and in the last position rewarding (6.8%). According to the OLS regression results, the model for the consumer engagement on brand’s posts measured as the sum of reactions per brand post i, shares per brand post i, and comments per brand post i, is significant as a whole (F-value = 10.41, p-value < 0.001) and explains the variance of the dependent variable reasonably well (adj. R2 = 14%). Providing information at a wine brand post is marginally significant and positive related to the consumer engagement on brand’s posts in support of hypothesis 1 (Table 1). Rewarding content is significant and positive related to brand post engagement, in support of hypothesis 2. Entertaining content is not significantly related to engagement, so we cannot support hypothesis 3. Relational content is significant and positive related to brand post engagement, in support of hypothesis 4. Network size is related positive and significantly to brand post engagement, in support of hypothesis 5. Media richness of a brand post is not found significant related to engagement, which is not in support with hypothesis 6. Posting on workdays is not significantly related to engagement while posting on working hours is significant and positive related to brand post engagement, in support of hypothesis 8.

Table 1 Supported and not supported hypotheses
Full size table

The models for the engagement measured by the reactions per brand post i, the engagement measured by the shares per brand post i, and the engagement measured by the comments per brand post i, are significant as a whole (F-value = 10.35, 13.22, 7.43, respectively, p-value < 0.001), and explains the variance of the dependent variable reasonably well (adj. R2 = 12%-18%). While providing information at a wine brand post has marginally influence related on the engagement, there is no influence in the case where engagement is measured by the number of reactions, shares and comments, separately. Whether a brand post is informative, rewarding or entertaining has no influence on the engagement measured by reactions. Only relational content of a brand post has influence on reactions, in support of hypothesis 4. Rewarding content is significant and positive related to the number of shares and to the number of comments, supporting hypothesis 2. In the case of engagement measured by the number of comments, relational content has significant influence on comments, in support of 4. Network size has influence only in the case of brand post engagement measured by reactions and shares. Richness of the brand post has positive and significant influence on reactions, negative and significant impact on shares and no influence on comments. Finally, posting on working hours is significant and positive related only to number of comments.

4 Discussion and Conclusions

This study provides three contributions. The first is the creation of a research model used in this study which takes into account important elements that have been identified as significant in the literature review. The second is the emphasis given on SMEs instead of multinational brands since effectively managing customer engagement in social media is even more crucial for smaller companies which need to make the most of their relatively smaller social media marketing budgets [2]. Finally, this study’s third contribution is the investigation of how various firm-determined characteristics of branded social media content affect different types of measurement of engagement of customer engagement on social media posts in a certain sector, Greek wine sector.

From the managerial perspective, this work offers some important practical implications for managers and marketers especially for wine brands with a small-medium fan base on how to manage their Facebook brand pages. Specifically, this research gives practitioners the first concrete evidence that social media content, particularly rewarded material, and the timing of postings contribute to increase customer involvement. Practically speaking, managers could adopt the strategy of enhancing customer connection by creating and delivering tailored experiences for various consumer engagement and preference segments. Rewarding content is the most consistent and promising variable for an increase in customers’ engagement. It seems that rewarding is the way to start a relationship with customers and after that brands should feed and maintain this relationship. Apart from rewarding content, managers should also invest on interesting, rich, and inventive postings. They should also be aware of the difficulties in creating informative, trend-setting content, and entertaining material through key market.

Future researchers should test additional approaches to measuring engagement and, more importantly, novel approaches to linking the engagement index to interactivity and customer satisfaction through a variety of targets. Second, investigations into engagement should focus on the unique characteristics of wine brands, such as brand experience, brand quality, and personality, as well as how these attributes are transmitted via social media. Finally, both researchers and practitioners would greatly benefit from a comparison examination of interaction utilizing more thorough and varied Facebook wineries’ brand pages from other nations.