A persuasive eWOM model for increasing consumer engagement on social media: evidence from Irish fashion micro-influencers

Zheng Shen (College of Media and International Culture, Zhejiang University, Hangzhou, China)

Journal of Research in Interactive Marketing

ISSN: 2040-7122

Article publication date: 31 May 2021

Issue publication date: 21 June 2021




This study aims to find how can fashion micro-influencers and their electronic word-of-mouth (eWOM) messages increase consumer engagement on social media, focusing on micro-influencers’ influence, typology, eWOM content and consumer engagement.


A total of 20,000 microblogs were collected from Irish fashion micro-influencers and analyzed through keyword classification and content analysis in NVivo. The determinants of eWOM persuasiveness for consumer engagement on social media were investigated based on Sussman and Siegal’s information adoption model.


The study finds that among the four types of micro-influencers, market mavens and their eWOM messages have the highest impact on consumer engagement on social media, and it presents a repetitive and persuasive eWOM model of market mavens to increase consumer participation. Also, the study discovers that micro-influencers’ occasion-related microblogs have an increasing impact on consumer interactions whereas microblogs with brands have a decreasing engagement with consumers on social media.


This study advances prior studies on the relationship between influencers’ eWOM messages and consumer participation on social media by the development of a persuasive eWOM model of micro-influencers to increase consumer engagement and fill in the lack of relevant literature. Also, findings provide actionable insights for marketing communication practitioners to persuade consumers to participate in eWOM communications and establish strong consumer-brand relationships on social media.



Shen, Z. (2021), "A persuasive eWOM model for increasing consumer engagement on social media: evidence from Irish fashion micro-influencers", Journal of Research in Interactive Marketing, Vol. 15 No. 2, pp. 181-199. https://doi.org/10.1108/JRIM-10-2019-0161



Emerald Publishing Limited

Copyright © 2021, Zheng Shen.


Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial and non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence may be seen at http://creativecommons.org/licences/by/4.0/legalcode


Electronic word-of-mouth (eWOM), derived from word-of-mouth, is a critical technique for online marketing communication and consumer interaction. In the past, WOM was widely accepted as an important determinant of consumer behaviour in marketing communication (Kozinets et al., 2010). For instance, both positive and negative WOM messages can drive consumers’ purchase possibilities (Sweeney et al., 2014); and interpersonal relationships (e.g. strong ties) can also affect the purchase decision of consumers in WOM communication (Voyer and Ranaweera, 2015). With the advance of technologies, eWOM becomes an augmentation of traditional WOM, and social media has revamped face-to-face communication into computer-mediated eWOM. Thus, eWOM is defined as “any positive or negative statement is made available to a multitude of people and institutions via the Internet” (Hennig et al., 2015, p. 39). Compared with WOM, typically, eWOM is characterized by a many-to-many electronic communication process, which provides high volumes of information to a large number of consumers promptly. Prior studies on eWOM concentrated on its persuasive role in consumer purchase decisions. They indicate that persuasive eWOM engagement can positively affect consumers’ attitudes and purchase intentions (Wang et al., 2015; Hayashi et al., 2017).

In spite of significant managerial and academic attention on persuasive eWOM communication in consumer–brand relationships, the past literature in the context of social media is still limited (Ismagilova et al., 2016). A few attempts have been made to investigate the role of social media in eWOM persuasiveness, but they have not considered the persuasiveness of eWOM for consumer engagement on social media (Ismagilova et al., 2016). For instance, it is reported that social media influencers have a high level of influence in targeting consumer groups and their purchase behaviour (García-de-Frutos and Estrella-Ramón, 2021). Particular examples are bloggers, Twitter users with a large number of followers, Youtube channel owners, etc. Their eWOM messages are thus critical to increasing consumer participation. Unfortunately, few studies looked into it, and previous studies focused on the context of SNS, forums and chatrooms for investigations (Hajli, 2018). As a consequence, the present study poses the following question:


How can micro-influencers and their eWOM messages increase consumer engagement on social media?

For this purpose, the study investigated 20,000 microblogs from Irish fashion micro-influencers, focusing on the relationship between their influence, typology, eWOM content and consumer engagement. The category of fashion was selected for the study because the existing literature focuses on the hospitality and tourism sectors (Verma and Yadav, 2021). However, past literature indicates that consumers are motivated to purchase fashion products because of their engagement in eWOM marketing on social media (Kim and Ko, 2012). Therefore, some brand companies hire micro-influencers for writing microblogs to interact with their consumers (Rettberg, 2014). Regarding the lack of eWOM analysis of fashion consumer engagement, this study further refines the research question: How can fashion micro-influencers and their eWOM messages increase consumer engagement on social media?

Overall, the findings of this study provide theoretical and managerial significance to the existing literature. Previous studies have not explained the relationship between influencers’ eWOM messages and consumer participation on social media (Chwialkowska, 2018; Lou and Yuan, 2019). As a result, this study develops a persuasive eWOM model to increase consumer engagement and to fill the lack of relevant literature. Findings show that fashion microblogs related to occasions have an increasing impact on consumer interactions while microblogs with brands have the least engagement with consumers on social media. More importantly, the study finds a repetitively persuasive eWOM model of market mavens, which has the most interaction among consumers on social media. For this reason, marketing communication practitioners are recommended to identify market mavens from different types of micro-influencers in social networks, and follow this model to spread eWOM messages and increase consumer engagement on social media.

The remainder of the paper is structured as follows. The next section reviews the relevant literature on eWOM and presents the propositions. It is followed by the section that details the methodology of the study. Subsequently, the section reports the results of the propositions and discusses the theoretical and practical implications of the findings respectively. Finally, the paper summarizes the findings and limitations of the study, and proposes future research directions.

