Leveraging digital sensory marketing in online wine sales – a moderated serial mediation model

In online shopping, the product presentation page functions as a digital storefront. An organized content presentation greatly enhances the perceived quality of information (Hasan, 2016), increasing consumers’ purchase intentions (Wang et al., 2017a, 2017b). However, this remains particularly challenging in categories such as wine, where the key determinant of consumer choice – flavor – is inherently sensory and thus difficult to convey through standard digital formats (Rahman and Reynolds, 2015). Most wine retail websites only display photos of wine bottles on their product presentation pages, offering minimal sensory information. Research suggests that concrete sensory descriptions evoke mental imagery, thereby enhancing purchase intentions (Yoo and Kim, 2014). Given the current technological limitations in transmitting gustatory experiences online, marketers increasingly turn to alternative sensory modalities, particularly visual and auditory cues, to evoke imagined consumption experiences.

Visual cues, represented by text and pictures, are widely used in marketing to convey product information and shape consumer perceptions. The frequently referenced saying “a picture is worth a thousand words” suggests the superiority of images. Indeed, pictures enhance emotional appeal and reduce ambiguity, increasing purchase intentions (Li et al., 2022; Wu et al., 2021). However, Pictures tend to evoke automatic processing, making people less engaged in deep cognitive processing (Murphy et al., 2010). In contrast, textual descriptions convey factual information, facilitate comprehension and support informed consumer decision-making (Baker et al., 2010). Auditory cues, such as music, are another sensory tool that is commonly adopted to craft the shopping atmosphere and induce mental imagery (Morrison et al., 2011). Literature has noticed that auditory elements (e.g. instruments) can influence consumers’ perceptions of flavors (Knoeferle et al., 2015). Sound effects that carry information, such as beverage pouring, influence taste perceptions while creating hedonic expectations for consumers (Spence and Wang, 2015a).

While visual and auditory cues each play a critical role in online marketing, consumer perceptions are inherently multisensory. Individual sensory cues do not appear in isolation (Zaidel and Salomon, 2023). What is considered a “unisensory cue” is, in fact, an outcome of multisensory processing. Spence (2020) indicates that multisensory experience is far richer than the simple sum of individual sensory cues. Consequently, research that reflects the dynamics of multisensory integration in consumption contexts is needed. However, most studies focus on the effect of one single sensory cue on consumers’ responses (e.g. Lowe et al., 2019), overlooking the integrated nature of the sensory inputs. Multisensory experience, such as visual-auditory integration, remains under-explored (Stead et al., 2022). An exceptional study (Spence and Wang, 2015b) investigated the interplay between music styles and lighting colors in a wine-tasting room. Despite the constructive findings, their study focused on the on-site context and treated visual and auditory cues as atmospheric cues. Thus, it remains unclear how visual and auditory cues that carry intrinsic information might affect consumers’ perception and consumption in online wine sales. This research gap inspires the following research questions:

To explore answers to the proposed research questions, this research draws on the grounded cognition theory to investigate how visual cues and their interaction with auditory cues on the product presentation page may affect consumers’ responses to online wine sales and the underlying mechanism. The research findings shed new light on digital multisensory integration research and provide applicable approaches to leverage digital sensory marketing for online food and beverage sales.

Visual cues are information visually perceived and processed by individuals. In the marketing field, visual cues contain graphical (e.g. colors and shapes) and spatial elements (e.g. product presentation and layout) that draw attention, provide sensory information and ultimately encourage consumers’ purchase behaviors (Escobar et al., 2022; Vermeir and Roose, 2020). In digital marketing literature, visual cues primarily refer to textual (e.g. consumer reviews) and pictorial cues (e.g. product displays), which have been shown to enhance the perceived helpfulness of information, thereby influencing individuals’ decision-making processes (Chrimes et al., 2022). As the most commonly used visual cues, texts and pictures serve as the building blocks for crafting informative and appealing shopping environments.

