Hedonic and informative values of influencer marketing for tap water: the driving role of identification, authenticity and credibility

Tap water (TW) is broadly regarded by global and local institutions as the best water to drink because it is of high quality and safer than bottled water (c.f. Abraham et al., 2024; Geerts et al., 2020). Despite public institutions’ efforts to communicate the safety of reclaimed TW as a solution to droughts, consumers remain hesitant. In doing so, consumers overlook environmental trade-offs between tap and bottled water, including the water used to produce plastic bottles, household cost savings from drinking TW and the rigorous processes reclaimed water undergoes to ensure organoleptic quality and health standards before reaching faucets. Outside short-term, drought-related emergencies, institutions likewise call for sustained public support to build water-smart economies and develop water resilience strategies that empower consumers for more efficient water decision-making (e.g. European Commission, 2025).

Marketing and communication efforts promoting the benefits of reclaimed water (or simply TW) for drinking, cooking and other household uses are still in their early stages (Rodríguez-Ardura et al., 2026; Xu et al., 2024). Although still nascent, these efforts represent a strategic opportunity to strengthen the competitiveness of water institutions, utilities and other sustainability-oriented stakeholders. Influencer-generated content can enhance consumers’ perceived value of TW, helping these actors build public trust and foster consumer engagement in water resilience strategies.

However, the increasing global interest in sustainable lifestyles makes this an opportune moment to promote TW consumption. Further, much remains to be explored regarding influencers’ role in encouraging healthy and environmental behaviors (Schmuck, 2021) that foster societal welfare (Aw and Agnihotri, 2024), including sustainable TW drinking practices. Research in sustainable consumption, although beginning to emerge, is still fragmented and mixed (see e.g. Kaňková et al., 2025; Munaro et al., 2024). It has also highlighted the potential role of influencers and social media in improving TW taste perceptions, in addressing health concerns, and encouraging positive habits (Zvěřinová et al., 2024). Through these focal marketing initiatives, influencers and institutions can cultivate a long-term and sustained water conservation culture (March and Saurí, 2026). Against this backdrop, this study seeks to advance understanding of how influencers can deliver value and promote the public good through increased TW consumption (Hudders et al., 2021; Hudders and Lou, 2023).

In parallel to this, marketing communication research calls for studies scrutinizing how influencers prompt consumers’ intended responses by simultaneously delivering hedonic and informative value in their content (Pan et al., 2025). By contrast, existing influencer marketing (IM) studies typically focus on emotional or utilitarian appeals, and the few that examine both hedonic and cognitive, instructional mechanisms yield contradictory results (Barari et al., 2025). Drawing on both the hedonic and informational dimensions of value initially conceived for studying shopping behaviors (Babin et al., 1994) and broadly accepted within marketing and consumer behavior literature (Babin and Krey, 2020; Picot-Coupey et al., 2021), the present study examines how Instagram influencers can evoke both hedonic and informative value perceptions in support of TW consumption. It shows that consumers’ value appraisals may help achieve more sustainable consumption patterns, thereby positioning influencers as strategic partners for water stakeholders in fostering a shared commitment with the 2030 Agenda for Sustainable Development and the Sustainable Development Goals of the United Nations (2015).

To do so, we consider three potential driving factors of both hedonic and informational benefits and IM effectiveness (Pan et al., 2025). Particularly, we investigate how Instagram influencers elicit the intended behavior of drinking TW through a novel combination of three underlying mechanisms of persuasion: identification (de Graaf, 2023; Huang et al., 2024), authenticity (Shoenberger et al., 2021) and source credibility (Breves et al., 2019). Together with authenticity and credibility – the main sources of influence within social media IM (Barari et al., 2025; Munaro et al., 2024) – we introduce the construct of identification for its significance in increasing consumers’ attachment to the meanings and values transmitted by influencers and demonstrate that the specific interplay of these three antecedents jointly influences both hedonic and informative values.

