Published on February 4, 2014
Eastern Michigan University Department of Marketing An Exploratory Investigation of a Brand Equity Model for an Internet Portal Website G. Russell Merz, Ph.D., Professor Department of Marketing Eastern Michigan University EMU 1
Eastern Michigan University Department of Marketing Presentation Agenda • Introduction and Literature Review – Website User Experiences – Website Brand Development – Building Branding Theories • Theoretical Framework – Brand Equity Model – Research Questions – Hypotheses • Methodology/Findings – – – – • Data Collection and Measurement Analysis Methods Sample Profile Structural Equations Modeling (SEM) Results Discussion – Contributions and Implications – Limitations and Directions for Future Research EMU 2
Eastern Michigan University Department of Marketing Introduction to the Study • • The value of branded websites in ecommerce is without dispute. Website brand research has evolved from: – Descriptions of user experiences (perceptions) and reactions (attitudes) to attributes of branded websites; to, – Prescriptions about how website brand strategies should be formulated; to, – Theoretical understanding of how branding outcomes are developed by websites. • This paper contributes to this research stream in two ways by: – Proposing a novel approach to measuring brand constructs such as image and preference; and then, – Modeling the constructs within a comprehensive consumer based brand equity framework (Keller 2003, 2008). EMU Branding Theories Prescriptions for Branding Strategies Descriptions of User Reactions 3
Eastern Michigan University Department of Marketing Previous Research: Descriptions of User Reactions to Branded Websites • Much of the earlier research described the experiences and reactions by users to various website design characteristics , such as: – Atmospherics of the website (Eröglu et al 2001, Mummalaneni 2005 and Richard 2005). – Attitudes toward the website resulting from the brand (Ward and Lee 2000, Balabanis and Reynolds 2001), website attributes (Chen and Wells 1999, Ho and Wu 1999, Newman et al 2004) and, services and functions (Van Riel et al 2003). – Perceived service or website quality (Janda et al 2002, Trocchia and Janda 2003, Zhang and van Dran 2001). – In addition, some studies have treated the measurement of website characteristics as predictors of user satisfaction (Mummalaneni 2005, Trocchia and Janda 2003, Yen and Gwinner 2003 and Zviran et al 2006) and shareholder value (Rajgopal et al 2001). EMU 4
Eastern Michigan University Department of Marketing Previous Research: Prescriptions for Website Branding Strategies • Other studies proposed brand strategy guidelines for websites based on: – Generalizations from branding practices based on the Brand Pyramid (Clarke 2001), benchmark databases (Flores 2004), and comparative case studies (Ibeh et al 2005, Bergstrom 2000, Simmons 2007); – Empirical investigations examining the interplay between brand image, trust and personality and factors such as website design (Chang et al 2002, 2003); shopping features (Ha 2004); customization (Thorbjornsen et al 2002) and levels of website usage (Thorbjornsen et al 2004). – Relationships between the summative constructs of brand attitudes, knowledge and experience and other attitudinal and behavioral outcomes such as perceived risk, search and intentions to adopt (Chen and He 2003); website involvement (Dahlen et al 2003); website satisfaction (Ha and Perks 2005; Na and Marshall 2005); and revisit and recommendation intentions (Toms and Taves 2004). EMU 5
Eastern Michigan University Department of Marketing Previous Research: Website Brand Performance Theories • Finally, a few studies provide theory-based models for understanding and explaining website brand performance. For instance: – The relationships between on-line brand experiences, familiarity and levels of brand trust (Ha and Perks 2005), – Website brand trust building features and on-line/off-line brand purchases (Lee et al 2005), – The relationship between branded communications tools (newsletters) and subsequent brand image (Müller et al 2008), – Brand image dimensions and brand loyalty (Alwi 2009), and – Brand personality congruence and brand loyalty (Magin et al 2003). • However, despite the breadth of the existing literature, a key component of brand strength missing from the many previous studies is the concept of brand equity. EMU 6
Eastern Michigan University Department of Marketing Theoretical Framework: Research Objective • • The objective of this study is to build and test a brand equity model based on the customer-based brand equity (CBBE) framework proposed by Keller (2003) using brand related survey data collected from an internet portal website. The intent is to develop a three stage model that relates two indices of brand meaning (brand performance and brand image) to a single index of brand judgment (brand preferences), and then to two indicators of brand resonance (brand loyalty and brand advocacy). EMU Keller’s Consumer Based Brand Equity Model 7
Eastern Michigan University Department of Marketing Research Questions and Hypotheses Given the findings from the previously reviewed literature, this study investigates the following three research questions: • First, is there evidence that the website brand image is differentially affected by the perceived brand performance of the website design? – H1: The index of website brand image is positively explained by an index of website brand performance. • Second, is there evidence that website brand image and perceived brand performance affects brand preference? – H2a, b: The index of website brand preference is positively explained by indices of website (a) brand performance and (b) brand image. • Third, as postulated in the CBBE theory is there evidence that brand judgment affects brand resonance? – H3a, b: The index of website brand preference positively explains the variation in two indicators of brand resonance (a) website brand loyalty and (b) website brand advocacy. EMU 8
Eastern Michigan University Department of Marketing Brand Judgment Brand Meaning Brand Resonance Information Seeking Content Tangible or Experiential Entertainment Functionality Security Socializing Brand Performance Music and Gaming Personalization Brand Loyalty H2a Advertising H1 Brand Preference H2b Customer Care Trustworthy Innovative Unapproachable EMU H3b Brand Advocacy Brand Image Quality Distinctive H3a Intangible or Inferred The proposed model is constructed from three second order latent variables and two first order variables.