Textual cues address information with clarity and efficiency. Different from reading printed materials, people tend to “scan” the online information (Shah et al., 2020). Displaying the information through concise texts allows consumers to gain a quick image of products. In wine sales, concrete sensory descriptions indicate the flavor layers, setting up the consumption expectation. Sreejesh et al. (2021) observed that consumers are primarily driven by utilitarian motives such as acquiring product information when browsing brand websites. Therefore, compared to pictorial cues, textual descriptions that clearly articulate product attributes enhance purchase intentions (Leung et al., 2017). Texts also tell open-ended stories, allowing consumers to interpret them in personal ways (Ceylan et al., 2024). Some wine brands include context-based information in pairing suggestions, leveraging the personal touch to appeal to diverse consumers.

In a digital sales environment, pictures are adopted to optimize website design for several reasons. First, the inclusion of pictures facilitates information processing. The picture superiority effect articulates that pictures foster information storage, recall and recognition (Stenberg, 2006). The conceptual distinctiveness of pictures (vs text) also supports information recollection, slowing down memory decay and contributing to greater recall (Hockley and Bancroft, 2011). Second, the “mere presence effect” of pictures enhances the overall appeal of the website by providing informational and aesthetic values (Li and Xie, 2020). Pictures contain diagnostic cues, helping elucidate abstract information that requires heavy textual explanation (Wu et al., 2021). The information presented by aesthetically pleasing pictures tends to be more vivid (Desrochers et al., 2019), contributing to stronger emotional responses, better product evaluations and higher purchase intentions (Li et al., 2022). Because most brands rely on textual descriptions for flavor information, including pictorial notes could help the businesses stand out from the competitors’ listings (Zinko et al., 2020).

Unlike other hospitality sectors, food and beverage experiences rely heavily on the intrinsic value of the products, namely, their flavors (Rahman and Reynolds, 2015). As shown in Table 1 of the supplementary materials, research on text–picture interplay in online marketing is limited in this sector, except for one work focusing on user-generated reviews of restaurants (Li et al., 2022). However, in such cases, the inclusion of pictures is random and user-initiated, limiting managerial control over how sensory cues are presented. Addressing this gap, the current study focuses on the product presentation page, investigating how businesses can combine visual cues (text and pictures) to communicate sensory information under their direct control.

Grounded cognition theory posits that cognition comes from individuals’ bodily actions, situated environment and perceptions (Barsalou, 2008). Barsalou (2011) further identified two core constructs of the theory: simulators and situated conceptualization. A simulator is a dynamical system distributed across the modalities that integrates perceptual attributes into an aggregated category representation. Simulators store knowledge in long-term memory and produce simulations in working memory. These simulations form situated conceptualizations, which are context-sensitive representations of particular situations.

Grounded cognition theory highlights that knowledge is grounded in the multisensory experiences arising from interactions with the environment (Lin and Lu, 2024). Food and beverage experiences are inherently multisensory, aligning closely with this framework (e.g. Lv et al., 2022). In the preconsumption stage, vivid product information triggers simulators, generating anticipated consumption scenarios that assist decision-making (e.g. Lick et al., 2017). In the postconsumption context, sensory and embodied experiences work as memory boosters, reinforcing the memorability and emotional impact of the experiences (Kim et al., 2025).

The current research taps into grounded cognition in the preconsumption stage. In online wine sales, visual descriptors serve as situational cues that activate simulators related to flavors. The activation of simulators manifests as mental imagery, which is frequently used in marketing to complement such missing sensory information as taste. The following hypothesis development section specifically discusses how visual cues embedded in the online context (i.e. Web page) might contribute to the activation of simulators.

Website entertainment, a key component of website quality (Li et al., 2017), reflects the extent to which a site provides engaging and hedonic experiences (Luo, 2002). Visually pleasing website designs enhance user enjoyment and exploratory behaviors (Jongmans et al., 2022), foster immersion in the shopping process and ultimately increase purchase intentions (Lee and Wu, 2017; Liu et al., 2022b).

According to the affect infusion model, emotional content is important for communicating hedonic products such as wine, particularly in low-involvement scenarios (Forgas, 1995). Online shopping is usually low-involvement because it reduces the time and effort required from consumers, allowing them to browse multiple alternatives with minimal cognitive investment (Beauchamp and Ponder, 2010). As a result, consumers rely less on systematic information processing but more on heuristic cues, which are processed intuitively through cognitive shortcuts (Niza Braga and Jacinto, 2022). In this context, entertaining website features act as salient heuristic signals of enjoyment.

