Sentiment Analysis - A Potent Tool for Evaluating Social Media Influencers
Dhun
Research Scholar, Department of Commerce, University of Delhi
Email: dhunsehrawat@commerce.du.ac.in
H. K. Dangi
Associate Professor, Department of Commerce, University of Delhi
Email:hkdangi@commerce.du.ac.in
Abstract
Social networking platforms provide opportunities for Internet users to create and share content and this led to the emergence of social media influencers. They integrate the brands' message in the form of reviews, recommendations, and usage in a real-life setting. The sphere of influence is dramatically shifting from traditional celebrity endorsements to social media content creators. Today, influencers have now become a valuable resource for marketers to disseminate the brand’s message to their target audience through highly personalized, relevant, and entertaining content, and this process is widely recognized as influencer marketing. This paper throws light on why and how influencer marketing has become the most preferred digital marketing technique among marketers. An extensive review of literature has been done for the present study to gain an in-depth understanding of this new concept, influencer marketing. To explore how users perceive influencer endorsements, sentiment analysis was conducted using R software on the comment section of an Instagram influencer to grasp users’ attitudes and emotions towards the influencer and the brand. The results demonstrate that followers have a positive emotion as they like and trust the influencer for their endorsements. This research contributes to the emerging concept of social media influencer marketing and its positive implication for marketing practitioners. The paper substantiates that this area is growing among researchers and shows huge potential for further studies on different platforms and industries.
Keywords: Social media, Social networking sites, Influencer marketing, Influencers, Social media influencers, Sentiment analysis, Instagram
Introduction
Globally, social media audience and the time spent on social media is rising each year. Social media is becoming ubiquitous. Companies have recognized the eminence of Social Networking Sites (SNS) such as Facebook, Instagram, YouTube, etc. that connect people, companies, institutions, etc., and disseminate content. Studies suggest that consumers place higher trust in social media as an information source over conventional mediums. However, over a period of time, consumers now find themselves ensnared with ads and pop-ups by advertisers on nearly all SNS. Innovations like ad-blockers pour cold water on conventional digital marketing strategies. This paves the way for influencer marketing, which serves as the perfect cure for “ad fatigue” and delivers authentic information. The digital population is developing ad blindness and is essentially becoming immune to marketing. Therefore, influencer marketing becomes rave of the moment. Factors like increasing penetration of the internet, the proliferation of mobile phones, the prevalence of SNS among millennials, and the advent of adblocking software make the case for influencer marketing on digital platforms.
Advertisers utilize social media to promote products or services through Electronic Word-Of-Mouth (eWOM) i.e. user’s sharing information, reviews, experiences, or feedback to their family and friends, who circulate it further, creating a type of viral-marketing effect. eWOM is an efficacious and sought-after method where people voluntarily talk about a brand, product, or service without getting paid for it. Influencer marketing, on the other hand, is like a paid version of the Word of Mouth (WOM) marketing with specific objectives like enhancing reach, engagement, trust, brand awareness, etc.(Mediakix, 2019). According to Tap Influence (2015), influencer marketing involves collaborating with social media influencers to publicize the brand's product or service to the target audience and impact their behavior in return for a payment or fee.Social media influencers are connoisseur content creators who create content on social platforms that woos audiences who are valuable to the brands. Influencers have the ability to persuade their followers regarding the information/message, attitude, or behavioral intentions and therefore can provide many benefits to the brands.
Influencer marketing goes inseparably with two different types of marketing, content marketing and social media marketing but they are not synonymous and often fit inside the concept of influencer marketing campaigns. There are multiple terminologies related to influencers like micro-celebrities, social media influencers, Instafamous, opinion leaders, bloggers, digital influencers, etc. They post several pictures or videos showcasing their personal lives such as doing daily work, hobbies, and advertising products or services of kindred nature. This is recognized as sponsored content (i.e. paid partnership with brands) where influencers review and recommend the products to their audience. They promulgate the brand messages, instigate new trends through hashtags, and engage with audiences to augment sales. Influencers may advertise in a brazen manner like a paid ad, or in a more subtle way, with tenuous reference to the product or service being advertised. Campaigns initiated by such influencers are enduring in nature, i.e. the message spreads among influencer’s followers and they successively circulate it further, creating a snowball effect. This ‘Evergreen Content’ creates a constantly rising Return on Investment (ROI) for brands.
Influencers are real people who are viewed as friends whom audiences feel a connection with, want to emulate, and invest as compared to scripted ads. Social media users follow influencers because they are popular and maintain a regular social connection with their followers, which is why users wish to resemble them and thus have developed a positive association with influencers. Influencers are integral to marketers as they enthrall the target audience through their tailored content that often takes the form of reviews and recommendations. Woods (2016) points out that influencer endorsements seem authentic to consumers because they admire and trust the influencers as they show the application of the product into real life. Thus, consumer’sskepticism gets allayed due to this realness factor and they favor this kind of endorsement.The greatest asset of influencer marketing is that people perceive influencers as their friends and look up to them, which has the utility of making the distinction between paid ads and genuine reviews nebulous (Boerman et al., 2017).It is the trust placed on these influencers by the consumers that is so imperative for brands because brand messages advocated through influencers are much more plausible as compared to online advertising which is circumvented by ad-blockers anyway.
The reason why influencers are courted by brands is that they can convert target customers' skepticism into credence. This premise is not spurious as substantiated by the research conducted by Twitter and Annalect (2016), where 49% of users counted on influencers for product recommendations. Further, a joint study conducted by TapInfluence and Nielsen Catalina Solutions (2016) indicated that influencer marketing can generate 11 times more ROI annually than other forms of conventional advertising(Smith,2016).The Annual Influencer Marketing Survey of Mediakix (2019) revealed that marketers feel confident about this channel as 80% of respondents found influencer marketing successful to attain their goals. Likewise, several research conducted by marketing and public relation companies have found favorable results for influencer marketing and convey that nearly all companies and brands are planning to expand their investments in this strategy, so this advertisement medium seems prospective to remain important for marketing communication.
