The Mediating Roles of Service Experience and Satisfaction: How Servicescape Influences Loyalty and Electronic Word-of-Mouth

Abstract

Servicescape, the physical and social environment of a service setting, is a critical strategic tool for creating competitive advantage. While its influence on customer loyalty and electronic word-of-mouth (e-WOM) is established, the underlying psychological mechanisms remain inadequately specified. This study addresses this gap by proposing and testing a dual-mediation model grounded in an integrated Stimulus–Organism–Response (S-O-R) framework, with cognitive evaluations informed by Expectancy-Disconfirmation Theory (EDT), distinguishing between affective (service experience) and cognitive (customer satisfaction) pathways. Data were collected from 420 patrons of nature-themed cafés in Nakhon Ratchasima, Thailand, and analyzed using structural equation modeling (SEM-PLS). The results confirm that servicescape significantly enhances both service experience (β = 0.805, p < 0.001) and customer satisfaction (β = 0.816, p < 0.001). However, its effects on customer loyalty and e-WOM are fully mediated through these parallel pathways. Customer satisfaction demonstrated a stronger influence on loyalty than service experience, while both were significant drivers of e-WOM. The findings suggest theoretical contributions by delineating the distinct affective and cognitive processes through which the service environment translates into digital advocacy and loyalty. For managers, this study suggests a strategic framework for allocating resources to foster both shareable experiences and satisfaction-driven loyalty.

1. Introduction

In the competitive landscape of experience-centric services, the physical and social environment—commonly referred to as the servicescape—is widely acknowledged as a critical determinant of business success. Substantial empirical evidence confirms that the servicescape influences customer loyalty and electronic word-of-mouth (e-WOM) (Bitner, 1992; Line et al., 2018). However, the specific psychological mechanisms through which these effects materialize remain theoretically underspecified, representing a significant research gap in servicescape literature. The proliferation of social media and digital platforms has further elevated e-WOM as a significant driver of business success, making it an important managerial priority to understand how designed environments translate into digital advocacy through distinct psychological pathways (Kandampully et al., 2023).

This research addresses these gaps within Thailand’s café sector, which presents a theoretically relevant context for investigating e-WOM dynamics. The selection of this context is academically justified as Thailand’s café industry, particularly the burgeoning nature-themed segment, epitomizes the experience economy. In this sector, the servicescape is a primary competitive tool and the creation of shareable experiences is central to the value proposition, making it an ideal setting to test the proposed dual-mediation model (Kandampully et al., 2023). Furthermore, the country’s high social media penetration and user engagement rates (Datareportal, 2024) enhance the ecological validity of examining e-WOM behaviors. The strategic emergence of nature-themed cafés in provinces like Nakhon Ratchasima exemplifies this phenomenon, where operators consciously design immersive servicescapes optimized for high visual appeal. This design strategy is informed by research demonstrating that the aesthetic quality of service environments significantly influences customers’ electronic word-of-mouth intentions (Lee & Kim, 2022; Zhang & Zhao, 2020). This setting, therefore, allows for a clear examination of how designed environments stimulate both affective experiences and cognitive evaluations. These processes are particularly salient in visually driven, experience-centric services where aesthetic consumption and the resulting motivation for digital sharing constitute integral aspects of the value proposition (Silanoi et al., 2022).

Bitner’s (1992) seminal servicescape framework offers a foundational understanding of environmental influences, with contemporary research recognizing its multidimensional nature, encompassing ambient conditions, spatial layout, signs/symbols, and social interactions (Dedeoglu et al., 2018). The Stimulus–Organism–Response (S-O-R) framework (Mehrabian & Russell, 1974) provides a theoretical basis for examining the mediating pathways between servicescape and behavioral outcomes. Despite this solid theoretical grounding, a fundamental consideration persists in the literature: the ‘Organism’ (O) component in the S-O-R model is often treated as a monolithic or underspecified construct. This has led to inconsistent empirical findings regarding the mechanisms linking servicescape to behavioral outcomes. Some studies report direct effects of servicescape on loyalty (Line et al., 2018), while others indicate fully mediated relationships through satisfaction (Inthasang et al., 2022) or customer engagement (Rini et al., 2021). These inconsistent findings may stem from methodological variations—such as differences in service contexts, measurement approaches, or model specifications—or from the under specification of the internal psychological mechanisms in the S-O-R framework. Specifically, prior studies have typically examined either affective or cognitive mediators in isolation, rather than modeling both pathways simultaneously within an integrated framework.

This study addresses this theoretical gap by proposing and testing a dual-mediation model that distinguishes between these parallel yet distinct psychological pathways. In this model, service experience is conceptualized as the immediate, multisensory, and affective response during the service encounter, grounded in affective response theory (Mehrabian & Russell, 1974). Customer satisfaction, in contrast, is defined as a post-consumption cognitive evaluation based on expectancy-disconfirmation theory (Oliver, 1980). These constructs represent distinct psychological mechanisms—affective processing versus cognitive judgment—that may operate as parallel mediators between servicescape and behavioral outcomes. Service experience represents the holistic, affective, and sensory response to the service encounter—the immediate emotional and sensory engagement with the environment. In contrast, customer satisfaction constitutes the cognitive, evaluative judgment resulting from the comparison between pre-consumption expectations and perceived performance. Understanding how these constructs differentially mediate servicescape effects provides valuable insights into the psychological sequence through which environmental cues transform into behavioral outcomes.