For hedonic products like wine, consumers expect sensory and emotional stimulation. Vivid design elements, such as color, music and product presentation, enhance these experiences by increasing sensory delights and perceived entertainment (Tan, 2008). Pictures, in particular, are more effective than text in capturing attention, enhancing aesthetic judgments and stimulating enjoyment (Jiang et al., 2024). Bufquin et al. (2020) further demonstrated that users’ enjoyment significantly increases with richer visual displays, such as an increased number of pictures on booking websites. Taken together, this suggests that websites combining text with pictures deliver stronger hedonic cues than text alone, thereby making the online wine-shopping experience more entertaining. Therefore, we proposed:

Grounded cognition theory suggests that sensory cues elicit mental imagery, which is defined as “a process by which visual information is represented in the working memory” (Schifferstein, 2009). In consumer behavior research, mental imagery has been recognized as a process enabling consumers to anticipate consumption experiences (Ghiassaleh et al., 2024). High-quality mental imagery enhances the persuasiveness of advertising, positively influencing consumers’ consumption-related behaviors, such as purchase intentions and word-of-mouth (Chen and Zhang, 2024; Cheng and Zhang, 2023).

The quality of mental imagery depends on vividness, which emphasizes the clarity and richness of mental representations (Li et al., 2023). Constructing vivid imagery requires significant cognitive resources since multiple details need to be formed (Dijkstra et al., 2017). Therefore, its effectiveness is shaped not only by sensory cues but also by consumers’ motivation and attention to go through the cognitive processing (Talamini et al., 2023). The amount of attention that individuals allocate to the sensory information determines the cues’ effectiveness on the mental imagery vividness (Li et al, 2023). Entertaining features can enhance this process by encouraging users to invest more attention and cognitive resources, thereby generating more vivid mental imagery and strengthening subsequent shopping behaviors (Park and Yoo, 2020). Therefore, in the current study, we argue that individuals’ perceived website entertainment, characterized by playful and enjoyable experiences, might be a key driver of consumers’ cognitive attention, which leads to enhanced mental imagery vividness and downstream shopping behaviors.

Based on the grounded cognition theory, we argue that perceived website entertainment positively affects mental imagery vividness, which influences consumption-related behavioral intentions such as purchase intentions and WOM. Integrating the aforementioned findings, we propose that:

Auditory cues are another key digital element in the online environment, ranging from naturalistic sounds produced by movements to non-naturalistic cues such as music designed to convey meanings (Stanton and Spence, 2020; Valzolgher et al., 2020). Music, as a rich integration of auditory cues, has long been shown to influence consumers’ emotions, perceptions and shopping behaviors, both through its physical features (pitch, tempo, timbre, volume) and as a whole in shaping store atmospheres (e.g. Ahlbom et al., 2023; De Luca et al., 2019).

In online retail, music is not only used as passive background but also increasingly as interactive music, where tempo, mode or volume adapts dynamically to user actions (Winkler, 2001). Such interactive music enhances enjoyment, novelty and flow experiences, thereby deepening cognitive involvement (Hwang et al., 2020). However, this richness may also generate unintended effects. Stimulus spillover research shows that arousal elicited by one cue can intensify responses to subsequent stimuli (Zillmann, 1971). In the current research, we argue that the arousal induced by interactive music may be carried away to other sensory stimuli, such as pictures, potentially causing unintended effects. Because interactive features like music require cognitive resources to process (Liu et al., 2022a), pairing them with other rich sensory cues, such as images, increases overall mental demands and risks sensory overload. Dual-coding theory suggests that combining text with a single nonverbal cue (e.g. images or music) facilitates processing by activating both verbal and nonverbal pathways (Paivio et al., 1978). However, when text is paired with multiple nonverbal cues (images and music), the heavy cognitive load may hinder processing and trigger sensory overload (Doucé and Adams, 2020). Such overload often results in frustration and withdrawal, negatively impacting individuals’ evaluation (Malhotra, 1984). This might be manifested as reduced perceptions of website entertainment and consequently undermined effects of visual cues on mental imagery and behavioral intentions. Therefore, we propose:

Before data collection, we obtained full approval from the Institutional Review Board and secured written informed consent from all participants. They were informed of the purpose of the study, the minimal risks involved, the assurance that their anonymity and confidentiality would be protected and that their data would be protected by the university and accessed only by the research team. Figure 1 demonstrates the conceptual model of the current research. We conducted two experimental studies to test our hypotheses, both adhering strictly to institutional ethical regulations and guidelines. In Study 1, we tested the relationship between visual cues and perceived website entertainment (H1). In Study 2, we examined the serial mediation effect (H2) and the moderating role of the auditory cues on the serial mediation effect (H3).

To examine how consumers perceive the entertainment level of different visual cues (H1), we conducted a two-condition between-subject experiment, which is a common methodology in consumer behavior research (Fong et al., 2016). Participants were randomly assigned to view one of the two wine sales Web pages with different visual cues: text only vs text and picture (see supplementary materials). In the text-only condition, the text description was displayed as “Bubbly with aromas of pear, lime, and lemon.” In the text and picture condition, in addition to the textual descriptions, icons of bubbles, pear, lime and lemon were displayed surrounding the wine bottle image.

After being exposed to the assigned scenarios, participants answered a questionnaire regarding their responses and reported their demographic information. To ensure the effectiveness of the experimental manipulation and data quality (Leung et al., 2024), responses that failed the manipulation recall check (i.e. “What types of visual cues were observed in the scenario?”) or attention checks (e.g. “Select agree for this item”) were excluded from further analysis. To assess the meaningfulness of the experiment (Leung et al., 2024), the scenario realism was checked via two questions:

(Ma et al., 2024). On the seven-point scale, participants rated this simulation as realistic (M= 5.48) and easy to understand (M = 6.29).

This research targets English-speaking adult participants residing in the USA. The eligibility requirements to participate in the study are above the legal drinking age of 21 years old and having consumed at least one glass of wine per week on average. After excluding 18 responses that failed the attention checks, 119 valid responses were collected from the online research panel of Prolific. The sample size exceeded the recommended threshold of 30 observations per experimental cell for experimental design studies (Fong et al., 2016; Kim et al., 2023). The participants in this study were predominantly 30–39 years old, held bachelor’s degrees and had an annual household income of more than $100k.

Perceived website entertainment was assessed by a five-item index (Gao and Bai, 2014) (e.g. “This website is fun to browse”; 1=Strongly disagree to 7=Strongly agree, Cronbach’s α = 0.95). General wine consumption pattern (i.e. “On average, how many glasses of wine do you drink per week?”; “To what extent are you familiar with sparkling wines.”) and information processing styles (information processing styles: visualizer, e.g. “I prefer to learn things visually,” Cronbach’s α = 0.91; verbalizer, e.g. “I prefer to learn things verbally,” Cronbach’s α = 0.87) were included as control variables (Massa and Mayer, 2006).

A one-way ANCOVA analysis revealed a significant main effect of visual cues, indicating that product presentation comprising visual cues in both text and picture (vs text only) leads to significantly higher perceived website entertainment [M_text = 3.94, M_text&picture = 4.67; F_{(1, 113)} = 8.02, p = 0.005, partial η² = 0.07], supporting H1. The analysis showed that general wine consumption patterns were not significant covariates [familiarity toward sparkling wine: F_{(1, 113)} = 1.76, p = 0.187; average wine consumption: F_{(1, 113)} = 0.92, p = 0.339]; while personal processing styles were significant covariates [visualizer style: F_{(1, 113)} = 5.77, p = 0.018; verbalizer style: F_{(1, 113)} = 7.62, p = 0.007].

Study 1 results support our theoretical estimation that multiple visual cues (text and picture) significantly improve the perceived website entertainment (H1 supported). The personal processing styles were significant covariates in this model, corresponding to the findings of Heller et al. (2019). The personal wine consumption patterns were not significant covariates. Study 2 was further conducted to explore the underlying mechanism explaining the positive effects of multiple visual cues (H2), as well as the moderating role of auditory cues (H3).