It is indispensable to understand the difference between traditional celebrity endorsement and influencer marketing. Although vague, research conducted so far points out few distinctions. Traditional celebrities popularly known as “stars” have created a name for themselves in their respective fields like films, TV shows, music, sports, etc., and later entered the advertising setting (Kamins et al., 1989). Celebrities are signed by companies as brand ambassadors of a particular brand or product and they endorse a brand not because they use it, but for a hefty compensation amount. On the other hand, Web 2.0 technologies marked the advent of social media influencers often called “digital influencers” or “micro-celebrities” (Marwick, 2011). These influencers are online personalities having a sizeable number of followers and have earned their fame through their active presence on social platforms like personal Blogs, Facebook, YouTube, Instagram, etc. Social media influencers are ‘grassroots individuals’ who shot to fame by creating and posting regular niche content on a particular subject or topic such as fashion and beauty, health and fitness, travel and lifestyle, food, etc. thus often known as content creators having high prominence and likeability among their audience. Gräve (2017) indicated that consumers perceive influencers and celebrities differently based on endorser effectiveness characteristics. Balaban and Mustățea (2019) confer that influencer marketing is a model of peer endorsement which is why customers trust influencers more than traditional celebrities. Jin et al. (2019) in their study found that influencers are perceived to be more trustworthy, consumers feel more envious and stronger social presence towards them and vis-à-vis traditional celebrities.Influencer marketingmelds the qualities of both eWOM and celebrity endorsement.The pattern of influence is dramatically moving away from traditional celebrity endorsements to social media content creators.
Performance appraisal of influencer marketing campaigns is imperative for brands and the most commonly used measurement metrics include engagement (likes, shares, comments, re-posts, etc.), impressions (reach i.e. the overall number of views and/or follower count), clicks (eg, giveaway prizes or traffic driven to a website), conversions(discount or checkout codes on the influencer's post), ROI, and sentiment analysis. Advanced software with built-in analytics tools helps marketers in accomplishing an otherwise convoluted task of appraising influencers. Sentiment analysis is now being progressively used to evaluate the success of influencer marketing activities and to assess audience attitudes, feelings, and opinions on SNS. Lou et al. (2019) acknowledged that influencer-generated ads stimulated higher engagement (i.e. likes and comments) and garnered positive sentiments from their followers as a contrast to brand promoted ads, as they were readily recognized as advertising.
The present study primarily examines influencer marketing as a digital marketing strategy and how sentiment analysis can help brands to measure the different sentiments of consumers towards a particular product endorsed by the influencer. The introduction part elucidated the emergence of influencer marketing, the difference between celebrity endorsement and influencer marketing, and measuring the performance of influencer campaigns. In the next section, extensive literature related to influencer marketing has been presented. The objectives of the study have been given in the third section. The methodology used will be discussed in the fourth section. Findings and discussion will be presented in the fifth section and the last section concludes and suggests the scope of further research.
Literature Review
Social media influencers - A personal brand
Khamis et al. (2017) describe social media influencers as “a particular kind of online micro-celebrity” who create authentic content, are trustworthy and accessible to their followers. They have an intimate relationship with their audience and brands partner with them for endorsing products and services for maximum message outreach. Duffy and Hund (2015) stress that influencers have turned their hobbies and passion into careers, where they are the self-entrepreneurs promoting themselves and different brands. Social media helps to gain a huge follower base, attract advertisers and promote themselves by projecting their unique selling proposition (for example, exhibiting a form of art through stories or content). Influencers indulge in self-branding where they highlight their specialties and uniqueness, to differentiate from others (Duffy & Hund, 2015). Influencers' personal brand is reflected in their number of followers, engagement, popularity, and endorsements (Khamis et al., 2017). Abidin and Ots (2015) posit similar views, authors discuss that influencers have over a while created a personal brand for themselves by delivering authentic content and creating personal relationships with their followers through regular interactions. Social media influencers have to maintain a balance and serve both their audiences and branding partners. However, it is difficult for them to differentiate between their personal and commercial interests.
Pedroni (2016) states that micro-celebrities (such as bloggers, vloggers, Instagrammers, etc.) are a form of mass celebrity who have evolved with the practice of web 2.0 technologies by generating content and interacting with their followers and audiences. They use their popularity (attention) as a tool to integrate with companies, media houses, fashion shows for marketing and product placement. Moreover, Jerslev (2016) states that micro celebrities on social platforms establish a temporality of permanent updating and connectedness. In other words, they keep posting fresh content and remain in close contact with their followers through social media, so users perceive them as accessible and authentic.
Isosuo (2016) conveys that social media users see influencers as their peers and that is why an influencer’s opinion is valued by their followers. The message content must look authentic and blend seamlessly with the influencer's social media account. Otherwise, if it is perceived by the users that the product or service is not used by the influencer or he/she does not hold an unbiased opinion about the brand, then users see the post as merely a paid endorsement, and influencer starts to lose credibility among its followers. Casaló et al. (2018) in their study highlighted that originality and uniqueness are important content characteristics that have been found as the antecedents of being perceived as an opinion leader. SanMiguel et al. (2018) admitted similar views, that social media users admire creative and authentic content from the influencers. Thus, creating original, innovative, authentic, and unique content has a positive impression on the user's discernment that the influencer is an opinion leader. The perceived opinion leadership by a user has positive effects on the user's intention to interact with the account, recommend the account to their friends, family, or peers, and follow the advice of the influencer. Brands desire to benefit from the "halo effect" of influencers' sense of charisma, personality, and good selfies (Abidin, 2016).