Accordingly, this investigation aims to (1) estimate the effects of servicescape on both service experience (affective pathway) and customer satisfaction (cognitive pathway); (2) assess the direct and indirect effects of these organismic states on loyalty and e-WOM; and (3) test the distinct mediating roles of service experience versus customer satisfaction in transmitting servicescape effects to behavioral outcomes. By doing so, this research aims to provide a nuanced, theoretically grounded explanation of how the service environment translates into digital advocacy and loyalty through distinct affective mechanisms (service experience) and cognitive evaluations (customer satisfaction).

2. Literature Review and Hypotheses Development

2.1. Theoretical Foundation: An Integrated S-O-R, EDT, and Affective-Cognitive Framework

This study is grounded in an integrated theoretical framework that combines the Stimulus–Organism–Response (S-O-R) paradigm (Mehrabian & Russell, 1974) with Expectancy-Disconfirmation Theory (EDT) (Oliver, 1980) to delineate distinct affective and cognitive psychological pathways. This integration addresses a key limitation in prior servicescape research where, although the S-O-R framework provides a robust processual structure, the specific nature of the internal ‘Organism’ (O) states often remains underspecified. The S-O-R framework serves as the overarching structure, positing that environmental stimuli (S) trigger internal psychological processes within an organism (O), which subsequently lead to behavioral responses (R). In this context, the stimulus is operationalized as the servicescape—the designed physical and social environment of the service setting (Bitner, 1992), modeled as a multidimensional formative construct encompassing ambient conditions, spatial functionality, signs/symbols, and employee interactions. The responses comprise customer loyalty and electronic word-of-mouth (e-WOM).

This study advances beyond prior research by theorizing that service experience and customer satisfaction constitute two distinct organismic states with different theoretical origins and behavioral consequences. Customer satisfaction is positioned as the cognitive pathway, informed by Expectancy-Disconfirmation Theory (EDT) (Oliver, 1980), which conceptualizes satisfaction as a post-consumption, evaluative judgment arising from comparing perceived service performance against pre-consumption expectations. In our model, the servicescape serves as a key component of perceived performance, and when this performance meets or exceeds expectations, it results in the cognitive state of satisfaction that provides justification for subsequent loyalty and advocacy behaviors.

In contrast, service experience is conceptualized as the affective pathway, informed by theories of affective response (Mehrabian & Russell, 1974) and experiential consumption (Pine & Gilmore, 2011). Unlike the cognitive evaluation of satisfaction, service experience represents the immediate, multisensory, and emotional response during the service encounter, capturing the holistic “feel” of the visit including mood, sensory engagement, and emotional reactions to the environment (Lemke et al., 2011). The servicescape dimensions in our model—ambient conditions, spatial functionality, aesthetic elements, and employee interactions—act as affective triggers for this experiential state. It is important to note that while measurement items for the servicescape stimulus may draw from established scales in servicescape and service quality literature (e.g., Bitner, 1992; Parasuraman et al., 1988), the affective pathway itself (service experience) is measured using scales specifically designed to capture emotional and sensory responses, ensuring theoretical and operational distinction from cognitive evaluation.

This theoretical integration creates a comprehensive model for testing two parallel yet distinct mediating pathways. It clarifies that servicescape simultaneously triggers both an immediate, affective response (service experience) and a deliberative, cognitive evaluation (customer satisfaction), with each pathway offering unique contributions to fostering behavioral responses. This dual-pathway approach provides a nuanced explanation for inconsistent findings in prior servicescape literature, as models omitting one pathway may present an incomplete picture of the psychological mechanisms linking environmental design to behavioral outcomes, potentially misattributing influences or overlooking complementary processes.

2.2. Servicescape as a Multidimensional Stimulus

The servicescape concept has evolved significantly from Bitner’s (1992) original framework, which emphasized physical dimensions. Contemporary research now conceptualizes servicescape as a holistic socio-physical construct, where social elements are integral components of the service environment (Gulertekin & Genc, 2021). This expanded view acknowledges that customers respond to the total environment, with human interactions often being as influential as physical design elements.

The social servicescape dimension, embodied by employee behaviors and customer-employee interactions, has emerged as particularly crucial in experience-centric settings. Employees are not merely service providers but active co-creators of the atmospheric experience. Their appearance, demeanor, professionalism, and interpersonal skills fundamentally shape customers’ perceptions (Ali et al., 2021). In café contexts specifically, baristas and service staff contribute significantly to the social atmosphere through their warmth, competence, and ability to create personalized connections (Mora & Ueasangkomsate, 2023). Research confirms that employee-customer interactions can evoke emotional responses that either enhance or detract from the overall service experience, making employees dynamic components of the servicescape (Gulertekin & Genc, 2021).