Value creation is central to social change (Zainuddin and Gordon, 2020) and helps explain why consumers adopt or reinforce drinking TW. Value not only provides immediate benefits or gratification but also offers insight into consumers’ decisions regarding products and lifestyles (Gkargkavouzi et al., 2019; Govaerts and Ottar Olsen, 2023), as well as end-consumer preferences for tap versus bottled water (Brouwer et al., 2021) and water versus other beverages (Onufrak et al., 2014). Accordingly, value theory can also help explain consumers’ acceptance of reclaimed water (Bolderdijk et al., 2013; Xu et al., 2024) and their preferences toward its uses (Li and Roy, 2021).

Shopping values research (Vieira et al., 2018) distinguishes between two types of value (hedonic and utilitarian) that guide individuals’ consumption decisions. Hedonic value refers to immediate satisfaction and other related pleasant feelings derived from engaging in a behavior (Rodríguez-Ardura et al., 2025). Influencers evoke captivating experiences by presenting their lifestyle and daily routines in visually appealing ways, encouraging consumers to emulate those positive and pleasing experiences. This approach captivates and maintains followers’ enthusiasm (Audrezet et al., 2020), provides entertainment and enjoyment (Barta et al., 2023), elicits affect (Vieira et al., 2018) and generates emotive rewards of the encounter with the post. Through original content, influencers engage consumers (Rosengren et al., 2020), stimulate their intellects about the benefits of adopting the intended behavior and project a vision and style that makes audiences feel satisfied with following the influencer (Leung et al., 2022).

Conversely, utilitarian value captures the perceived usefulness and functional benefits that individuals obtain through or after consumption experiences. We draw here on M. R. Lee et al.’s (2014) concept of information value in social networks, which is rooted in value perception theory. Driven by task-oriented motivation, the value of information (i.e. perceived informativeness) increases as it more effectively satisfies the audience’s cognitive needs. By addressing these needs, influencers position themselves as attractive and trusted sources of knowledge (Belanche et al., 2021a) who provide valuable content to their audiences. When the content is novel and interesting (Sánchez-Fernández and Jiménez-Castillo, 2021), as well as substantive [i.e. relevant and complete (Zhang et al., 2014)], consumers acquire knowledge that can readily apply in their daily lives, enabling them to make more proficient consumption decisions (Farivar et al., 2023).

Health and environmental communication studies (e.g. Bullock and Shulman, 2021; Guenther et al., 2021; Ropret Homar and Knežević Cvelbar, 2021) have shown that influencers are more persuasive when audiences processes and interpret their content as intended. This suggests that influencers should carefully design their posting strategies to engage audiences in a pleasant manner while fostering the perception that they have acquired valuable information (Pan et al., 2025). The underlying premise is that consumers derive greater value when the content promoting the cause is both hedonically appealing and cognitively stimulating, thereby supporting problem-solving processes that encourage TW consumption.

Identification entails experiencing the same feelings as the referents and aligning with their style, interests and actions (Leung et al., 2022). Audiences develop strong connections with influencers by placing themselves in various aspects of the character’s life, even in intimate situations (Chung et al., 2023; van Laer et al., 2019). Identification evokes aspirational feelings that enable consumers to project themselves as being, or becoming, the ideal or desired character (Koay and Lim, 2025). Consumers who identify with referents vicariously participate in influencers’ daily lives for the hedonic appeal those experiences convey (Escalas and Bettman, 2017).

As observed by Dekoninck and Schmuck (2025), consumers are more likely to feel identified with green lifestyle ideals when influencers showcase attainable pro-environmental behaviors, whereas distant, unrealistic or overly demanding ideals discourage sustainable choices (Buvár et al., 2023). In this context, consumers’ gratification stems not primarily from the promoted behavior itself but from perceiving that they share a similar stance on a specific sustainability issue with the influencer (Huang et al., 2024). This mechanism may be particularly relevant for TW drinking behavior, as drinking TW has no inherent hedonic appeal (Xu et al., 2024) and is often a private action with limited need for public display (Bearden and Etzel, 1982). This challenges influencers to reframe TW consumption as a captivating, relatable and desirable drinking behavior. By closely resembling influencers, consumers adopt the influencer’s perspective as their own. Accordingly, identification may contribute to self-construction, represent an affective merging of the self and the character and strengthen consumers’ attachment to the hedonic meanings and values conveyed by influencers:

On Instagram, authenticity refers to the coherence between what the influencer feels, says and does (Ibarra, 2015). Being true to oneself and to one’s audience forms the foundation of authenticity (Lee and Eastin, 2021; Wellman et al., 2020). Hence, authenticity becomes a necessary element of every post and an essential characteristic of every influencer (Hudders et al., 2021). In this vein, influencers who produce direct and organic content that is consistent with their image may evoke emotional attachment and other pleasant responses (Hofstetter and Gollnhofer, 2024) that make consumers feel closer to them. Authenticity is also associated with the veracity and transparency of the information, as well as the significance that consumers attach to honest and non-biased opinions (Lee and Eastin, 2021). Thus, influencers’ candid presentation of the advantages and disadvantages of their experiences, together with the sharing of reliable, truthful and genuine content, enhances authenticity of Instagram stories (Kowalczyk and Pounders, 2016; Kreling et al., 2022).

Audiences value authenticity in any pro-environmental mise-en-scène (Lu, 2024), appreciating down-to-earth representations that combine emotional resonance with substantive content (Dekoninck et al., 2023) and persuade precisely because they acknowledge imperfections (Buvár et al., 2023). A similar pattern can be expected in the context of TW consumption. When consumers perceive an influencer’s content as authentic, they are more likely to experience it as a means of achieving personal and intrinsic gratification (Audrezet et al., 2020). In turn, they may attribute symbolic meanings to the influencer’s narrative that align more closely with their self-concept, bringing them closer to their ideal self-image and strengthening emotional attachment. Accordingly, stories that authentically promote TW consumption are expected to generate greater hedonic value for consumers.

Furthermore, high-quality content that is congruent with the consumer’s existing self-schemas should better enable them to make value-based decisions (Ibarra, 2015). Hence, we hypothesize that acquiring relevant and genuine knowledge from a direct referent also enhances informative value:

Credibility refers to consumers’ willingness to accept information based on the congruence between the message and the characteristics of its source (Belanche et al., 2021b). When influencers share messages that align closely with their established image, consumers are more likely to perceive them as knowledgeable, valid and trustworthy sources of information (Breves et al., 2019). Perceived similarity between influencers and consumers (Lou and Yuan, 2019) together with the influencer’s stance toward the promoted topic, fosters credibility, particularly when influencers communicate non-professional expertise (Wellman, 2024). It remains unclear, however, how influencers establish their expertise (Munaro et al., 2024) and how they transfer their credibility to the messages they post (Boerman et al., 2022). This issue is particularly relevant because credibility serves as a mental shortcut for evaluating pro-environmental initiatives (Schorn et al., 2022).

We suggest that when consumers perceive trustworthiness and honesty in the source, credibility enhances both the utility and persuasiveness of the information (Belanche et al., 2021a, 2021b). Since credible individuals are considered reliable information providers, influencers who deliver credible messages strengthen consumers’ belief that it is worthwhile to acquire or reaffirm the shared knowledge. Accordingly, a positive and confident belief in the information source might enhance consumers’ perceptions of the message’s informative value:

Hedonic and informative appeals can elicit positive responses in social media when the audience processes and interprets the content as intended (Guenther et al., 2021; Ropret Homar and Knežević Cvelbar, 2021). In the context of IM, this requires a careful selection of posting strategies that engage the audience in a pleasant manner while fostering consumers’ perceptions of acquiring beneficial insights (Pan et al., 2025). Indeed, Leung et al. (2022) and Pan et al. (2025) recommend developing content that is both engaging and insightful – that is, combining hedonic and informative qualities. However, the few studies that have simultaneously considered hedonic and informative values for mediating consumption behaviors report mixed and sometimes contradictory evidence (c.f. Hughes et al., 2019; Kim et al., 2023; Lu et al., 2024; Zhang et al., 2024).