Significance of social media influencers
Uzunoğlu and Kip (2014) in their research developed a model “brand communication through digital influencers model” to uncover the persuasive role of influencers, based on a two-step flow of communication (Katz & Lazarsfeld, 1966), where the message flows from the brand to influencers, and from influencers to the audience through blogs or other SNS. The receivers (audience) also play a vital role by spreading the message to their own networks, and this process is continuous. Authors indicate that influencers have the ability to digitally integrate all the social platforms by interacting with their audiences on all mediums which increases their audience size and builds friendly relations. Gretzel (2018) indicates the need for marketers to adopt a two-step flow of information, wherein digital influencers receive and process the brand’s message and disseminate it to a specific audience on social media space. The author talks about the importance of bloggers and vloggers on the travel decisions of the audience and how travel influencers created a name for themselves.
Sudha and Sheena (2017) found that bloggers have a significant impact on women’s decision-making stages in fashion-related purchases. Users trust the reviews provided by bloggers, as the blog helps them to gather information about the product, tips to use, allows them to have a look at attractive pictures, and provides them direct and secure links for purchasing the item online. Xiao et al. (2018) studied different factors that affect the consumer’s perception of information credibility on YouTube and consumer attitude towards the video and brand endorsed using the Heuristic-Systematic model (HSM). As per the results, it can be concluded that viewers like to socialize with influencers and they trust and accept YouTubers as experts in their fields.
Lou and Yuan (2019) proposed an integrated Social Media Influencer Value (SMIV) model to understand how influencer marketing impacts customers via social media. This model provides a holistic perspective encompassing two main factors namely advertising content value (i.e. advertising informativeness and entertainment) and influencer credibility which includes expertise, trustworthiness, attractiveness, and similarity. These two factors affect follower's trust in influencers' branded content which in turn impacts brand awareness and purchasing intentions. Balaban and Mustățea (2019) indicated similar results, where expertise, trustworthiness, attractiveness, and similarity were identified as crucial elements of the user’s perspective on the credibility of social media influencers.
Paid partnership between influencers and brands
The disclosure of paid endorsement by digital influencers has been one of the most important aspects of previous studies. Couto and de Brito (2020) acknowledged that brand instructions (contract terms) are the decisive element for disclosure of the paid partnership between influencers and brands. Therefore, when brands cite in the contract to explicitly declare the sponsorship using relevant tags and/or hashtags, the influencers will obey the same. However, when brands do not issue any specific instructions, it is up to the influencer’s discretion to disclose or not. The concealment of a paid partnership is an effort to delude consumers and this may have harmful consequences such as negative attitudes, brand reputation, loss of trust, followers, etc for both brands and influencers.
Lu et al. (2014) stress that when bloggers clearly disclose the paid connection between brands while recommending any product in their blogs, in that case, consumers do not feel misinformed or duped and tend to have no negative attitudes towards such blog posts. This research indicates that when products with high brand awareness are endorsed in the blog posts, consumers have positive attitudes towards both the blogger and sponsored post and enhances their willingness to purchase the product. Contrarywise, the experiment conducted by Colliander and Erlandsson (2015) shows that when bloggers themselves reveal about the monetary partnership, it led to a lower attitude towards the blog and its credibility, however, there was no negative impact on brand attitude and purchase intention. They also discussed that when consumers are informed about the paid connection i.e. sponsored endorsement by a third-party (such as a media house or newspaper) it leads to a decrease in parasocial interaction, which mediates a negative attitude toward the blog and its credibility. Bloggers thus lose their trustworthiness, as consumers feel that bloggers suppressed critical information.
SNS like Instagram and Facebook have included the disclosure of influencer marketing in their policies, and require influencers to tag the products and brands shown if the content ‘features or is influenced by a business partner for an exchange of value’. But a stringent implementation of these guidelines is still missing. This practice of concealing monetary partnership between influencers and brands is not only unethical but illegal in some countries and may call for legal penalties. Influencers should have a moral responsibility towards consumers and convey truthful information so that followers do not feel hoodwinked. To uphold their self-brand and trust among users, influencers must be cautious about unethical situations. Likewise, brands should also be vigilant while choosing influencers. So, for the betterment of consumers, influencers, and brands transparency must be encouraged in paid social media content.
Selecting the right influencer
Gräve (2019) in his research pointed out that identifying the right influencers on social media and measuring the campaign outcomes represent the two key challenges confronted by brands while incorporating influencer marketing strategy. Marketers generally use Key Performance Indicators (KPIs) to recognize and select appropriate influencers for their campaigns and to measure their performance (Gräve, 2019). The reach of an influencer (number of followers) and engagement (number of interactions such as likes and comments) are commonly used metrics due to easy availability and comparability across social media platforms.
Pang et al. (2016) proposed a “social media influencer engagement model” to facilitate organizations to build effective relations with social media influencers. The model states that it is advantageous for both influencers and companies to maintain healthy and long-lasting relations with each other. Authors state that brands must achieve a fit between their marketing goals, the type of content produced by influencers, and the platform espoused by the influencer. In other words, brands must take into account the target audience of influencers and influencers should have the required capabilities (example, persuasive power) for achieving fruitful results. Likewise, Woods (2016) identified similar factors which should be considered while deciding the influencer such as matching up the voice and image with the brand and alignment of the target audience. He indicated that influencers originated as a need for brands to advertise their products and services on SNS like Instagram, which earlier did not permit advertisers to fuse their paid content onto the platform.