In the highly competitive café industry, servicescape design has evolved beyond functional aspects to become a strategic tool for competitive differentiation. This study conceptualizes servicescape as a second-order formative construct comprising four interconnected dimensions: Ambient Conditions: Background sensory elements such as lighting, music, temperature, and scent that subconsciously influence mood and dwell time (Wang et al., 2025).

• Space and Functionality: The layout, furniture arrangement, and facility design that affect comfort, privacy, and movement (Jin & Xiao, 2016).
• Signs and Symbols: Intentional design elements, branding, and “Instagrammable moments” that communicate value and encourage social sharing—a critical element in café environments where visual appeal drives digital marketing through user-generated content (Situmorang et al., 2018).
• Employees: Frontline staff whose demeanor, professionalism, and interactions directly shape the social atmosphere and perceived service quality (Ali et al., 2021).

While the S-O-R framework emphasizes mediated pathways, some previous research has suggested direct links between servicescape and behavioral outcomes. Line et al. (2018) found that servicescape directly influenced loyalty and word-of-mouth in hospitality settings. However, the literature reveals significant contradictions that highlight the theoretical gap this study addresses. Inthasang et al. (2022) and Rini et al. (2021) found that servicescape’s effect on loyalty was fully mediated by customer satisfaction and customer engagement, respectively. This study posits that in experience-centric cafés, any direct influence is likely insignificant once internal psychological states are accounted for.

Hypothesis 1a.

Servicescape has a significant positive effect on Electronic Word-of-Mouth.

Hypothesis 1b.

Servicescape has a significant positive effect on Customer Loyalty.

Substantial evidence supports servicescape as a significant driver of both experiential and evaluative states. The integrated effect of physical environment and social interactions creates the foundational context that shapes customers’ affective experiences and satisfaction judgments (Kandampully et al., 2023). Research in café settings specifically demonstrates that servicescape elements strongly predict both the affective experience and cognitive satisfaction judgments (Mora & Ueasangkomsate, 2023). Ambient conditions like comfortable temperature and pleasant music create a calming context that elevates customers’ affect (Justin et al., 2016), while spatial layout and employee interactions fundamentally shape the service encounter (Jin & Xiao, 2016).

Hypothesis 1c.

Servicescape has a significant positive effect on Service Experience.

Hypothesis 1d.

Servicescape has a significant positive effect on Customer Satisfaction.

2.3. Service Experience as Affective Organism

Service experience, grounded in affective response theory (Mehrabian & Russell, 1974) and experiential consumption theory (Pine & Gilmore, 2011), represents the customer’s immediate, multisensory, and emotional response to the service encounter—capturing the holistic “feel” of the visit through affective reactions, sensory engagement, and atmospheric immersion (Lemke et al., 2011; Mehrabian & Russell, 1974). In experience-centric services such as nature-themed cafés, where competition transcends core product quality, service experience emerges as a critical differentiator that creates emotional connections beyond functional service delivery. As an affective organismic state within the S-O-R framework, service experience differs fundamentally from cognitive satisfaction by emphasizing real-time emotional and sensory processing rather than post-consumption evaluation.

The relationship between service experience and e-WOM is particularly salient in visually driven café environments designed for social sharing. Positive service experiences generate compelling “experience-driven content sharing” that motivate customers to disseminate content through digital channels. This affective-driven sharing is supported by research showing that service experience directly influences the tone and content of online reviews (Pai et al., 2013), while memorable experiences predict positive e-WOM through emotional pathways (Juliana et al., 2025). Furthermore, service experience shapes online review behavior with emotions serving as key moderators (X. Xu, 2020). In café contexts where aesthetic consumption and digital sharing constitute part of the value proposition, experiential aspects become primary drivers of electronic word-of-mouth.

Hypothesis 2a.

Service experience has a significant positive effect on Electronic Word-of-Mouth.

Service experience also serves as a foundational element for customer loyalty in café settings, where emotional connections often surpass transactional considerations. Positive experiences foster affective bonds that encourage repeat patronage and brand commitment. Research across service contexts consistently demonstrates that customers reporting positive service experiences exhibit stronger loyalty intentions (J. Xu, 2022; Saputra & Yuniarinto, 2023). Even in digital contexts, experience elements cultivated through design and functionality strengthen loyalty through mediated pathways (Saputra & Yuniarinto, 2023). For cafés targeting experience-seeking demographics, creating memorable affective experiences becomes essential for building sustained behavioral loyalty.

Hypothesis 2b.

Service experience has a significant positive effect on Customer Loyalty.

2.4. Customer Satisfaction as Cognitive Organism

Customer satisfaction represents a post-consumption evaluative judgment concerning a specific service encounter (Oliver, 1996). Rooted in Expectancy-Disconfirmation Theory (EDT), satisfaction arises when perceived service performance meets or exceeds pre-consumption expectations. This cognitive evaluation serves as a critical link between service perceptions and behavioral outcomes, particularly in competitive markets like cafés where switching costs are low.