Theoretical perspectives and model development

Electronic word-of-mouth on social media

Studies of eWOM on social media primarily focus on consumer behaviours (Godey et al., 2016), strategic perspective (Williams et al., 2014), consumer loyalty (Kandampully et al., 2015) and typology (Weisfeld-Spolter et al., 2014). In brief, the studies are divided into two main directions: emotion-oriented and content-oriented research. In the emotion-oriented exploration of eWOM, researchers try to figure out the positive, neutral and negative attitudes of people towards eWOM marketing. For example, Hayashi et al.’s (2017) experiment certifies that negative eWOM is good at developing trust, supporting product recommendations and motivating product buying behaviours. Regarding content-oriented studies, scholars are more interested in collecting online consumers’ reviews. Through reviews, scholars intend to show that eWOM benefits persuasion in marketing. It is reported that 70% of consumers trust online reviews on products and services (Nielsen, 2012). Recently, this group of researchers is inclined to discuss textual features of eWOM messages. Wang et al. (2015) further analyzed what kind of characteristics in eWOM messages led to convincing consumers online. This offers the basic premise of the present study in that the content of eWOM messages affects consumer behaviour in marketing communication.

Although many studies have considered content-oriented eWOM persuasiveness, they have focused on the hospitality and tourism industries. Verma and Yadav (2021) reviewed the literature on eWOM from 2000 to 2020, discovering that hospitality and tourism is one of three clusters as important subthemes of eWOM research. However, eWOM messages may have different effects on people depending on contexts and settings (Sussman and Siegal, 2003; Kyu Kim et al., 2020). Wolny and Mueller (2013) analyzed the participation of fashion consumers in eWOM communication on social media, extending the eWOM framework to other contexts, and indicated that they would like to see more extensive eWOM research conducted in the fashion context, which paved the way for future studies. Therefore, the category of the fashion industry is selected for the present study to expand the literature on eWOM persuasiveness.

In addition, social media context in eWOM is another area that holds the potential for interesting future research (Mishra and Satish, 2016). Verma and Yadav (2021) pointed out that social media has become a popular basic theme of eWOM research since 2017, whereas eWOM and user-generated content have become transversal themes. For example, Lever et al. (2017) shed light on the rise of eWOM for influencing travellers’ shopping patterns on social media based on Anahita Khazaei’s (2016) data analysis process, including the application of NVivo 10 software, axial coding through mind mapping and between-group analysis. However, most of these studies focus on the exploration of eWOM content on websites, and other sources of eWOM content from online communities can be considered in future research (Mishra and Satish, 2016). In a similar vein, Patsiotis and Kapareliotis (2016) propose that future research on digital marketing needs to explore eWOM and social network marketing in various contexts and focus on the primary marketing texts. As a consequence, the present study shifts the investigation of eWOM messages on websites to social media content in microblogging, and develops models for the understanding of eWOM communication and consumer engagement in the fashion context.

Factors of electronic word-of-mouth engagement models and propositions

Many previous studies are interested in the determinants of eWOM persuasiveness for consumer behaviour (Jeong and Koo, 2015; Tsao and Hsieh, 2015). Most studies have borrowed the building blocks for the foundation of eWOM models from a number of fields and theories such as information adoption model, elaboration likelihood model, social exchange theory, multistep flow model and more. Among them, Sussman and Siegal’s information adoption model provides the basic premise for the present study. As such, factors that affect consumers’ adoption of eWOM messages can be understood clearly. The model indicates that consumers’ adoption depends on two main factors – argument quality and source credibility.

Argument quality.

Argument quality refers to the persuasive strength of arguments to convince message recipients (Cheung et al., 2009). Prior studies have argued that argument quality affects message recipients’ attitude in a social media context (Hsu et al., 2013; Teng et al., 2014). Hsu et al. (2013) identified the factors that influence users’ intent to adopt and disseminate eWOM messages, including complexity, relative advantage, compatibility, informativeness, social interactivity and playfulness. Also, Teng et al. (2014) found that eWOM messages with negative valences have a greater impact on persuading consumers than those with positive valences. In this study, argument quality denotes the persuasiveness of eWOM content in fashion microblogging. Prior studies have further identified word count, detailed information, percent of negative words and persuasive words as determinants of eWOM persuasiveness (Cheung et al., 2009; Cheng and Ho, 2015). Cheng and Ho (2015) show that a high word count makes readers feel that the review is more useful. Based on their findings, it can be considered whether consumers on social media are inclined to adopt fashion microblogs if there is a high word count. Because keywords are the most important words related to the content of documents (Rossi et al., 2014), the high word count associated with fashion words in microblogging may affect eWOM persuasiveness and fashion consumer participation. Thus, the following propositions were proposed:


Fashion microblogs with a high word count have a positive impact on the eWOM persuasiveness for fashion consumer engagement on social media.


High-frequency words related to fashion (e.g. fashion brands and products) in microblogging have a positive impact on the eWOM persuasiveness for fashion consumer engagement on social media.

Source credibility.

Source credibility is the intermediary between eWOM quality and consumers’ purchase intention (Tsao and Hsieh, 2015). Existing studies concentrate on two determinants of source credibility – expertise and trustworthiness (Cheng and Ho, 2015; Lou and Yuan, 2019). Expertise concerns source’s competence or qualification, including knowledge and skills related to a certain topic (Cheng and Ho, 2015). Trustworthiness is the receivers’ perception of a source as truthful (Lou and Yuan, 2019). In the context of social media, several studies have tested the impact of source credibility on consumers (Djafarova and Rushworth, 2017; Kim et al., 2020; Bao and Wang, 2021). Bao and Wang (2021) show that source credibility of a brand microblog can enhance consumers’ community commitment, trust towards the microblog and promote their participation. Djafarova and Rushworth (2017) further described that Instagram consumers’ trust in eWOM messages was formed by celebrities’ expertise and their relevance to consumers. Prior studies indicate that influencers’ credibility affects consumer reactions (Barney-McNamara et al., 2020; Cuevas et al., 2021). However, the question of whether factors in influencers’ marketing content and influencer credibility affect consumer behaviour needs to be further discussed (Lou and Yuan, 2019). To find the answer, the study propositioned the following:


Microbloggers who have a high active influence on fashion have a positive impact on the eWOM persuasiveness for fashion consumer engagement on social media.


Fashion micro-influencers’ types of expertise in microblogging have a positive impact on the eWOM persuasiveness for fashion consumer engagement on social media.

In summary, the study proposed determining the factors of eWOM persuasiveness and consumer engagement in social media marketing based on Sussman and Siegal’s information adoption model. It developed four propositions from the perspectives of argument quality and source credibility. Figure 1 visualizes the proposed model.


