“This is my favorite beauty product… Do you like it too?”

(1)

“This is my favorite beauty product…

Do you like it too?”

A quantitative study on the user's interaction on beauty brands' Instagram account and how the presence of an influencer effect it.

En kvantitativ studie om användares interaktion på skönhetsmärkens Instagram konto och hur en influencer’s närvaro påverkar den.

Fanny Evensson and Erika Jansson

Faculty of Humanities and Social Science

Media and Communication: Digital Media and Analysis 15 HP

Michael Karlsson 9/6/2020

(2)

Abstract

Looking at earlier studies and previous research in influencer marketing and parasocial relations between influencers and their followers, many of them are studying influencers in other countries such as Singapore and not in Sweden. This is a research gap we want to fill, which is why this study is focusing on Swedish influencers. By investigating the effects of interaction between consumer and producer might give a better understanding of how marketing strategies can develop in the online environment by using the interactive features of the social media platform Instagram.

For this study, the objects of analysis are three Swedish companies selling beauty products. One of the purposes of this study is to illustrate what effect an influencer has on a company’s interaction in terms of likes and comments on Instagram. Another one is to create new knowledge in this research field by studying a specific industry, such as the beauty market. In light of previous research that confirms that influencer marketing is reaching several markets and are quickly expanding, we want to examine how this new way of marketing takes place among Swedish brands. Based on the purposes of the study, two research questions have emerged:

➢ What effect does the participation of an influencer have on a company's Instagram account in terms of interaction?

➢ What features on Instagram can be seen as influential factors on the interaction in terms of likes and comments?

The analytical framework is based on the theory of parasocial interaction, which has the purpose to give an understanding of how the relationship between influencers and followers are established in today’s social media environments. In addition, we have used relevant previous research in perceived interconnectedness, influencer marketing, and how the interaction online from the audiences have developed.

The empirical material that was examined in the study was content in forms of posts on the social media platform Instagram. The objects of analysis are three Swedish beauty product companies, CAIA Cosmetics, Indy Beauty, and Kicks. Two well-established influencers have created the first two companies; Bianca Ingrosso is the creator of CAIA Cosmetics, and Therese Lindgren is the creator of Indy Beauty. The collected data, 300 posts in total, has been analyzed through a quantitative content analysis.

The results from the study show that there is an extensive difference in interactions, in terms of likes and comments on average, when the companies are posting content including the influencer, in contrast to the opposite. Even if the number of posts with influencers is considerably low on two of the Instagram accounts compared to posts without influencers, they still manage to achieve more interaction in both likes and comments. For CAIA Cosmetics, the results show that interactions on average share approximately the same amount of numbers throughout the analysis. For Indy Beauty, this was not the case. The number of comments on average when not including Therese Lindgren was distinguishable higher compared to when including her.

(3)

The reason behind this result might be a product-oriented campaign Indy Beauty launched at the beginning of December 2019. The empirical material of this study included data from this specific time and caused an effect on the results in this case. The campaign only included posts of products, and one of the requirements to win the Christmas gift was to comment on the posts, which led to a higher number of comments of posts without an influencer.

To summarize the analysis, the involvement of an influencer in its marketing strategy gives a certain effect in interaction. However, it can also be efficient to apply the features of Instagram and launch product-oriented campaigns on your social media platforms to gain interaction in terms of likes and comments among your followers. Finally, it is difficult to generalize the conclusion since the outtake of this study only includes three Swedish companies and their content.

Keywords: Influencer, interaction, parasocial interaction and beauty market.

(4)

Sammanfattning

I tidigare studier och forskning inom influencer marketing och parasociala relationer mellan influencers och deras följare, studeras influencers i andra länder såsom Singapore och inte i Sverige. Detta är ett forskningsområde vi vill bidra till genom att i denna studie studera svenska influencers. Genom att undersöka effekterna av interaktion mellan konsument och producent kan det ge en bättre förståelse för hur marknadsstrategier kan utvecklas i dagens sociala mediemiljöer och dessutom genom användandet av de interaktiva funktionerna på den sociala plattformen Instagram.

Ett av syftena med denna studie är att illustrera vilken effekt en influencer har på interaktionen när det gäller likes och kommentarer på Instagram. En annan är att skapa ny kunskap inom detta forskningsområde genom att studera en specifik marknad, till exempel skönhetsmarknaden. Mot bakgrund av tidigare forskning som bekräftar att influencer marketing når flera marknader och snabbt expanderar, vill vi undersöka hur detta nya sätt att marknadsföra sker bland svenska märken. Baserat på syftet med studien har två forskningsfrågor framkommit:

➢ Vilken effekt har deltagandet av en influencer på företags Instagram-konto när det gäller interaktion?

➢ Vilka funktioner på Instagram kan ses som inflytelserika faktorer för interaktionen när det gäller likes och kommentarer?

Det analytiska ramverket är baserat på teorin om parasocial interaktion, som har som syfte att ge en förståelse för hur förhållandet mellan influencer och följare är etablerat i dagens sociala mediemiljöer. Dessutom har vi använt relevant tidigare forskning i perceived interconnectedness, influencer marketing och hur interaktionen från publiken har utvecklats online.

Det empiriska materialet som undersöktes i studien var innehåll i form av inlägg på den sociala medieplattformen Instagram. Analysobjekten är tre svenska skönhetsproduktföretag, CAIA Cosmetics, Indy Beauty och Kicks. Två väletablerade influencers har skapat de två första företagen; Bianca Ingrosso är skaparen av CAIA Cosmetics, och Therese Lindgren är skaparen av Indy Beauty. De insamlade uppgifterna, totalt 300 inlägg, har analyserats genom en kvantitativ innehållsanalys.

Resultaten från studien visar att det finns en omfattande skillnad i interaktioner, i termer av likes och kommentarer i genomsnitt, när företagen publicerar innehåll inklusive influencers, i motsats till att inte inkludera influencers. Även om antalet inlägg med influencers är avsevärt lågt på två av Instagram-kontona jämfört med inlägg utan influencer lyckas de fortfarande uppnå mer interaktion i både likes och kommentarer.