Satisfaction is a significant driver of e-WOM behaviors, especially in service environments where social proof influences consumer decisions. Satisfied customers are motivated to share positive evaluations online, serving as organic brand advocates. Kuo and Nakhata (2019) established that satisfaction significantly predicts both the volume and valence of e-WOM, while Abidin et al. (2025) confirmed that customer satisfaction is a primary antecedent of favorable e-WOM, which meaningfully shapes brand reputation and reinforces consumer trust. In the café industry, where online reviews heavily influence venue selection, satisfaction-driven e-WOM becomes a valuable marketing asset.

Hypothesis 3a.

Customer satisfaction has a significant positive effect on Electronic Word-of-Mouth.

The relationship between satisfaction and loyalty represents one of the most empirically supported relationships in marketing literature, though it demonstrates contextual variations across industries. Satisfaction provides the fundamental rationale for repeat patronage and resistance to switching. As Thakur (2019) notes, while satisfaction alone may not guarantee loyalty, it provides a necessary foundation without which sustained loyalty is unlikely to develop. Ginting et al. (2022) empirically demonstrated that customer satisfaction has a positive and significant effect on consumer loyalty in coffee shop settings. However, in competitive markets with abundant alternatives, the satisfaction-loyalty link may require additional experiential elements to translate into sustained behavioral loyalty.

Hypothesis 3b.

Customer satisfaction has a significant positive effect on Customer Loyalty.

2.5. The Loyalty-e-WOM Relationship

Loyalty and e-WOM represent interconnected yet distinct behavioral responses that form a virtuous cycle for business growth. Loyal customers naturally evolve into brand advocates who voluntarily promote the business through digital channels. Walean et al. (2025) demonstrated that loyalty significantly predicts e-WOM behaviors, as committed customers seek to validate their choices and recruit others to the brand. This relationship underscores that true loyalty extends beyond repeat purchases to include active advocacy—a particularly valuable outcome in the café industry where personal recommendations carry substantial weight. The transition from loyal customer to brand advocate represents the culmination of successful customer relationship management in experience-driven businesses.

Hypothesis 4.

Customer loyalty has a significant positive effect on Electronic Word-of-Mouth.

2.6. The Mediating Pathways

The core theoretical contribution of this study lies in testing the mediating mechanisms through which servicescape influences behavioral outcomes. The S-O-R framework provides the theoretical rationale for expecting that environmental stimuli must first shape internal psychological states before influencing behavior.

2.6.1. Customer Loyalty as Mediator

The literature presents competing perspectives on loyalty’s role in the servicescape-eWOM relationship. While loyalty is commonly positioned as an outcome variable influenced by environmental and psychological factors (Thakur, 2019; Putra et al., 2020), some conceptual models suggest it may also function as a mediating mechanism. Research in related contexts indicates that environmental attributes can foster loyalty, which in turn motivates advocacy behaviors (Walean et al., 2025). However, empirical evidence for this mediating pathway remains inconsistent, particularly in café settings.

Recent studies in digital environments suggest alternative configurations where environmental stimuli and behavioral outcomes may operate through parallel rather than sequential pathways. Koçan and Yıldız (2025) found that e-servicescape and e-WOM can influence consumer behaviors independently, without loyalty necessarily mediating their relationship. This raises questions about whether similar patterns might exist in physical service environments like cafés, where experiential elements are particularly salient.

In light of these divergent perspectives, this study examines loyalty’s potential mediating role as an empirical question. Testing this pathway allows for a more comprehensive assessment of the mechanisms linking servicescape to e-WOM, acknowledging that multiple theoretical configurations are plausible based on existing literature.

Hypothesis 5.

Servicescape has a significant positive effect on Electronic Word-of-Mouth, with Customer Loyalty as a mediator.

2.6.2. Service Experience as Mediator

Service experience is positioned as a critical mediator that translates environmental cues into behavioral responses, particularly in experience-centric services. Research confirms that servicescape shapes the affective experience, which, in turn, drives both loyalty and advocacy behaviors. Xiong and Kim (2025) found that servicescape in sports contexts influences word-of-mouth through the mediating role of flow experience. Similarly, Suanchimplee et al. (2023) demonstrated that servicescape affects e-WOM intention through customers’ overall service experience in hospitality settings. In café environments, where atmospheric elements are deliberately designed to create specific moods and experiences, this mediating pathway is expected to be particularly strong.

Hypothesis 6a.

Servicescape has a significant positive effect on Customer Loyalty, with Service Experience as a mediator.

Hypothesis 6b.

Servicescape has a significant positive effect on Electronic Word-of-Mouth, with Service Experience as a mediator.