Sampling decisions

To create sample data, the study investigated the influence of Irish fashion microbloggers. For measuring influence, Owyang and Lovett (2010) recommend the use of key performance indicators (KPIs), including active advocates, advocate influence and advocacy impact. Among them, this study chose advocate influence to measure the influence of fashion microbloggers, because it can show “the unique advocate’s influence across one or more social media channels” (Barker et al., 2013, p. 291). It is measured based on the number of comments, reach, relevant contents and shares. Therefore, active influence is calculated by dividing the influence of a single advocate by the total number of advocates. Accordingly, ten most influential Irish microbloggers were identified on the basis of their active influence. The final results are shown in Table 1. In the past literature, influencers were defined as a “third party who significantly shapes the customer's purchasing decision” (Brown and Fiorella, 2013, p. 2). Representative influencers include industry analysts, consultants, celebrities, microbloggers and more. They are further divided into micro- and meso-influencers according to the number of followers (Boerman, 2020). As such, the study describes the ten most influential Irish fashion microbloggers as fashion micro-influencers, who have a certain active influence on consumer behaviour in social media marketing.

Coding procedure

A total of 20,000 microblogs were collected from these fashion micro-influencers. According to the research questions and propositions, they were analyzed through keyword classification and content analysis in NVivo. Many prior studies on eWOM have used NVivo to analyze research data (Blas and Buzova, 2016; Ferreira et al., 2020). Blas and Buzova (2016) used NVivo to count the word frequency of the text corpus in the eWOM analysis. In a similar vein, Ferreira et al. (2020) used NVivo to analyze textual and content data to study eWOM effects, which followed the procedures indicated by Veal (2017). Previous studies showed that NVivo is a suitable method for the investigation of eWOM content. Therefore, it was selected as the coding method for keyword classification and content analysis in the study.

Correspondingly, the word frequency of microblogs was analyzed, and high-frequency words were identified as keywords. Keyword is a set of significant words in an article, which provides readers with a high-level description of content (Lee et al., 2008). Namely, it gives a compact representation of the content of the document (Rose et al., 2010). As a result, keywords can be used for text classification (Verma and Yadav, 2021). For this reason, the study classified the microblogs into different types based on keywords. Before classification, the raw keywords were refined. The useless but high-frequency words were cleared, such as definite articles, indefinite articles and prepositions. The rest keywords were categorized into groups on the ground of their content. Subsequently, content analysis was carried out referring to these groups. With this aim, fashion micro-influencers and their microblogs were observed for three years from 2015 to 2018. Drawing on past literature (Blas and Buzova, 2016; Ferreira et al., 2020), influencers’ microblogging content was finally analyzed and coded according to Table 2.


Keyword classification and types of fashion micro-influencers

In consequence, a total of 3,010 keywords were extracted from the microblogs. The range of frequency is from 0 to 250. As mentioned above, keywords were refined by clearing definite articles, indefinite articles and propositions. Also, words with low frequency (<50) were not taken into account because they were insufficient to understand microblogs as a whole compared with the overall frequency. Finally, 100 keywords were identified, considering the usefulness of interpreting fashion micro-influencers’ eWOM content in microblogging. Words with high frequency include “fashion”, “got”, “best”, “like”, “look”, “new”, “today”, etc. Among them, keywords such as “best”, “like”, “love”, and “favourite” further reveal the positive attitudes of Irish fashion micro-influencers towards eWOM marketing in microblogging. Figure 2 shows the high-frequency words in the microblogs.

As a result, the word frequency shows three types of frequent words in the eWOM messages:

  1. Brands. Brands are one of the most frequently mentioned words in eWOM messages. The brands are various, ranging from luxury brands (e.g. Gucci, Armani) to affordable brands (e.g. Topshop, H and M). They are generally emphasized by capital and bold letters in the microblogs.

  2. Products. Irish micro-influencers introduce a variety of fashion products in the eWOM messages, including bag, dress, shoes and look. These words always come together with positive verbs and adjectives (e.g. best, favourite, love) in the microblogs.

  3. Occasions. There are three categories of occasions in the microblogs. They are holidays (e.g. Christmas, New Year, Payday), seasons (e.g. winter, summer, autumn, spring) and places/locations (e.g. Ireland, Dublin).

Accordingly, Irish fashion micro-influencers are classified into four groups: brand adopters, product leaders, occasion leaders and market mavens (see Figure 3):

  1. Brand adopters are eager to be the first for spreading brand news and using branding strategies to strengthen brand identities in fashion consumers’ minds through eWOM messages. The brand strategies involve brand awareness, brand loyalty and brand recognition. They are familiar with varieties of fashion brands as well as branding strategies. However, they do not often comment on fashion brands. Instead, they prefer to adopt and spread the latest fashion news to consumers. Brand names are the core of brand adopters’ microblogs.

  2. Product leaders are found to avoid mentioning brand names directly in the eWOM messages compared with brand adopters. Instead, they review fashion information and give suggestions on fashion trends for specific fashion products. More concretely, they offer photos of fashion products, links and product names in the microblogs. Therefore, this type of micro-influencers is highly interested in collecting and analysing fashion news, especially concerning products.

  3. Occasion leaders are good at reviewing fashion information and giving opinions in their microblogs according to varieties of occasions. Their fashion recommendations for consumers follow different situations. Thus, they are not only knowledgeable about fashion, but also keep up with news firmly. They incline to share the news with other fashion consumers in the network as soon as they receive them.

  4. Market mavens are very knowledgeable about the whole fashion industry. They can make integrated use of information collected from fashion marketplaces in microblogs. For this reason, they are followed by other micro-influencers in the consumer network. They assist companies in spreading fashion trends through eWOM messages. At the same time, they help other microbloggers and followers to grasp fashion trends using microblogging. Other microbloggers are capable of re-spreading these trends and continue to influence other micro-influencers’ followers. In such a case, market mavens’ eWOM messages are more influential than other microbloggers on social media.

Fashion micro-influencers’ electronic word-of-mouth engagement models

On the basis of Table 2, fashion micro-influencers and their microblogs were observed, analyzed and coded to explore the influence of different micro-influencers’ eWOM messages on social networks and how followers respond. Specifically, the number of likes, shares and comments for each post was also examined in addition to keyword extraction and content analysis. Previous studies support the use of likes, shares and comments as engagement measures in Facebook brand posts (Quesenberry and Coolsen, 2018; Chwialkowska, 2018). Afterwards, the average number was calculated depending on different types of micro-influencers for understanding the effect of different influencers’ eWOM persuasiveness on consumer behaviour. Consequently, the study identifies three eWOM models of Irish fashion micro-influencers (Table 3).