För CAIA Cosmetics visar resultaten att interaktioner i genomsnitt delar ungefär samma antal siffror under hela analysen. För Indy Beauty var detta inte fallet. Antalet kommentarer i genomsnitt när Therese Lindgren inte inkluderades var högre jämfört med när hon inkluderades. Anledningen bakom detta resultat kan vara en produkt-orienterad kampanj Indy Beauty lanserades i början av december 2019. Det empiriska materialet i denna studie inkluderade data från denna specifika tidpunkt och påverkade i detta fall resultaten. Kampanjen

(5)

inkluderade bara inlägg av produkter, och ett av kraven för att vinna en julklapp var att kommentera inlägget, vilket ledde till ett högre antal kommentarer på inlägg utan influencern.

Slutsatsen av studien visar genom att involvera influencers i sin marknadsföringsstrategi kan ge en viss effekt i följares interaktion. Det kan även vara effektivt att tillämpa funktionerna på Instagram och lansera produkt- orienterade kampanjer på dina sociala medieplattformar för att öka interaktionen när det gäller likes och kommentarer bland dina följare. Slutligen är det svårt att generalisera slutsatsen eftersom undersökningen från denna studie endast innehåller tre svenska företag och deras innehåll.

Nyckelord: Influencer, interaktion, parasocial interaktion och skönhetsbranschen.

(6)

Preface

This essay is written together by Fanny Evensson and Erika Jansson. We have both taken an equal responsibility in terms of commitment and participation during the joint work process.

We want to thank our supervisor Susanne Almgren for her advice, knowledge, feedback, and support throughout this semester and study.

We also want to thank each other for great collaboration during the work process!

Fanny Evensson & Erika Jansson Karlstad, June 2020

(7)

Table of figures

Table 3:1 33

Table 4:1 37

Table 4:2 38

Table 4:3 40

Table 4:4 41

Table 4:5 42

Table 4:6 43

Table 4:7 44

Table 4:8 45

Table 4:9 46

Table 4:10 47

(10)

1. Introduction

The first chapter will begin by introducing the background of this study, it will give a presentation of the research problem, the limitations of the study, and definitions.

Social platforms online such as Facebook, Instagram, and YouTube offer various infrastructures, where users of social media can join social networks, read blogs, or watch videos (Kaplan & Haenlein, 2010). The increasing availability of high-speed internet has led to creations of social networking sites such as Facebook and Instagram (Kaplan & Haenlein, 2010). By investigating the effects of interaction between consumer and producer might give a better understanding of how marketing strategies can develop in the online environment by using the interactive features. The features of the social media platform allow the users to interact with each other, show appreciation, and be supportive. This study focuses on Instagram, where the features likes and comments are most common. Liking something is a way for the users to show support on the posts since it is a positive feature in the form of a heart (Instagram, 2020). When users comment on the post, all comments are collected in a comment section beneath the post. In the comment section, all users can read, like, and comment on other comments, which makes it possible for the users to connect.

Subsequently, the features give the company a way of seeing their audience and measure what they like, which can help them grow their business and reach a bigger audience (Green & Jenkins, 2020).

In Abidin and Ots’s (2016) study, one specific group of stakeholders in the transforming marketing landscape is more focused on; influencers. The everyday Internet users who have become the new social media micro- celebrity. They carry a persona that feels authentic to their audience, and their success establishes on the management and growth of their personal brands that are grounded in a deep and intimate relationship with their social media followers (Abidin & Ots, 2016).

As influencers can increase the number of followers on Instagram, they become valuable for marketing and brand endorsers as influencers can produce adverts on their social media accounts. The influencers advertise companies and, through their accounts, accumulate customers to the company in exchange for payments and sponsored products (Abidin and Ots, 2016). By implementing influencer marketing to market products or services of a company, one may reach an incredibly high amount of people that might represent a new market for the company. De Veirman, Cauberghe, and Hudders (2017) mention that companies have been discovering this type of marketing strategy recently and understood the impact one specific influencer might have on his/hers followers on for example social media platforms. Companies who target the right influencer can gain high credibility, grow their social media audience, and have an advantage within influencer marketing, which for an e-commerce business can be crucial (De Veirman, Cauberghe, & Hudders, 2017).

Previous research shows that influencer marketing has proliferated the last couple of years (Vettese, 2019), which has led us to investigate if the presence of an influencer has any effect in interaction on Instagram.

Interaction for this study is some of the features of Instagram, such as likes and comments. What lay behind

(11)

the action to interact then? Horton and Wohl’s (1956) theory parasocial interaction (PSI) defines as the illusion of a seemingly face-to-face relationship with the performer that mass media allows. It describes the feeling of an intimate connection that an audience member feels towards a performer in media. It is one-sided, nondialectical, and only controlled by the performer (Horton & Wohl, 1956). The more the performer adjusts the performance towards the supposed response, the more the audience makes the predictable response, and this creates an illusion of intimacy (Horton & Wohl, 1956).

According to Abidin (2015), Horton and Wohl (1956) oppose that the parasocial interaction is the created one-sided relation that mass media produces with media personalities. However, Abidin (2015) implies that this is what the influencer does as well when mediating intimacy through social media platforms, though they are using another structural way of doing it. Abidin's (2015) communication model perceived interconnectedness looks at how the intimate relationship between influencers and their audience is created. This model refers to the interactive features that social media platforms allow, which makes it easier for influencers and audiences to have a tangible give- and take relationship. Since social media platforms allow the influencer to control their self-presentation and interactions with their audience directly, it extends beyond the PSI in some ways (Abidin, 2015).

We are interested in studying how Swedish beauty brands relate to the theory of PSI (Horton & Wohl, 1956) and to examine what effect it can generate on the interaction of the companies' Instagram account. This research area is interesting to investigate since influencers are a new way of marketing beauty products on social media (Gerdeman, 2019). It is also essential to study this area since the online environment has changed marketing strategies (Vettese, 2019), and this analysis can be applied to social media platforms. Companies can apply this knowledge when developing their marketing strategies to reach their target group in the right way to get the highest amount of impact. We want to fill a research gap where analysis shows influencer's impact on the beauty industry by looking at how much interaction is generated on social media platforms. Can a pattern in the interaction be detected by looking at an influencer's presence?