De Veirman et al. (2017) proposed in their study that brands should not only consider the reach of an influencer but also other factors like “follower/followee ratio” and type of product while choosing an influencer. The results of the study indicate that the higher the number of followers, the higher is the perceived likeability of an influencer which reflects a true opinion leadership. Thus, it can be said that audience size is an indicator of popularity. According to Oliveira et al. (2020) for judging the influencer’s impact, companies should not only employ quantitative measures like search engine rank, number of page views, likes, followers, or engagement rate but also use qualitative measures such as personality, the authenticity of influencer, or quality of content.
Sentiment analysis as a performance indicator
Despite huge popularity, the reach and engagement metrics are regarded as inadequate measures by De Veirman et al. (2017). As various unethical practices are adopted by influencers or agencies to inflate their numbers such as buying fake followers, likes, etc. through paid websites, apps, and chatbots and such exercises negatively impact the public reaction and brand campaign. Nevertheless, the results of the conjoint analysis conducted by Gräve (2019) show that managers attach greater importance to sentiment measures while trading-off multiple KPIs. Sentiment analysis encompasses more information than simple heuristics (like the number of interactions) and this is the reason why sentiment KPI is being progressively used to evaluate the success of influencer marketing activities. Brands target positive net sentiment, where net sentiment index means positive minus negative comments.
Puschmann and Powell (2018) describe sentiment analysis as an objective tool to judge audience attitudes, feelings, and frame of mind on social sites. It reduces the large amount of intricate data into quantifiable sentiment scores to assist brands in marketing decisions. In simple words, sentiment analysis is the process of identifying users’ opinions expressed in text and categorizing them as positive, negative, or neutral. De Veirman et al. (2019) opine that sentiment analysis can be an advantageous tool to provide insights into consumers' perception towards brands, products, and influencers and to analyze the interactions between influencers and their followers. However, despite the appreciation received from various researchers, the use of sentiment analysis to study consumer's perspectives towards influencer marketing is still unexplored.
Objectives of the Study
Research Methodology
An exploratory research design has been adopted to conduct the study. The main purpose of exploratory research is to further investigate or probe a phenomenon. As the research topic is a new concept and is still budding, this approach will help to enhance the understanding of influencer marketing from the perspective of previous researches. An extant review of literature has been applied to the subject to achieve an in-depth understanding of this digital marketing strategy. Secondary data from various marketing and consumer behavior related journals, articles, and websites have been referred to accomplish the objectives of the present study. Research spanning from 2014-2020 from reputable publishing firms like Elsevier, Springer, Sage, Wiley, etc., were identified and analyzed to gain insights about influencer marketing as a digital marketing strategy.
To examine how social media users perceive influencer endorsements, a qualitative approach has been adopted, wherein sentiment analysis was conducted on the comment section of a beauty influencer to learn about consumer's opinions, emotions, and feelings towards a particular influencer and the brand endorsed. Previous research has majorly focused on Instagram and the fashion and beauty industry. Likewise, the present research intends to explore the scope of sentiment tool on fashion and beauty influencer on the Instagram platform. Chen (2018) acknowledges that convenience and contextuality are the key advantages of Instagram over other social sites. He discussed that many companies have found success through Instagram campaigns, such as hashtags being promoted by the influencer are made popular among the users which increase the brand awareness and following (i.e. audience size) of the brand’s official page. Guarda et al. (2020) express that Instagram is the foremost popular platform for influencers and brands to sponsor products or services and it has flourished rapidly due to its differentiating features such as aesthetics, sharing, and instantaneity. According to Influencer Marketing Hub survey report (2020), Instagram is used in nearly 90% of all influencer marketing campaigns. Correspondingly, many other reports and studies find Instagram to be the most preferred SNS. Hughes et al. (2019) identified that drivers of engagement on sponsored campaigns differ across platforms such as Blogs and Facebook. Thus, it becomes important to conduct different studies for each SNS because the nature of the application and content varies across all platforms. Watching the trend, this research chooses Instagram as the platform to study about influencers and their relationship with the followers.
There are plenty of advanced sentiment analysis tools that can break down any written message and quickly determine its sentiment. The present study used open-source software “R” to understand the sentiments of followers towards influencer promoted endorsements. R allows data processing and visualizations of the information using various techniques. The sentiment tool has been applied to the comments made by social media users on the influencer’s post on Instagram. The selected influencer post had a total of 63,677 likes and 642 comments. All the comments from the post were imported in an excel file using “ExportGram.com” and then emoticons and slangs were removed as they cannot be processed by the software. After data cleaning, a total of 487 comments were imported into the R-software. Stop words function was used to eliminate irrelevant words such as now, also, will, can, etc. NRC lexicon has been used in the study to categorize the comments into different sentiments.
Findings and Discussion
The paper analyzed past researches to gain an understanding of this newly emerged strategy, influencer marketing. The study explicated the emergence of influencer marketing, the significance of social media influencers, paid partnership between influencers and brands, and varied metrics used by brands to measure their performance. With burgeoning influencers across various platforms, specialized influencer marketing agencies came into being that act as a middleman between brands and influencers such as Viral Nation, TapInfluence, Mediakix, etc. Yet, measuring the results of influencer marketing campaigns i.e. how well the influencers have performed according to the company’s marketing goals remains the biggest challenge. It is extremely crucial for marketers to analyze the data generated on social media platforms as consumers tend to share their reviews, feedbacks, and opinions about anything and everything. Apart from the most common metrics like reach, engagement, and conversions, sentiment analysis is also being used as an emerging metric to measure the influencer's impact on consumers. The total reach or engagement of the message might not be a sufficient criterion for brand decision making, it is also important to analyze these comments in terms of their emotions and opinions, i.e. how the audience is feeling about the influencer and the brand endorsed. Brands ought to understand not only the content but also the context of what users are saying online.