To reconcile this assorted evidence, we suggest that both forms of value may function jointly in driving consumers’ responses. When influencers present visually appealing content alongside practical and useful information, they can achieve an effective balance between experiential gratification and informational utility. Thus, a positive perception of an influencer’s value proposition that simultaneously evokes hedonic and informative value can more effectively lead consumers toward the intended behavior of adopting TW. Ultimately, we posit that the higher the experienced emotional appeal of the content toward the promoted cause and the greater the consumer’s acquisition of knowledge for problem-solving in support of TW intake, the greater the perceived value acquired:

Figure 1 depicts the proposed research model.

To ensure content validity, we exclusively used measurement scales that previous studies had verified for reliability and validity, with psychometric properties exceeding recommended thresholds. The selected scales were adapted to the context of TW consumption. We relied on de Graaf et al. (2012) for identification; Shoenberger et al. (2021) for authenticity; Belanche et al. (2021a) for credibility; Barta et al. (2023) and Mathwick et al. (2001) for hedonic value; Zhang et al. (2014) for informative value; and Moon and Kim (2001) and Nunes et al. (2021) for behavioral intentions. Responses were based on seven-point Likert scales (ranging from 1, “completely disagree,” to 7, “completely agree”). To minimize common method biases, we refined the questions and provided clear definitions for unfamiliar concepts. After ensuring equivalence between English and Spanish and performing parallel back-translation procedures for content validity, we circulated the questionnaire in the target languages.

The questionnaire was preceded by a fictitious Instagram post, created ad hoc to ensure all participants shared the same understanding of TW. In the post, a picture of an influencer drinking TW from a glass was accompanied by a text that highlighted influencer’s reasons to drink TW (e.g. high quality) and the benefits of switching to TW (e.g. cost savings). We targeted adult Instagram users in the greater metropolitan areas of Phoenix and Barcelona who actively use the platform for non-work purposes, as both regions experience recurring droughts and periods of water stress. We included four attention-check questions to reduce random responding and ensure that participants careful consider and carefully read all questions. We recruited participants from an online panel. After sending a friendly reminder to those who did not finish the survey, we eliminated incomplete questionnaires and those with inconsistent responses, retaining 800 valid ones.

We examined the sample’s composition by age and gender (Table 1), concluding that there were no non-response biases, and that the sample accurately represented the target population.

We used Harman’s single-factor test and the correlation matrix procedure to verify that the data was free from common method variance. All factors have unrotated variance explained values smaller than 50%, and all pairwise correlations between constructs are below 0.90 (see Table 2).

Building on the work of Rodríguez-Ardura et al. (2025), we used a novel approach that makes it possible to bring together the advantages of using PLS as an effective causal-predictive method with the capacity of artificial neural networks (ANNs) to account for non-linear relationships. While PLS is particularly useful for testing measurements and assessing predictive relevance, ANNs complement these advantages by detecting hidden patterns in the data, capturing potential non-linear dependencies and minimizing prediction errors.

First, we conducted a conventional assessment of the measurement model and, next, we trained unsupervised ANNs to capture the non-linear relationships between the measurement items and the constructs. Then, we used the scores generated to test the significance of the causal paths in the PLS structural model.

Finally, we examined the potential non-linear dependencies between behavioral intentions and the other constructs in the model by training supervised ANNs with five inputs (i.e. identification, authenticity, credibility, hedonic value and informative value), one output (behavioral intentions) and one hidden layer (see Figure 2). Through a trial-and-error process, we identified that two hidden nodes in the hidden layer yielded the best results, satisfying Blum’s (1992) requirement, as it falls between the number of inputs and the number of outputs.

Data was scaled to a range between 0 and 1 using the min-max scaling method. The three-layer perceptron training algorithm used the logistic function as the activation function and the resilient backpropagation algorithm with weight backtracking. The sum of squared errors was the differentiable error function to minimize.

The measurement model demonstrates strong internal consistency reliability. Cronbach’s α and Dillon-Goldstein’s ρ values range from 0.88 to 0.95, exceeding the minimum accepted threshold of 0.70 (Hair et al., 2019). Furthermore, the first eigenvalues for each construct are greater than 1, while the second eigenvalues are all below (Sarstedt et al., 2014). The discriminant validity of the measurement model was first established by the cross-loading matrix, which revealed that each indicator’s loading on its designated construct was higher than its loadings on any other constructs. Next, we applied Fornell and Larcker’s criterion, confirming that the square root of the average variance extracted (AVE) for each latent variable exceeded its correlations with other constructs (Hair et al., 2019). In addition, the heterotrait-monotrait (HTMT) ratio matrix verified that all values were well below the threshold of 0.9 (see Table 2).