Abidin (2015) reports that since the influencers attract a considerable number of followers on social media, the competition for the consumer's attention is tough. Influencers are improving their way of marketing, specifically through Instagram, where they can market their products and services with high-quality photos and visual effects, making the consumers have an at the fingertip experience with their content. Many consumers reject traditional marketing forms, such as TV and magazine ads (Abidin, 2018). This study chooses to analyze the social media network Instagram since it is the second most popular social media network in Sweden (Internetstiftelsen, 2019) and because of how influencer marketing has increased on the platform (Schomer, 2019). The global spending in influencer marketing has been rising from $2 billion in 2017 to about

$8 billion in 2019. It is estimated to be $15 billion by 2022 (Vettese, 2019).

(12)

1.1 Research problem

Looking at earlier studies and previous research in influencer marketing and parasocial relations between influencers and their followers, many of them are studying influencers in other countries such as Singapore and not in Sweden. This is the research gap we want to fill, which is why this study is focusing on Swedish influencers on Instagram. For this study, the objects of analysis are three Swedish companies selling beauty products.

With this study, we want to contribute to research by comparing Instagram accounts to see if presenting influencers for marketing reasons can make a difference in interactions. Marketing beauty products on social media have expanded rapidly in the last couple of years, which makes it relevant to study this market and its influencers. It is relevant since it fills a gap in the research field yet also relevant for companies in this industry to understand the importance of the relationship between an influencer and its followers to be able to make an effect and reach the company's marketing goals.

1.2 Purposes of the study and research questions

One of the purposes of this study is to illustrate what effect an influencer has on a company’s interaction on Instagram. Another is to create new knowledge in this research field by studying a specific industry, such as the beauty market. In light of previous research that confirms that influencer marketing is reaching several markets and are quickly growing, we want to examine how influencer marketing is shown for Swedish companies.

For us to examine this, an analysis of three Instagram accounts will be conducted by applying a quantitative content analysis. We will study content from each Instagram account and measure the features in forms of likes and comments.

Based on the purposes of the study, a research question emerged:

➢ What effect does the participation of an influencer have on a company's Instagram account in terms of interaction?

During the process of conducting this study, we came across the different features that the social media platform Instagram allows, such as tag and hashtag, which can generate interactions as well. Subsequently, a second research question emerged:

➢ What features on Instagram can be seen as influential factors on the interaction in terms of likes and comments?

(13)

1.3 Limitations of the study

Instagram is just one of several social media platforms available in Sweden for influencer marketing. Facebook, Twitter, and YouTube are other popular networking sites. Instagram is the second most popular social media in Sweden (Internetstiftelsen, 2019) and will be the only object of analysis for this study.

Another limitation is the difference in the timeline of each account's content. It may differ in seasons, depending on how active the account is with posting content on Instagram. The collection of data will start from the same date yet not finish within the same month, which can have an effect on the results.

1.4 Definitions

Influencer

Young people that monetise their lives, someone that often is an ordinary internet user but has a relatively high number of followers on social media. Influencers show their lives and lifestyles together with highly personal promotions of products and services, for a fee (Abidin, 2015).

Interaction

In this study the interaction refers to features Instagram allows such as liking and commenting posts (Instagram, 2020).

Advertorials

Advertorials are the advertisements that are written as editorial opinions (Abidin, 2016). In this study it refers to the posts with commercial and sponsored content posted by influencers on their social media accounts.

Posts

A post in this study refers to a picture, video or an InstagramTV posted on Instagram. The posts are all gathered as a timeline on a specific feed that is connected to the company's account (Instagram, 2020).

Followers

The definition of followers in this study refers to the people that use the social media platform feature

“follow”. It allows someone to connect their account with another one's account to show support and get updates on the account in order to interact continuously (BigCommerce, 2020).

CAIA Cosmetics/ Bianca Ingrosso

CAIA Cosmetics is a beauty product brand. It was created by Bianca Ingrosso who is a blogger, social influencer and singer. She has 346 000 subscribers on YouTube (Youtube, 2020) and 1 100 000 followers on Instagram (Instagram, 2020). The products can only be bought on their E-commerce store, caiacosmetics.se (CAIA Cosmetics, 2020).

(14)

Indy Beauty/ Therese Lindgren

Indy Beauty is a cruelty-free and vegan beauty brand that is owned and operated by Therese Lindgren, who is a YouTuber that fights for animal rights and mental illness. She is one of Sweden's biggest YouTubers with 1 000 400 subscribers (YouTube, 2020) and an Instagram Influencer with over 900 000 followers. Therese has won awards for YouTuber of the year, Business of the year and has released non-fiction books that became best-selling 2016 (Indy Beauty, 2020).

Kicks

Kicks is one of the biggest beauty chains in the Nordic countries, it was formed in 1991 and they now have 250 stores in Sweden, Norway and Finland which they combine with an E-commerce store. Their staff are trained beauty professionals, make-up artists and beauticians (Kicks, 2020).

1.5 Disposition

Altogether, the essay consists of seven chapters, of which the first one is the introduction. In chapter two, the theories and previous research on which the study is based on are presented. In chapter three, the chosen method of the study to help investigate the research area and help fulfill the purpose and answer the research questions of the study is presented. The variables, operalization, the collecting procedure, ethical issues of the study, and what makes the study reach high reliability and validity, is also presented in chapter three. Chapter four presents the results of the collected empirical material in tables. Subsequently, a discussion and an analysis of the results are presented in chapter five. In chapter six, the conclusions of the study are presented and also suggestions for further research. The essay finish with chapter seven, where implications for society are presented. Subsequently, references and appendixes of code scheme, code instructions, and other tables are displayed.

(15)

2. Analytical framework

This section will explain theories and previous research on parasocial interaction and its development in the social media environment, influencer marketing and the development of active audiences online.

2.1 Parasocial interaction theory

This study applies Horton and Wohl's (1956) theory about parasocial interaction (PSI) to identify what patterns in interactions of influencer marketing on social media can be found by studying Instagram accounts.