Eastern Michigan University Department of Marketing Methodology: Measurement and Data Collection • • • The website that was the subject of this study is a branded internet portal that provides a full range of services to users such as shopping assistance, access to information sources such as news and sports, online games, email, etc. The use of one website was considered sufficient for an exploratory examination of the hypothesized relationships. The data was collected during a two week period from website visitors who were randomly selected and asked to participate through on-network invitations in a web survey hosted by a data collection firm. – There were usable 3,127 responses to the invitation. – All participants were required to be at least 13+ years and have visited the website in past 30 days. – No check for non-response bias was done, but because of the randomness of the survey trigger, it is not believed that the responses are substantially biased. EMU 10
Eastern Michigan University Department of Marketing Findings: Profile of the Sample • • Examination of the demographics shows that most respondents are female (54.4%), middle aged (3554, 42%), with household incomes of between $35K to $110K (51.2%), and education that includes some college education through the bachelor’s level (59.9%). The usage characteristics revealed that 56.5% of users have used the brand for 5 years or more, and 55.5% were classified as heavy users of the website. EMU 11
Eastern Michigan University Department of Marketing Methodology: Analysis and Modeling First, since brand preferences were measured using a set of ordinal ranking measures, the preference dimensions were defined by using principal components analysis with varimax rotation. Second, the entire first and second order theoretical brand equity model was estimated in a single step using a latent variable partial least squares (LV-PLS) to test the hypothesized causal structure. Third, the measurement model quality is assessed by examination of the construct and discriminant validity. Fourth, the significance of the hypothesized paths is evaluated by a bootstrapping routine. EMU 12
Eastern Michigan University Department of Marketing Measurement: EMU Index of Brand performance: A composite second order latent measure comprised of five first order latent variables that capture the perceived performance of the website. 13
Eastern Michigan University Department of Marketing Measurement: EMU Index of Brand image: A composite second order latent measure comprised of six BAV brand image factors specified by Lebar et al (2005) developed from twenty four multiple response items. 14
Eastern Michigan University Department of Marketing Measurement: Index of Brand preference: A composite second order latent measure comprised of four dimensions of brand preferences derived using factor analysis from ordinal preference rankings of the branded website compared to other websites. Brand loyalty: A first order latent containing three items measuring the likelihood that the respondent will: (1) use the website again, (2) choose the website as primary internet portal in future, and (3) whether the website brand has earned the respondent’s loyalty. Brand advocacy: A single item scale asking respondents to rate the likelihood that they will recommend the website to someone else. EMU 15
Eastern Michigan University Department of Marketing Findings: Structural Equations Modeling • • • • • To test hypotheses a structural equations model (SEM) with latent variables was estimated using a latent variable partial least squares (LV-PLS) algorithm (Ringle, et al 2005). The measurement model in PLS is assessed in terms of item loadings and reliability coefficients (composite reliability), as well as the convergent and discriminant validity. Measures with loadings onto underlying latent variables of greater than 0.7 possess acceptable levels of association with a component (Fornell and Larcker 1981). Interpreted like a Cronbach’s alpha for internal consistency reliability, a composite reliability of 0.7 or greater is considered as an acceptable level of reliability (Fornell and Larcker 1981). The average variance extracted (AVE) measures the variance captured by the indicators relative to the measurement error, and it should be greater than 0.5 to justify using a construct (Barclay, Thompson and Higgins 1995). EMU 16
Eastern Michigan University Department of Marketing Findings: Measurement Model Discriminant Validity The square roots of AVE values exceed inter-correlations of the latent variables. All Cronbach’s Alphas are > 0.7. All AVEs exceed minimum of 0.5. EMU 17
Eastern Michigan University Department of Marketing Findings: Indicators of Model Quality Cross Loadings: If the cross loadings are smaller and the discriminant validity test is met, then the structure is acceptable for exploratory analysis. Bootstrapping of Path Coefficients: To obtain the standard errors of the path coefficients a bootstrapping routine was used. The results are based on 500 cases resampled 500 times. EMU 18