We conducted a 2 (visual cues: text vs text and picture) by 2 (auditory cues: with vs without music) between-subject experiment to test H2 and H3. Participants were randomized to one of the four scenarios with combinations of visual and auditory cues. Each participant watched a video stimulating an online wine shopping experience, where they were navigating a wine product presentation using their mouse. The visual setup of the wine presentations was consistent with that used in Study 1. In the condition with interactive music, participants heard interactive sound effects of bottle uncorking, wine pouring and glass clinking when the mouse sensor hovered over the wine bottle; no such sound effects were in the without-music condition. To ensure consistency, all participants were required to enable their device’s audio before watching the videos. After watching, participants answered a questionnaire with manipulation check questions similar to those in Study 1. The realism check results confirmed that participants considered the scenarios as realistic (M = 5.34) and easy to understand (M = 6.31).

Following the same study participation requirements in Study 1, 43 responses that failed the attention checks were excluded. The final sample consisted of 157 valid responses collected from Prolific, with demographics predominantly 30–39 years old, holding bachelor’s degrees and reporting an annual household income of over $100k.

Perceived website entertainment was assessed using the same measurement scale from Study 1 (Cronbach’s α = 0.94). Purchase intentions were measured using a four-item scale adapted from Fan et al. (2022) (e.g. “I’m likely to buy this wine”; Cronbach’s α = 0.96). Word-of-mouth (WOM) was measured using a three-item scale adapted from Hwang et al. (2019) (e.g. “I’d like to say positive things about this wine to others”; Cronbach’s α = 0.95). Mental imagery vividness was gauged using a five-item scale following a seven-point bipolar scale (Lee and Gretzel, 2012) (e.g. “The images that came into mind while I experienced the shopping in the scenario were…”; Vague to vivid, Cronbach’s α = 0.96). The same covariates were included (visualizer: Cronbach’s α = 0.91; verbalizer: Cronbach’s α = 0.84).

A 2 (visual cues: text vs text and picture) by 2 (auditory cues: with vs without music) MANCOVA on purchase intentions and WOM was conducted with personal general wine consumption patterns and individual processing styles being controlled. The results showed that all of the control variables were not significant [familiarity toward sparkling wine: F_{(3, 147)} = 0.43, p = 0.732; average wine consumption: F_{(3, 147)} = 0.39, p = 0.764; visualizer style: F_{(3, 147)} = 1.57, p = 0.199; verbalizer style: F_{(3, 147)} = 1.12, p = 0.343].

MANCOVA–Purchase intentions. The MANCOVA results revealed significant main effects of visual cues [F_{(1, 149)} = 6.09, p = 0.015, partial η² = 0.04] and auditory cues [F_{(1, 149)} = 9.85, p = 0.002, partial η² = 0.06] on purchase intentions. The interaction effect of visual and auditory cues was insignificant [F_{(1, 149)} = 0.231, p = 0.632, partial η² = 0.02]. The simple main effect tests showcased that when the wine presentations came without music, those with descriptors in both text and picture (vs text) produced significantly higher purchase intentions [M_text&picture = 4.41, M_text = 3.78; F_(1,149) = 4.21, p = 0.042, partial η² = 0.03]. When the wine presentations came with music, those with visual cues in both text and picture resulted in a similar level of purchase intentions to those with visual descriptors in text only [M_text&picture = 4.96, M_text = 4.54; F_(1,149) = 2.06, p = 0.154, partial η² = 0.01].

MANCOVA–WOM. The main effects of visual cues [F_{(1, 149)} = 8.54, p = 0.004, partial η² = 0.05] and auditory cues [F_{(1, 149)} = 7.96, p = 0.005, partial η² = 0.05] on WOM were also significant. The interaction effect of visual and auditory cues was insignificant [F_{(1, 149)} = 0.11, p = 0.739, partial η² = 0.01]. The simple main effect tests revealed that when the wine presentations were developed without music, those comprising visual cues in both text and picture (vs text) led to significantly higher WOM [M_text&picture = 4.17, M_text = 3.33; F_(1,149) = 5.13, p = 0.025, partial η² = 0.03]. When the wine presentations were designed with music, those with visual cues in both text and picture resulted in a similar level of WOM to those with visual descriptors in text only [M_text&picture = 4.80, M_text = 4.14; F_(1,149) = 3.49, p = 0.064, partial η² = 0.02].