Each item’s loading on its designated construct exceeds 0.70 with communalities above 0.50. In addition, the AVE for each construct exceeds the recommended threshold of 0.50. Accordingly, all criteria recommended by Hair et al. (2019) are satisfied, confirming the convergent validity of the measurement model.

Next, following Rodríguez-Ardura et al. (2025), we used an unsupervised ANN to measure the constructs non-linearly. This ANN model included a single hidden layer with one node, using the hyperbolic tangent as the activation function.

Given that the R² values for the three PLS regressions are all clearly greater than 0.30 (Figure 3), the model demonstrates moderate predictive accuracy (Hair et al., 2019). Stone−Geiser’s Q² values exceed 0.35, indicating that the predictive relevance of the path model for the endogenous latent constructs is high (Henseler et al., 2009). Furthermore, the SRMR of the structural model is 0.05. These results support the validity of the linear structural model (Hair et al., 2019).

Since the sample does not follow a multivariate normal distribution, we assessed the significance of the PLS path coefficients using the bootstrapping procedure, with 5,000 resamples. The results presented in Table 3 show that all p -values are substantially below 0.05 and below the Benjamini-Hochberg α correction values (Benjamini and Hochberg, 1995). Consequently, all hypothesized causal paths are statistically significant from a linear perspective, supporting H1–H6.

Because each regression includes multiple predictors, we measure their effect size on the dependent variables using f ². Based on Cohen’s (1988) cut-off values, we assert that identification has a large impact on hedonic value (f ² = 0.604); authenticity has a small influence on this construct (f ² = 0.060), but a medium effect on informative value (f ² = 0.238); credibility has a small impact on informative value (f ² = 0.142); and behavioral intentions receives a medium influence from hedonic value (f ² = 0.183) and a small effect from informative value (f ² = 0.123). These effect sizes indicate that hedonic value is the primary mechanism that elicits behavioral intentions towards TW consumption.

For the ANN model, we conducted 10-fold cross-validation, using a 90:10 split between training and testing data. Specifically, the data set was divided into ten equal and mutually exclusive folds. In each iteration, the model was trained on nine folds and tested on the remaining fold. The mean absolute error (MAE) values for both the training and testing data sets were below 20% (0.185 and 0.190, respectively), an acceptable range in social science research (Sonkar et al., 2026). These results indicate that the model achieves accurate predictions, and that input variables predict behavioral intentions by capturing potentially non-linear relationships.

Following Garson’s (1991) algorithm, we performed a sensitivity analysis to assess the significance of each input factor on the prediction of behavioral intentions (Table 4) and check for unexpected patterns. The average importance (AI) of the input factors was calculated by averaging the importance values obtained from each iteration of the ten-fold cross-validation process. We then derived the normalized importance (NI) by expressing each factor’s relative importance as a proportion of the maximum relative importance among all factors.

The findings of the ANN analysis strongly reinforce the results of the PLS: all input variables contribute to explain behavioral intentions, being the hedonic value the most influential. However, the NI values reveal differences between the PLS and the ANN. In the linear analysis, informative value is the second most important factor (80.66%), whereas in the non-linear, it becomes the fourth (68.74%). The other constructs maintain the same order: identification is more important than authenticity, and authenticity more than credibility.

This research shows that influencers who simultaneously deliver hedonic and informative value through their posts can shape followers’ behavioral intentions toward TW consumption. The findings indicate that hedonic value is strengthened when consumers identify with the influencer and perceive the content as authentic (H1, H2), supporting the view that self-resemblance and perceived genuineness make pro-environmental content experientially appealing. Informative value, in turn, is enhanced when the influencer’s communication is perceived as both authentic and credible (H3, H4), suggesting that consumers derive greater knowledge-related value from content they perceive as genuine, trustworthy and reliable. Overall, the results confirm that identification, authenticity and credibility are distinct yet complementary antecedents of consumers’ value perceptions.