PSI is the illusion of a face-to-face relationship a member of the audience has with the performer (Horton &

Wohl, 1956). The interaction is characterized by being one-sided and controlled by the performer who adjusts the performance to get the supposed response of the audience. The PSI is comparable to and, in many ways, resembles the social interaction that happens between people in real life. The interaction encourages the audience to feel a sense of intimacy with the performer as it would towards a real person. However, the image of the performer that is presented is, to some extent, a construct with little resemblance to the performer's private character. The private life of the performer is not to be made an absolute secret since it creates an interest for the audience but it should be made into an acceptable facade that the audience finds desirable and interesting (Horton & Wohl, 1956). The influencer who is the performer has the control over the relationship but there are ways in which the follower can make their feelings known to the performers and technicians of the show. Whoever finds the experience unsatisfying has the option to withdraw, which in this case means unfollow the account on Instagram (Horton & Wohl, 1956). It is argued that what the media persona offers above all is a continuing relationship and that the appearance is regular and integrated into the routines of daily lives. Those who follow, share a small part of the performer's public life. During the time, the devotees will come to believe that he knows the performer (Horton & Wohl, 1956).

PSI can be recognized as the relationship one can develop with media personalities on TV when, for example, the viewer agrees with the TV-host. The more the performer seems to adjust his performance to the supposed response of the audience, the more the audience tends to make the response anticipated (Horton & Wohl, 1965). Since influencers can adapt their content on Instagram towards their followers and the followers can in return like and comment to support them, the give and take actions are related to what is needed in a PSI.

Furthermore, by having a back and forth interaction in the comment section or when the “ask a question”

feature is used a give and take interaction is also made (Horton & Wohl, 1956). Further on, this study will apply the communication model perceived interconnectedness by Abidin (2015), who argues is a model that describes the intimate relation that goes beyond PSI between an influencer on social media and their audience.

(16)

2.2 Perceived interconnectedness

The internet allows social platforms where people, often young, monetize their personal lives by posting and sharing content about themselves to their followers (Abidin, 2015). Some of them have created accounts so that many people want to follow and interact with, making them attractive for commercial use. Abidin (2015) mentions in her article that she conceptualizes these internet microcelebrities as influencers regardless of the platform they are using. Through the textual and visual narration that social media platforms allow, these ordinary users can accumulate a large number of followers just by showing their personal lives and lifestyles.

At the same time, they can integrate and market adverts and commercial content into their account for a fee.

The industry is highly personalized, as the promotion of products and services the influencer promotes are personally experienced. Previous research mentions that influencers can be seen as celebrities of the internet (Abidin, 2018). A traditional celebrity, as Abidin (2018) argues, is someone that is associated with an achievement, talent, or position. Nevertheless, having fame is not necessarily tied to these sources. When the public takes an interest in a person for their personal lives rather than for what they have accomplished, they stop being just a public figure and become a celebrity with a public persona. The study objects that have been chosen for this study are not only influencers, since they have extended their careers to reality shows, TV, and have become entrepreneurs with their beauty brands. This makes a gap from previous research since they are not only mainstream celebrities or influencers; they are a mix between these two categories.

Abidin (2015) refers to Horton and Wohl's (1956) work on PSI and argues that influencers use and mobilize intimacy in different ways on their accounts and towards their followers, which the author connects to a model of communication between the influencer and followers, called perceived interconnectedness. Influencers have now gone beyond the web, and their impact and status have even been recognized by the mainstream industries, which is challenging the media space where regular celebrities and public figures once were dominant. A difference compared to celebrities, is, however, that the influencers refer to their audience as followers instead of fans to avoid the hierarchical naming that "fans" implies. The influencers want to maintain the intimacy they have with their followers, which they do by emphasizing that they are ordinary people and not above their followers. The intimacy that plays out between influencers and followers is understood as to how familiar and close the followers feel to an influencer. It seems to be possible that this intimacy can be motivated by commerce as long as the followers feel close and attached to the influencer even if they are critically aware of the situation (Abidin, 2015).

This critical awareness goes both ways as the influencers seem to be aware of the underlying profit-orientation behind their actions, but even though they have an underlying commercial interest, they also experience personal attachments toward their followers. PSI is constructed to be more theatrical and has a production backend; in contrast, the perceived interconnectedness is a more democratic infrastructure where influencers and followers produce and shape the conversation amongst each other (Abidin, 2015). As Abidin (2015) claims, influencers can interact with their audience in a more intimate way, which creates an illusion that they

(17)

are connected, familiar, and closer with the audience. Unlike mainstream celebrities, influencers are ordinary internet users that happen to have many followers. That is why this study explores, with the help of perceived interconnectedness, how this intimacy between the audience and influencer is used commercially.

In this study, two of the companies that are created by influencers that are both entrepreneurs and have extended careers from being influencers on social media are investigated. The reason for this is that their audience originates from different mediums and can be seen as expecting different types of intimacies and connections towards the performers. Previous research has shown how audiences are influenced by PSI (Labrecque, 2014), and we believe that this study can fill a research gap by analyzing how the PSI is related to influencer marketing on social media in Sweden.

2.3 Interactive audience

In this section, previous research will be looked at to explain what an active audience means. It explores interactivity and how this phenomenon enables the audience to be active, which makes it easier for them to become fans. The negative effects of the interactive platforms are further discussed, and what problems it brings to the fan-created communities (Olsson, 2013). With interactive media, audience behaviour has changed. They are no longer expected or seen as being only able to recover the author's meanings and reworking it to make it fit in the context of lived experiences. Interactivity makes it possible for audiences to create new texts from the original stories, viewing a Tv-series and seeing it as a springboard for conversation and debate among the fans (Olsson, 2013). Interactivity makes the media gain resonance and significance.

How the media has evolved, allows social media to be viewed as a tool for community formation, as online communities strengthen the offline ties of fandoms (Olsson, 2013). Fans can now, thanks to being online, discuss their favourite TV-series, film or idol in real-time as they are broadcasted (Duffett, 2013). The characteristics of social media therefore means that a new audience can be formed. As Duffett (2013) mentions, the new pace of interaction that is possible between the members of the audience, means that practices of the media industry had to change and adapt with it.

Users can now upload, download, stream and share content (Duffett, 2013). They can take part in a new way and interact with the media content, producers and stars. Often the fans know more about a character or TV- series than the actors or people related to the actual production of the content, and the development of new media, create possibilities of new forms of fandom that are even more involved (Hills, 2002).