2.6.3. Customer Satisfaction as Mediator

Customer satisfaction similarly functions as a transmission mechanism for servicescape effects, operating through cognitive-evaluative rather than affective pathways. Multiple studies across service contexts have demonstrated that servicescape influences loyalty and e-WOM indirectly through satisfaction. Ginting et al. (2022) found customer satisfaction mediates the relationship between servicescape and loyalty in coffee shop contexts. Kazungu and Kubenea (2023) confirmed that customer satisfaction serves as a mediating mechanism connecting the service facility to word-of-mouth in higher education institutions. This pathway represents the cognitive route through which environmental quality judgments translate into behavioral intentions.

Hypothesis 7a.

Servicescape has a significant positive effect on Customer Loyalty, with Customer Satisfaction as a mediator.

Hypothesis 7b.

Servicescape has a significant positive effect on Electronic Word-of-Mouth, with Customer Satisfaction as a mediator.

In summary, while the servicescape–behavior relationship is well-established, the precise mediating mechanisms remain contested. The present study addresses key limitations in prior research by (1) conceptually and empirically distinguishing between affective (service experience) and cognitive (customer satisfaction) pathways; (2) testing a comprehensive dual-mediation model within the S-O-R paradigm; and (3) grounding these distinct pathways in established theories (SERVQUAL and EDT, respectively). This approach provides a more nuanced understanding of how environmental cues are transformed into behavioral outcomes through parallel psychological processes.

Furthermore, the proposed model (Figure 1) suggests the existence of more complex serial mediation pathways beyond the simple mediations tested in H5, H6a,b, and H7a,b. For instance, servicescape might influence electronic word-of-mouth through a sequence of mediators, such as by first enhancing service experience, which then fosters customer loyalty, ultimately leading to e-WOM (SS → SE → CL → EW). A parallel cognitive-evaluative pathway (SS → CS → CL → EW) is also theoretically plausible. Although testing these serial mediation effects is not the primary focus of this study, they will be explored post-hoc to provide a more comprehensive understanding of the overall model structure.

3. Materials and Methods

3.1. Research Design and Conceptual Framework

A quantitative, cross-sectional survey design was employed to examine the proposed structural relationships within an integrated S-O-R framework. The research model posits servicescape as a stimulus that influences two distinct organismic states—service experience (affective pathway) and customer satisfaction (cognitive pathway)—which subsequently drive behavioral responses in the form of customer loyalty and electronic word-of-mouth. The design is explanatory in nature, aiming to test the hypothesized mediation effects and clarify the psychological mechanisms through which environmental design translates into behavioral outcomes. The unit of analysis is defined as the individual café customer. Data were collected from patrons of nature-themed cafés in Nakhon Ratchasima, Thailand—a context selected for its relevance to experience-centric services where servicescape serves as a key competitive differentiator.

3.2. Population, Sample, and Data Collection

The study population comprised café customers in Nakhon Ratchasima province who had visited a nature-themed café at least once in the past three months. A non-probability sampling method, specifically purposive sampling, was employed to target eligible respondents. The minimum sample size was determined using Cochran’s formula for finite populations (95% confidence level, 5% margin of error), indicating a target of 430 respondents. Data were collected between October and December 2024 through a combination of intercept sampling at participating cafés and online dissemination to relevant community groups, with a total of 430 questionnaires distributed to account for potential data cleaning. The survey instrument was initially developed in English and translated into Thai through a translation and back-translation procedure to ensure conceptual equivalence, with content validity established through expert review and a pilot test with 30 respondents. All participants provided informed consent, and the study protocol was approved by the Institutional Review Board of Suranaree University of Technology (Protocol Code: COE 144/2567; Date of Approval: 8 October 2024) in accordance with the Declaration of Helsinki.

The survey consisted of two sections: (1) demographic and behavioral information, where gender was measured with a multiple-choice question (Male, Female, LGBTQIA+), along with other standard demographic variables, and (2) core constructs measured using established scales adapted from literature (full measurement details are provided in the Methodology section). Servicescape was measured with 15 items (Bitner, 1992; Gulertekin & Genc, 2021; Situmorang et al., 2018), service experience with 5 items (Lemke et al., 2011; Kandampully et al., 2023), customer satisfaction with 4 items (Oliver, 1996; Kuo & Nakhata, 2019), customer loyalty with 4 items (Oliver, 1996; Zeithaml et al., 1996), and electronic word-of-mouth with 4 items (Hennig-Thurau et al., 2004; King et al., 2014). All perceptual items used five-point Likert scales. Data validation included assessment of missing patterns and outlier detection using Mahalanobis Distance (p < 0.001) in IBM SPSS Statistics (Version 26). After removing 10 invalid records, the final cleaned dataset comprised 420 valid responses, which remains above the generally accepted threshold of 400 for SEM analysis and provides adequate statistical power for our model.

3.3. Data Analysis

Data analysis was conducted using IBM SPSS Statistics (Version 26) and AMOS (Version 24). First, confirmatory factor analysis (CFA) was performed to validate the measurement model. Construct validity and reliability were assessed with the following criteria: standardized factor loadings ≥ 0.70, composite reliability (CR) ≥ 0.70, Cronbach’s alpha ≥ 0.70, and average variance extracted (AVE) ≥ 0.50 (Hair et al., 2019). Discriminant validity was established, as the average variance extracted (AVE) for each construct was greater than its maximum shared variance (MSV) and average shared variance (ASV).