Sentiment analysis allows brands to measure the different sentiments of consumers, i.e. their feelings, emotions, reactions, and attitudes. It is a sophisticated tool that uses machine learning algorithms to comprehend the true motive of a statement. There are many benefits associated with social media sentiment analysis. For instance, it helps to track the brand's perception and reputation, adds depth to the raw KPIs like reach and engagement, builds long-lasting customer relationships by tailoring the content according to audiences’ needs, and helps to spot and prevent potential crisis and take adequate measures when required. With the profusion of influencers on all SNS, it becomes necessary to research in-depth about influencers before launching any campaign. A proper exploration about influencers can help comprehend if the influencer has been involved in any negative social media backlash lately which precludes the possibility of selecting a wrong influencer. Sentiment analysis can help brands to look for the best influencers in the industry and create a good match for their products. So undoubtedly, sentiment analysis can come in handy to brands or marketers dealing with influencers on SNS.
The present study has used “R” software for conducting sentiment analysis to understand the followers’ opinions about the influencer and the endorsed brand on Instagram. The comments were collected from the official account of an influencer “SHAYLA” based in the USA. Shayla Mitchell is a fashion and beauty blogger with a YouTube channel having 741,000 subscribers and an Instagram account with 2,777,682 followers (approximately 2.7 million)(figures as on date of post extracted) and both accounts by the name @makeupshayla are verified. She makes in-depth makeup tutorials, fashion styling, and fitness videos. She goes by the notion that there’s a place for every skin tone with makeup and uses her platforms to push the industry forward by making brands more diverse. A post on 15th May 2019 was extracted from her Instagram account where she was recommending a skincare product by “Clinique”.
Cliniqueis a high-end American brand that manufactures skincare, cosmetics, toiletries, and fragrances. It is a subsidiary of the Estée Lauder Companies and has its presence globally in over 188 countries. Clinique has its influencer marketing game very strong since the very beginning of 2016 and has roped in numerous top beauty influencers in different countries for various campaigns like new product launches, event documentation, sponsorship, etc. Estée Lauder has teamed up with “SHAYLA” on many occasions, including several sponsorships to endorse their wide range of products. Here, we select a post which is a paid sponsorship between the influencer “makeupshayla” and the brand “clinique”, and the caption read as follows:
“Sometimes it feels good to give your skin a break! I use the @clinique id custom moisturizer everyday before my workout. I love that it’s lightweight and I can customize my hydration by changing the cartiridge. I’ve been using the orange cartridge because it has ingredients to energize & brighten my skin. Try it out at the Clinique counter in Sephora #ad”
Figure 1: Screenshot of the post utilized in the research
The post utilized in the data analysis (Figure 1) was available in public on the Internet. The post mentioned that it is a “paid partnership with Clinique” and the influencer has used the hashtag “#ad” in the description section. The software used a predefined “NRC lexicon” to infer whether most users are talking positively or negatively about the influencer and the product.
The results of sentiment analysis are presented in figure 2 below:
Figure 2: Sentiment analysis of the comments of an Instagram influencer
Figure 2 displays the overall sentiment score of the comments collected. NRC lexicon has grouped the comments into different emotions i.e., Positive, Negative, Anger, Anticipation, Joy, Trust, etc. The R software generates a quantifiable score of sentiments. It is noticeable from the above bar diagram that users are talking positively about the influencer and have a joyous and trustworthy feeling. For instance, one user shared “Oh they got the right one for this campaign!! Your skin is amazing” which implies the acceptance and appreciation for the influencer. The use of words expressing positive emotions is highest among the followers. Such positive image of an endorser is advantageous for brands. As per McCracken’s (1989) meaning transfer model, the characteristics of the endorser are transferred to the endorsed product or brand and thereby from the product or brand to the end-user through consumption which is like an image transfer effect, so the positive image of the influencer can be carried over to the brand an disultimately transmitted to the consumer which leads to improved brand image and behavioral intentions.
Anger, fear, disgust, and sadness are low scoring emotions, and very few users found it negative. The surprise is a neutral emotion which is also scoring low in this case. Coates et al. (2020) state that influencers regularly interact with their followers through unique features of social media such as comments, live stories, Q&A sessions, etc and this interpersonal communication creates a sense of familiarity with influencers which leads to a favorable attitude towards both influencers and product recommendations from them. The findings of the present research confirm with the results of Coates et al. (2020) and indicate that followers identify with the influencer and have a positive attitude towards the influencer, recommendations, and the partnered brand. It means that users are appreciating how the influencer is sharing reviews and recommendations for the skincare product. For example, few followers mention that “Omg @makeupshayla your skin looks amazing!! I’m going to get me some Clinique”, “@clinique ID is my favorite right now!”, “You use other Clinique products I’m going to try this” and “Thanks for sharing! Most definitely going to try this!!!”. These comments suggest users' liking and positive attitude towards the brand, purchase intentions, and efficacy of the influencer campaign.
The influencer has used the build-in standardized Instagram disclosure i.e. “Paid partnership with [brand]” on the post. Boerman (2020) reports that this standardized disclosure was found effective as users were able to recognize the post as advertisement, i.e. increased ad recognition and brand recall. Identically, the present results also deduce that followers were generally aware of the nature of paid endorsement collaboration, as comprehended by a user comment i.e. “I love that you actually use that product because you show it all the time! You deserve that ad girl”. Thereupon, we can articulate that clear disclosure about the sponsorship leads consumers to have a positive outlook and trust both the influencer and brand being promoted. So, brands must persuade influencers to use the standardized disclosure built into social media applications. The present results are in support of the study conducted by Dhanesh and Duthler (2019) who found that awareness of paid advertisement leads to ad recognition among the followers and they develop a relationship with an influencer based on trust and honesty.
