“A dark side of retargeting? How advertisements that follow users affect post-purchase consumer behavior: Evidence from the tourism industry in Saudi Arabia”

This study aims to explore the complex effects of post-purchase retargeting ads on consumer behavior, with a focus on expectation confirmation, satisfaction, and repurchase intentions. Additionally, it examines the influence of time spent online on these effects. Anchored in expectation confirmation theory (ECT), the study analyzes responses from 396 Saudi Arabian e-tourism customers who encountered competitive retar-geting ads after purchasing an e-tourism package. The analysis employs partial least squares structural equation modeling (PLS-SEM) and multigroup analysis (MGA) to test the hypotheses. A notable finding is the direct negative impact of retargeting ads on expectation confirmation: increased exposure to such ads post-purchase seems to diminish the perception that initial expectations of the product or service are being met. The negative effect of these ads also indirectly influences satisfaction and repur-chase intentions. Furthermore, the MGA results indicate variations in this negative impact based on the time spent online. Specifically, the more time consumers spend on-line, the stronger the negative impact, leading to a significant decrease in satisfaction and repurchase intentions. These insights reveal the complex nature of post-purchase retargeting ads and underscore the importance of accounting for consumers’ online behavior. They offer valuable direction for marketers to refine retargeting strategies to better resonate with consumer expectations.


INTRODUCTION
In today's digitized world, retargeting has become a crucial digital marketing tool.Businesses aim to enhance sales and engage more effectively with their audiences by creating messages tailored to users' past online behaviors.Yet, this modern technique comes with its intricacies.With its significant reliance on e-platforms for bookings and the transient nature of its offerings, the tourism industry presents unique challenges and opportunities for retargeting.
The well-documented efficacy of retargeting in driving initial purchases starkly contrasts with the relatively unexplored post-purchase realm.After finalizing a purchase, consumers frequently encounter similar adverts, potentially influencing their satisfaction with their decision.This ongoing exposure can reshape their perception of the initial provider and influence their likelihood to repurchase or recommend the service to others.This scenario suggests a potential dark side to retargeting, which might not only affect consumer satisfaction but also long-term brand loyalty and trust.
Using the e-tourism industry in Saudi Arabia as a backdrop, this study delves into this critical area of post-purchase retargeting.With Saudi Arabia emphasizing tourism as a significant component of its broader future vision, grasping the nuances of digital strategies like retargeting becomes paramount.

LITERATURE REVIEW AND HYPOTHESES
Dynamic retargeting targets past website visitors using customized adverts, reflecting products or services they have previously viewed online.This digital marketing strategy has become prevalent because it is technologically easy to implement, reaches a vast target audience in real time, and allows marketers to identify consumer interests and match their needs with available products.Lambrecht and Tucker (2013) outline a retargeting system based on three main entities: retargeters, advertisers, and consumers.Retargeters aggregate advertising space across social sites like Facebook, LinkedIn, WordPress, and YouTube, connecting advertisers to consumers.This advertising space is then sold to advertisers who wish to promote their products by showing adverts to consumers.The effectiveness of this strategy has attracted much attention from researchers in terms of the importance of advert personalization, positive and negative effects, timing and frequency, advertising exposure, and variations of retargeting ad effects across different stages of the consumer journey.
Building upon the effectiveness of this strategy, Sahni et al. (2019) identified positive effects of retargeting, discovering that retargeting leads to a 14.6% increase in users revisiting a website within the first four weeks of targeting, with a 33% efficiency rate in the first week.Their study found complementarity in retargeting, where users exposed to advertising in the first week were significantly influenced by retargeting in the second week, indicating a positive effect on purchasing decisions.Notably, the study did not show decreased retargeting effectiveness with increased exposure.Building on the post-purchase stages, Villas-Boas and Yao (2021) investigated the optimal retargeting strategy for firms.When consumers search for product information, researchers highlighted the limited control firms have over retargeting ads.They pointed out that consumers might continue to receive these ads even after completing a purchase, which could affect the post-purchase impact of retargeting ads.
The existing research has acknowledged the gap in understanding the post-purchase effects of retargeting ads.Johnson et al. (2017) stated that if retargeting ads lead to reactance in consumers, it can undermine the effectiveness of advertising.Therefore, as discussed by Baek and Morimoto (2012), it is essential to acknowledge that the in-fluence of these ads goes beyond their role in driving initial purchases.They also have the power to shape consumers' post-purchase behavior, impacting their satisfaction levels and shaping their intentions for future purchases.
This study focuses on the negative effects of retargeting on the post-purchase stage of the buying process.To study this effect, retargeting is framed within the expectation confirmation theory (ECT), as this makes it possible to link retargeting with expectation confirmation, satisfaction, and repurchase intentions.ECT was created to clarify how satisfaction is influenced by expectations, perceived experiences, and the disconfirmation of those expectations (Oliver, 1977(Oliver, , 1980)).ECT compares consumer notions regarding a service or product before purchase (expectation) with their post-purchase opinion of the product (experience).It, therefore, measures whether expectations are met.
The association between expectation confirmation, satisfaction, and repurchase intentions is well-documented in existing research and has also been tested in several contexts, including e-tourism.Zhong et al. (2015) aimed to understand Chinese user behavior regarding mobile travel booking services utilizing the expectation confirmation theory.They found that expectation confirmation is a significant driver of satisfaction, and users' intention to continue using mobile travel booking services primarily hinges on how satisfied consumers are.
The type of retargeting ads, whether for less or more competitive offers than the initial purchase, is assumed to influence the consumer's expectation confirmation and satisfaction by altering the reference point.Consumer satisfaction after purchasing an offer (A) depends on the quality of the post-purchase competitive offer (B) and the offer (C).Offers by competitive websites using retargeting ads may be better or worse than the purchase (offer A).However, this study only considers the case where the post-purchase retargeting offer is considered more competitive than the original purchase from the respondent's perspective.
Competitive retargeting ads are assumed to modify a consumer's original expectations (Pinquart et al., 2021) and are also assumed to indirectly affect satisfaction and repurchase intentions.
Chen and Lin (2019) examined the impacts of online advertising on user satisfaction and future buying behavior through social media.The article delved into the role of e-word of mouth -a notable form of social media marketing -and its influence on user intentions such as continued usage, participation, and purchases.Through a survey of 502 social media users, they identified that social identification and perceived value are significant mediators.These factors not only impact user satisfaction directly but also shape their broader intentions on social media platforms.
To understand the correlation between using digital platforms and making online purchases, H2: Satisfaction has a positive association with repurchase intentions.
H3: Retargeting is negatively related to expectation confirmation.
H4: Retargeting is negatively and indirectly related to satisfaction and repurchase intentions.
H5: The daily time spent online by users has a significant impact on their perception of retargeting ads, leading to differences in expectation confirmation, satisfaction, and repurchase intentions.

