Analysis of the relation between online visual merchandising stimuli and consumer marketing funnel
Nowadays, the way an eCommerce is presented, in terms of visual stimuli, plays a key role in the consumers' online experience. Therefore, it is becoming increasingly important to focus on the phenomenon of the “online visual merchandising” (OVM), defined as the way in which e-tailers (online sellers) display the assortment on their online points of sale or e-stores (Wessels, 2019), using aesthetic and visual elements, such as colours, lettering and images.
More specifically, previous studies have classified the elements that characterise the online visual merchandising in two main macro-categories: on the one hand, the high task-relevant cues (HTRE), namely the stimuli of informative nature, linked to the assortment offered by the website. Among these elements, there are aspects related to products’ presentation (e.g. images and descriptions) and navigation features. On the other hand, the low task-relevant cues (LTRE), which are the elements of a more aesthetic nature, not specifically linked to the products but still able to create a pleasant atmosphere, including stimuli linked to the visual attractiveness of the site itself (e.g. background and font colour) and those linked to the presence of registration forms and/or promotional messages.
By taking clothing web sites as a setting, the study aims to investigate the impact that online visual merchandising elements have on Net Generation consumers, meaning those belonging to Gen Y and Gen Z (18-41 years old). Specifically, as illustrated in Figure 1, by adopting a psychological model defined as Stimulus-Organism-Response (S-O-R), the aim is to understand the impact that online visual merchandising stimuli have on both the individuals' emotional and cognitive reactions (pleasure and interest-organism, respectively), and on the consumers' satisfaction and purchase intention (in terms of response behaviour). Following this approach, the study introduces the “position in the funnel” as a moderating variable, in order to understand how these relationships change according to the three different stages of the marketing funnel (i.e. a funnel model summarising the stages of the consumer purchasing process), in which users may find themselves while browsing a specific site.
Figure 1: Conceptual research framework
The importance of introducing this moderation variable derives from the fact that, depending on the phase of the marketing funnel (awareness, consideration, purchase), both needs and attitudes of an online consumer change: the user, in fact, moving from the top to the bottom of the funnel, goes from a phase of pure exploration and familiarisation with the website (awareness), to a stage of browsing through the catalogue and evaluating the different product alternatives (consideration), and lastly to a phase of real purchase (purchase).
Therefore, this study not only aims to investigate the relationship between online visual merchandising stimuli, emotional and cognitive reactions and response behaviours in general terms, but it mainly focuses on the analysis of these variables in relation to the different stages of the marketing funnel. Nowadays, it is essential to understand what key characteristics a website page should have in order to design an online selling space as customised as possible, according to the stage of the purchasing process in which users find themselves.
To carry out this study, a remote experiment has been conducted, preceded by a sample recruitment process via a screening survey. The resulting 250 subjects were randomly and equally divided into three groups, one for each stage of the marketing funnel (awareness, consideration, purchase). Then, each participant received a specific invitation email, according to the randomly assigned stage. Each email contained: the context in which consumers had to imagine themselves and the related tasks to be performed (e.g. people assigned to the purchase phase were asked to pretend they had to spend a coupon to buy products on the web site); the link to the experiment setting (bershka.com) and the link to the post-experiment survey. After determining the final dataset (consisting of 228 respondents, 76 for each stage of the funnel) and ensuring that all participants had actually understood their assigned stage of the funnel (manipulation check), the structural equation model (SEM) was applied in order to analyse the causal relationships between variables and test the conceptual framework on real data.
Therefore, applying SEM involved carrying out firstly an exploratory factor analysis (EFA), with the aim of identifying latent variables, and secondly a confirmatory factor analysis (CFA), in order to validate the measuring model. As shown in Figure 2, the obtained outcomes appeared to be consistent with the previously hypothesised framework, with the exception of the factors 'Satisfaction' and 'Purchase Intention', which resulted in a single latent variable, named 'Reaction'. However, the high and positive correlation observed between these two factors is entirely justifiable considering that they both represent two possible positive consumers' response behaviours.
Figure 2: Causal model
Once the causal model was structured, the hypothesis test was conducted, which first included a global test, to verify the model's fit to the data, and then a local test, with the aim of examining the significance (p-value) and sign (β-weight) of each specific relationship.