To test H2 and H3, two analyses using PROCESS Model 83 were conducted to examine the proposed moderated serial mediation effects. In both models, visual cues served as the independent variable, with perceived website entertainment (M1) and mental imagery vividness (M2) as serial mediators. The first model used purchase intentions as the dependent variable, while the second model used WOM as the dependent variable.

Moderated serial mediation–Purchase intentions. The moderating effect of auditory cues on the indirect effect from visual cues to perceived website entertainment to mental imagery vividness to purchase intentions was significant (b = -0.10, 95% CI [−0.2164, −0.0123]). Specifically, when wine presentations did not feature interactive music, the moderated serial mediation effect was significant (b = 0.14, 95% CI [0.0248, 0.2819]). The mediation analysis showed a full mediation effect (Baron and Kenny, 1986): visual cues significantly influenced the perceived website entertainment (b = 1.48, p = 0.000), which in turn affected mental imagery vividness (b = 0.51, p = 0.000) and subsequently purchase intentions (b = 0.19, p = 0.0065). After including the two mediators, the direct effect from visual cues on purchase intentions became insignificant (b = -0.23, p = 0.15). On the other hand, when wine presentations came with interactive music, the moderated serial mediation effect was not significant (b = 0.04, 95% CI [-0.0057, 0.1303]). This model explained 59.28% of the variance in purchase intentions [R² = 0.5761, F_{(7, 149)} = 28.93, p = 0.000].

Moderated serial mediation–WOM. A similar pattern was observed for WOM – the moderating effect of auditory cues on the indirect effect from visual cues to perceived website entertainment to mental imagery vividness to WOM was significant (b = -0.11, 95% CI [−0.2454, −0.0091]). Specifically, when wine presentations were designed without interactive music, the moderated serial mediation effect was significant (b = 0.16, 95% CI [0.0195, 0.3132]). The mediation analysis showed a full mediation effect (Baron and Kenny, 1986): visual cues had a significant influence on perceived website entertainment (b = 1.48, p < 0.000), which in turn affected mental imagery vividness (b = 0.51, p < 0.000) and subsequently WOM (b = 0.21, p = 0.013). After including two mediators, the direct effect from visual cues on WOM became insignificant (b = −0.14, p = 0.46). On the other hand, when the wine presentations were developed with music, the moderated mediation effect was insignificant (b = 0.05, 95% CI [−0.0088, 0.1432]). This model explained 58.13% of the variance in WOM [R² = 0.5813, F_{(7, 149)} = 29.55, p = 0.000].

Study 2 revealed significant moderation and moderated serial mediation effects. Compared with the single visual cue of text only, multiple visual cues (text and picture) significantly enhanced consumers’ purchase intentions and WOM through the serial mediation effects of perceived website entertainment and mental imagery vividness, supporting H2. While these positive effects became insignificant when the auditory cue of music was added, supporting H3. Additionally, personal processing styles no longer showed significant effects as covariates after the auditory cue was added, implying that mismatched sensory cues may disrupt information processing.

This research underscores the importance of multiple visual cue combinations, providing valuable insights into how such cues can be leveraged to positively shape consumers’ perceptions and behaviors in online wine sales. Specifically, the current findings suggest that the inclusion of pictures alongside text contributes to a more emotionally engaging shopping environment than text alone, enhancing consumers’ entertainment perceptions of the websites. While Sreejesh et al. (2021) emphasized the importance of utilitarian motives (e.g. exploring product information) in driving consumers’ intentions to navigate brand websites, the current findings suggest that the hedonic and aesthetic value of the website also plays a critical role for hedonic products such as wine. Visually appealing pictures enhance the aesthetic value and elucidate the information presented by the text, making the information more vivid. This finding corresponds with previous literature suggesting that pictures enrich textual communication (e.g. Desrochers et al., 2019), reflecting the proposition of dual-coding theory (Paivio et al., 1978), which indicates that verbal-visual information facilitates information processing, thereby improving the effectiveness of the marketing communication (Ceylan et al., 2024).

