The effect of web advertising visual design on online purchase intention: An examination across gender

Highlights

• A model of the effects of web advertising visual design (WAVD) is proposed.

• Structural equation model (SEM) is adopted to test the model.

• WAVD positively influences online purchase intention via attitudinal effects.

• Gender moderates the effect of WAVD on consumers' attitudes and purchase intention.

Abstract

With web advertising growing to be a huge industry, it is important to understand the effectiveness of web advertisement. In this study we investigate the effects of web advertising visual design (WAVD) purchasing intention within the framework of an integrated model. Nine hypotheses were developed and tested on a dataset of 316 observations collected via a questionnaire survey. The results of structural equation modeling (SEM) indicate that while web advertising visual cues influence consumers' purchasing intention through advertising attitudes and brand attitudes, they do not have direct effects on purchasing intention. Further results on the moderating role of gender suggest that web advertising visual cues have direct effect on consumers' purchasing intention for male groups but not for female groups. This study contributes to the understanding the role of visual dimensions in forming online purchase intentions.

Introduction

The use of web advertisements began in 1994, when the first banner advertising was displayed on Hotwired.com for AT&T (Holis, 2005), and since then the internet has become an important medium through which companies advertise their products and services. According to a recent report released by the Interactive Advertising Bureau (IAB, 2014), web advertising revenue surpassed $23 billion in the first half of 2014, up 15% on the same period in 2013. In the coming years, web advertising expenditure is projected to overtake all advertising media, including TV advertising. Advertising is often considered as the main marketing tool in terms of influencing consumer purchase decisions (Kiang, Raghu, & Shang, 2000). In an internet context, most scholars agree that more careful consideration needs to be given to web advertising visual design (WAVD) in order to achieve its goals (Cho, 1999, Duffett, 2015, Méndez and Leiva, 2015, Pieters et al., 2010). This may be because of the sheer numbers of site stimuli competing for consumers' visual attention. Online users have been found to spend on average only 6.4 s on each search engine results page (Hotchkiss, 2006), and they usually decide to stay or leave the website within the first 2 min (Dahal, 2011). Thus, it is increasingly essential to measure the efficiency of web advertising and its design because the beauty of advertising is in the eye of the beholder.

Against this backdrop, web advertising has gained a great deal of attention in recent years due to its potential effect on online shoppers' responses (Almendros and García, 2014, Ching et al., 2013, Flores et al., 2014, Goodrich, 2011, Kuisma et al., 2010, Saadeghvaziri et al., 2013, Sajjacholapunt and Ball, 2014, Sokolik et al., 2014). While previous studies have identified the impacts of web advertising on basic consumer reactions such as click-through rates and consumer recall, a significant gap remains in the theoretical understanding of how WAVD influences online purchase intention (Cyr et al., 2009, Goodrich, 2011). This study sets out to fill in the gap by investigating effects of WAVD on consumers' purchase intention. In addition to the potential direct impact of WAVD on purchase intention (Goodrich, 2011), this study also establishes the role of attitudinal effect as an important mediator in the relationship between WAVD and purchasing intentions (Ha and Janda, 2014, Pavlou and Fygenson, 2006).

Moreover, we examine the moderating role of gender in the relationships between WAVD, consumers' attitude and purchasing. Gender is considered a key segmentation variable in the field of marketing (Darley & Smith, 1995), and plays a key role in moderating consumers' evaluative judgments (Holbrook, 1986). Based on gender-based research, differences between the sexes have been uncovered related to attitudes toward web advertising, attitudes toward online shopping (Hasan, 2010, Rodgers and Sheldon, 1999), information searching and processing styles (Krugman, 1996, Richard et al., 2010), visual design preferences (Cyr and Head, 2013, Mahzari and Ahmadzadeh, 2013), attention to web advertising (Goodrich, 2014, Park, 2015), satisfaction with online shopping (Rodgers & Harris, 2003), online communication strategies (Holmberg & Hellsten, 2015), and purchase intention (Davis, Lang, & Diego, 2014). These differences between men and women may moderate the effects of web advertisements (Goodrich, 2014). Although some studies focus on the importance of gender in the online shopping context, empirical evidence regarding the moderating role of this variable remains scarce. According to Goodrich (2013), such investigation may be beneficial with regard to providing a link between web advertising design elements and marketing goals. Thus, this investigation also attempts to fill this gap and to help marketers better understand the impact of WAVD on online shoppers across genders.

This study contributes to the advertising literature in two ways. First, the study advances our understanding of the role of WAVD in online purchase intention formation by integrating the direct and indirect effects of WAVD on online purchase intention into a single model. Second, unlike previous research that proposes that gender only moderates the effect of site stimuli on attitudes (Goodrich, 2014, Richard et al., 2010, Tsichla et al., 2014), our study discovers that gender plays a significant moderating role with behavioral intentions as well. This study therefore is the first of its kind to provide such empirical evidence in a web advertising context.

The remainder of the paper is organized as follows. First, we provide a theoretical background of the variables in our model. Next, we develop the hypotheses proposed in our model. After that, we discuss the method used for data collection and analysis. Then, we conclude the paper with a discussion and some directions for future research.