Certainly, the positive attitude of users indicates that they appreciate how the influencer is rightly disclosing the advertisement and not deceiving them. Yet, the present results stand contradictory to the experiment conducted by Colliander and Erlandsson (2015), where disclosure about the monetary partnership by bloggers, led to a lower attitude towards the blog and its credibility. One possible reason for the deviation could be that users are now conversant with the influencer marketing strategy, and due to the plethora of social media influencers on various platforms, consumers feel that influencers have a variety of products to endorse, however, they promote only those products which they have tried and tested, which matches with their personality and when they genuinely carry a positive opinion about them. Thus, influencers are perceived as honest and trustworthy sources of information.
Summarising the results of the present study, we can say that largely followers are aware of the paid partnership of influencers with brands, thus it is of utmost importance for influencers to disclose this association to gain followers' trust and to build enduring relations with them. Social media users carry a positive opinion towards influencers and value their advice and recommendations. Thereupon, it is in the interest of brands to collaborate with social media influencers to accomplish their marketing goals like increasing brand awareness, ad recognition, higher sales, etc. The above findings convey that the social media sentiment analysis tool is useful for both marketers and influencers. Brands can use this technique to select the right influencer and to measure the results of influencer campaigns. Likewise, influencers can see their audiences’ opinions, emotions, and feelings and what their followers are liking or not liking and modify their content accordingly.
Figure 3: A word cloud representing the most frequently used words in the comments
A word cloud has been prepared using R software (Figure 3) to examine the frequency of words appearing in the comments. Word cloud is created from a text frequency count and the size of words shows their importance in the data set. Accordingly, the most frequently used words are “skin”, “flawless” and “beautiful”, these words convey that users love influencer’s skin and the results of the skincare product were quite visible to users. All other words appearing in the figure 3 above talk positive about the influencer such as “gorgeous”, “goals”, “glowing”, which means that followers admire the product’s results on the influencer’s skin and wish to have the same flawless skin.
Conclusion
The paper sheds light on why influencer marketing is slowly replacing other traditional methods and rising as the star of the show. Influencer marketing is a powerful strategy in the social media age. Marketing practitioners are progressively teaming up with influencers to get customer’s attention in the social media clutter. The field of influencer marketing seems promising in the near future, is undergoing constant change, and is set to be more challenging. This can be contributed to the changes that are taking place in social platforms like the development of new features such as Instagram TV (IGTV), social commerce (eg. Instagram storefront), andthe emergence of new SNS.With the convergence of e-commerce and social media, influencers encourage impulse purchasing among customers thus providing immediate returns to brand campaigns. There is an increased professionalization of social media influencers and the rise of specialized agencies and intermediaries of influencer marketing. Hence, all these reasons point that researchers must continue to extend knowledge and best practices in this field.
Consumers voice their opinions and feedback on social media sites about brands, products, and influencers. Apart from the product or service endorsement by the influencer on their personal accounts, the two-way aspect of social platforms provides an opportunity for the influencer to interact and maintain intimate relations with their followers, which gives the campaign a whole new dimension. Thus, personal conversations and recommendation can convert a follower into a brand consumer. Conventional marketing focused on mass audiences, but influencers have the distinguishing ability to target niche audiences that have so far been inaccessible. Traditional marketers used to focus on the 4P’s. Now, marketers have shifted their focus to the 3E's i.e. Engage, Equip, Empower. The right influencer can effectively target these E's and engage the audience to talk about the product, have positive associations about it, which will ultimately lead to increased sales. As influencers become more bountiful and proven, brands are flooding dollars into this space. Several types of influencers have thrived and each of these offers diverse benefits and trade-offs for brands.
Marketers employ a combination of metrics to select influencers and to measure the effectiveness of influencer marketing campaigns. While engagement is one of the most useful metrics to gauge a post's (content) relative popularity, sentiment analysis is also gaining quick recognition among marketing practitioners. It allows brands to enhance content strategies to effectively maximise ROI. In the present study, R software was used to analyze the language in the comment section to get a picture of how the influencer endorsement has been received by social media users. Above mentioned results of sentiment analysis conveys that users are talking positively about both the influencer and the brand/product mentioned in the post. The sentiment tool is expedient for both influencers and brands. Instagram is the most popular app for both social media users and influencers. Instagram also has a huge potential for brands to promote and reach their audiences through social media influencers.
Implications of the study
The present study throws light on influencer marketing as the most recognized digital strategy in the social media realm and how sentiment analysis can overcome the shortcomings of other KPIs. This study is of great relevance to academicians as it adds to the scarce pool of knowledge on influencer marketing and solicits further research into this phenomenon. There is also a need for developing countries like India to conduct more research since this strategy is gaining prominence among Indian influencers and brands. This paper has managerial implications for firms to embrace the sentiment metric to measure the results of an influencer campaign. The sentiment analysis tool can help brands and marketers to plan their social media marketing strategies and assess their effectiveness. This analysis can also be performed on reviews of other influencers, products, or services. The research is also valuable for influencers as they can figure out the opinion and predilection of the audience and create personalized content.