These favorable emotional perceptions, in turn, trigger individuals’ motivation to engage in cognitive processing by generating vivid mental imagery, thereby leading to increased purchase intentions and positive WOM. Different from previous research that investigates a direct relationship between sensory cues and mental imagery (e.g. Jeong and Yeu, 2023), this study suggests that the mental imagery vividness is not granted by sensory cues alone. Instead, consumers’ motivation and attentional engagement are essential, aligning with the findings of Li et al. (2023). In today’s competitive digital landscape, consumers are exposed to abundant information and facing choice overload (Wang et al., 2021). Therefore, consumers tend to allocate selective attention (Goto et al., 2017). Entertaining website features capture attention and sequentially motivate consumers to mentally simulate the vivid consumption scenarios, which is critical in influencing consumption-related behaviors.

While online retailers often aim to mirror the immersive and multisensory experience of physical stores, this study suggests that the intensity of sensory cues should be carefully considered before implementation. The positive effect of visual cues on behavioral intentions is attenuated when the interactive music is included, suggesting that a proper combination of sensory features should be employed to achieve effective communications. Contrary to the notions of “the more, the better” and “three is a charm” in previous marketing research (e.g. Helmefalk and Hultén, 2017), the current study notes that the impacts of sensory inputs on consumer behavioral intentions are not cumulative with the increase in the number of sensory cues. Instead, the accumulation of information-rich sensory cues may backfire, limiting the consumer’s experiences. This conclusion is consistent with the findings of Doucé and Adams (2020), which revealed that the third high-arousal atmospheric cue resulted in avoidance behaviors due to sensory overload in offline stores. The current study suggests that this “third-cue effect” may also occur in online sales. Given that consumers tend to expend less cognitive effort when shopping online, their reduced engagement may lower the threshold for sensory tolerance, increasing the likelihood of overload. Wang et al. (2022) also supported this argument by noticing that the “charm of three” effect occurs only when consumers do not engage in mental imagery, which requires significant cognitive resources.

This study contributes to digital marketing and sensory marketing research by exploring how multisensory website design influences customers’ decision-making in online wine shopping. Unlike previous research mainly focusing on isolated sensory experiences (e.g. Lowe et al., 2019), the current research extends the knowledge by investigating the integration of multiple digital sensory cues within the same modality (visual texts-visual pictures) and between modalities (visual and auditory cues), responding to the call of more studies on multisensory stimuli needed in the research stream (Stead et al., 2022).

When discussing the communication appeal, previous literature has typically focused on the richness of verbal rhetoric (e.g. metaphorical or descriptive texts) or the use of pictures as contextual cues (e.g. Yao and Shao, 2021). However, these studies have largely overlooked the potential of visual cues that convey the same intrinsic product information. Wine contains intrinsic flavors that are typically conveyed through concrete descriptors (e.g. fruits and spices), benefiting uniquely from the use of both verbal and pictorial cues without compromising informational and aesthetic value. Drawing on grounded cognition theory, this study provides empirical support for such a multisensory approach, shedding new light on how visual cues can be strategically combined to enhance communication effectiveness in this niche category.

Multisensory experience is far richer than the linear sum of individual sensory cues (Spence, 2020). By revealing the moderating role of the auditory cue, this research underscores the nuanced way consumers respond to multisensory environments. In online communication, trying too hard to capture attention can overwhelm consumers. Prior literature has shown that disruptions in time (e.g. skippable ads in a video), visuals (e.g. pop-up banners) or flow (e.g. irrelevant ads) are often perceived as intrusive, as consumers prefer to maintain control over the information they engage with (Riedel et al., 2018). In the current research, auditory cues act as unexpected elements embedded within a visually dominant product presentation, interrupting the information processing experience. Focusing on multisensory integration in webpage design, the current research enriches the discussion of online marketing intrusiveness by placing a specific sensory cue combination on the “don’t-do list” for digital design.