To what extent has the internet transformed fan cultures? The internet has placed the power of fan activity and distribution of media content in the hands of the audience itself, now having the possibility of accessing the information themselves and even communicating directly with the performers and producers (Duffett, 2013). However, Duffet (2013) argues that a complicated aspect of the active audience is that the online communities, in some cases without the audience intention, make a profit for site owners and sponsors (Duffet, 2013). Platforms often hide that all user activity, whether it is active by posting, commenting or

(18)

passive observation, contributes to the platform or site’s economy. This means that platform architecture pushes users towards being active and connecting, to create groups and communities (Olsson, 2013).

2.4 The role of parasocial interaction in social media environments

Previous research has shown how PSI can be recognized from traditional mediums such as TV (Horton &

Wohl, 1956). However, digitalization is changing the media audience's behavior and how they interact with digital devices. This creates a new field for researchers to study since the shape and form of PSI might have changed over the last period of time. Abidin (2015) argues that because the social media platforms allow the performer to interact with the audience in more ways than what other medium has allowed, influencers can create and have an even more intimate bond with the audience than other performers can. Which is one of the reasons why marketing on social media is an important subject to investigate together with PSI. As Abidin (2015) states, the audience on social media feel a more intimate and trustworthy relationship towards influencers than they do with mainstream celebrities. Their interaction with the influencer is more organic and similar to being friends with someone (Abidin, 2015). Even though TV and magazine are mediums that have commercials of celebrities, they tend to be less credible, since the consumers want to ask questions and get personal responses which a social media platform allows for. At the same time, the marketing of the product gives the consumer a feeling of being a regular person. Subsequently, the established beauty brands are losing their credibility as they cannot offer the same direct-to-consumer relation without using influencers’ perceived intimate relationship with the consumers (Gerdeman, 2019).

Labrecque's (2014) study examines how to foster consumer-brand relationships in social media environments and what role PSI has in it. Labrecque (2014) also brings up the importance of having active social media accounts for companies since social media platforms are a massive part of people's everyday life. The study shows how historical theoretical frameworks can be related to today's online environment and how companies can use these theories in order to build relationships with their customers (Labrecque, 2014). The theory that the scholar is referring to is PSI. It describes as being "used to help explain a brand's success in developing strong ties with consumers through social media and provides insights on how to preserve intimate relationship feelings in light of the increasing movement towards response automation" (Labrecque, 2014, p.

135).

So, how can PSI be placed within an online environment? Since it is not limited to traditional mass media, it can be shown through messages online that are designed to bring the customers closer to a specific brand or celebrity, or as Labrecque uses the term media persona. (Labrecque, 2014). While social media can be a successful choice for marketing, Labrecque shows some negative aspects of the automation of messages and their outcomes. Through technological developments, it is now possible to have direct two-way communication between the media persona and its audience. However, it might end up being a one-way conversation to the audiences since many companies use pre-written scripts, and the responses are automated, which causes a loss of the two-way communication (Labrecque, 2014). She says that companies "can even

(19)

scan message content to inform dynamically generated responses" (Labrecque, 2014, p. 136), which makes it harder for the audience to know the difference between what is real or not. They might question their loyalty to the company.

Labrecque’s (2014) study can be used as a foundation for companies who value communication very highly and want to reduce the distance between them and their customers. Continuing with the example of messages, Labrecque (2014) discusses two message components that ought to be transferred from traditional settings of parasocial interaction to the online environment. The first one is “the perceived interactivity of the persona with the viewer” (Labrecque, 2014, p. 136) and can be seen through signs of messages that indicate responsiveness and listening. The second message component is openness in communication, which imitates the brand's self-disclosure, and can be seen through the content of the message (Labrecque, 2014). Research shows that what makes the perceived interactivity as efficient as possible, is that the speed of the response from the brand is crucial. Basically, how quick they are to respond to their customers but also that they communicate the most accurate message that matches the customer's needs. In this case, the term interactivity does not focus on the technological features of the social media account, but on the content and symbolic of the message as this is what really matters (Labrecque, 2014).

PSI can be related to the feelings one might have for a real relationship since it involves revealing personal or private information about you, which builds trust and intimacy (Labrecque, 2014). Sharing this information creates openness between the participants, even if it is only the media persona who is sharing; it still feels for the audience as if they have something special together. Research also shows that audiences who engage in PSI feel the desire to learn about personal details of the persona, and that is similar to having a real relationship (Labrecque, 2014). The power of the Internet has changed how communication looks like, and it provides consumer-brand relationships with the ability to grow and to succeed (Labrecque, 2014).

2.5 Marketing through Instagram influencers

Influencers can be seen as the popular kids in school, as they are more likely to be found as more likable and popular based on their high number of followers (De Veirman, Cauberghe & Judders, 2017). Since they have a high number of followers, they are perceived as tastemakers, and as brands need to evolve with the new media, they focus more on these influencers as they can promote their products among their followers who subsequently can spread it further. With the new media, people are rejecting the traditional media and paid to advertise, and the influencers make the message more genuine, authentic, and credible (De Veirman, Cauberghe & Judders, 2017). However, even though brands that want to stay relevant need to be active on social media and use influencers in their marketing, they are up for a challenge when the brand needs to identify and select the right influencers. The influencers need to have a substantial impact on the brand's target group and be able to influence them to become consumers. A high number of followers can result in a more extensive reach of consumers, but to increase the impact of the message, the influencers need to be likable,

(20)

credible and be seen as an opinion leader by the audience. This means that there is a challenge for brands and producers to find suitable influencers that can, through their social media channels, be both representatives for the brand but also be able to influence their follower's attitudes, decisions, and behaviors. This means that influencer marketing from the beginning is based out of brands identifying and targeting who is an influential user (De Veirman, Cauberghe & Judders, 2017).

2.6 Influencer marketing in beauty markets

In Gerdeman’s (2019) article, it is discussed how influencers are changing the marketing of beauty products online, and a study made by a Harvard Business School MBA graduate named Alessia Vettese is presented.