Limitations and Scope for Future Research
There are several limitations of this study that offer scope for future research. The comments from respondents have been drawn from a single influencer in the fashion and beauty industry. To further generalize the results, a wider sample of different influencers can be undertaken from diverse fields such as sports and fitness, travel, etc. Another drawback is the sample size as 487 comments were used for analysis. Future studies can increase the sample size to extrapolate the results further. Third, the study obtained comments from a specific platform, i.e. Instagram. Other SNS such as YouTube or Facebook could be explored to get a broader view of opinions. Moreover, a comparison study of few SNS can also be an advantageous contribution. Fourth, sentiment analysis might not be the most sophisticated technique to study user’s opinions. So, future researchers can adopt more advanced tools and techniques to analyze user behavior.
References
Abidin, C. (2016). “Aren’t these just young, rich women doing vain things online?”: Influencer selfies as subversive frivolity. Social Media+ Society, 2(2).https://doi.org/10.1177/2056305116641342
Abidin, C., &Ots, M. (2015, August). The influencer’s dilemma: the shaping of new brand professions between credibility and commerce. In AEJMC 2015, annual conference.
Balaban, D., &Mustățea, M. (2019). Users’ perspective on the credibility of social media influencers in Romania and Germany. Romanian Journal of Communication and Public Relations, 21(1), 31-46.
Boerman, S. C., Willemsen, L. M., & Van Der Aa, E. P. (2017). “This post is sponsored”. Effects of sponsorship disclosure on persuasion knowledge and electronic word of mouth in the context of Facebook. Journal of Interactive Marketing, 38, 82-92. http://dx.doi.org/10.1016/j.intmar.2016.12.002
Boerman, S. C. (2020). The effects of the standardized instagram disclosure for micro-and meso-influencers. Computers in Human Behavior, 103, 199-207.https://doi.org/10.1016/j.chb.2019.09.015
Casaló, L. V., Flavián, C., & Ibáñez-Sánchez, S. (2018). Influencers on Instagram: Antecedents and consequences of opinion leadership. Journal of Business Research, 117, 510-519.https://doi.org/10.1016/j.jbusres.2018.07.005
Chen, H. (2018). College-aged young consumers' perceptions of social media marketing: The story of Instagram. Journal of Current Issues & Research in Advertising, 39(1), 22-36.https://doi.org/10.1080/10641734.2017.1372321
Coates, A. E., Hardman, C. A., Halford, J. C. G., Christiansen, P., &Boyland, E. J. (2020). “It’s Just Addictive People That Make Addictive Videos”: Children’s Understanding of and Attitudes Towards Influencer Marketing of Food and Beverages by YouTube Video Bloggers. International Journal of Environmental Research and Public Health, 17(2), 449. https://doi.org/10.3390/ijerph17020449
Colliander, J., &Erlandsson, S. (2015). The blog and the bountiful: Exploring the effects of disguised product placement on blogs that are revealed by a third party. Journal of Marketing Communications, 21(2), 110-124.https://doi.org/10.1080/13527266.2012.730543
Couto, A. F., & de Brito, P. Q. (2020). Tactical Approaches to Disclose Influencers’ Advertising Partners. In Marketing and Smart Technologies (pp. 88-100). Springer, Singapore.
De Veirman, M., Cauberghe, V., &Hudders, L. (2017). Marketing through Instagram influencers: the impact of number of followers and product divergence on brand attitude. International Journal of Advertising, 36(5), 798-828.https://doi.org/10.1080/02650487.2017.1348035
De Veirman, M., Hudders, L., & Nelson, M. R. (2019). What Is Influencer Marketing and How Does It Target Children? A Review and Direction for Future Research. Frontiers in Psychology, 10, 2685.https://doi.org/10.3389/fpsyg.2019.02685
Dhanesh, G. S., &Duthler, G. (2019). Relationship management through social media influencers: Effects of followers’ awareness of paid endorsement. Public Relations Review, 45(3), 101765.https://doi.org/10.1016/j.pubrev.2019.03.002
Duffy, B. E., & Hund, E. (2015). “Having it all” on social media: Entrepreneurial femininity and self-branding among fashion bloggers. Social Media+ Society, 1(2).https://doi.org/10.1177/2056305115604337
Gräve, J. F. (2017, July). Exploring the Perception of Influencers vs. Traditional Celebrities: Are Social Media Stars a New Type of Endorser?. In Proceedings of the 8th International Conference on Social Media & Society (pp. 1-5).
Gräve, J. F. (2019). What KPIs are key? Evaluating performance metrics for social media influencers. Social Media+ Society, 5(3).https://doi.org/10.1177/2056305119865475
Gretzel, U. (2018). Influencer marketing in travel and tourism. In Sigala, M. &Gretzel, U. (Eds.), Advances in Social Media for Travel, Tourism and Hospitality: New Perspectives, Practice and Cases, pp. 147-156. New York: Routledge.
Guarda, T., Lopes, I., Victor, J. A., & Vázquez, E. G. (2020). User Behavior: The Case of Instagram. In Marketing and Smart Technologies (pp. 38-48). Springer, Singapore.
Hughes, C., Swaminathan, V., & Brooks, G. (2019). Driving brand engagement through online social influencers: An empirical investigation of sponsored blogging campaigns. Journal of Marketing, 83(5), 78-96.https://doi.org/10.1177/0022242919854374
Isosuo, H. (2016). Social media influencer marketing. (Bachelor Thesis, JAMK University of Applied Sciences, Jyväskylä, Finland). Retrieved from https://www.theseus.fi/bitstream/handle/10024/114035/thesisheliisosuo.pdf?sequence=1.
Jerslev, A. (2016). Media Times| in the time of the microcelebrity: Celebrification and the YouTuber zoella. International Journal of Communication, 10, 19.
