MASTER THESIS
“Analyzing brand personality communication and its impact on consumer engagement” An online investigation on Twitter and Instagram
Author: Laima Hamel-Jolette (10826858) MSc. in Business Administration, Marketing Track University of Amsterdam, Faculty of Economics and Business
June 29th, 2015
ABSTRACT
The aim of this study is to explore the use of brand personality by brands on social media and its impact on consumer engagement. Brand personality has been extensively researched in an offline context and many authors already argued that brand’s personality plays an important role in generating consumer engagement but no significant research has been done online. This study uses a qualitative and quantitative approach to investigate the impact of the communication of brand personality online and which personalities are used by brands. Findings show that not all personalities can be expressed in an online context. The most frequently employed brand personalities are the ones leading to a higher level of consumer engagement online. The two platforms are used differently by the brands to express their personalities. Finally, the explanatory research suggests that the use of brand personality online is beneficial for brands to reach a higher level of consumer engagement. Therefore, the usage of brand personality could help marketers achieve higher consumer engagement online using some specific personalities as ruggedness, activity and aggressiveness in their communication.
Statement of Originality
This document is written by Student Laima Hamel-Jolette who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in
ACKNOWLEDGEMENTS
I would like to express my gratitude towards my supervisor Dr. Meg Lee Hsin-Hsuan for her guidance through out the whole writing process, her useful feedback and her help with developing on an interesting topic.
A very special thank you is an order to Arturo Manzaneda for helping me developing a web-scraper to collect my data specifically for that research and showing me that coding can be accessible to anyone. I would also like to say thank you to my dear friend Susann Krüger for helping me with SPSS and making the software less complicated than it actually is.
I would also like to thank my parents, my brothers and sister for supporting me financially and mentally through my whole Master in Business Administration. Finally, a special thank you to my friend Lizbeth Gutierrez for being my thesis partner, my moral support and encouraging me in harder times.
TABLE OF CONTENT
1. INTRODUCTION... 6
2. LITERATURE REVIEW ... 11
Social media ... 12
Consumer engagement ... 14
Types of Brand Personality ... 17
Comparison of Brand personality scales ... 22
Conclusion ... 25 3. METHODOLOGY ... 27 Research Design... 27 Sample... 28 Platforms ... 31 Independent Variables ... 32 Personalities ... 32 Dependant variables ... 39 Engagement... 39
Coder training and reliability ... 41
Analysis methods ... 42
Control variables ... 42
4. RESULTS ... 42
Brand personality communication ... 45
Qualitative results ... 48
Twitter and Instagram ... 51
Consumer engagement ... 52
Analysis of covariance of brand personality on engagement (ANCOVA)- Twitter ... 53
Analysis of covariance of brand personality on engagement (ANCOVA)- Instagram ... 56
Personalities and engagement ... 59
5. DISCUSSION ... 63
Managerial implications... 67
6. LIMITATIONS AND SUGGESTIONS FOR FUTURE RESEARCHES ... 69
7. CONCLUSION ... 71
REFERENCES ... 73
LIST OF TABLES
Table 1: Resemblance of Brand personality dimensions within the 4 scales selected ... 23
Table 2: Brands overview ... 29
Table 3: Examples of coded types of brand personality ... 34
Table 4: No brand personality coding scheme ... 39
Table 5: Engagement Coding scheme ... 41
Table 6: Descriptives of the brands online behavior ... 43
Table 7: Descriptives of brands on Twitter ... 44
Table 8: Descriptives of brands on Instagram ... 44
Table 9: Brands and Primary Category ... 46
Table 10: Frequencies of brand personalities on Instagram ... 47
Table 11: Frequencies of brand personalities on Twitter ... 47
Table 12: Communication of brand personality by brands on Twitter and Instagram ... 52
Table 13 Descriptives for retweets and Favorites under brand personality ... 53
Table 14: ANCOVA results for retweets ... 54
Table 15: ANCOVA results for favorites ... 55
Table 16: Estimated marginal means for favorites ... 56
Table 17: Estimated marginal means for retweets ... 56
Table 18 Descriptives of likes and comments under brand personality ... 56
Table 19: ANCOVA results for likes (1000) ... 57
Table 20: ANCOVA results for comments ... 58
Table 21: Estimated means for likes (1000) ... 59
Table 22: Estimated means for comments ... 59
Table 23: Descriptives for brand personalities on Instagram ... 60
Table 24: ANOVA for difference of personalities on Instagram ... 61
Table 25: Descriptives of brand personalities for Twitter ... 62
1. INTRODUCTION
Over the past few years, the media world changed introducing social media for companies as a marketing tool. Literature shows that marketers have been embracing social media marketing for different objectives as branding, customer engagement, customer service, promotion and research (Ashley and Tuten, 2015). Social media platforms create an additional marketing communications channel for companies to promote their brands and brand personality in addition to traditional marketing tools (Michaelidou, Siamagka, Christodoulides, 2011; Burton and Soboleva, 2011). Multiple aspects of social media are more attractive than traditional marketing channels because of its free, personal and social nature characteristics (Coursaris, van Osch, Balogh, 2013; Coursaris, van Osch and Brooks 2014). Counting 150 Social Networking Sites (SNS) in 2009, they are at the top of many business executives’ agenda today (Kaplan and Heanlein, 2010; Michealidou et al,. 2011; Nielsen Wire, 2010). Businesses have multiple options to work with; Pinterest, Facebook, Twitter, Instagram, Google+, LinkedIn, Youtube1. Social media platforms are a growing way to communicate with their target audience segments and definitely a mainstay as a communication channels for marketers to use (Ashley and Tuten, 2015).
Brands have now a very important place on those platforms because of their number of followers and the amount of money firms invests into their social media presence. Although, social media platforms were initially created to link people together, brands
present and participate (Fournier and Avery, 2001; Michealidou et al,. 2011). As an example of the brands on social media, Nike has 15,7 millions followers on Instagram (Instagram, 2015) and 4,9 millions on Twitter (Twitter, 2015)2. With these large amounts of followers, businesses need to make a profitable use of those applications to obtain the best engagement from their consumers and interact with their consumers (Li and Li, 2014; Kaplan and Haenlein, 2010). Enhancing trustworthiness, brand attitude and customer commitment are motives for companies to use social media (Dijkmans, Kerhof, Beukeboom, 2015). Mostly, companies use social media to reach out to target groups and create consumer engagement with the help of messages, images and videos to expose their identity.