Second, structural equation modeling (SEM) was employed to test the hypothesized relationships using maximum likelihood estimation. Model fit was evaluated using multiple indices: chi-square statistic, root mean square error of approximation (RMSEA), comparative fit index (CFI), and Tucker–Lewis index (TLI). Excellent model fit was determined by RMSEA < 0.06, CFI > 0.95, and TLI > 0.95 (Hu & Bentler, 1999).

Finally, mediation analysis was conducted to test the indirect effects hypothesized in H5 through H7b. The significance of mediation effects was determined by examining the statistical significance of the indirect path coefficients using the product of coefficients approach (MacKinnon et al., 2002).

4. Results

4.1. Demographic of Customers

The demographic characteristics of the 420 respondents are presented in

Table 1. Demographic Profile of Nature-themed Café Patrons in Nakhon Ratchasima.

Items	Details	Frequency	Percentage
Sex	Male	101	24.05
	Female	291	69.29
	LGBTQIA+	28	6.67
Age	20–29 Years	201	47.86
	30–39 Years	116	27.62
	40–49 Years	61	14.52
	More than 50 Years	42	10.00
Marital Status	Single	271	64.52
	Marry	130	30.95
	Divorce	19	4.52
Education	Primary education	4	0.95
	Secondary education	52	12.38
	Bachelor’s degree	317	75.48
	Master’s degree	25	5.95
	Doctor of Philosophy	8	1.90
	Diploma	14	3.33
Occupation	Student	70	16.67
	Employee	123	29.29
	Civil servants/State enterprise employees	90	21.43
	Housekeeper/Househusband	23	5.48
	Private business/freelance/trading	114	27.14
Monthly income	Less than 285 USD	66	15.71
	286–570 USD	167	39.76
	571–857 USD	86	20.48
	More than 857 USD	101	24.05

Source: Author.

. The sample consisted of 24.05% males, 69.29% females, and 6.67% identifying as LGBTQIA+. Regarding age distribution, 47.86% were 20–29 years old, 27.62% were 30–39 years, 14.52% were 40–49 years, and 10.00% were over 50 years. Most participants were single (64.52%), with 30.95% married and 4.52% divorced. Educational attainment was predominantly at the bachelor’s degree level (75.48%), with 12.38% having secondary education, 5.95% holding master’s degrees, 1.90% doctoral degrees, 3.33% diplomas, and 0.95% primary education. Occupationally, 29.29% were salaried employees, 27.14% were in private business/freelance/trading, 21.43% were civil servants/state-enterprise employees, 16.67% were students, and 5.48% were housekeepers/househusbands. Monthly income distribution showed 39.76% earning 286–570 USD, 24.05% earning more than 857 USD, 20.48% earning 571–857 USD, and 15.71% earning less than 285 USD.

Usage behavior patterns (

Table 2. Usage Behavior Patterns of Nature-themed Café Patrons in Nakhon Ratchasima.

Items	Details	Frequency	Percentage
On average, how many times per month do you visit a coffee shop?	1–2 Time/month	140	33.33
	3–4 Time/month	111	26.43
	5–6 Time/month	63	15.00
	More than 6 Time/month	106	25.24
What time of day do you visit the coffee shop most often?	5:00–8:59 a.m.	32	7.62
	9:00–12:00 a.m.	125	29.76
	12:01–13:00 p.m.	104	24.76
	13:01–16:00 p.m.	134	31.90
	After 16:00 p.m.	25	5.95
How long do you spend in total at a coffee shop per visit?	Less than 30 min	141	33.57
	31 min–1 h	162	38.57
	1–2 h	89	21.19
	More than 2 h	28	6.67
The amount of money spent at a coffee shop per visit	Less than 2.85 USD/time	127	30.24
	2.86–8.57 USD/time	225	53.57
	8.58–14.28 USD/time	37	8.81
	More than 14.28 USD/time	31	7.38

Source: Author.

) indicated that 33.33% of respondents visited cafés 1–2 times per month, 26.43% visited 3–4 times, 15.00% visited 5–6 times, and 25.24% visited more than 6 times. The most common visiting times were 13:01–16:00 (31.90%), followed by 09:00–12:00 (29.76%), 12:01–13:00 (24.76%), 05:00–08:59 (7.62%), and after 16:00 (5.95%). Visit duration was typically 31 min to 1 h (38.57%) or less than 30 min (33.57%), with 21.19% staying 1–2 h and 6.67% staying more than 2 h. Expenditure per visit was primarily 2.86–8.57 USD (53.57%), with 30.24% spending less than 2.85 USD, 8.81% spending 8.58–14.28 USD, and 7.38% spending more than 14.28 USD.

4.2. Reliability Testing

The measurement model demonstrated excellent psychometric properties across all constructs, confirming their reliability and validity for subsequent structural analysis. As presented in

Table 3. Convergent Validity and Discriminant Validity.