Hedonic and informative values are found to mediate behavioral responses. This supports the notion that social media can influence consumer preferences toward TW (Zvěřinová et al., 2024). From a marketing perspective, it suggests that social media influencers can foster transformative consumption patterns by promoting sustainable habits and gently guiding consumers toward more environmentally responsible choices.

The findings further show that Instagram can effectively convey value-driven messages that integrate hedonic and informative appeals, thereby helping to clarify the mixed empirical evidence reported in previous IM studies. Compared with informative value (β = 0.317, p  = 0.000), hedonic value (β = 0.393, p  = 0.000) is the strongest predictor of TW consumption intentions – a pattern that resonates with the emphasis placed on the experiential and affective value facets proposed in personal shopping value research.

The importance of activating hedonic value perceptions is confirmed by both the linear and the non-linear analyses. However, the ANN provides a more fine-grained assessment of the relative influence of the persuasive mechanism leading to TW consumption intentions. It shows that authenticity, which has already been recognized in prior IM research as an important antecedent of consumption intentions (Li et al., 2025) operates alongside consumers’ identification with the influencer as the most salient drivers of behavioral intentions. This finding is particularly relevant because identification has received comparatively little attention in IM research, yet it appears to play a decisive role in making pro-environmental content personally meaningful and emotionally engaging.

This research provides evidence that Instagram influencers can effectively persuade consumers to adopt the sustainable behavior of drinking TW when audiences interpret and internalize the influencers’ value-based appeals. These appeals, both hedonic and informative, enable consumers to acquire information in a pleasurable manner. The study’s primary contribution to value theory lies in demonstrating the mediating role of hedonic and informative values in enhancing behavioral intentions toward TW consumption. These dual appeals serve as a meaningful avenue for influencers to connect with their audiences and guide them toward the intended behavior.

The delivery of hedonic and informative value arises from the combined influence of three underlying persuasive stimuli: identification, authenticity and credibility. Therefore, the study introduces a new theoretical perspective to persuasion theory. By clarifying the factors that strengthen persuasive impact, it also contributes to the growing literature on the effectiveness of IM (Pan et al., 2025) and IM communication (Leung et al., 2022). Furthermore, it highlights identification as a primary persuasive mechanism that deserves recognition alongside authenticity and credibility as a major source of influence on social media (Barari et al., 2025; Munaro et al., 2024). The study also illuminates how these persuasive stimuli interweave marketing content with hedonic and informative purposes (Isaac and Calder, 2025), thereby advancing understanding of the antecedents within the nomological network of value.

This research also extends marketing frameworks and IM beyond traditional commercial focus. Specifically, it contributes to the emerging literature on non-business-oriented IM aimed at sustainable causes (Boerman et al., 2022; Schmuck, 2021). By demonstrating how influencers can encourage individuals to adopt beneficial behaviors – such as drinking TW as a value-based choice – the study expands the bright side of IM (Hudders and Lou, 2023). Moreover, by focusing on this individual enacted yet socially significant behavior, it contributes to research that applies a marketing perspective to global challenges, supporting the transformative and heroic potential of marketing (Plangger et al., 2025).

Finally, this study introduces IM, largely overlooked in water research, as a promising approach for designing value-driven messages that can mitigate negative perceptions and concerns about TW consumption. It also serves as a counterpoint to the growing body of research favoring bottled water over TW (Olatunde et al., 2025), challenging the paradoxical consumer tendency to pay for bottled water when TW is safer, cheaper and of higher quality (Abraham et al., 2024; Moya-Fernández et al., 2021).

Individuals’ first-hand experiences during water restriction periods, coupled with growing awareness that water reaches faucets after reclamation and purification, have changed perceptions of TW. TW is no longer regarded as a product noticed only in its absence (Brouwer et al., 2019). The growing adoption of sustainable consumption practices and pro-environmental lifestyles presents an ideal moment for institutions to promote TW consumption and reverse persistent consumer reluctance.

IM emerges as a valuable tool for policymakers and institutions seeking to encourage TW consumption. Influencers can advance this societal goal by overcoming consumers’ apathy and increasing the perceived personal relevance of sustainable water consumption. The interpersonal nature of influencer communication positions influencers as effective opinion leaders in the field of water-related communication. To leverage this potential, water stakeholders should ensure that influencer campaigns foster perceptions of identification, authenticity and credibility. Accordingly, institutions face a dual challenge: selecting credible influencers who can effectively advocate for TW while ensuring that the content they create is authentic, trustworthy and resonates with consumers, ultimately encouraging them to emulate the influencers’ behavior.

Influencers should not be regarded as an isolated means of reaching audiences, but rather as an integral part of companies’ and organizations’ broader integrated communication strategies, with the potential to guide consumers toward behaviors that serve the public interest. Accordingly, institutions and policymakers should aim not only to raise consumers’ awareness in specific sustainability debates (Deb et al., 2023), but also to foster audiences’ engagement with environmentally friendly lifestyles. Such an approach can strengthen the competitiveness of water-related institutions within an increasingly sustainability-driven landscape by positioning them as credible, transparent and environmentally responsible organizations. To this end, these institutions should collaborate with influencers who consistently demonstrate a commitment to sustainability across different aspects of their lives and can naturally integrate TW consumption into a broader sustainability-oriented lifestyle, rather than portraying it as an isolated behavior. This approach is likely to strengthen perceptions of authenticity and sincerity while reinforcing the halo of goodwill surrounding both the institutions and the influencers. In turn, influencer communication can help reduce resistance to TW, strengthen the legitimacy of water resilience strategies and enhance consumers’ trust in water-related institutions.

Furthermore, consumers’ commitment to water resilience strategies requires influencer content that combines accessible, credible and engaging information about the price, quality and safety of TW with a relatable narrative style. For example, messages highlighting household cost savings, safety assurances and environmental benefits through humorous, attractive or personally resonant storytelling can make TW consumption appear both relevant and resource efficient. Narratives showing influencers overcoming common barriers, such as concerns about taste, safety or social appropriateness, may help followers imagine themselves making similar choices. Likewise, creative and participatory formats, including challenges, memes and calls to use public drinking fountains or request TW in restaurants, can translate sustainability goals into simple, actionable behaviors that consumers can readily adopt.

Policymakers and institutions should draw on evidence from both commercial and water communication contexts. Accordingly, influencers whose content naturally aligns with sustainability and pro-environmental themes are likely to be more effective at reaching and building trust with relevant consumer segments – including young consumers, heavy social media users and audiences with the potential to drive social change, even if they are less aware of water conservation issues (March and Saurí, 2026). By delivering meaningful content, influencers can reinforce positive attitudes among existing TW consumers and inspire behavioral change that positions TW intake as a value-based choice. Contrary to Claassen et al. (2022), we argue that habit formation in this context does not require an intermediary stage of increased bottled water consumption. Instead, increases in TW consumption should occur alongside reductions in bottled water use.

This study did not account for consumers’ prior experiences and learning processes, which might influence their responses to influencer-generated content. Future research could address this limitation by examining how influencer-generated content shapes behavior in relation to individuals’ awareness of themselves, society and the environment in everyday consumption practices. Future research could also examine whether the effects of influencer-generated content differ across private, social-signaling and system-changing forms of pro-environmental behavior, as well as how environmental value orientations shape consumers’ responses to sustainability-focused influencer communication.

Because the study examines consumers’ behavioral intentions toward TW in regions affected by droughts and water restrictions, future research should replicate the model in areas without such pronounced water constraints to better understand the role of situational factors, neglected in this study. In addition, the absence of cultural variables limits the generalizability of the findings but opens promising avenues for exploring how cultural scripts embedded in influencers’ lifestyles motivate audiences to act (Lu, 2024).

Finally, Instagram proved to be a suitable visual platform for communicating value-laden narratives that promote TW consumption. However, because this study focused exclusively on Instagram, it cannot determine whether the platform is more effective than other social media for encouraging consumers to choose TW over alternative beverages. Cross-platform research would provide valuable insights into the role of platform-specific experiences as contextual influences.