Being one of the core dimensions of brand identity and brand image, brand personality can be use to project the actual and aspirational self-images of the brand (De Chernatony, 1999; Freling and Forbes 2005). Aaker (1997, p. 347) defines the concept of brand personality as “the set of human characteristics associated with a brand”. In the literature, brand personality has been extensively researched in an offline context but scholars have been given little attention to its use in an online context and specifically the use of brand personality on social media platforms. Because of the importance of social media and presence of brands on them, it is important to understand the impact brand personality communication online. Furthermore, there is a general agreement that brands with personalities are associated with positive consequences (Freling and Forbes, 2005) as consumer preference, usage and levels of trust and loyalty (Sirgy 1982; Okazaki 2006;
2 For more information, http://instagram.com/nike on January 29th and retrieved from https://twitter.com/Nike on January 29th 2015
Aaker 1997). Therefore, it is necessary for marketers to know how to trigger consumer’s attention and engage them on social media by using brand personality. Companies use brand personality to distinguish their brands from competitors to perhaps achieve growth or gain customer loyalty (Ghodeswar, 2008). Keeping in mind those advantages for marketers, the use of brand personality is a must to create a strong brand identity and therefore it is the basis of a long-term relationship with a brand (Hawkins et al, 2001; Okazaki 2006). Finally, it is pressing to understand how brand personality can be expressed and communicated online because of the popularity of the tool and the opportunities brands can benefit from it.
Different social media platforms offers to brands to the possibility to reach out to different target audience and express themselves in different ways. They can use them differently to gain engagement from consumers. Consequently, challenges arise for branding in online environments to integrate and brand their marketing communications across different channels (Rowley, 2004). Brands are now using text, images and videos to display their brand personality that existing users association with their organizations on different platforms e.g. Twitter and Instagram (Rowley, 2004). It is the scope and the functionality of the currently existing rich and diverse ecology of social media site that makes it interesting to analyze two different platforms (Keitzmann, Hermkens, McCarthy, Silvestre, 2011). Therefore, the impact of the use of brand personality on online platforms has not been the center of a research yet and could provide insights on
personality and the way its communicated online. Specifically, this research will focus on the communication of brands from the fashion industry on Twitter and Instagram and the way those brands express brand personality in images or text and its impact on the type and level of engagement on consumers.
To begin to address this gap in the literature, an explanatory research of the use of brand personality through out messages and images on two different social platforms was conducted. Using a sample of global brands highly ranked in brand equity and brands recognized for excellence in social media engagement, brand personality messages were analyze to answer the following the following question:
How can brand personality be communicated online by brands and what are the consequences on engagement?
This explanatory qualitative and quantitative research will be focusing on answering those two more specific questions on the brands selected in the fashion industry:
RQ1: How can brand personality be communicated online? a. What are the different types?
b. In which format does brand personality appear online?
RQ2: How different personality will lead to different level of engagement and to different responses?
The research paper finds its relevance in the relation of concept brand personality and its way of being communicated on online platforms precisely because this link has not been
established yet in any previous literature perhaps brand personality has been studied in an offline context.
The brand personality subject will be approach with a different angle than in the previous literature; its communication on social media, which will explore new academic fields. Being an actual subject and also a rising topic in marketing, the research of the use of brand personality across social media platforms needs to be elaborated and will open path for futures researches. The aim of this study is to research brand personality online and to research how the usage of different personality or types of personality will lead to different level of engagement from consumers analyzing their behaviours online and different types of responses from consumers. The engagement for this study will be analyze with behavior of the followers only. Finally, this paper will also do a platform comparison of the use of brand personality between Instagram and Twitter.
The managerial contribution remains in the impact of the communication of brand personality for social media marketing strategy. New positions as content manager have been developed in the last few years and brand personality is an aspect of that content shared on social media. Brands need to know which brand personality strategy to use to obtain more engagement from consumers or which personality stress the highest reactions in an online environment because the brand personality scales weren’t specifically created for an Internet or an online context (Müller and Chandon, 2003). Thus, this study
since the literature already proved that social media is a must stay in marketing communication (Ashley and Tuten, 2015). Therefore, companies need to have a clear strategy and plan to develop their brand equity and project the right brand personalities to consumers to create a bond with them because researches showed the stronger the congruity between them, the brand and the brand personality, the greater the preference for the brand (Okazaki 2006; belk 1988; Hong and Zinkhan 1995; Malhotra, 1981).
In the following sections, this paper will consist of first a literature review summarizing and investigating the findings made on the topics of social media, consumer engagement online and brand personality. Following the literature review, the research methodology is then explained in detail and will be followed by the research results. The conclusion will consist of the implications of the findings and managerial suggestions. Finally, the paper will conclude with the limitations of this research and suggestions for future research.
2. LITERATURE REVIEW
This chapter will review the relevant literature with regard on academic research focusing on social media, consumer engagement and on the different types and scales of brand personality. Finally, brand personality scales will be compared and specific brand personality types will be selected to perform the explanatory research with brands from the fashion industry.
Social media
Social media categorize by marketers as the number one branding channel can be defined as “a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0, and that allow the creation and exchange of User Generated Content” (Kaplan and Haenlein, 2009, p. 61; emarketer, 2013). The social dimension of social media give the opportunity for companies to use the concept of self-presentation which states that in any type of social interaction people have the desire to control the impression other people form of them (Goffman 1959; Kaplan and Haenlein, 2010). This concept is driven by the desire to create an ideal image that is consistent with their personal identity. This concept can be related to how a brand wants to control the projection of their brand identity and portrait their ideal self through the content they share on social media. As an example, sport apparel brand will share that they are courageous, bold, true champions and perseverant as their ideal personality. Finally, information about a brand, as their brand personality, needs to be significant to the consumer if marketers want the consumer to engage with their brand. (Schmitt 2012; Ashley and Tuten, 2015).
Following on the image creation concept, social media is also the channel for business to express their personality with sharing content in different ways with the creation of images or text to engage their customer in today’s fragmented and visual-driven media environments where both brands and individuals are in. Social media is a powerful tool,
2015). In addition, marketers will benefit from understanding whether their brand personality applies to branding in social media and which brand personality strategies are the most effective in achieving consumer engagement (Ashley and Tuten, 2015). Consumer engagement is defined by Hollebeek (2011,p.6) as ‘ the level of a customer’s motivational, brand related and context-dependant stated of mind characterized by specific levels of cognitive, emotional and behavioural activity in brand interactions’. Therefore, social media allows companies to engage with their customer at a relatively lower cost and more efficiently than to traditional media (Kaplan and Heanlein, 2010). Finally, multiples platforms offers different avenues for brands but this research will focus on the comparison of two platforms online because of the way companies express themselves differently on both.