Constructs	Items	Factor Loading	CR	Cronbach’s Alpha	AVE	r2	MSV	ASV
Servicescape	SS1: Ambient Conditions	0.919	0.948	0.950	0.819	0.845	0.638	0.603
	SS2: Space and Functinionality	0.918				0.842
	SS3: Signs and Symboe	0.886				0.784
	SS4: Employees	0.896				0.802
Service Experience	SE1: Aappealing Design & Atmosphere	0.801	0.906	0.906	0.658	0.641
	SE2: Relaxing Environment & Service	0.835				0.697
	SE3: Perceived Value & Revisit Intention	0.817				0.668
	SE4: Positive Social Media Sharing	0.786				0.617
	SE5: Positive Social Interaction	0.816				0.666
Customer Satisfaction	CS1: Pleasant Ambiance	0.802	0.887	0.887	0.663	0.643
	CS2: Functional Layout & Amenities	0.832				0.693
	CS3: Clear Signage	0.813				0.661
	CS4: Competent Staff	0.809				0.655
Customer Loyalty	CL1: Positive Service Experience	0.823	0.898	0.900	0.687	0.677
	CL2: Revisit Intention	0.821				0.674
	CL3: Price Insensitive Loyalty	0.818				0.669
	CL4: First-Choice Advocacy	0.854				0.730
Online Word of Mouth	EW1: Positive Online Reviews	0.814	0.898	0.895	0.689	0.662
	EW2: Online Brand Awareness	0.822				0.675
	EW3: Visual Sharing Motivation	0.826				0.683
	EW4: Influence by Online Content	0.857				0.734

Source: Author.

, all standardized factor loadings ranged from 0.786 to 0.919, substantially exceeding the 0.70 threshold for indicator reliability. Construct reliability was firmly established with both composite reliability (CR: 0.887–0.948) and Cronbach’s alpha values (0.887–0.950) well above the 0.70 benchmark, while convergent validity was confirmed through average variance extracted (AVE) values ranging from 0.658 to 0.819, all surpassing the 0.50 criterion. The servicescape construct showed particularly strong measurement properties with factor loadings up to 0.919, CR of 0.948, and AVE of 0.819.

Discriminant validity was established by examining the relationships between the AVE, maximum shared variance (MSV), and average shared variance (ASV). The results confirmed that each construct’s AVE was greater than its corresponding MSV and ASV. For instance, the servicescape construct demonstrated an AVE of 0.819, which was greater than its MSV (0.638) and ASV (0.603). This pattern held for all other constructs, providing evidence that they are empirically distinct. The comprehensive validation confirms that all constructs collectively provide a solid psychometric foundation for proceeding with the structural model analysis and hypothesis testing in the subsequent sections.

4.3. Structural Equation Modeling Analysis—Testing the Integrated S-O-R Framework

The structural equation model, visually represented in Figure 2 and conceptualized within the Stimulus–Organism–Response (S-O-R) paradigm, showed good fit to the empirical data. The model fit indices (χ² = 186.109, df = 160, p = 0.007; RMSEA = 0.020; CFI = 0.997; TLI = 0.995) met or exceeded recommended thresholds, suggesting adequate alignment between the proposed theoretical framework and observed data patterns.

4.3.1. Stimulus to Organism Pathways: Dual Mechanisms Confirmed

The analysis revealed that the servicescape stimulus activated both hypothesized organismic pathways in parallel. For the affective pathway, servicescape exerted a significant influence on service experience (β = 0.805, p < 0.001), demonstrating that environmental cues directly shape customers’ emotional and sensory engagement. Simultaneously, for the cognitive pathway, servicescape showed an equally strong effect on customer satisfaction (β = 0.816, p < 0.001), confirming that physical and social environmental elements drive post-consumption evaluative judgments.

4.3.2. Organism to Response Pathways: Distinct Behavioral Influences

Both organismic states significantly predicted behavioral responses yet exhibited distinct influence patterns. The affective pathway demonstrated that service experience significantly influenced both customer loyalty (β = 0.405, p < 0.001) and electronic word-of-mouth (β = 0.205, p = 0.004). Comparatively, the cognitive pathway revealed that customer satisfaction exerted stronger effects on both loyalty (β = 0.474, p < 0.001) and e-WOM (β = 0.250, p = 0.002).

4.3.3. Hypothesis Testing Summary: Direct Effects

The hypothesis testing results, comprehensively summarized in

Table 4. Hypothesis Testing Results for Direct Paths.

Hypothesis	Paths	Path Coefficient	p-Value	Relationship
H1a	Servicescape → Online Word of Mouth	0.125	0.148	Not Supported
H1b	Servicescape → Customer Loyalty	0.042	0.638	Not Supported
H1c	Servicescape → Service Experience	0.805 ***	<0.001	Supported
H1d	Servicescape → Customer Satisfaction	0.816 ***	<0.001	Supported
H2a	Service Experience → Online Word of Mouth	0.205 **	0.004	Supported
H2b	Service Experience → Customer Loyalty	0.405 ***	<0.001	Supported
H3a	Customer Satisfaction → Online Word of Mouth	0.250 **	0.002	Supported
H3b	Customer Satisfaction → Customer Loyalty	0.474 ***	<0.001	Supported
H4	Customer Loyalty → Online Word of Mouth	0.349 ***	<0.001	Supported

Source: Author. Note: ** Sig at 0.01 level, *** Sig at 0.001 level.