METHODOLOGY
The sample consisted of e-tourism consumers in Saudi Arabia.Each participant purchased an online tourism service and was exposed to more competitive post-purchase retargeting ads.
Participants were administered a questionnaire with measurement items regarding the four constructs: expectation confirmation, satisfaction, repurchase intentions, and retargeting.The survey was drafted in English and then translated into Arabic using the back-translation procedure (Brislin, 1976).A form's hyperlink was distributed through social media, and individuals willing to participate in the survey provided their email addresses.Subsequently, an automated email containing the survey link was dispatched to those respondents.850 emails were sent; 396 valid responses were collected after removing incomplete questionnaires, yielding a 46.5% response rate.This rigorous approach ensured data collection efficiency and minimized potential biases, providing a comprehensive dataset for analysis.
The questionnaire responses were scored according to a 7-point Likert scale from 1 ("strongly disagree") to 7 ("strongly agree").The degree to which perceptions match (confirmation) or differ from (disconfirmation) expectations.
EC1.The online service of the e-tourism website meets my expectations.
EC2.The e-tourism website provides me with all the essential info to decide.EC3.The sales service and payment process provided by the e-tourism website meet my expectations.An assessment of the probability that a customer will conduct another transaction through the e-tourism website or app.
RI1.I want to continue purchasing from the e-tourism website rather than discontinuing it.IRI2.I intend to continue using the service of the e-tourism website I used rather than use another e-tourism website.2).Both Cronbach's alpha and composite reliability exceeded the minimum threshold of 0.7, AVE exceeded the minimum threshold of 0.5 in all cases, and all loadings surpassed the suggested value of 0.7 (Hair et al., 2019) and were significant according to the bootstrap intervals.Also, discriminant validity was assessed by the heterotrait-monotrait (HTMT) ratio, as proposed by Franke and Sarstedt (2019) (Table 3), resulting in values lower than the conservative threshold of 0.85 in all cases.Furthermore, common method bias (CMB) was checked using the full collinearity test approach (Kock, 2015).The results are presented in Table 4.
Since the variance inflation factor (VIF) was below the 3.3 threshold for all latent variables, according to Kock (2015), CMB could be ruled out.
Results of the inner model are detailed in Figure 1 and Table 5 (path coefficients, significance, and model goodness-of-fit) and Table 6 (model predictive power).Expectations confirmation had a positive impact on satisfaction (b = .666),and satisfaction had a positive impact on repurchase intentions (b = .670).As expected, retargeting had a negative impact on expectation confirmation (b = -.324).All coefficients were significant according to the confidence intervals, supporting H1-H3.Support for H4 was also found since retargeting al-  Before running the MGA for the online time groups (<1 hour, 1-5 hours, and >5 hours), the measurement invariance of composite models (MICOM) procedure (Hair et al., 2017) was applied to check measurement invariance.This process comprised three consistent stages: (1) establishing configural invariance, (2) ensuring compositional invariance, and (3) confirming the equality of composite mean and variance values.According to Hair et al. (2017), steps 1 and 2 are prerequisites for running the MGA.
Configural invariance was secured by explicitly defining each latent variable -namely retargeting, expectations confirmation, satisfaction, and repurchase intentions -consistently across all three online time groups in the PLS-SEM.Compositional invariance was then established by comparing the correlations of latent scores between groups to a reference distribution of correlations generated through permutation of the groups.Acceptance of the null hypothesis of a theoretical correlation of 1, indicating composite invariance of the construct, occurs if the observed correlation is within the upper 95% of the distribution.Subsequently, to assure complete measurement invariance, tests were conducted to compare mean values and variances of latent scores across groups against the reference distribution obtained by permutation of the groups.Note: EC = expectations confirmation; RI = repurchase intentions; ST = satisfaction.
The results of the MICOM procedure are displayed in Table 7, indicating that step 2 was verified in all but one instance, while step 3 was only partially supported.Consequently, while configural and compositional invariance were assumed, full invariance was rejected.
MGA results for the three online time groups are reported in Table 8.The impact of retargeting ads on expectations confirmation was significantly greater for the >5 hours group than for the 1-5 hours group (p = .002).Also, the direct impact of retargeting ads on expectations confirmation for the <1 hour group was insignificant, unlike the other groups.Another indirect effect was that the retargeting effect on satisfaction and repurchase intentions increased and became significant for the >5 hours group (b = -0.318and b = -0.222,respectively) and the 1-5 hours group (b = -0.154and b = -0.099,respectively), but not for the <1 hour group (b = -0.184and b = -0.135,respectively).As such, H5 was supported.

DISCUSSION
Digital marketing strategies aim to influence buying behavior in a particular direction (Omar & Atteya, 2020).The literature indicates that digital marketing, including dynamic retargeting (Lambrecht & Tucker, 2013;Sahni et al., 2019; Villas-Boas & Yao, 2021), can affect pre-purchase (from need awareness to purchase decision) but also post-purchase steps (satisfaction and repurchase), given that the goal is to have the consumer purchase and repurchase a particular product or service.This study, framed in ECT, investigated the retargeting effect for e-tourism in the post-purchase period, as tourists are susceptible to this effect.
Concerning the extent to which post-purchase retargeting ads influence consumer expectation confirmation, satisfaction, and repurchase intentions, findings indicate that more competitive retargeting ads negatively affect consumer expectation confirmation and indirectly affect their satis-  Findings suggest a crucial message for firms concerning their differentiation strategy: while differentiation is important, comparative advantage is even more critical.Thus, retargeting focusing on differentiation based mainly on competitive pricing may erode expectation confirmation, satisfaction, and repurchase intentions and induce the consumer to switch to another firm, i.e., the consumer's loyalty is undermined, and a process of mutual cannibalization is launched between firms.However, retargeting focused on differentiation based on product or service benefits may lead consumers to self-assign themselves to the right competitive offer.In this case, consumer satisfaction is less likely to be affected by retargeting ads, which, even if they offer better prices, may not provide the benefits sought by a particular segment of consumers.Future studies could examine the roles of digital skills and customer degree of involvement on the influence of retargeting ads.Time spent online could serve as a proxy for digital skills and might explain variable retargeting effects.At the same time, differences between consumer groups could be explored through consumer involvement in terms of time spent seeking products and services online.Finally, future research may also analyze the effect of retargeting on consumers when offers are differentiated by price compared to when offers are distinguished by the value delivered to specific consumer segments.

CONCLUSION
The purpose of this study was to assess the effect of competitive retargeting ads on consumer post-purchase behavior in the e-tourism sector, especially in the context of expectation confirmation, satisfaction, and repurchase intentions, framed within the expectation confirmation theory.Findings derived using PLS-SEM indicated a distinct negative effect of retargeting on these parameters.Additionally, MGA results showed this adverse impact of retargeting grew more pronounced as consumers spent more time online.Alongside the negative impact of retargeting, there was a positive effect of expectation confirmation on consumer satisfaction, and likewise, a positive effect of satisfaction on repurchase intentions was observed.
From the thorough analysis of these results, a compelling and clear conclusion emerges.It is imperative for businesses to reconsider and pivot their retargeting strategies effectively.Instead of primarily focusing on and competing based on price, there is a significant need to shift the emphasis towards highlighting the overarching value that is delivered to consumers.This strategic change enables consumers to align more organically and intuitively with brands, based on their personal preferences and the perceived value.Such an approach fosters a deeper, more authentic, and loyal relationship between consumers and brands in the dynamic digital marketplace.This realignment not only benefits consumers by providing them with choices that resonate more closely with their needs and values, but it also aids businesses in establishing a more sustainable and meaningful connection with their customer base.

Bhattacherjee
am satisfied by the choice to utilize the service from the e-tourism website.ST2.I made a wise decision by choosing the service from the e-tourism website.ST3.I am happy with my previous choice to utilize the service from the e-tourism website.ST4.Using the services provided by the e-tourism website was the right thing to do.

Table 1 .
Table 1 provides the definitions, list of items, and references for each construct in this study.Scales introduced by Bhattacherjee (2001) were used to measure expectation confirmation, satisfaction, and repurchase intentions.A scale based on Dubrovski's (2001) consumer buying-decision model was created to assess the impact of retargeting.Construct definitions and measurements

Table 4 .
Evaluation of common method bias

Table 5 .
Path coefficients, bootstrap confidence intervals, and standardized root mean square residuals

Table 7 .
Steps 2 and 3 of the MICOM procedure Note: *Confirmed.