The first model is the brand adopters and product leaders’ eWOM model [see Figure 4(a)]. Brand adopters and product leaders convey fashion news on brands and products to their followers through eWOM in the microblogs. The following two typical examples illustrate the microblogs of brand adopters and product leaders. These posts focus on standing out brand information and leave strong brand images on consumers. For this reason, the length is small, and content keeps simple to emphasize brand identities. The brand names in the form of capital letters and bold letters are eye-catching:

Example 1. Beauty Drop: Giorgio Armani Beauty Face the Runway arrives at Brown Thomas […] and we've got a first look! #Beauty (Help My Style, Facebook, July 15, 2018).

Example 2. TOPSHOP’s latest darling and gorgeous designs-(Help My Style, Facebook, July 15, 2018).

In general, the influence of these messages is a single direction from fashion micro-influencers to followers. Namely, followers hardly respond to the messages. The study finds that the total reactions of followers (likes, comments, reposts) are less than 1,000/per post. Respectively, the number of reactions to branding messages is from 0 to 300/per post, and reactions to product messages are from 300 to 1,000/per post. The most frequent responses are likes, and there are seldom comments and reposts. As mentioned by Reijmersdal et al. (2016), consumers turn to resist the persuasion from sponsored blogs because of their obvious marketing intentions. Specifically, consumers incline to have the freedom of choice for fashion and refuse to be manipulated by micro-influencers in microblogging. “Consumers today are not susceptible to one-way advertising. Besides, consumers have more autonomy and product options, so the advertising effectiveness of most advertisements is unsatisfactory” (Ho et al., 2015, p. 359). That is to say, followers feel threats to the freedom of choice when they recognize marketing information in the microblogs such as brand and product names, which further cause followers’ resistance to respond to micro-influencers’ messages in the microblogs. For this reason, brand adopters and product leaders’ eWOM communication maintain the lowest influence and followers’ responses compared with other micro-influencers. In other words, it is challenging to change micro-influencers’ followers to potential consumers through eWOM in social media marketing.

The second model is the occasion leaders’ eWOM model [see Figure 4(b)]. The study shows that occasion leaders send many eWOM messages on different occasions. Different from brand adopters and product leaders, occasion leaders’ eWOM model is two-way communication with their followers. Specifically, followers engage with occasion leaders using comments, likes and reposts. In the study, followers’ responses to eWOM messages range from 1,000 to 1,600 reactions for each post roughly. The following example is a representative instance of occasion leaders’ eWOM messages:

Example 3. |NEW POST| |Treat Yourself […]| It’s Pay Day! ||| Check out our top fashion picks with the autumn/winter season in mind including these fab boots here www.pippa.ie/treat-yourself-its-pay-day-august/ (Pippa O’Connor, Facebook, September 1, 2017).

In the example, Pippa introduces boots and other top fashion picks following occasions in the microblogs. This microblog has 1.5k responses, 304 comments and 10 shares from consumers. The good traffic from followers’ reactions implies that the eWOM messages on occasions have a strong influence on followers. Ramya and Ali (2016) indicate that consumer buying behaviour is affected by their motivations. The motivations drive them to take purchase actions and satisfy individual needs. The life style is one of these motivations, which influences buyer behaviours. Consequently, the occasions in the microblogs offer followers motivations to achieve their demands for life styles. In the microblog, Pippa mentions two occasions – Pay Day and autumn/winter season to motivate followers. In detail, for instance, the seasons of autumn and winter are the right timing for followers to look for new boots and other fashion products. At the same time, Pay Day is another occasion for followers to purchase new fashion products. In response to followers’ requirements, Pippa enhances the influence of fashion products in the microblogs corresponding to occasions so as to stimulate followers’ purchase behaviours. As a result, occasion leaders’ eWOM messages have significant influence and engagement with their followers. Followers in the social network tend to be persuaded by occasion leaders and become fashion consumers eventually.

Last but not least, the third model is the market mavens’ eWOM model [see Figure 4(c)]. The study finds that market mavens convey various marketing messages such as brands, products, occasions and other eWOM messages to their followers in the microblogs. The influence of these messages is also two-way communication. Followers reply to these messages actively. The responses are more than 1,600 reactions for each post roughly. Similar to occasion leaders’ eWOM, the reactions to market mavens’ messages also include likes, comments and reposts.

Different from the previous models, market mavens’ eWOM messages not only affect followers in the market mavens’ social network, but also influence followers’ followers. In other words, they have more significant effects on the communities of followers and turn them into consumers. Specifically, market mavens’ followers consist of general followers who are probably potential consumers and secondary micro-influencers who are also leaders but with lower influence. The general followers in the market mavens’ network are affected by secondary micro-influencers and market mavens at the same time through eWOM messages. The repetitive marketing from different voices, especially influencers’ opinions, makes eWOM messages more persuasive for followers. Also, the secondary micro-influencers have followers and the third level of micro-influencers in their network. Market mavens’ eWOM messages can affect them as well.

The communication of eWOM messages in this model is repetitive spontaneously, from market mavens, secondary micro-influencers, third level of micro-influencers to nth level of micro-influencers and other followers. That is to say, the effect of first micro-influencers is kept affecting the next lower level of followers automatically, which finally results in the considerable influence of eWOM messages and micro-influencers in the communities online. For example, the study identifies So Sue Me as a market maven, and discovers that Just Jordan is one of So Sue Me followers, and at the same time, she is also a micro-influencer. She has reposted So Sue Me’s microblogs over 14 times every month, and inclines to imitate So Sue Me’s style of microblogging to increase her influence. Here are some instances of Just Jordan’s reposting about So Sue Me’s microblogs:

Example 4. This is on fleek @SoSueMe_ie https://t.co/UyhpWoW01B (Just Jordan, Facebook, February 5, 2016).

Example 5. This palette is amazing!! Well done Suzanne xxx So Sue Me-Blog www.sosu.ie (Just Jordan, Facebook, February 5, 2016).