This study also extends the application of Grounded Cognition Theory by underscoring the critical role of consumers’ voluntary attention in generating vivid mental imagery – particularly in online shopping environments where cognitive engagement tends to be lower. By focusing on the product presentation Web pages, this study brings a concentrated yet impactful perspective on how website design can shape consumers’ cognitive engagement and subsequent behavioral responses. Collectively, while previous literature emphasizes the mere presence of sensory cues in inducing mental imagery, this research introduces a refined perspective – the effectiveness of the sensory cues depends not only on their presence but also on their ability to induce attentional engagement.

Online shopping enables consumers to browse as many alternatives as they wish before making purchase decisions, which provides consumers with great convenience but also makes digital marketing communication more challenging. Businesses need to work within limited resources to stand out. Compared to high-involvement consumption contexts that require significant acquisition costs including high-value payment and lots of effort (e.g. an overseas vacation), purchasing wines online is characterized as a low-involvement context because it is less risky, less expensive and less likely to generate long-term impact (Jain, 2019). Consumers tend to respond more to emotional rather than rational appeals and make low-effort decisions (Akbari, 2015). Therefore, eye-catching advertising messages are particularly important for low-involvement products (Lang et al., 2022).

As the pivotal digital showcase, the product presentation serves as the foremost touchpoint for shaping customers’ impressions, bearing the dual responsibility of conveying product information and capturing attention. In online wine sales, as well as the broader food and beverage businesses, many brands feature wine bottle images as the profile photo but fail to communicate the intrinsic qualities that may drive purchase intentions (Bruwer et al., 2011). Drawing on the insights of this study, we recommend incorporating vivid cues, such as high-quality and contextually relevant pictures, to pair with text to convey flavor information. Perceived website entertainment should serve as a benchmark for not only evaluating the effectiveness of the pictures but also the overall design quality. Given the flexibility of online retail layout, brands should integrate such visual cues for each product, fully leveraging this affordance to optimize the presentation and enhance customer experience.

Due to the reduced human touch and limited sensory availability, developing immersive shopping experiences presents unique challenges for online retail. While interactive and dynamic features offer opportunities, practitioners should exercise caution and assess both the intensity and quality of these elements before integrating them. Interactive music, as a convenient but emotionally strong feature, should be carefully included when vivid text and images are already in place. Due to its emotional strength and richness, music may demand substantial cognitive resources and diminish the positive effects produced by visual cues.

Beyond online retail environment design, this strategy can also be applied to other digital marketing materials, such as social media posts and advertising banners. After extended browsing of social media platforms with sensory-rich content, consumers may experience fatigue and disengagement (Dhir et al., 2018). To enhance communication effectiveness without adding to this burden, brands should prioritize using vivid pictures that evoke high-quality mental imagery. Online banners are often perceived as intrusive, with animated banners considered more disruptive than static ones due to greater demand on cognitive resources (Pasqualotti and Baccino, 2014). Based on this study, we recommend that brands improve the perceived entertainment of their banners by including vivid pictures to visualize the intrinsic cues, catching users’ attention but maintaining the information at a comfortable level to process.

While this study focuses on wine, the implications can be generalized to broader food and beverage marketing, benefiting products such as cocktails and artisanal desserts that carry rich intrinsic cues. The findings are particularly valuable for businesses that lack the financial resources to implement extended-reality technology for communications. Leveraging the key sensory cues can encourage impactful consumer behavior.

We acknowledge the limitations of this study and would like to offer suggestions for future research to build upon these findings. This study aimed to explore a general strategy that the industry may benefit from sensory cues, so we didn’t consider the influences of sensory cues on brands with different positioning. While the adoption of vivid sensory cues improves the consumers’ perceived website entertainment, its effectiveness on consumption-related behavioral intentions may vary based on the different images that brands want to showcase. Future research is recommended to explore the nuanced effectiveness of sensory cues by examining their interaction with different brand images.

The supplementary material for this article can be found online.