In particular, regarding the relationship between the stimuli and the organism (pleasure and interest), it emerged that elements of an aesthetic nature (LTRE) have a significant and positive impact on pleasure and even more on consumers’ interest (LTRE→ Pleasure with β-weight: 0.748 and p-value < 0.001; LTRE→ Interest with β-weight: 0.733 and p-value < 0.001), regardless of the stage of the funnel taken into consideration. Therefore, the findings were consistent with both the S-O-R model and the previous studies on this topic (Eroglu et al. 2001, 2003; Ha and Lennon, 2010). On the other hand, as far as information-related stimuli (HTRE) are concerned, there is no evidence that these product-related elements have a significant impact on consumers’ pleasure and interest (p-value > 0.05). However, the variable HTRE represents the weakest factor in the model for multiple reasons, related, for example, to the characteristics of the dataset and to the limited sample size. Therefore, as there was no evidence of HTRE's impact on the other variables due to a lack of significance, it was taken as an assumption what the literature had affirmed so far (Eroglu et al. 2001, 2003), namely that this construct significantly impacts both consumers' pleasure and interest.
Regarding the relationship between consumers' emotional and cognitive reactions and their response behaviours, although it is not possible to test direct effects on satisfaction and purchase intention, it was found that the more the consumers feel stimulated (interest) and gratified (pleasure), the more they will adopt a positive response behaviour, in terms of 'REACTION' (INTEREST→ REACTION = β-weight: 0.307 and p-value < 0.001; PLEASURE→ REACTION = β-weight: 0.468 and p-value < 0.001). This outcome not only confirms the validity of the S-O-R paradigm, but is also consistent with findings from previous researches, especially those focusing on Net Generation (Thomas, 2019).
Figure 3: General assumptions (regardless of funnel stages)
With reference to consumers' position in the marketing funnel, it was found that the closer a user gets to the purchase phase, the more the positive relationship between aesthetic elements (LTRE) and the consumer's response behaviour becomes weaker. Therefore, it is evident that the “position in the funnel” has a negative moderating effect between OVM stimuli and response behaviours, but only as far as the aesthetic factors (LTRE) are concerned (in fact: LTRE = p-value < 0.10; HTRE = p-value > 0.10). This finding is consistent with what was expected, i.e. that LTRE, aesthetic elements, have a progressively lower impact on the other variables, in favour of HTRE, product-related elements, as the consumer proceeds along the funnel and gets closer to the purchase phase.
Comparing the impact (β-weight) of the two types of stimuli in each stage of the funnel, it emerged that the aesthetic stimuli (LTRE) have a greater impact or at least not lower than the product-related triggers (HTRE), with reference to pleasure and interest and to the consumer's behavioural reaction. This relationship is valid not only for users in the awareness stage, as hypothesised, but also for those in the consideration and purchase phases. Therefore, without denying the role of assortment-related stimuli, the results gave evidence of a general influence of aesthetic elements on users in the whole marketing funnel. This finding can be explained by considering that consumers’ attention to stimuli related to items presentation, such as descriptions, may be lower for fast fashion products (e.g. those on bershka.com) than for luxury products. Furthermore, one must also take into consideration that the Net Generation is strongly attracted by visual and interactive contents.
Figure 4: Specific assumptions (related to the funnel stages)
In conclusion, the findings not only confirm the applicability of the theoretical framework (Figure 1) on real data but also the validity of the S-O-R paradigm when a new moderating variable is added. Thus, the study provides a conceptual model illustrating how the relationships between stimuli, organism and response change according to the stage of the purchasing process, highlighting that the marketing funnel phase is able to weaken the positive relationship between aesthetic stimuli (LTRE) and the consumers’ behavioural reactions. In addition, considering the three stages of the funnel individually, the study emphasises that aesthetic stimuli (LTRE) play a relevant role in the whole purchasing process.
Based on these statistical evidences, it is possible to obtain concrete guidelines in order to optimise the structure of eCommerce pages. In addition to product-related stimuli, web sites should include numerous visual elements, namely those related to the atmosphere. It is essential, however, to modulate these aesthetic elements according to the stage of the marketing funnel in which the consumer is at that given moment. For example, to capture customers’ attention and to create a sense of trust in the awareness phase, it would be advisable to ensure sufficient attractiveness of the homepage on a visual level, through the presence of videos and the use of colours and fonts.
Therefore, by providing a framework of the different impacts of the online visual merchandising stimuli, depending on the stage of the marketing funnel, the study strongly emphasises the importance of personalising an eCommerce, in order to obtain both seller-side and consumer-side benefits.