Eastern Michigan University Department of Marketing Brand Meaning Brand Judgment Brand Resonance Information Seeking Content Entertainment Functionality Socializing Security Brand Performance Music and Gaming Personalization Brand Loyalty .220 Advertising .377 (4.41) (10.45) .504 Brand Preference (14.46) .279 Customer Care (5.60) Trustworthy Innovative R2=.142 R2=.188 .417 (11.61) Brand Advocacy R2=.174 Brand Image Quality Distinctive R2=.254 Unapproachable EMU Path coefficients are standardized β (t-statistics in parentheses: all are p ≤ 0.01)
Eastern Michigan University Department of Marketing Discussion: Contributions There are three contributions to the website brand development literature: • First, this study takes a holistic approach to understanding the brand equity of websites. It does this by modeling facets of brand meaning together with brand preference and brand resonance in a predictive cause and effect theoretically-based framework. This represents an unique empirical addition to the website brand literature. • Second, the model was built using a novel mixed measurement (interval, nominal and ordinal) approach. This method reduces the potential for common methods bias, and provides a way to incorporate a wider and richer source of brand measures into a comprehensive modeling framework. • Third, several of the findings within the model support the results of earlier studies reported in the literature; and because this study uses a comprehensive brand equity framework, the value of the earlier studies is enhanced by a better understanding of their role in brand theory development . EMU 20
Eastern Michigan University Department of Marketing Discussion: Implications The implications for pure-play websites, such as the one used in this study, are the following: • The design characteristics of the website can be considered a strategic activity that builds brand equity (as reflected in the brand resonance measures) rather than simply as a media or service delivery vehicle. • Since perceptions of brand performance influence the overall formation of brand image, website designers and managers of websites need to consider if the experiences of users with the website enhance or detract from desired brand images for the website. • The effects from both the tangible aspects of website performance as well as intangible aspects of brand imagery onto the brand preferences formed by users suggest that website managers must be concerned with both facets of brand meaning to build preferences and positive evaluations for their website. EMU 21
Eastern Michigan University Department of Marketing Discussion: Limitations and Directions for Future Research • • • • This exploratory study utilized limited and rather abstract measures of the customer-based brand equity framework espoused by Keller (2003). Future studies need to consider additional brand measures especially regarding judgment and emotional responses to brand meaning. In addition, a fuller set of brand resonance indicators including community and engagement aspects are needed. Regarding the respondents to the survey, this study did not examine how user differences might affect the measurement of and the relationships between the perceptions and judgments of the brand examined. Although many of the users in the current sample have been using the website for some time, their objectives or motivations might be quite diverse. Examining different segments in the sample may reveal alternative brand equity configurations reflecting variations in usage rates, conditions and frequency of use. Finally, only a single pure-play website was used in this study to examine the hypothesized relationships. Subsequent studies of website brand equity need to expand the range of websites as study objects. EMU 22
Eastern Michigan University Department of Marketing Thank You for Your Attention Are There Any Questions? EMU 23
Brands are more invested today than ever before on curating and distributing paid,...
Marketers need to be creating and publishing original content across many differen...
As content marketing continues to increase in popularity in every industry, more m...
Il Direct Email Marketing (DEM) è una tipologia di marketing diretto che usa la po...
This presentation contains all 120 rules from Part 1 of the 2nd edition of "Email ...
A Brand Equity Model for an Internet Portal ... of an exploratory empirical investigation into ... Equity Model for an Internet Portal Website ...
A Brand Equity Model for an Internet Portal ... of an exploratory empirical investigation into how ... Corporate Websites and Brand Equity ...
Conference Programme . ... An Exploratory Investigation of a Brand Equity Model for an Internet Portal W ... Building Brands with Interactive Websites?
BRAND EQUITY IN INTERNET BUSINESS: AN EXPLORATORY ... we develop a model of brand equity ... we find that product/technology innovation and Website ...
1 1 The Ability of Branded Online Games to Build Brand Equity: An Exploratory ... increases in brand equity for the ... portal in Malaysia ...
Consumer engagement in a virtual brand community: An exploratory ... with the brand as personified by the website or other ... brand equity ” (Keller ...
An Exploratory Investigation of Reverse Causality ... Exploring the Dynamic Relationship between Brand Equity and Sport ... Prof. Dr. Maik Hammerschmidt
... and a model for the management of brand equity ... Relationships among Brand Equity Components: An Exploratory ... Innovative use of the Internet ...