, revealed a clear pattern. H1a and H1b proposing direct servicescape effects on behavioral outcomes were not supported (β = 0.125, p = 0.148 for e-WOM; β = 0.042, p = 0.638 for loyalty). In contrast, all hypotheses concerning the organismic pathways received strong empirical support. H1c and H1d, positing servicescape’s influence on the organismic states, were strongly supported (β = 0.805 and 0.816, p < 0.001). The organismic-behavioral pathways (H2a, H2b, H3a, H3b) and the loyalty-eWOM relationship (H4) all demonstrated significant effects.

4.3.4. Mediation Pattern Evidence

The consistent pattern of non-significant direct paths from servicescape to behavioral outcomes, coupled with strong indirect pathways through organismic states, provides initial evidence for mediation effects. This pattern sets the foundation for the comprehensive mediation analysis presented in the following section.

4.4. Mediation Analysis—Testing Indirect Pathways

The mediation analysis, presented in

Table 5. Mediation Testing Results for Indirect Effects.

Hypothesis	Paths	Path Coefficient	p-Value	Mediation	Relationship
H5	Servicescape → Customer Loyalty → Online Word of Mouth	0.016	0.614	No	Not Supported
H6a	Servicescape → Service Experience → Customer Loyalty	0.348 ***	0.000	Full	Supported
H6b	Servicescape → Service Experience → Online Word of Mouth	0.181 *	0.022	Full	Supported
H7a	Servicescape → Customer Satisfaction → Customer Loyalty	0.413 ***	0.000	Full	Supported
H7b	Servicescape → Customer Satisfaction → Online Word of Mouth	0.224 *	0.013	Full	Supported

Source: Author; Note: * Sig at 0.05 level, *** Sig at 0.001 level.

, tested the hypothesized indirect pathways through which servicescape influences behavioral outcomes. The results are consistent with the dual mediation mechanism proposed in the integrated S-O-R framework.

4.4.1. Service Experience as Mediator: Affective Pathway Confirmed

The analysis confirmed service experience as a significant mediator in both behavioral pathways. For the servicescape-loyalty relationship (H6a), the indirect effect through service experience was substantial and statistically significant (β = 0.348, p < 0.001), indicating full mediation. Similarly, for the servicescape-eWOM relationship (H6b), service experience demonstrated a significant mediating role (β = 0.181, p = 0.022), confirming that the affective pathway transmits environmental influences to both behavioral outcomes.

4.4.2. Customer Satisfaction as Mediator: Cognitive Pathway Established

Parallel to the affective pathway, customer satisfaction functioned as a significant mediator in both behavioral relationships. For the servicescape-loyalty link (H7a), the indirect effect through customer satisfaction was particularly strong (β = 0.413, p < 0.001), demonstrating full mediation. The servicescape-eWOM relationship (H7b) likewise showed significant mediation through customer satisfaction (β = 0.224, p = 0.013), establishing the cognitive pathway as a fundamental transmission mechanism.

4.4.3. Loyalty as Mediator: Non-Significant Pathway

In contrast to the organismic mediators, the hypothesized mediation through customer loyalty (H5) was not supported. The indirect effect of servicescape on eWOM through loyalty was trivial and statistically non-significant (β = 0.016, p = 0.614), indicating that loyalty does not function as a mediating mechanism in the servicescape-eWOM relationship.

4.4.4. Total Indirect Effects and Mediation Patterns

The combined indirect effects reveal compelling patterns: servicescape influences customer loyalty primarily through both service experience and customer satisfaction (total indirect effect = 0.713), while its influence on eWOM operates through multiple pathways including both organismic states and their subsequent effects on loyalty (total indirect effect = 0.632). These findings confirm that the relationship between servicescape and behavioral outcomes is fully mediated by the parallel affective and cognitive pathways, with no significant direct effects remaining once these mediating mechanisms are accounted for in the model.

5. Discussion

This study proposes and tests a dual-mediation model that delineates the distinct psychological mechanisms through which servicescape influences customer loyalty and electronic word-of-mouth. The findings offer empirical support for the full mediation effects of service experience (affective pathway) and customer satisfaction (cognitive pathway), thereby presenting a potential resolution to theoretical ambiguities in the servicescape–behavior relationship.

5.1. Reconciliation of Contradictory Empirical Evidence

The non-significant direct effects from servicescape to e-WOM (H1a) and loyalty (H1b) may provide a plausible explanation for the mixed findings in prior research. While some studies reported direct servicescape effects (Line et al., 2018), others found fully mediated relationships (Inthasang et al., 2022; Rini et al., 2021). Our results suggest that these apparent contradictions may stem from model specification differences. