This chapter discusses the most important findings with which the research question can be answered.
Secondly, the theoretical and managerial implications will be laid out. In the last section, the limitations of the thesis and related directions for further research are provided.
6.1 Answering the research question
The aim of this thesis is to answer the main research question:
How does the type of device (mobile vs fixed) influence the level of cross-channel free riding intention in a webrooming situation and how is this relation moderated by customer loyalty, product type and
product price?
With the use of both a survey and an experiment, important findings in relation to this question have been found. By combining the two studies, both stated and revealed preferences of consumers could be discovered. Both studies confirmed that cross-channel free riding intentions are higher when people switch to a more fixed device. In the context of this thesis, this means that consumers who search on either a laptop or desktop for product information, have higher intentions to use more websites when they plan to purchase the product in a physical store. Because this finding is supported in both studies, the external validity is increased. Chin et al. (2012) confirmed that a fixed device provides advantages over a mobile device when it comes to information search on products online as it provides more convenience and security. Especially when a customer is not time-constrained and has little experience with the specific brands or stores, he or she is probably more likely to use a more fixed device like a laptop or desktop. Consequently, there is an observed relation found in this thesis with using a more fixed device and higher cross-channel free riding intentions.
Secondly, study 1 provided convincing evidence for the negative moderating effect of customer loyalty on the relation between device switching and cross-channel free riding intentions.
The observed relation where customers have higher cross-channel free riding intentions on a more fixed device eroded when a customer has more loyalty towards a specific brand or store. Such customers are indifferent in choosing a specific device for extensive product research because most likely they are already very familiar with the company and the product. The advantages of using a fixed device in online search therefore become redundant for such customers. The results of study 2 did not provide evidence for the existence of this relation however. Nevertheless, because the results of study 1 are observed as more credible because actual preferences are revealed here, this thesis accepts the presence of the negative moderated relationship of customer loyalty.
48
Thirdly, both studies did not provide any confirmation for the moderated effects of product type and product price. Therefore, it can be assumed that both product type and price do not significantly influence higher cross-channel free riding intentions when using a more fixed device.
While study 1 found coefficients which are on average in the predicted directions, study 2 fails to do so. Still, even though the hypotheses are not accepted, it does not mean that they were completely wrong. The assumptions on which the hypotheses were formed are partly correct. Both studies find evidence that more hedonic products and more expensive products result in higher cross-channel free riding intentions. For that reason, it cannot be ruled out that the hypotheses are completely false, but that the studies have not been designed properly. Nevertheless, in this thesis, product type and price did not inflate the effect of higher cross-channel intentions on a more fixed device.
6.2 Theoretical implications
Omnichannel literature is still only at its inception (Verhoef et al., 2015). Still, with the use of the existing research, this thesis is able to build on to this literature. Cross-channel free riding is a specific problem within omnichannel marketing and has raised increasingly more questions over the last years.
Much research within the topic has revolved around the different motives of customers to engage in the behaviour and which products are more prone to it (Maggioni et al., 2020; Heitz-Spahn, 2013;
Chiu et al. 2011). On the other side, the discussion on different devices in an omnichannel context has also started very recently and has mainly looked into different conversion rates across devices (Xu et al. 2017; De Haan et al., 2018). This thesis extents on the current literature by providing answers to questions that relate to these topics combined, which has not been done before.
Firstly, by answering the question how specific devices influence the level of cross-channel free riding intentions in a webrooming situation, this thesis provides evidence that different devices matter. The integration of omnichannel marketing across the different devices has become more sophisticated as consumers are using increasingly higher number of different devices interchangeably (Verhoef et al., 2015). As this thesis shows that cross-channel free riding intentions on a more fixed device are higher and consequently lower on a mobile device, it adds to the literature of Shankar et al.
(2016) who argued the latest rise of mobile devices in omnichannel marketing. This rise combined with lower free riding intentions is an interesting dynamic which is an intriguing addition to the existing literature.
Webrooming is a relatively new phenomenon and received increasingly more academic attention in the last decade. This attention has mainly focused on antecedents of the behaviour and its relation to showrooming (Kleinlercher et al. 2020; Viejo-Fernández et al., 2018). Now, this thesis relates to that literature by focusing on the intentions during the search stage rather than the purchase intentions inside the store. This thesis finds that customers are more likely to use multiple different web shops when using a laptop, both in the scenario when they maybe do not plan on buying the product in a physical store (study 1) and when they are forced to buy the product in an offline store
49
(study 2). This is an important distinction which has not been researched before and is a valuable addition to the current literature on webrooming.
The cross-usage of different devices has been extensively researched by De Haan et al.
(2018), who among other things show that risk reduction is an important motivation to switch channels. Conversion rates on more fixed devices are larger than on more mobile devices and this effect is more pronounced when product risk is higher. The outcomes of this thesis are in line with these outcomes as customers tend to switch to a more fixed device when they are in greater need of better information search which is often the case when product risks are higher. Interestingly, this thesis finds that customer loyalty erodes this effect fully. This moderating effect of customer loyalty is a new addition to the literature of cross-device usage and product risks.
Even though, this thesis did not find any evidence for the moderating effect of product type and price, it does add to the existing literature by providing evidence for the direct effect of these variables on cross-channel free riding intentions. Like previous research from Heitz-Spahn (2013), cross-channel free riding behaviour is more likely to occur when product prices are higher. Such transactions are considered riskier and therefore people search through more web shops to reduce this risk. This thesis also finds that for hedonic products cross-channel free riding intentions are higher.