Jin, S. V., Muqaddam, A., & Ryu, E. (2019). Instafamous and social media influencer marketing. Marketing Intelligence & Planning, 37(5), 567-579. https://doi.org/10.1108/MIP-09-2018-0375
Kamins, M. A. (1989). Celebrity and noncelebrity advertising in a two-sided context. Journal of advertising research, 29(3), 34-42.
Katz, E., &Lazarsfeld, P. F. (1966). Personal Influence, The part played by people in the flow of mass communications. Transaction publishers.
Khamis, S., Ang, L., & Welling, R. (2017). Self-branding,‘micro-celebrity’and the rise of Social Media Influencers. Celebrity studies, 8(2), 191-208.https://doi.org/10.1080/19392397.2016.1218292
Lou, C., & Yuan, S. (2019). Influencer marketing: how message value and credibility affect consumer trust of branded content on social media. Journal of Interactive Advertising, 19(1), 58-73.https://doi.org/10.1080/15252019.2018.1533501
Lou, C., Tan, S. S., & Chen, X. (2019). Investigating consumer engagement with influencer-vs. brand-promoted ads: The roles of source and disclosure. Journal of Interactive Advertising, 1-18.https://doi.org/10.1080/15252019.2019.1667928
Lu, L. C., Chang, W. P., & Chang, H. H. (2014). Consumer attitudes toward blogger’s sponsored recommendations and purchase intention: The effect of sponsorship type, product type, and brand awareness. Computers in Human Behavior, 34, 258-266.https://doi.org/10.1016/j.chb.2014.02.007.
Marwick, A. (2011). To see and be seen: Celebrity practice on Twitter. Convergence: the international journal of research into new media technologies, 17(2), 139-158.
McCracken, G. (1989). Who is the celebrity Endorser? Cultural Foundations of the Endorsement Process. Journal of Consumer Research, 16(3), 310-321.https://doi.org/10.1086/209217
Mitchell, S. [@makeupshayla]. (2019, May 15). Sometimes it feels good to give your skin a break! I use the @clinique id custom moisturizer everyday before my workout.[photograph].Instagram.https://www.instagram.com/p/BxduZ69Anbv/?igshid=73re9ccbdl81
Oliveira, M., Barbosa, R., & Sousa, A. (2020). The Use of Influencers in Social Media Marketing. In Marketing and Smart Technologies (pp. 112-124). Springer, Singapore.
Pang, A., Yingzhi Tan, E., Song-Qi Lim, R., Yue-Ming Kwan, T., & Bhardwaj Lakhanpal, P. (2016). Building effective relations with social media influencers in Singapore. Media Asia, 43(1), 56-68.
Pedroni, M. (2016). Meso-celebrities, fashion and the media: How digital influencers struggle for visibility. Film, fashion & consumption, 5(1), 103-121. https://doi.org/10.1386/ffc.5.1.103_1
Puschmann, C., & Powell, A. (2018). Turning words into consumer preferences: How sentiment analysis is framed in research and the news media. Social Media+ Society, 4(3).https://doi.org/10.1177/2056305118797724
SanMiguel, P., Guercini, S., &Sádaba, T. (2018). The impact of attitudes towards influencers amongst millennial fashion buyers. Studies in Communication Sciences, 18(2), 439-460.https://doi.org/10.24434/j.scoms.2018.02.016
Smith, E. (2016, May 4).TapInfluence and Nielsen Catalina Solutions Launch Collaboration for CPG Brands to Measure Sales Impact of Influencer Marketing Campaigns [Blog post]. Retrieved from:https://www.tapinfluence.com/blog-tapinfluence-and-nielsen-catalina-solutions-launch-collaboration-for-cpg-brands-to-measure-sales-impact-of-influencer-marketing-campaigns/
Sudha, M., & Sheena, K. (2017). Impact of influencers in consumer decision process: the fashion industry. SCMS Journal of Indian Management, 14(3), 14-30.
Tapinfluence. (2015, February 6).Influencer Marketing. Retrieved march 16, 2020 from https://www.tapinfluence.com/blog-what-is-influencer-marketing/
The State of Influencer Marketing 2019: Benchmark Report. Retrieved march 15, 2020 fromhttps://mediakix.com/influencer-marketing-resources/influencer-marketing-industry-statistics-survey-benchmarks/#gs.cwtjta
The State of Influencer Marketing 2020: Benchmark Report. Retrieved march 20, 2020 from:https://influencermarketinghub.com/influencer-marketing-benchmark-report-2020/
Uzunoğlu, E., & Kip, S. M. (2014). Brand communication through digital influencers: Leveraging blogger engagement. International Journal of Information Management, 34(5), 592-602.https://doi.org/10.1016/j.ijinfomgt.2014.04.007
Woods, S. (2016). # Sponsored: The emergence of influencer marketing. (Honours Thesis, University of Tennessee, Knoxville, USA). Retrieved from http://trace.tennessee.edu/utk_chanhonoproj/1976
Xiao, M., Wang, R., & Chan-Olmsted, S. (2018). Factors affecting YouTube influencer marketing credibility: a heuristic-systematic model. Journal of Media Business Studies, 15(3), 188-213.https://doi.org/10.1080/16522354.2018.1501146
ANNEXURE
Table 1: Word frequency of comments:
|
Skin |
206 |
|
Flawless |
53 |
|
Beautiful |
40 |
|
Yes |
38 |
|
Goals |
30 |
|
Gorgeous |
25 |
|
Wow |
23 |
|
Like |
20 |
|
Glow |
19 |
|
Glowing |
16 |
|
Beauty |
14 |
|
Shayla |
14 |
|
Come |
13 |
|
Look |
12 |
|
Amazing |
11 |
|
Good |
11 |
|
Perfect |
10 |
|
Clinique |
10 |
|
Makeupshayla |
9 |
|
Love |
9 |