These microblogs are further reposted, liked and commented by Just Jordan’s followers. Namely, So Sue Me’s eWOM messages influence Just Jordan corresponding to market mavens’ eWOM model, and then Just Jordan further influences her followers. For instance, Pippa, another micro-influencer, has reposted these microblogs and her other microblogs (see examples below):

Example 6. Lisa Jordan from Just Jordan looks so beautiful ||| #PMVIPStyleAwards

Niamh x (Pippa O’Connor, Facebook, April 21, 2017).

To be noted, the number of times for the repetition in this model relies on the influence of primary micro-influencers and the number of followers. The more influence they have, the higher the number of times this model is duplicated. Consequently, market mavens’ eWOM messages have the highest impact and engagement with the followers.

Discussion and conclusion

Model testing

To investigate the determinants of micro-influencers’ eWOM persuasiveness for consumer engagement in social media marketing, the study proposed testing four propositions through keyword classification and content analysis in NVivo. First, P1 posits that microblogs with a high word count have a positive impact on the eWOM persuasiveness for fashion consumer engagement on social media. The result [Figure 2(b)] shows that the average number of words in a sentence ranges from 18.90 to 58.90. The minimum value is from Sosueme, which belongs to a market maven, and the maximum value is from The Style Fairy, which is categorized as an occasion leader. The findings do not reveal the correlation between the word count of eWOM messages and eWOM persuasiveness for consumer engagement because market mavens’ eWOM messages have the highest impact and engagement with followers, whereas brand adopters and product leaders’ eWOM communication maintain the lowest influence and followers’ responses compared with other micro-influencers. Therefore, P1 is not supported.

Subsequently, P2 posits that high-frequency words related to fashion in microblogging have a positive impact on the eWOM persuasiveness for fashion consumer engagement on social media. Initially, the study identified 3,010 keywords from fashion micro-influencers’ microblogs, and finally selected 100 keywords for data analysis, considering the usefulness of interpreting fashion micro-influencers’ eWOM content in microblogging. These words are further divided into three groups – brands, products and occasions, and found in different types of micro-influencers’ microblogs. The findings indicate the following:

  1. Brand-related microblogs have the least reactions from followers.

  2. Occasion-related microblogs are popular in the online social network.

  3. A mixed-use of fashion-related words has the most reactions from followers in eWOM communication.

As a result, high-frequency words related to fashion in microblogging increase the eWOM persuasiveness for fashion consumer engagement on social media, which means that P2 is supported.

Concerning Propositions 3 and 4, the findings of influencer types and eWOM engagement models support that eWOM persuasiveness for fashion consumer engagement is related to micro-influencers’ active influence and types of expertise in microblogging. Specifically, the study identifies four types of fashion influencers, brand adopters, product leaders, occasion leaders and market mavens. Accordingly, the study divides them into three eWOM models:

  1. a single-way brand adopters’ and product leaders’ eWOM;

  2. a two-way occasion leaders’ eWOM; and

  3. a multiway market mavens’ eWOM.

Consequently, the study determines three factors (content keywords, active influence and influencers’ expertise type) for eWOM persuasiveness on consumer engagement in social media marketing regarding the research question.

Research implications

Theoretical implications.

This study answers how fashion micro-influencers and their eWOM messages can positively influence consumer engagement on social media by developing a persuasive eWOM engagement model. Previously, few studies have examined the empirical link between messaging dimension and consumer participation (Chwialkowska, 2018). Besides, previous literature indicates that future research on the persuasiveness of eWOM on consumer behaviour should consider influencers’ effect (Lou and Yuan, 2019). As such, the development of the eWOM engagement model contributes to the existing literature on how to understand the role of micro-influencers and their eWOM messages in consumer-brand marketing interactions. Findings show that eWOM messages with eye-catching brand names from micro-influencers, especially brand adopters, have the lowest level of participation among consumers, which contradicts the past literature that believes consumer engagement is positively affected by brand attitudes (Ha-Brookshire and Norum, 2011). The possible reason for this difference is because of the obvious brand marketing intentions, which has been explained above. On the contrary, findings indicate that market mavens’ eWOM messages have the highest level of participation among consumers because of the spontaneously repetitive mechanism, which is not shown in previous studies and is worthy of further study.

Additionally, the eWOM engagement model is conceptualized on the ground of the determinants of eWOM persuasiveness. Based on Sussman and Siegal’s information adoption model, the study identifies three determinants: content keywords, active influence and influencers’ expertise type. Prior studies have identified word count, detailed information, percent of negative words and persuasive words as determinants of eWOM persuasiveness in consumer–brand relationships (Cheung et al., 2009; Cheng and Ho, 2015). Findings argue that a higher word count cannot enable micro-influencers’ eWOM messages to interact with more consumers. The present study does not reveal the correlation between the word count and eWOM persuasiveness on consumer engagement, which is inconsistent with the previous claims (Cheng and Ho, 2015). Moreover, findings add content keywords, especially occasion-related keywords, as a new determinant of persuading fashion consumers to participate in consumer-brand marketing. Furthermore, regarding debates in the existing literature on whether the influence or content of influencers positively influences consumer behaviour in interactive marketing (Djafarova and Rushworth, 2017), findings confirm that both of them increase consumer engagement in the social network. Therefore, this study contributes to existing research on the persuasiveness of eWOM for consumer engagement and advances our knowledge of consumer–brand interactive marketing on social media.

Managerial implications.

From a practical perspective, this study also provides important insights for consumer-brand interactive marketing strategies. Marketing communication practitioners should note the different effects of micro-influencers on consumers’ eWOM engagement. Because of a large number of followers of influencers, brand companies tend to take advantage of them for marketing interactions with consumers (Barker et al., 2013). However, findings present that influencers such as brand adopters cannot help marketers interact with consumers significantly. Hence, marketers should identify market mavens from different types of micro-influencers in social networks and encourage them to spread eWOM messages given that they extend their reach beyond their immediate circles of consumers. Namely, market mavens not only have a large number of consumer followers, but also have influencer followers. Also, marketers should pay attention to occasion leaders that have an increasing impact on consumer engagement. Occasion leaders can be an alternative helper for active interaction with consumers on social media.