This has not been previously found by other research and therefore is an interesting outcome of the study. It means that transactions for hedonic products are considered riskier than transactions of utilitarian products and customer want to decrease this risk by searching more thoroughly online.
6.3 Managerial implications
The phenomenon of cross-channel free riding produces problems for practitioners worldwide. Now that worldwide shipping and cross-device usage have become main stream, people are better able to switch between web shops and channels (Watson et al., 2015). Every person in the world can and will use the free services of any company for its own personal benefit and this comes at the cost of the companies. Tackling this problem should therefore be a major concern for every firm and this thesis provides valuable insights into how to approach this problem more efficiently.
Based on the outcomes of this thesis, practitioners should tackle their marketing campaigns differently across the different devices. If reducing the amount of free riding is a priority, then trying to retain customers on more fixed devices will be more costly, as such devices show larger cross-channel free riding intentions in this thesis. However, because more mobile devices show lower intentions, marketing costs can be reduced on that side. Consequently, marketing efforts can be divided more efficiently between the two different types of devices to increase total retention of customers. For example, if a marketing campaign is executed using discounts which can be used in store, the discounts should be set differently depending on which device you are using. Heitz-Spahn (2013) concluded similarly in her research that using different discounts on different devices can
50
benefit customer retention. Based on the findings in this thesis, discounts can be lower on mobile devices and higher on fixed device to also retain more of those customers in your own channels.
Alternatively, practitioners could try and lure more customers towards their mobile devices as customers have lower intentions to switch retailer when buying in a physical store on such devices. In this scenario the importance of an organized internal marketing campaign is again emphasised.
Different divisions have to understand where the possibilities are to increase customer retention and this would mean less attention would be spend on the fixed device webpage. Providing (price) discounts to lure customers towards the mobile devices would however be contradictory to the previous advice. Therefore, other longer-term benefits of downloading the application of the retailer on a mobile device should be used for this purpose like providing perks for returning to the
application.
An important mechanism to reduce cross-channel free riding in general is by increasing customer loyalty. Increasing customer loyalty will also erode the higher cross-channel free riding intentions for more fixed devices. This would mean that marketing campaigns can be streamlined across different devices and would still be as efficient, and would consequently reduce costs while not hurting benefits. However, this thesis does not provide answers how to specifically increase customer loyalty for this purpose. Still, increasing customer loyalty can be achieved through various different mechanisms. Ultimately, customers become more loyal when they buy your products, therefore combining the different discounts across the devices is an efficient way to tackle the problem of cross-channel free-riding while also improving customer loyalty. Also, downloading a retailer’s mobile application increases customer loyalty (Molinillo et al., 2021) while Ask et al. (2011) found that offering mobile storefronts increases customer loyalty. Therefore, the two previously mentioned strategies can be used effectively in improving customer loyalty, moving more people to using the mobile channels and allocating discounts more effectively and efficiently across the devices.
6.4 Limitations and recommendations for future research
There are specific limitations that characterize this thesis, which ultimately suggest recommendations for future research. Firstly, with regard to the study design, the thesis focuses on webrooming
behaviour only. This was a rational choice as it is the most common cross-channel behaviour (Chiu et al., 2011) and provided a more precise framework to analyse. Still, it does limit the implications of the results as they can only be considered in a webrooming context. Future research could therefore either focus on a similar framework in a showrooming context, where the purchase is done on different devices, or alternatively through a larger variety of touchpoints. This would require a more complex model, however, with a larger sample size and a more specific target group with more experience and knowledge of omnichannel retailing this could be constructed and analysed.
51
Secondly, both the survey and experiment where unable to extract the moderated effect of product type and product price. This may be the reason because there is not actually a moderated effect or because of a problem in the research design. The survey only allowed a binary response for the dependent variable, which meant the respondent either engaged in cross-channel free riding or did not. As a result, there is very little variation in the dependent variable which could also make it harder to extract the effect of multiple moderators. The hypothesis with regard to the moderation of price specifically focused on risks associated with transactions and not search intentions. Therefore, it may be the case that people do not experience higher risks when only searching for more expensive products online.
The direct effect of the moderator variables could be extracted with the expected directions.
This shows that including the variables in the model is not a wrong idea itself. Still, more information on specific customer journeys would be required to answer the more detailed effects of multiple moderators in the relationship. Therefore, in future research a more sophisticated dataset should be gathered which includes more details on a single customer journey. As a result, more detailed effects of specific variables can be laid out and investigated more thoroughly.
The design of the experiment also had several limitations. Not much research previously used an experiment to investigate cross-channel free riding intentions, because stated actions are not the same as revealed actions. Still, the experiment adds to the robustness of the results of this study and resulted in a lot lower dropout ratio, which meant a more efficient response collection. However, designing a scenario in which a respondent has to visualize going through an online search process and consequently buy a product in that store is very restrictive. Especially because a customer journey is a very complex process which involves many different decisions along the way. This cannot be all accounted for in an experiment. Therefore, just like the previously mentioned limitation, it would be advised to use a larger dataset of revealed actions. By collecting survey responses of people with a better memory of their last engagement in webrooming a more refined model can be constructed that is better able to analyse the complex nature of a customer journey.
Lastly, with regard to specific directions for future research, it will be interesting to research empirically how practitioners can benefit from the observed outcomes of this thesis. The moderating effect of customer loyalty on the relation between device switching and cross-channel free riding provides a lot of potential opportunities for marketers but this thesis does not provide empirical answers how to take advantage of this. The current framework can also be extended with other relevant moderators to provide even more specific answers as to how and why people have higher cross-channel free riding intentions on a more fixed device and how to retain more customers inside the firms’ channels.
52