Since there are hundreds of social media platforms and new ones appearing on the horizon every day, choosing the right medium for your brand depends on the target group and the message to be communicated (Kaplan and Heanlein, 2010). For this specific research two different channels were selected, Twitter and Instagram were selected because companies might use the two platforms differently and also because of their popularity. Both platforms are in the top 10 most used platforms at the moment3 (Adweek, 2015). Twitter allows expressing yourself with 140 characters, which creates very short intense messages that are mostly real-time status updates (Keitzmann, hermkens, McCarthy, Silvestre 2011). It is undoubtedly the most popular member of the micro-blogs (Kaplan and Heanlein 2011) allowing companies to also share images but
companies are mostly using it for customer service, invitation, keeping up-to-dates customers. Instagram, a newer platform in comparison of Twitter, allows companies to share images that represent them and to inspire their followers or engage them (Instagram blog, 2015).
Consumer engagement
Having a distinctive brand personality has already proven advantages for long term relationship with consumers (Kim, Han and Park. 2001) and by having a distinctive personality, consumers can see themselves through the brand and use it as representing themselves and expressing their own values, which bring consumers to feel more attached to a brand. In addition, previous researches suggest that the greater the congruity between the human characteristics of how an individual’s actual or idea self and the characteristics describe by a brand will lead to a greater preference for that particular brand (Sirgy 1982; Aaker 1997). Therefore, the way different brand personality types influence consumer preference needs to be understood better (Aaker 1997). Finally, marketers are seeking more opportunities to expose their target audience to the brand message as reported by the 2013 Social Media Industry Report and 78% of marketers report using social media to enhance their customer engagement (Ashley and Tuten, 2015).
Customer/Consumer engagement has been defined by relatively few authors in the marketing academic literature but it provides a platform for a definition of “consumer
connection with the organization’s offerings and activities initiated by either the customer or organization. While Hollebeek (2011) views “customer brand engagement” as the level of a customer’s motivational, brand related and context-dependent state of mind characterized by specific level of cognitive, emotional and behavioural interactions (Brodie et al. 2011). Furthermore, Ashley and Tuten (2015) state that engaged consumers participate and share. As an explanatory study, this research will use the behavioural interactions mostly used to evaluate engagement on social media.
Consumer engagement (CE) is an important concept for online brand communities but lets first look at Coursaris, van Osch and Balogh (2013) who dived into the assessing typologies for online messages on Facebook to understand the relevant messaging strategies to attempt strategic CE. The traditional marketing theories are not able to classify those messages and are not relevant for social media (Coursaris et al. 2013). To asses this gap, the authors categorized the messages in seven categories: “Brand Awareness, Corporate Social Responsibility, Costumer Engagement, Product Awareness, Promotional and Seasonal” on Facebook and where each of these categories triggered different level of engagement where customer engagement is relevant for this research (Coursaris et al. 2013, p. 10). Furthermore, in social media literature, the terms engage and engagement are used to describe the nature of participants in special interactions or interactive experiences (Brodie et al., 2011). Neff (2007) sees consumer engagement as one of the first drivers for sales growth and Voyles (2007) sees engagement as a growth for profitability.
Furthermore, customer engagement is also described as a “psychological process” including cognitive and emotional aspects (Bowden, 2009). Finally, Brodie, Llic, Juric and Hollebeek (2011) propose a conceptual model to illustrate the consumer engagement, which is divided in 3 categories: emotional, cognitive and behavioural and perhaps theses three dimensions are influence by the online community of a brand. Consumer brand engagement is demonstrated by involvement in Hollebeek (2011) conceptual model also show a positive impact on the relationship quality. Relationship quality is defined by three variables: trust, commitment and customer satisfaction, which will lead to customer loyalty (Hollebeek, 2011). Since, customer brand engagement is one of the most powerful predictor of customer loyalty it is interesting to research which brand personality lead to the higher behavioural engagement online (Hollebeek 2011, Appelbaum 2001)
For this explanatory study, the behavioural engagement category is most interesting because we want to see the impact of the communication of brand personality on behaviours of consumers and it seems the most appropriate choice for a first study. Future researches could go further by analyzing the cognitive and emotional responses on brand personality messages. In this study, reaching a higher level of engagement will be described as retweeting a post from a brand, liking the image or commenting on a post where brand personality is used. Finally, based on the brand personality literature, the research will explore if the use of brand personality will effectively influence the online brand engagement and specifically which personality lead to a higher engagement.
Types of Brand Personality
Formally defined, “brand is a name, term, sign, symbol, or design, or a combination of them which is intended to identify the goods and services of one seller or group of sellers and to differentiate them from those of competitors” (Kottler 1994. P 442; Keller 1993). Focusing on brand image for the purpose of the research, brand image is one if the two components of brand equity which is the “perception about a brand as reflected by the brand associations held in consumer memory” (Keller 1993). The brand personality is one of the dimensions of brand image (Aaker, 1991, Kapferer 1995; Plummer 2000). Plummer (2000) considers that brand image is composed of 3 major dimensions: product attributes, consumer benefits and brand personality. The quality of the brand image depends of the favourability consumers have of the brand’s associations as their uniqueness, their strength, cohesion and level of abstraction (Keller 1993). Kapferer (1995) presents the idea that when a brand communicates, it acquires a character as the same way every individual has its own personality (Koebel and Ladwein 1999).
Kim, Han and Park (2001) studied the effect of brand personality and brand identification on brand loyalty. It was proved in that study that developing brand identification through the brand personality encourage consumers for a long-term relationship with the brand. In addition, Van Rekom, Jacobs and Verlegh (2006) mention that a strong brand personality is also invaluable to build brand equity. Based on previous literature, it was confirmed that the higher the distinctiveness and the self-expressive value, the more the brand personality will be appealing to consumers for brand identification (Kim et al. 2001). To emphasize the importance of using brand personality in communication, brand
personality will directly affect the positive word-of-mouth, which will have an indirect affect on brand loyalty (Kim et al. 2001). Finally, Plummer (2000) argues that brand personality could be a crucial element to understand consumer’s brand choice. This research speculate that some of the brand personality may lead to a higher level of engagement as per example being sophisticated showing charm and upper class could trigger more like from consumers than competence as being reliable and intelligent (Aaker, 1997).
Building on previous literature, Aaker (1997) defines the concept of brand personality as “the set of human characteristics associated with a brand”. This definition is often use in the literature and brand personality has already been a subject largely exploited and studied by multiple researchers but little attention was giving to develop a systematically a reliable, valid and generalizable scale to measure brand personality before Aaker (1997). Since then, Multiples authors have add new categories to Aaker’s work and presented new personalities.