In addition, marketers are encouraged to consider the determinants of eWOM persuasiveness and develop personalized marketing microblogs to attract consumer participation. For example, marketers need to avoid mentioning brand names in the microblogs because findings show that even microblogs with brands from market mavens have the least interaction with consumers. Instead, it is recommended that marketers conceal brands in hashtags and hyperlinks, thereby indirectly marketing brands to consumers and avoiding decreasing consumer participation. More importantly, marketers are advised to adopt a variety of microblog content. For instance, they can associate microblogs with various occasions such as Christmas gifts, Payday treats and more to arouse consumer interest. The findings of the present study indicate that content related to occasions has a positive impact on consumer interaction. In summary, this study contributes to helping marketers increase the persuasiveness of consumer engagement and build strong consumer–brand relationships.

Limitations and future research directions

The study is not without its limitations. First, we believe that the factors determined in the study help to better understand the persuasive power of eWOM for consumer engagement; however, we recognize that other related factors may also affect persuasiveness, which are not included in this study; for example, the role of influencers’ visual eWOM. Like most previous studies, this study focuses on written eWOM (Baek et al., 2012; Teng et al., 2014); however, findings indicate that many influencers use photos and short videos in eWOM messages. Because the persuasiveness of visual eWOM is an under-researched area (Ismagilova et al., 2016), future research could explore any differences in consumer engagement between influencers because of the persuasiveness of visual eWOM. Also, the study focuses on text content instead of text style because most texts are statements rather than questions. However, a question or controversial microblog statement may elicit greater reactions from followers than a positive statement. Therefore, text style may be one of the persuasive factors for consumer participation in future research.

Second, findings reveal the increasing influence of occasion leaders’ eWOM and a new repetitive eWOM model of market mavens in consumer networks on social media. It would be worth further investigating theoretical and managerial applications of eWOM persuasiveness to build strong consumer–brand relationships in the future. Finally, the study applies keyword classification and content analysis in NVivo to explain the mechanism of micro-influencers’ eWOM persuasiveness and consumer engagement. As mentioned above, the determinants of eWOM persuasiveness may vary because of different technical foundations (Ismagilova et al., 2016). Emerging techniques such as text mining and natural language process have opened up new dimensions (Verma and Yadav, 2021). For instance, Tirunillai and Tellis (2014) analyzed user-generated content based on natural language processing. In consequence, future research can use these methods to study variables of eWOM persuasiveness for consumer engagement, and the research question may have different explanations.


The proposed model

Figure 1.

The proposed model

Keywords and word frequency

Figure 2.

Keywords and word frequency

Types of fashion micro-influencers

Figure 3.

Types of fashion micro-influencers

eWOM models of fashion micro-influencers

Figure 4.

eWOM models of fashion micro-influencers

Ten most influential Irish fashion microbloggers

Irish micro-influencers No. of comments No. of reach No. of relevant contents No. of shares Active influence
Sosueme 4,016 605,207 11,974 5,663 0.88
Thunder and Threads 3,512 56,9453 11,855 4,540 0.82
Pippa 3,244 493,959 11,547 4,505 0.81
Help my style 3,303 424,584 9,255 4,132 0.78
Anouska 3,453 389,824 8,090 3,949 0.70
Fluff and Fripperies 3,371 390,842 7,998 3,988 0.70
The Style Fairy 3,200 276,472 6,367 3,323 0.63
What she wears 3,169 276,228 6,163 3,585 0.63
Just Jordan 2,179 195,426 5,222 2,340 0.57
Love Lauren 2,112 174,200 5,230 2,255 0.53

Coding fashion micro-influencers’ microblogs

Dimensions of factors Measure factors Definition Coding strategy
Argument quality Word count Total words of each microblog The average length of microblogs
Content keywords High-frequency words in the microblogs Word frequency
Source credibility Expertise type Content types of microblogs Keyword classification
Active influence Microbloggers’ influence on consumer behaviour Number of likes, shares and comments

Descriptive statistics

Descriptions Variables Minimum Maximum Mean SD Examples
Brand adopters Likes 7.00 113.00 38.8571 31.10673 Help my style
Shares 0.00 5.00 0.4286 1.16496
Comments 0.00 19.00 1.1429 4.12657
Product leaders Likes 2.00 300.00 51.4500 78.83325 Just Jordan, Love Lauren, Fluff and Fripperies
Shares 0.00 14.00 1.4000 3.06766
Comments 0.00 56.00 7.9500 14.61605
Occasion leaders Likes 44.00 1041.00 215.9524 140.26635 Anouska, The Style Fairy, Thunder and Threads, What she wears
Shares 0.00 35.00 4.2857 9.05617
Comments 0.00 67.00 8.8095 7.13876
Market Mavens Likes 90.00 2400.00 1083.8125 571.28583 Pippa, Sosueme
Shares 4.00 314.00 57.6875 99.88374
Comments 13.00 1900.00 263.8125 540.29748


Baek, H., Ahn, J. and Choi, Y. (2012), “Helpfulness of online consumer reviews: readers' objectives and review cues”, International Journal of Electronic Commerce, Vol. 17 No. 2, pp. 99-126.

Bao, Z. and Wang, D. (2021), “Examining consumer participation on brand microblogs in China: perspectives from elaboration likelihood model, commitment–trust theory and social presence”, Journal of Research in Interactive Marketing.

Barker, M., Barker, D., Bormann, N. and Neher, K. (2013), Social Media Marketing: A Strategic Approach, South-Western, Cengage Learning, Boston.

Barney-McNamara, B., Peltier, J., Chennamaneni, P.R. and Niedermeier, K.E. (2020), “A conceptual framework for understanding the antecedents and consequences of social selling: a theoretical perspective and research agenda”, Journal of Research in Interactive Marketing.

Blas, S.S. and Buzova, D. (2016), “Guided tour influence on cruise tourist experience in a port of call: an eWOM and questionnaire-based approach”, International Journal of Tourism Research, Vol. 18 No. 6, pp. 558-566.

Boerman, S.C. (2020), “The effects of the standardized Instagram disclosure for micro- and meso-influencers”, Computers in Human Behavior, Vol. 103, pp. 199-207.

Brown, D. and Fiorella, S. (2013), Influence Marketing: How to Create, Manage, and Measure Brand Influencers in Social Media Marketing, Que Publishing, Seattle.

Cheung, M.Y., Luo, C., Sia, C.L. and Chen, H. (2009), “Credibility of electronic word-of-mouth: informational and normative determinants of on-line consumer recommendations”, International Journal of Electronic Commerce, Vol. 13 No. 4, pp. 9-38.

Cheng, Y.-H. and Ho, H.-Y. (2015), “Social influence's impact on reader perceptions of online reviews”, Journal of Business Research, Vol. 68 No. 4, pp. 883-887.

Chwialkowska, A. (2018), “The effectiveness of Brand- and customer-centric content strategies at generating shares, ‘likes’, and comments”, Journal of Promotion Management, Vol. 25 No. 2, pp. 270-300.

Cuevas, L., Lyu, J. and Lim, H. (2021), “Flow matters: antecedents and outcomes of flow experience in social search on Instagram”, Journal of Research in Interactive Marketing.

Djafarova, E. and Rushworth, C. (2017), “Exploring the credibility of online celebrities’ Instagram profiles in influencing the purchase decisions of young female users”, Computers in Human Behavior, Vol. 68, pp. 1-7.

Ferreira, B.M., Abrantes, J.L., Seabra, A. and Rubio, I.M. (2020), “Digital technology and eWOM in the context of higher education: a study from Portugal and Spain”, Journal of Marketing for Higher Education, doi: 10.1080/08841241.2020.1834488.

García-de-Frutos, N. and Estrella-Ramón, A. (2021), “You absolutely (don’t) need this! Examining differences on customer engagement components for (anti)haul youtubers’ videos”, Journal of Research in Interactive Marketing.

Godey, B., Manthiou, A., Pederzoli, D., Rokka, J., Aiello, G., Donvito, R. and Singh, R. (2016), “Social media marketing efforts of luxury brands: influence on brand equity and consumer behaviour”, Journal of Business Research, Vol. 69 No. 12, pp. 5833-5841.

Ha-Brookshire, J.E. and Norum, P.S. (2011), “Willingness to pay for socially responsible products: case of cotton apparel”, Journal of Consumer Marketing, Vol. 28 No. 5, pp. 344-353.

Hajli, N. (2018), “Ethical environment in the online communities by information credibility: a social media perspective”, Journal of Business Ethics, Vol. 149 No. 4, pp. 799-810.

Hayashi, Y., Orita, R. and Kajiwara, S. (2017), “Negative electronic word-of-mouth can support product recommendation: experimental investigation”, Proceeding CSCW 17 Companion Companion of the 2017 ACM Conference on Computer Supported Cooperative Work and Social Computing, pp. 191-194.

Hennig, T.T., Wiertz, C. and Feldhaus, F. (2015), “Does twitter matter? The impact of microblogging word of mouth on consumers’ adoption of new movies”, Journal of the Academy of Marketing Science, Vol. 43 No. 3, pp. 375-394.

Ho, C.H., Chiu, K.H., Chen, H. and Papazafeiropoulou, A. (2015), “Can internet blogs be used as an effective advertising tool? The role of product blog type and brand awareness”, Journal of Enterprise Information Management, Vol. 28 No. 3, pp. 346-362.

Hsu, C., Chuan‐Chuan Lin, J. and Chiang, H. (2013), “The effects of blogger recommendations on customers’ online shopping intentions”, Internet Research, Vol. 23 No. 1, pp. 69-88.

Ismagilova, E., Slade, E., Williams, M., et al. (2016), “Persuasiveness of eWOM communications: literature review and suggestions for future research”, in Dwivedi, Y.K., Mäntymäki, M., Ravishankar, M.N., Janssen, M., Clement, M., Slade, E.L., Rana, N.P., Al-Sharhan, S. and Simintiras, A.C. (Eds), Social Media: The Good, the Bad, and the Ugly, Springer, Cham, pp. 354-359.

Jeong, H.-J. and Koo, D.-M. (2015), “Combined effects of valence and attributes of e-WOM on consumer judgment for message and product. The moderating effect of Brand community type”, Internet Research, Vol. 25 No. 1, pp. 2-29.

Kandampully, J., Zhang, T.T. and Bilgihan, A. (2015), “Customer loyalty: a review and future directions with a special focus on the hospitality industry”, International Journal of Contemporary Hospitality Management, Vol. 27 No. 3, pp. 379-414.

Khazaei, A. (2016), Immigrants’ Engagement in Tourism Planning: The Case of the Rouge National Urban Park, University of Guelph, Ontario.

Kim, J.A. and Ko, E. (2012), “Do social media marketing activities enhance customer equity? An empirical study of luxury fashion brand”, Journal of Business, Vol. 65 No. 10, pp. 1480-1486.

Kim, J.M., Kim, M. and Key, S. (2020), “When profile photos matter: the roles of reviewer profile photos in the online review generation and consumption processes”, Journal of Research in Interactive Marketing, Vol. 14 No. 4, pp. 391-412.

Kozinets, R.V., De Valck, K., Wojnicki, A.C. and Wilner, S.J. (2010), “Networked narratives: understanding word-of-mouth marketing in online communities”, Journal of Marketing, Vol. 74 No. 2, pp. 71-89.

Kyu Kim, Y., Yim, M.Y.-C., Kim, E.A. and Reeves, W. (2020), “Exploring the optimized social advertising strategy that can generate consumer engagement with green messages on social media”, Journal of Research in Interactive Marketing.

Lee, S., Lee, S., Seol, H. and Park, Y. (2008), “Using patent information for designing new product and technology: keyword based technology roadmapping”, R&D Management, Vol. 38 No. 2, pp. 169-188.

Lever, M.W., Mulvey, M.S. and Elliot, S. (2017), “From hashtags to shopping bags: measuring the rise of eWOM through social media and its impact on travelers’ shopping patterns”, Tourism Travel and Research Association: Advancing Tourism Research Globally, Vol. 12, pp. 1-6.

Lou, C. and Yuan, S.P. (2019), “Influencer marketing: how message value and credibility affect consumer trust of branded content on social media”, Journal of Interactive Advertising, Vol. 19 No. 1, pp. 58-73.

Mishra, A. and Satish, S.M. (2016), “eWOM: extant research review and future research avenues”, Vikalpa: The Journal for Decision Makers, Vol. 41 No. 3, pp. 222-233.

Nielsen (2012), “Consumer trust in online, social, and mobile advertising grows”, available at: www.nielsen.com/us/en/insights/news/2012/consumer-trust-in-online-social-and-mobile-advertising-grows.html (accessed 28 November 2018).

Owyang, J. and Lovett, J. (2010), “Social marketing analytics: a new framework for measuring results in social media”, Web Analytics Demystified and Altimeter Group, available at: www.slideshare.net/jlovett/social-marketing-analytics (accessed 3 October 2019).

Patsiotis, A. and Kapareliotis, I. (2016), “Digital marketing trends and the curriculum”, Journal of Management and Financial Sciences, Vol. 23 No. 9, pp. 1-122.

Quesenberry, K.A. and Coolsen, M.K. (2018), “What makes Facebook brand posts engaging? A content analysis of Facebook brand post text that increases shares, likes, and comments to influence organic viral reach”, Journal of Current Issues and Research in Advertising, Vol. 40 No. 3, pp. 229-244.

Ramya, N. and Ali, S.M. (2016), “Factors affecting consumer buying behaviour”, International Journal of Applied Research, Vol. 2 No. 10, pp. 76-80.

Reijmersdal, E.A., Fransen, M.L., van Noort, G., Opree, S.J., Vandeberg, L., Reusch, S., Van Lieshout, F. and Boerman, S.C. (2016), “Effects of disclosing sponsored content in blogs: how the use of resistance strategies mediates effects on persuasion”, American Behavioral Scientist, Vol. 60 No. 12, pp. 1458-1474.

Rettberg, W.J. (2014), Blogging, Polity Press, Medford, MA.

Rose, S., Engel, D., Cramer, N. and Cowley, W. (2010), “Automatic keyword extraction from individual documents”, in Berry, M.W. and Kogan, J. (Eds), Text Mining: Applications and Theory, John Wiley and Sons, Hoboken, NJ, pp. 1-20.

Rossi, R.G., Maracini, R.M. and Rezende, S.O. (2014), “Analysis of domain independent statistical keyword extraction methods for incremental clustering”, Learning and Nonlinear Models, Vol. 12 No. 1, pp. 17-37.

Sussman, S.W. and Siegal, W.S. (2003), “Informational influence in organizations: an integrated approach to knowledge adoption”, Information Systems Research, Vol. 14 No. 1, pp. 1-125.

Sweeney, J., Soutar, G. and Mazzarol, T. (2014), “Factors enhancing word-of-mouth influence: positive and negative service-related messages”, European Journal of Marketing, Vol. 48 Nos 1/2, pp. 336-359.

Teng, S., Wei Khong, K., Wei Goh, W. and Yee Loong Chong, A. (2014), “Examining the antecedents of persuasive eWOM messages in social media”, Online Information Review, Vol. 38 No. 6, pp. 746-768.

Tirunillai, S. and Tellis, G.J. (2014), “Mining marketing meaning from chatter: strategic brand analysis of big data using latent Dirichlet allocation”, Journal of Marketing Research, Vol. 51 No. 4, pp. 463-479.

Tsao, W.C. and Hsieh, M.T. (2015), “eWOM persuasiveness: do eWOM platforms and product type matter?”, Electronic Commerce Research, Vol. 15 No. 4, pp. 509-541.

Veal, A.J. (2017), Research Methods for Leisure and Tourism, Pearson, London.

Verma, S. and Yadav, N. (2021), “Past, present, and future of electronic word of mouth (EWOM)”, Journal of Interactive Marketing, Vol. 53, pp. 111-128.

Voyer, P.A. and Ranaweera, C. (2015), “The impact of word of mouth on service purchase decisions: examining risk and the interaction of tie strength and involvement”, Journal of Service Theory and Practice, Vol. 25 No. 5, pp. 636-656.

Wang, S., Cunningham, N.R. and Eastin, M.S. (2015), “The impact of eWOM message characteristics on the perceived effectiveness of online consumer reviews”, Journal of Interactive Advertising, Vol. 15 No. 2, pp. 151-159.

Weisfeld-Spolter, S., Sussan, F. and Gould, S. (2014), “An integrative approach to eWOM and marketing communications”, Corporate Communications: An International Journal, Vol. 19 No. 3, pp. 260-274.

Williams, A.S., Rhenwrick, I., Wright, B., Choi, W., Kim, D.Y. and Vickey, T. (2014), “Building viable fitness brands: importance of brand communication strategies in attracting potential health club members”, International Journal of Sport Management, Recreation and Tourism, Vol. 15, pp. 49-68.

Wolny, J. and Mueller, C. (2013), “Analysis of fashion consumers’ motives to engage in electronic word-of-mouth communication through social media platforms”, Journal of Marketing Management, Vol. 29 Nos 5/6, pp. 562-583.

Further reading

Berezan, O., Raab, C., Tanford, S. and Kim, Y.S. (2013), “Evaluating loyalty constructs among hotel reward program members using eWOM”, Journal of Hospitality and Tourism Research, Vol. 39 No. 2, pp. 198-224.

Luo, C., Wu, J., Shi, Y. and Xu, Y. (2014), “The effects of individualism-collectivism cultural orientation on eWOM information”, International Journal of Information Management, Vol. 34 No. 4, pp. 446-456.

Munzel, A. and Kunz, W.H. (2014), “Customer-to-customer interactions within online review sites: a typology of contributors”, in Geyer-Schulz, A. and Meyer-Waarden, L. (Eds), Customer and Service Systems, KIT Scientific Publishing, Karlsruhe, pp. 69-75.

Pedersen, S.T., Razmerita, L. and Colleoni, E. (2014), “Electronic word-of-mouth communication and consumer behaviour: an exploratory study of Danish social media communication influence”, LSP Journal, Vol. 5 No. 1, pp. 112-131.


The author would like to thank the editors, Prof Cheng Lu Wang, Prof John Schibrowsky and Dr Morgan Yang, and anonymous reviewers for their valuable comments and suggestions on this study. This work was supported by the Fundamental Research Funds for the Central Universities.

Corresponding author

Zheng Shen can be contacted at: janeshen0916@126.com

Related articles