The article starts with a segment where the British makeup guru Katie Jane Hughes shows beauty products for her 336 000 followers on Instagram and gives beauty tips on how to apply the products. Hughes has a YouTube channel with tutorials where her followers can see her applying all the beauty products from the start with a scrubbed clean face, with the same spots that one might recognize by themselves, until the last finishing product. She is giving her best beauty advice and sharing different experiences in the video, which makes it feel like a conversation more than an advertisement. She is not only talking and creating a bond with her followers, but she is also selling the beauty products she uses. Gerdeman (2019) is saying, Hughes is not the typical celebrity IT-girl, yet her social media posts convert many followers into customers, which increases the selling of products.

Gerdeman (2019) mentions that new research shows that influencers like Hughes are changing beauty marketing on social media, especially on Instagram and YouTube. To get the consumer’s attention nowadays, influencers have found the winning concept of packaging photos and videos. According to Vettese (2019), consumers tend to think that the traditional forms of marketing like TV commercials and magazine ads are unmodern and give less credibility and less trustworthy compared to the new way of marketing on social media. Vettese (2019) also discusses that the personal relationship between influencers and consumers is a much more critical aspect compared to before, and that is why people want to go online and get advice instead of reading it in magazines. The consumers value the option to ask questions to the influencers and get personal responses back (Vettese, 2019). Many of the well-established brands face a difficult time ahead with changing the way they advertise their products. Vettese (2019) also mentions that the global spending of influencer marketing has skyrocketed in recent years. For example, the beauty giant Estée Lauder has revealed that 75%

of its marketing budget is on influencers.

In the article by Gerdeman (2019), the mantra is “consumers listen to influencers, not company ads” which is central throughout the text. In Vettese’s (2019) study, 520 women were surveyed and said that Instagram was the social media they visited the most and used daily by 82% of the respondents. Vettese (2019) says that the interface and features of Instagram and the ability to get direct responses through comments, which makes the communication easy are parameters to being the most used social media. Many of the videos influencers

(21)

post are focusing on selling products which the surveyed women are aware of. There is a difference in about half of the respondents who would purchase from an influencer ad, and the other half is not too sure about it. The fact that it feels trustworthy from the influencer's perspective matters and also that the influencer is clear about whether they are being paid or not to promote the products. Vettese (2019) agrees with what Abidin (2015) discusses in her article, which is the importance of creating and maintaining a relationship between the influencer and its followers. Vettese (2019) says that consumers are drawn to influencers with the same skin tone or same skin sensitivities as themselves and are paying close attention to what decisions the influencers make and “the slightest slipup can jeopardize their sense of trust”. Another sign of legitimacy is if the influencer can give negative critique to products if they do not like it, which is appreciated by consumers.

This fosters the relationship and the reputation of the influencer if they can show honesty and transparency.

Vettese (2019) advises not only beauty companies, but to other industries as well, that they, for example, need to make active decisions in what social media channels suit their audience to keep it updated and modern. Just as other scholars (Abidin, 2015) state, the relationship, and trust towards influencers are one of the essential parts of your marketing strategy.

2.7 Spreadability of content online

Lastly, the interactive media and the world wide web has not only changed the way the audience behave and interact. It has also changed the producers' way of marketing and creating. It is now crucial to create texts that can be portable as audiences do not want to be stuck in one place. It has to be created so that it is easy for consumers to spread it and circulate it, and since the new media allows content to "stumbled upon" by new audiences, it should be relevant to multiple audiences (Jenkins, Ford, & Green, 2013).

Subsequently, when having spreadable content, which is essential for producers who want their content to grow and become viral, it is also essential to be able to measure the interaction and be aware of what happens with the content. As argued by Green and Jenkins (2011), a new consumer within the media landscape is a new producer; with the transforming audience, the way they are valued and understood is changing. The audience is said to have a new value where they no longer are only part of the end of the one-way communication chain. They still have an essential work as audiences and too not only be producers that they are now said to become. However, since the audience is aware that they now can be part of and contribute to meaningful participation, it is argued that there are still those who only listen and watch without actively participating. However, they do it differently with knowledge in the process.

With social media platforms and the world wide web, every move a user does is logged. Even if it is by an inactive member, the actions are generating data, which is useful when producers want to drive up popularity online. Subsequently, it is also a downside to the audience being active and having the opportunity to spread content. All are now distributors and appraisers, which means that when a large number of people are actively passing along content, the spreadability quickly gets large. However, for content that is not suitable to be spread, this is not a good thing. However, by using the technical affordances of spreadability, the social

(22)

networks, and the diverse motives that make people share content, the producers can make their content become more spreadable than others (Green & Jenkins, 2011).

Green and Jenkins (2011) use the example of the popular television show American Idol, and Britains got the talent to explain the purpose of using the spreadability online. They produce content that can be used on different media platforms and interests those who casually watch it by glancing as well as the dedicated watchers who glaze at the content. If the producers want their content to be relevant and spreadable, they must take into consideration that users only share content that they think is interesting enough and easy to share on their favourite platform (Green & Jenkins, 2011).

2.8 Summary of analytical framework

When companies such as the ones within the beauty market use influencers, they can commercialize on the intimate relationship that the influencer has with its audience. It is crucial for companies that want to reach young consumers to use social media platforms as a marketing tool, since influencers have changed the industry to become more intimate and the consumer wants a face-to-face experience (Abidin, 2018). Previous research shows that the relation between influencers and their audience can be seen as even more profound than the PSI since the influencer can interact with their followers on a more intimate level. As Abdin (2015) argues, with PSI, the performer is more controlled than influencers are. The features of Instagram allow influencers to interact and regularly share an aspect of their personal life (Abidin, 2015). Some people have a more significant impact on others and that even before the internet, audiences sought a relationship with performers (Abidin, 2018).

This study will focus on how the presence of an influencer has an impact on the interaction, which will be done by looking at companies' usage of influencers in their social media strategies. Research clearly shows that the audience members can feel a connection towards the performer and since Instagram allows features that contribute to an even more intimate relationship between the performer and the audience, all different sub- chapters in this study are linked to each other.

Since research has primarily been focused on the relationship between the audience and the performer it will be interesting to see how a third party, in this case, the beauty company, is achieving interaction on their account by using an influencer’s presence. To be able to answer the purpose of the study, relevant theories have been selected that will further help explain how the relationship between a performer and an audience can be used for marketing purposes on Instagram. This study will look further into this development of automation and see how it can be both used in a negative way and in a positive way. It depends on how the company works with it and how much they value their two-way conversation with their customers.

(23)

3. Method and material

This section will explain the reason for choosing a quantitative content analysis as the methodology approach for this study, the methodological problems that can be encountered by using that method, the population and sample, the collecting process, operalization of the chosen variables that were used for the code scheme and a discussion of the study’s relation to validity, reliability, generalizability and ethics.

3.1 Quantitative content analysis

To answer the research question, “What effect on the interaction on a company’s Instagram account can be found through the participation of an influencer in the posts?”, this study uses quantitative content analysis to find patterns in posts on Instagram. If you are interested in media content and want to achieve a study within that research field, to find out to what extent, what similarities or differences exist or to find patterns in media content, quantitative content analysis is the most suitable method to use (Karlsson & Johansson, 2019). Media content has been produced and analyzed in centuries. Karlsson and Johansson (2019) mention the book “Content analysis in communication research” by Bernard Berelson from 1952 as the starting point in the modern age for the method. Content analysis and its foundations are still relevant and up to date. For example, for this study, it is still meaningful to study and to quantify the manifest content, manifest content such as sentences, letters, and images that are published through various forms of media channels differ from the qualitative research methods, that usually put more weight into the context and the researcher’s interpretation of the content (Karlsson & Johansson, 2019). We seek to answer how companies are presenting influencers and what effect it might have on the interaction and not the feelings or thoughts behind the interaction. Since the study objects of this study are Instagram profiles, a quantitative method such as content analysis is the most suitable method to use. If the study’s direction were toward understanding the reasons why an influencer is being presented and how it stimulates the feelings, a qualitative method such as interviews would be more suitable.

According to Karlsson and Johansson (2019), to quantitatively media content is essential to translate the content’s linguistic and formal expressions into figures to describe various patterns and make general conclusions. How often and how much something is occurring, also called frequency, is a central part of the foundations of the method and can help answer the research question. Since the study is starting from a quantification of the manifest content, specific demands characterize the collection of the data; it should be collected systematically and objectively. Systematic collection of material means that all relevant material shall have an equal chance of being included in the study and also to be evaluated from the same definitions and models, which is the codebook and code instructions. Objectively in this context, is that the interpretation of the content shall be the same regardless of who is coding the data. One example of keeping a study objective is to conduct an inter-coder reliability test between the researchers before the actual coding. As discussed earlier, the intention is to eliminate the impact from the particular researcher in quantitative content analysis in contrast

(24)

to qualitative research where the researcher’s interpretation is central. Quantitative content analysis makes it possible to process a high volume of data, which enables one to find comparisons over time and between different mediums and, therefore, offer potential to generalize the results from the research to a larger population (Karlsson & Johansson, 2019).

3.2 Methodological problems

One problem with using quantitative content analysis as a method for this study is the risk of human error in the analysis process when coding the variables since it is two persons who will collect the data and not a computer. Karlsson and Johansson (2019) mention this problem as one of the critiques of the method. They argue that the main focus of the method is to get an (objective) interpretation, which can be problematic since people are interpreting texts differently (Karlsson & Johansson, 2019). That is why the variables of this study are required to be as descriptive as possible and easy to understand for the researchers.

3.3 Units of analysis, population and sample

The selection for this study is three different Instagram accounts that share a common interest in selling beauty products. The chosen accounts are caiacosmetics, indybeautysweden, and kicks.se.

The reason for choosing the three specific accounts was to strengthen the connection between the parasocial interaction and the influencer. We choose two big Swedish influencers that both have created brands within the beauty business, Bianca Ingrosso, that is the creator behind CAIA Cosmetics, and Therese Lindgren that is the creator behind Indy Beauty. These two units were chosen since they both have created brands and careers successfully outside of their influencer careers, which makes it interesting to study to see how much influence they have on their brands. By choosing them, we can see if their audience interacts more when they are present on the companies Instagram, which also makes it easier for us to see if there is a difference in interaction depending on if it is the influencer behind the brand that is present in the posts or if another person generates more interaction. Kicks.se was chosen as a brand to compare the brands created by influencers with since kicks are one of the Nordic countries’ biggest beauty brands (Kicks, 2020) it is suitable to compare with. The brand is highly known and can generate interaction without using influencers.

The accounts were created at different times and, therefore, do not share the same amount of posts in total.

Kicks were created in August 2015 with up to this day, 2738 posts, CAIA Cosmetics was created in October 2018 with up to this day, 740 posts, and Indy Beauty was created in January 2018 with up to this day 304 posts in total. The object of analysis will be a hundred posts posted from each Instagram account, and the sample size will be three hundred posts in total. In order to gather information about the interaction in posts, a selection has been made to start analyzing the posts. The selection made is a total of the most current Instagram posts.

(25)

According to Karlsson and Johansson (2019), when deciding what selection suits your study the most, you want to aim at different forms of a random sample. Even if quantitative content analysis is used to draw general conclusions from the population, it is not always possible and especially in digital mediums. This study will have a selection of the most current sample since the posts we will collect are posted in a chronological timeline, starting from six days old until the oldest of the one hundred. We start the collection with posts that have the minimum of being six days old, since we want posts where the interaction can be seen as non-active, meaning that the posts do not generate as much interaction as when they were first posted (Green & Jenkins, 2011). Using this type of sample will not create as good results in the generalizability of the study. Yet, it will represent the beauty companies’ way of marketing in today’s era.

Instagram is a social media platform where users can create accounts, follow each other, upload images or videos, go live, interact by commenting, liking, and sharing posts (Instagram, 2020). It is a platform that allows for interaction in different ways, which are the reason why we have chosen to use it as our object of analysis in this study. Since we want to explore the amount of interaction an influencer can generate, Instagram is the most appropriate platform as it allows us to measure interaction. The features of Instagram makes it easy to visualize the amount of both likes, comments, and views on posts. There are several options for posts on Instagram. Picture is a still photograph with no movement and video is a short film uptil 60 seconds. IGTV, short for Instagram TV, is long-form, immersive video. It is not limited to one minute, and takes up your entire screen (Instagram, 2020). In an early stage of the study, a comparison between the companies’ different accounts in social network services was made, and it was clear that Instagram was the platform that each account was most active on.

3.4 Data collection & procedure

To collect the data from Instagram, we had to log in to be able to view all the accounts. We used our private profiles, and since the Instagram accounts look the same way for everyone who is logged in, it would not be a problem as if it might have differed depending on what account you were logged into. One problem we encountered before the actual data collection was the fact that depending on what device you were using, the view and interface of Instagram differed. The differences laid in not being able to see the links to the products in the posts when using a computer compared to using a smartphone. One of our variables was to see if links were present in posts, but after that encounter, we decided to remove the variable from our code scheme since it would not help us answer our research question. As the collection of the data for this study was made manually, we decided to collect it from a computer to make sure to get the same features included in all the posts and for the appropriateness of the study. The collection was made by two individuals together, one that analyzed the posts and one that coded the posts to figures into the statistics software SPSS Statistics. This was to get as accurate information as we could and prevent one person from dealing with 300 posts by themselves at the same time as coding everything.

(26)

When collecting the data such as the number of likes, the number of comments, what type of post, and what the post included, it was manually coded into an SPSS Statistics file. The collection of data was thoughtfully done on the same day, which was Wednesday 29/4 2020. All the posts had a separate URL and documented to an Excel file with a given Post ID that furthermore was used in the SPSS file to keep the posts apart. The posts collected from CAIA Cosmetics were from April 23rd until March 4th. The posts from Indy Beauty were from April 21st until November 26th and the posts from Kicks were from April 23rd until March 11th.

The accounts were not updated during the data collection to keep them static when calculating the number of likes and comments. The date for each post was also collected and documented in the SPSS file. After that, the results were analyzed through frequency tables in SPSS Statistics.

During the collection of data, we noticed that Indy Beauty was not as frequently active on posting content to their Instagram compared to the other two accounts. Because of that, an issue with the content was inevitable since most of the analyzed posts belong to a Christmas campaign launched at the beginning of December 2019 and ended on Christmas Eve. The empirical material of this study includes data from this specific time and caused an effect on the results in this case. The campaign only included posts of products and one of the requirements to win the Christmas gift each day was to comment on the posts, which led to a higher number of comments of posts without an influencer present.

3.5 Operalization of variables

This study focuses on the social media platform Instagram and uses the features that the platform allows to justify how the research question will be answered. During the first stages of the pilot study, several of the social media features that Instagram allows was identified that later on could be operationalized into variables in the coding scheme of the quantitative content analysis to help answer the research questions of this study (Karlsson & Johansson, 2019). The variables that were created were identified as different stages of importance in relation to how they helped answer our research question. The posts were divided into different important parts: the content of the posts, the text of the post and the interaction on the post.

A generalization was also made on the different content that was identified in the posts, which was a major part of the operalization. To identify if there was a difference in interaction depending on what the posts contained a generalization of it had to be made. Such as type of post, If the post contains a product or if it is a post of an influencer, contains another person than an influencer or a tag of a company or person.

Some variables were created for the text of the post to help identify if there was a hashtag of an influencer or another company and to see if the company had shared a URL.

Furthermore, to separate the data that was collected from our units of analysis and to more easily access it in SPSS Statistics two identification variables were created, Post ID and Source. As each post has been collected it gets a Post ID to help separate all 300 posts and make them individual. The variable Source refers to which

(27)

of the beauty companies Instagram accounts the post was collected from, the values of this variable is the three units of analysis caiacosmetics, indybeauty and kicks.se as values (Karlsson & Johansson, 2019).

The variables that were chosen concerning the phenomenon influencer are related to the influencer that has created the beauty companies that this study has as study objects. Since influencer can be displayed and used in different ways we wanted to separate other influencers from the influencers that have created the companies. The reason is to see if there is any difference in interaction depending on that as well as the variables created for products and other companies to see if they earn more interaction by not having any persons or influencers in the content. The interaction was based on the scale variables amount of likes and amount of comments to see how much difference there is depending on what the posts contain. All chosen variables are important for this study since they help answer the research question that separates this study from the earlier research that has been done on the subject. We can see that earlier research focuses on the relationship between mainstream celebrities and the audience (Horton & Wohl, 1956) and influencer and their intimate relation with their followers that is created on their accounts (Abidin, 2018). However, this study focuses on what the relation between influencers that have generated an audience from different mediums and how they generate interaction on a third parties Instagram account, in this case, we focus on the beauty brand industry. As we can see the variables that concern interaction features on social media platforms have not had much attention in studies related to the PSI between the audience and the performer.

3.6 Variables

V1. Post ID

This variable will help us keep the posts organized. The number is only an identity and will not be relevant for the analysis but rather help us separate all the posts and keep them individual. It is a variable on the scale level ratio.

V2. Source

1. CAIA Cosmetics 2. Indy Beauty 3. Kicks

Which of the accounts is the post posted on?

This variable will help us to see the importance of what company has the most interaction and connects to the theory of PSI (Horton & Wohl, 1956). It is a nominal variable.

(28)

V.3 Type of post 1. Picture 2. Video 3. IGTV

What type of post is posted?

This is a nominal variable and it is connected to the theory of PSI (Horton & Wohl, 1956) since it is important to know what type of post is most likely to engage followers. The audience might feel more connected to the influencer and the company if they see the influencer talking about the promoting products in a video rather than on a photograph.

V4. Date of the post

Which day is the content posted on?

This is a ratio scale variable and it will help us see how active the companies are with posting on Instagram.

The coding will be conducted April 29th with a systematic sample of all the latest 100 posts from at least six days old posts.

V5. Number of likes on the post How many likes does the post have?

This is a ratio scale variable and it is important for the study since it can help us answer the research question.

It relates to previous research in PSI (Abidin, 2015) and research in the development of an active audience (Olsson, 2013).

V6. Number of comments on the post How many comments does it have?

This is a ratio scale variable that is important for the study since it can help us answer the research question.

It relates to previous research in PSI (Abidin, 2015) and research in the development of an active audience (Olsson, 2013).

V7. Post of influencer 0. No

1. Yes

Is the post showing the influencer?

This is a nominal variable and one of the most important variables for this study since the research question is built around the phenomenon influencers. Horton and Wohl’s (1956) theory of PSI with a celebrity in traditional mass media can be related to this new type of relationship between influencers and the audience.

“This is my favorite beauty product… Do you like it too?”