Firstly, Aaker (1997), the most cited author for brand personality, created a theoretical Framework of brand personality dimensions “the big five”, which was reused in multiple other studies but not within any relevant social media studies. Aaker (1997) elaborated the “Big Five” based on the creation of a list of 309 personality traits. The results of the study show a five-dimension model composed of 44 items (Appendix I). The “Big Five”
1. Sincerity: Domestic, honest, wholesome, cheerful, down-to-earth, family-oriented, small-town, sincere, real, original, cheerful, sentimental, friendly;
2. Excitement: daring, trendy, exciting, spirited, imaginative, up-to-date, cool, young, imaginative, unique, independent, contemporary;
3. Competence: reliable, intelligent, successful, hard-working, secure, technical, corporate, successful, leader confident;
4. Sophistication: upper class, charming, glamorous, good looking, feminine, smooth; and
5. Ruggedness: though, outdoorsy, masculine, western, rugged.
Most criticism about the scale proposed by Aaker (1997) focus on the fact that the scale encircle some traits that have no equivalent in terms of human personality as sophistication and ruggedness and some traits who are more about psychosocial values as being seen outdoorsy, up-to-date or upper class (Louis and Lombart, 2010). Because of those critics, it leads us to present and evaluate other type of brand personality scale to find which personality can be expressed online by brands and which one should be included in the analysis. Perhaps, certain brand personality type can lead to higher level of engagement from consumers as sophistication for luxury fashion brands or ruggedness for a sport apparel brand (Kim, 2000)
Koebel and Ladwein (1999) builded on Aaker’s scale adapting it to a more French context see appendix II. Koebel and Ladwein (1999) mention that as the same way every individual has its very own personality, brands also have a unique personality. Every
brand can be described with specific traits and which defines their brand identity. Following on that way of thinking and including a French context, their research result was a six-dimension scale composed of 30 items:
1. Domination: unique, independent, self-confident, contemporary, leader; 2. Ability: reliable, robust, technical, safe;
3. Conscientious: realistic, hard-working, organized, intelligent, honest; 4. Masculine: masculine, virile, rough;
5. Expensive: bold, imaginative, ardent, trendy, winning, cheerful; and
6. Seduction: bewitching, attractive, good-looking, sentimental, feminine, exciting, distinguished.
Following on adapting Aaker’s scale to the French context, Ferrandi et al. (1999) in developed a five-factor model with 33 items:
1. Sincerity: sincere, honest, natural, trustworthy, safe, conscientious, healthy, realistic, intelligent, authentic;
2. Excitement: trendy, modern, young, imaginative, up-to-date, positive attitude, bold;
3. Sophistication: charming, attractive, elegant, gentle, feminine, sentimental, fascinating;
4. Robustness: robust, sturdy, charismatic, western; and 5. Liking: friendly, cheerful, family, nice, provincial.
Geuens, Weijters and De Wulf (2009) developed a new brand personality measure build on Aaker’s work and multiple other authors but with only using personality items in Appendix II. Geuens et al. (2009) also put forward a new definition of brand personality define by Azoulay and Kapferer (2003, p.151): “brand personality is the set of human personality traits that are both applicable to and relevant for brands”. Gueunes et al (2009) results present a new brand personality measure with five-factor and 12 items:
1. Responsibility: down to earth, stable, responsible; 2. Activity: active, dynamic, innovative;
3. Aggressiveness: aggressive, bold; 4. Simplicity: ordinary, simple; and 5. Emotionality: romantic, sentimental.
Geuens et al. (2009) criticize Aaker’s work because it includes some non-personality items in its framework. The new scale built with only personality traits is very important because authors defend that consumers use brands with a strong personality to build relationship with and also show their own personality with it (Fournier, 1998). Geuens et al (2009) also mentions that the five-factors defined by Aaker’s scale is not cross-cultural and defend their new scale presented above is. Using Geuens et al. (2009) scale is interesting and highly relevant for this research because multinational brands are used and they are active in a cross-cultural context leading to a more global scale and solving the problem of country-specific brand personality scale. The new scale can be therefore used for any product category and any country. Moreover, global companies can use the scale to determine to what degree their brand has a true global personality (Geuens et al. 2009).
Comparison of Brand personality scales
A table is presented to show the attributes of the brand personality identified by those multiple authors. Drawing from the table, we can notice that some attributes and types between the different scales are similar which will lead to combining them in the analysis of the data and not using the four scales to evaluate the communication of brand personality online.
Table 1: Resemblance of Brand personality dimensions within the 4 scales selected
Author (s) Aaker’s “Big
Five” 1997 Ferrandi and Valette-Florence, 1999
Koebel and Ladwein 1999
Geuens, Weijters and De Wulf, 2009
Country USA France France Cross cultural
Competence Competence: reliable, intelligent, successful; Conscientious: realistic, hard-working, organized, intelligent, honest; Sophisticated Sophistication: upper class, charming; Sophistication: charming, attractive, elegant, gentle, feminine, sentimental, fascinating; Seduction: bewitching, attractive, good-looking, sentimental, feminine, exciting, Emotionality: romantic, sentimental. Excitement Excitement: daring, spirited, imaginative, up-to-date; Excitement: trendy, modern, young, imaginative, up-to-date, positive attitude, bold; Sincerity Sincerity: honest, wholesome, cheerful, down-to-earth; and, Liking: friendly, cheerful, family, nice, provincial; and, Responsibility: down to earth, stable, responsible; Ruggedness Ruggedness: though, outdoorsy. Robustness: robust, sturdy, charismatic, western. Masculine: masculine, virile, rough; Aggressiveness: aggressive, bold; Other dimensions Domination: unique, independent, self-confident, contemporary, leader; and Expensive: bold, imaginative, ardent, trendy, winning, cheerful; and Simplicity: ordinary, simple; and Activity: active, dynamic, innovative.
Looking back at the four scales presented in table 1, some personality traits are present in more than one scale and some presents more human traits easier defined in messages or images as Aaker (1997)’s scale. In order to build a new scale for online social media message or images more than one scale will be used. Because, most of the global brands are international companies, they it is important to have scales that can use cultural difference and across product categories in Geuens et al. (2009)’s scale. Ferrandi et al. (1999)’s and Koebel and Ladwein’s scale will be left aside because both has high similarity with Aaker (1997)’s scale and to avoid redundancy in the analysis.
Finally, limited amount of studies have shown how brand personality affects important marketing variables such as engagement on social media and which marketing strategy of brand personality is more effective how to communicate the brand personality online. Using the Aaker (1997)’s scale and Geuens et al (2009)’s scale together to determine which personality creates the most engagement from followers on social media, which will lead to the creation of a new scale for social media purposes. Certain personality types of two scales are overlapping which we will lead us to compare together some personality for a better analysis. The brand personality trait Ruggedness is consistently over the 4 scales expressed in a different name, it will be compare for the analysis using Items from the Aaker (1997)’s and Geunes et al. (2009)’s scale. The sophistication personality trait from Aaker (1997) is also very similar in the other 3 scales.
Prior preliminary research showed that some personality can expressed online, the author Okazaki (2006) with a qualitative study examined the “intended” brand personality online and found some personalities were better and most expressed online by global brands as sophistication, competence and excitement from Aaker (1997)’s dimensions. Following on that idea, the current study will also identify the most recurrent expressed personalities used on social media by fashion global brands and their efficiency on engagement of consumers measured as the most favourable responses, which will be determines as the number of reactions and likes; behavioral reactions. Certain personalities might lead to a higher level of engagement as emotionality from Geuens et al. (2009) when romantic and sentimental traits are expressed or aggressiveness where boldness or being aggressive is expressed creating higher reactions from customers.
Conclusion
Marketers will benefit from understanding whether the usage of brand personality common in traditional marketing is also applying to social media. Therefore, this paper will argue that some personalities are easier communicated online and particularly easier expressed with the use of an image rather than a text or vice versa and that some types of personality are creating and leading to more engagement from the consumers and activating more responses from the consumers. Consumer engagement is crucial for brands and there are different ways to evaluate them and interpret them online. It is stressed that on online marketing certain responses are more valuable and measurable for companies and brands, where content managers counts the amount of likes and comments created by their online actions. Therefore, it is important to identify which personalities
are more likely to result in the types of actions that can traced by managers; likes, comments, favorites, retweets. Perhaps some personality cannot be expressed with words and images. Finally, as said before we propose that use of brand personality in general will lead to more favourable responses and a higher consumer engagement.
3. METHODOLOGY
Research Design
In this study, thematic coding is adopted to analyze the data based on 2 different social media platforms. In particular, the manual coding with an interpretive approach is use to code the data crawled based on the brand personality categories elaborated above. The brands’ posts will be looked at individually and the responses from each posts will be examined. An example of images and text below is giving to show how the coding was done.
The social media content analysis will be focus on specific brands from the fashion industry present on both on Twitter and Instagram selected from the list of 100 brands of 2014 by Interbrand’s Best Global Brands evaluation and from Interbrand’s Best Retail Brands. Those global brands selected are consequently doing major social media efforts to express their brand personality with images and text and engage their consumers with a lot of activity online. Furthermore, the fashion industry has been selected because they are very active on social media and fashion brand online has been studied before (Smith, Fischer, Yongjian, 2012). Consequently, constant new collection and season change can also explain their high presence on social media and their high level of content creation to inspire their followers to buy more (Mohr, 2013) Their presence in the Global brand Interbrand top 100 is a proof that they are doing major branding effort and therefore are all very active on social media (Interbrand, 2014).
Sample
The sample for this research consists of 9 brands in different categories of the fashion industry as Haute Couture, Sports apparel and Prêt-à-porter to have a global image of the fashion industry and help draw conclusions comparing to other research using fashion brands before (Morh, 2013; Kim and Ko 2012; 2010). Moreover, multiple categories lead to more various brand personalities between the brands. Based on those criteria Louis Vuitton, Burberry, Gucci, Nike, Adidas, Zara, H&M, GAP and Victoria Secret were selected. Looking at similar researches small samples were used for qualitative social media research as seen in Ashley and Tuten (2015) with 28 brands and 10 brands in Lee, Van Dolen, Kolk (2013). Finally, the following table presents an overview of each brand of the sample with a small description and when they joined each platform.
Table 2: Brands overview
Twitter Instagram
Brands Description Name Account creation Name Account creation
H&M Swedish ready-to-wear clothing brand. Fast affordable fashion.
@hm April 15th, 2008 @hm February 18th, 2012 Zara Spanish ready-to-wear clothing
brand
@zara August 1st, 2011 @zara_worldwide February 11th, 2014 Gap American ready-to-wear
clothing brand #dressnormal
@gap December 29th, 2008
@gap September 14th, 2011
Victoria Secret American lingerie and sports apparel brand
@VictoriaSecret September 8th, 2008
@victoriasecret May 11th, 2011 Nike American sports apparel brand
#justdoit
@Nike November 18th, 2011
@nike December 11th, 2011 Adidas German sports apparel brand
and making athletes better since 1949.
@adidas May 17th, 2011 @adidas March 2nd, 2012
Gucci Founded in Florence in 1921, Gucci is one of the world's leading luxury brands.
@gucci October 22nd, 2008 @gucci March 9th, 2011
Louis Vuitton French luxury brand. A world of elegance, inspiration and innovation
@LouisVuitton June 2nd, 2009 @louisvuitton September 26th, 2012
Burberry English luxury brand. A 159 year-old global brand with a distinctly British attitude.
@Burberry June 15th, 2009 @burberry February 11th, 2011
The social media content for each brand was captured on Instagram and Twitter the 19th of may 2015 with an amount of 100 latest Tweets and 60 latest images posted on Instagram. The data was gathered with a Web Scraper developed with a Google add–on, personalized for that research and each platforms. Because of the nature of the application Web Scraper and the natural restrictions of the Instagram platform, only 60 images were allowed to scrap, which explain the reason of the odd number 60. Twitter and Instagram were selected because companies might use the two platforms differently and also because of their significant popularity. Both platforms are in the top 10 most used platforms at the moment4 (Adweek, 2015). Facebook was not selected because we wanted to use two platforms that present distinctive characteristics and Facebook is both text and photo-oriented.
On Instagram, brands use images and on Twitter brands mostly use text even though images are also possible. The Data collected will be all the images and posts with the respective reactions by platforms (retweet, Favorites, Likes, comments), the time-span is not an important factor in this study because we focus on brand personality of the brands, which does not usually change over time. The reactions for each post are scraped only once simultaneously with the post, the timespan of the reactions is not a focus in this particular research. The focus is on how brand express their brand personality and if some personalities are easier to express via different platforms. After looking at those, we will look at the impact of each personality on the consumer engagement. The consumer
counting the numbers of like, comments, retweets or favorites and not in a cognitive or emotional way but could be interesting for further researches.
Platforms
Twitter, founded in 2006, is a micro-blogging site where people are allowed to tweet their thoughts in 140 characters. Posts may include photos, blogs, stories and they are mostly publically available (Smith, Fischer, Yongjian, 2012). Jansen et al (2009) find that 19% of the Tweets are brand-related which implies that users express their opinions about and seek or provide information on brand. Twitter’s behavior mostly consistent of retweeting posts, adding to favorites and reading an enormous flow of content. Brands mostly use tweeter to share information, news, opinions, answer to complaints or details about daily activities (Smith et al, 2012).
Instagram, founded in 2010, is an online photo-sharing and social networking platform that allows users to take photos and apply filters over them and after share them to their followers. The platform is used by general users and by brands that want to inspire their followers with images and inspired lifestyle, which is used by 300 million people at the moment (Instagram, 2015).
On twitter and on Instagram, the brands have very specific account, for that research the global brand was used but multiple options could have been used as for example; Nike Sportswear, Nike Women, Nike Football, etc. Furthermore, each brand has different accounts per specific country on Twitter and Instagram to answer the need of their
customers better and customer service related purposes. This phenomenon maybe implies that they share different things with different countries and perhaps maybe different personalities. For the purpose of this research the global brand was used for each of them.
As this review shows, Twitter and Instagram represents different types of social media and each site have its own norms, culture and function (design). Consumers visit these platforms for different reasons, intentions and also they interact in different ways. Twitter is recognized for being a more customer service/ retention platform and Instagram more an inspiring platform (Burton and Soboleva, 2011). Consequently, consumers are also engaged in different ways with brands. However, yet we still do not know how brand personality is used differently through out those different platforms and which type of personality can be better expressed on each platform because no literature has been developed on the matter. The following analysis will look into which type can be expressed online.
Independent Variables Personalities
In order to answer the research question and based on the previous brand personality studies summarized in the literature. The work of two authors, Aaker (1997) and Geuens et al (2009) have been selected for the matter to have a bigger image and compare the two different scales developed. The first one, more American focused and the second one
an image only for Twitter’s platform because Instagram only allows posts with an image. If the brand personality is expressed with an image by a brand on Twitter, it will be coded as “1” and “0” for a text. Furthermore, the images or text will be coded if there is a brand personality displayed or nothing expressed because some images or text can be publicity or information or customer service messages. When brand personality is expressed in a post it will be coded “1” and if no brand personality is displayed “0” will be coded. Finally, the images or text that express a personality will be coded by there personality based on 10 personality traits of two scales mentioned above and 27 sub-categories from each specific items. The sub-categories will be noted but for the findings, they will be referred to their categories for a better global evaluation because some sub-categories are very close to each other and hard to distinguished. Thus, in order to reduce bias and because perhaps some sub-categories are overlapping, which makes it difficult to evaluate each of them precisely. Each category will be coded with a number from 1 to 10 as expressed in the following table 2. Finally, each sub-categories are coded by a letter of the alphabet.
A code sheet was developed for use in recording the brand personality traits used in the branded content over Instagram and Twitter of the nine fashion brands. Table 2 identifies the personality traits with their specific numbers measured in the content analysis and the definitions used to identify the presence of that specific brand personality. Social media channels are also identified in the coding, either the post is made on Twitter or Instagram to be able to compare if a brand express the same brand personality on both platform and or either one of them.
To be able to code all those personalities, the first impression felt was used when looking at an image. If there was an impression of brand personality, more attention was put into it and deciding which category the image belongs into with the help of the sub categories but when nothing was felt or nothing was really expressed, the coder would pass on to the other image.
Table 3: Examples of coded types of brand personality
Category of codes of Brand personality
Sub categories that can be included
Instagram or twitter posts including a brand personality
Sincerity (1) Down-to-earth (a)
Honest (b)
Wholesome (c)
Cheerful (d)
« Make sure to schedule
#HMConscious sustainability report 2014 LIVE tomorrow at 09.00 GMT » (H&M, Twitter, 2015)
(Victoria Secret, Twitter, 2015)
Up-to-date (h)
Dare to be @NeymarJr. #riskeverything
pic.twitter.com/1zNjZXIWY8
(Nike, Instagram 2015) Competence (3) Reliable (i)
Intelligent (j)
Successful (k)
- no examples (personality not found)
Sophistication (4) Upper class (l)
Charming (m)
Ruggedness (5) Outdoorsy (n) Tough (o) (Nike, Instagram, 2015) Responsibility (6) Down-to-earth (p) Stable (q) Responsible (r)
.@oliviawilde & @babs take a moment to show off their conscious style at our Conscious Commerce shop in #NYC! pic.twitter.com/ly6yLAf8Hk ( H&M, Twitter, 2015) Activity (7) Active (s) Dynamic (t) Innovative (u)
Innovative : « Switch up the lunchtime routine. We take any excuse to discover new music on the run with #adidasgo »
pic.twitter.com/Ae6qAqWGf0
(adidas, Twitter, 2015) Aggressiveness (8) Aggressive (v)
Bold (w)
#Justdo it (Nike, Twitter, 2015) “The game has changed. Prepare for the Football revolution.
Simplicity (9) Ordinary (x) Simple (y) (Gap Instagram, 2015) Emotionality (10) Romantic (z) Sentimental (@) (Gap, Instagram, 2015)
In order to understand how the coding was done, a brief explanation of how the categories were thought is required from Aaker (1997) and Geuens et al. (2009) scales. Sincerity was selected when happy and cheerful people were present in the images and honesty about the product or the company was showed. Excitement could be interpreted when the post was daring as ready to accept a challenge or looking for one. Competence would be selected if reliability, intelligence or successfulness were displayed. Sophistication was selected when red carpet photos with elegant and famous celebrities were showed and when the post expressed high class or high social status. Ruggedness was coded when the post showed an individual or a group of people doing outdoor activities, working out outside and being though and strong. Responsibility was coded
when the posts were showing interest in being environmentally responsible and conscious about the society and the planet. Activity was coded when the post was showing high level of dynamism and/or physically active individuals. Aggressiveness was coded when the posts showed aggressive and bold people chasing their goals and dream. Simplicity was coded when ordinary people or simple behavior were expressed. Finally, emotionality was coded when romantic behavior as a wedding proposal or love was expressed.
Some posts are simply not displaying brand personality; they will be coded as “no personality” to evaluate the amount of content shared is expressing brand personality.
Table 4: No brand personality coding scheme
Brands Examples of Posts Displaying no
personality H&M (Instagram, 2015) Burberry (Twitter, 2015) Dependant variables Engagement
As previously mentioned before, for this explanatory research consumer engagement will be analyzed with behavioural engagement. Per brand personality type, reactions will be counted and which brand personality type leads to the highest reactions from consumers. On Instagram likes and comments will be reported. For Twitter, the re-tweets and the mention favourites will be compile. The engagement part will be the second step of the
coding and of the analysis on SPSS. The brand personality type mentioned will firstly be evaluated; the engagement will be following that first step. The personality will compare next to each other with the average of engagement created when those personalities are express by brands.
The engagement was measured only once when the whole data was collected on the 19th of may 2015. Engagement is represented by the numbers of retweets on Twitter where you have the number next to the two arrows and the number of favorites is next to the star (Table 4). Retweets are when a Twitter user directly repost on their own page the tweet of another user, consequently sharing it to all their own followers. Favorites are the version of liking a post on Twitter. Instagram uses different engagement measures. The number next to the heart represents the number of times users liked the photo and the number next to the dialogue bubble represents the number of comments users made on that specific image (Table 4). The web-scraper developed, automatically selected those specific numbers for each posts and the numbers were collected for each post. As earlier explained those numbers represent the engagement measures used in this particular research because behavioural engagement is used.
Table 5: Engagement Coding scheme
Engagement Instagram Twitter
Examples of codes
Coder training and reliability
To ensure reliability of the result and improve the quality of the research, an external coder was asked to code over 100 posts, equally divided between the Instagram and Twitter data collected to verify the encoded posts. First, the coder was explained and presented the two brand personality scales with each categories to gain a better understanding of the purpose of the research and how to code the images or text looked at. Secondly, the external coder was asked to identify which sub categories of personality on the text and images of the posts to see if they correspond with the ones who were already stated by the first coder. Finally, agreement between the coders was 89% and when disagreement or ambiguities was encountered, the coding was discussed and rectified. Finally, the 11% left was coded right with both coders agreeing.
Analysis methods
Following on the quantitative method, a qualitative analysis is added to the explanatory research in order to add some value to the results section and better understand relationship between brand personality and consumer engagement online. Using SPSS with our data coded, several frequencies were performed with the brand personalities category. Finally, in order to evaluate if the presence of brand personality in a post affects the consumer engagement while controlling multiple variables, ANCOVAs were performed. SPSS and the analysis performed help us confirm propositions from the literature stressing that brand personality lead to a higher engagement.
Control variables
The research presents four ANCOVAS to test whether there is a difference in engagement based on brand personality while controlling for the number of followers for each brand, the numbers of people each brand are following. The numbers of days a brand have had an account on the Platform and the number of posts they did on the platform are also included as control variables.
4. RESULTS
This chapter reports on the results of the study and includes separated sections of brand personality, platforms and engagement. First, an overview of the brands on both
show a detailed overview of the brands evaluated, their numbers of followers, following, days online, number of posts and their posts frequency on both platforms and the percentage of post expressing any brand personality on Twitter and Instagram from Aaker (1997) and Geuens et al. (2009) scale.
Table 6: Descriptives of the brands online behavior Brands Twitter followers Instagram Followers Mean weekly frequency post (Twitter) Mean weekly frequency post (Instagram) % of posts that communicates a certain personality (Instagram and Twitter) H&M 5.62M 6.80M 11 15.30 13.10 Zara 850K 4.30M 9 * 9* 4.30 Gap 619K 840K 17.3 8.30 13.10 Victoria Secret 6.65M 15.40M 19 15.60 21.80 Nike 4.93M 15.70M 3 1.60 60.60 Adidas 2.06M 4.20M 25.6 5.30 24.30 Gucci 1.66M 3.90M 16.5 16.30 15.60 Louis Vuitton 4.37M 4.80M 9 3.60 15.60 Burberry 3.94M 3.40M 13.6 4.30 17.50
* Same number of posts every week
Average calculated over 3 weeks of posts (www.twitter.com), 11th of June 2015 Average Calculated over 3 weeks of posts www.instagram.com), 11th of June 2015 Brand personality: calculated on the 160 posts per brand examined
The table 6 shows that some brands are expressing are displaying brand personality way more than others as Nike with the highest percentage; 60.6% of its content analyzed compare to 4.3% for Zara. Furthermore, some brands as Zara are very consistent on their posts, using the same number for each platform and also the same content on both platforms. In addition, the table 6 shows that the amount of followers highly differs between the brands. Thus, it could lead to a different impact on responses to brand personality in our analysis as the amount of followers and other factors could influence the reactions on posts. For that specific reason, ANCOVAS were calculated over
Instagram and Twitter controlling the following variables in table 7 and 8: number of followers, number of following, number of posts and the number of days each brand have an account on the platform. The table 7 and 8 illustrate how active each brand are on the platform and how often they share content to their followers. It also shows that brands have a upload content on Twitter more often than on Instagram.
Table 7: Descriptives of brands on Twitter
Brands Followers Following Total posts Days online Post per day since online H&M 5.62M 261 8988 2604 3.45 Zara 850K 6700 973 1387 .70 Gap 619K 159 22700 2541 8.93 Victoria Secret 6.65M 121 97100 2360 41.14 Nike 4.93M 158 19800 1295 15.28 Adidas 2.06M 152 2852 1479 1.92 Gucci 1.66M 345 3198 2421 1.32 Louis Vuitton 4.37M 199 2386 2178 1.09 Burberry 3.94M 229 8886 2191 4.05
Table 8: Descriptives of brands on Instagram
Brands Followers Following Total posts Days online on Instagram
Post per day since online H&M 6.80M 241 1298 1187 1.09 Zara 4.30M 41 630 463 1.03 Gap 840K 387 1272 1343 .94 Victoria Secret 15.40M 400 2358 1464 1.61 Nike 15.70M 130 785 1238 .063 Adidas 4.20M 59 705 1174 .06 Gucci 3.90M 132 745 1532 .48 Louis Vuitton 4.80M 4 886 969 .91 Burberry 3.40M 119 1669 1346 1.2
Brand personality communication
We start by addressing the first research question mentioned in the introduction related to the types of brand personality communicated by the 9 fashion brands selected; What are the different types of brand personality communicated online? Furthermore, frequencies for brand personalities were performed in SPSS and are presented for each platform. The Table 9 presents the overall 10 personality traits elaborated by Aaker (1997) and Geuens (2007) and for which brand they appear in their posts. The personality traits are evaluated only on the posts where brand personality was identified. The table presents the overall personalities used amongst the 9 brands developed using a cross tabulation from SPSS. Most of personalities are recurrent in Instagram and Twitter for the same brand but some only showed a personality on either one of the platforms. Looking at each personality (Table 9), Competence was not coded at all and responsibility appears only for one brand on one platform. Competence had the sub-categories of reliable, intelligent and successful and Responsibility was determined with down to earth, stable and responsible. Responsibility was coded with H&M brand when the brand was expressing their concern about environment and their new conscious clothing line: ‘@oliviawilde & @babs take a moment to show off their conscious fashion and responsible style at our Conscious Commerce shop in #NYC!’ Twitter 2015) with an image of the two celebrities wearing the concious collection H&M developed, ‘green line’. Finally, in the following table 6, x were placed where the personality category was coded for each platform.
Table 9: Brands and Primary Category
Brands Platform Sincerity Excitement Competence Sophistication Ruggedness Responsibility Activity Aggressiveness Simplicity Emotionality
H&M Instagram x x x x Twitter x x x x x Zara Instagram x x x Twitter x x Gap Instagram x x x x x Twitter x x x x x x VS Instagram x x x x x x Twitter x x x x x x Nike Instagram x x x Twitter x x x x Adidas Instagram x x x Twitter x x x x Gucci Instagram x x Twitter x x LV Instagram x x Twitter x x Burberry Instagram x x x Twitter x x
Table 10: Frequencies of brand personalities on Instagram
Scale Primary Category Frequency % Cumulative %
Aaker (1997) Sincerity 14 9.9 9.9 Excitement 5 3.5 13.5 Sophistication 30 21.3 34.8 Ruggedness 43 30.5 65.2 Geuens et al. (2009) Activity 30 21.3 86.5 Aggresiveness 4 2.8 89.4 Simplicity 6 4.3 93.6 Emotionality 9 6.4 100 Total 141 100 Total 540
Table 11: Frequencies of brand personalities on Twitter
Scale Primary Category Frequency % Cumulative %
Aaker (1997) Sincerity 13 8.2 8.2 Excitement 7 4.4 12.6 Sophistication 52 32.7 45.3 Ruggedness 14 8.8 54.1 Geuens et al (2009) Responsibility 2 1.3 55.3 Activity 41 25.8 81.1 Agressiveness 23 14.5 95.6 Simplicity 3 1.9 97.5 Emotionality 4 2.5 100 Total 159 100 Total 891
In table 10, 141 posts were coded with brand personality on a total of 540, showing that only 26.1% of the posts on Instagram are expressing brand personality. In table 11, 159 posts were coded with brand personality on overall 891 posts, which shows that overall less brand personality was coded on Twitter than on Instagram, 17.2% of the posts on Twitter compare to 26.1% on Instagram. The posts with no brand personality 73.6% on Instagram and 82.2% on Twitter for the fashion brands were showing the products, the clothes or other informational content. Furthermore, only 8 of 10 personalities were
present on Instagram compared to 9 on Twitter where images were analyzed with text next to it. The personality ‘competence’ from Aaker (1997) is not present in either of the platforms and the personality responsibility doesn’t appear on Instagram. The personalities who are the most frequent on Twitter and Instagram are Sincerity (9.9%; 8.2%), Sophistication (21.3%; 32.7%), Ruggedness (30.5%; 8.8%), Aggressiveness (2.8%; 14.5%) and Activity (21.3%; 25.8%). The less frequent personalities are Excitement (3.5%; 4.4%), Simplicity (4.3%; 1.9%) and Emotionality (6.4%; 2.5%).
Qualitative results
In order to answer the research question how can brand personality be communicated online and on which format some examples will be shown and discussed by overall brand categories. It is interesting to point out that in the 9 sample brands used multiple can be put together as demonstrating the same kind of personalities. H&M, Zara, Gap and Victoria Secret could be put together as ready-to-wear brands mainly expressing sincerity, simplicity and activity. In the other hand, Nike and Adidas mostly express activity, ruggedness and aggressiveness. Finally, the luxurious brands, Gucci, Louis Vuitton and Burberry, mainly express sophistication and a very little of other personalities. The following paragraphs present three examples with the most popular personalities to demonstrate how they are expressed by brands and illustrated by images.
frequencies gave us significant results to draw on which brand personality can be express and which one were not in this particular research context of fashion brands online and how. As expressed earlier, not all personalities were coded because the fashion brands didn’t express themselves with being competent or responsible. First, most fashion brands aimed for sophistication in posts where models or celebrities were wearing their luxury clothes and associating themselves with high-end and exclusive events as Gucci on Twitter; ‘#Gucci at the #Oscars: Chris Evans’ (Gucci on Twitter 2015). Louis
Vuitton, Burberry, Victoria Secret even ready-to-wear brands as H&M expressed sophistication using celebrities, models and displaying upper class context. H&M aim for sophistication because they have occasionally partnerships with luxury brands.
Figure 2: Nike
Furthermore, Nike and Adidas the two main sport apparel brands and Victoria Secret with its sports line are very focused on expressing strong personalities as ruggedness, activity and aggressiveness to inspire and encourage the consumers to do the same and go over their limit wearing their brand. Nike was expressing multiple brand personalities as the same time making those three personalities very similar and sometimes hard to distinguish. As example, figure 2 shows the personality ruggedness as a tough man, outdoorsy but also active from the activity personality and bold to surf in icy water from the aggressive personality. We can mention that some personalities are very similar to each other between the two scales and they can overlap each other leading to expressing more than one personality at the same time. As a second example, Adidas can be quoted on Twitter ‘Spend less making playlists. Discover music as you run, matched to your pace http://a.did.as/60187VN0 #adidasgo’ and combine with an image (figure 3). The tweet express the personality ‘activity’ as the brand is active and very innovative combining sports apparel with technological apps for smartphone.
The ready-to-wear brands, H&M, Zara, Gap and Victoria Secret expressed more personalities as sincerity demonstrating happiness, cheerfulness and being wholesome in their posts as those examples from H&M’s Instagram, see figure 4. H&M displays its brand as sincere with happy and cheerful people focusing on moments and not necessarily on the clothes. Seen on multiple posts the
personality sincere can be often associated with the personality activity of the second scale where someone cheerful can also be perceived as dynamic and active.
Twitter and Instagram
Twitter uses 140 characters to express any messages but photos are also allowed and because of the Web Scraper developed specifically for this research, the data scraped was mostly images on Twitter and after looking into the brands account, most brands are only using images with their tweets on Twitter, which lead to a very similar usage of Twitter as Instagram. Fashion Brands are using Twitter as the same as Instagram, using a combination of short characters and hastags in combination of an image and very rarely brands will simply use the 140 characters allowed to communicate to their followers and express their brand personality. Multiples brands are also using the same content on both Figure 4: H&M (Sincerity)
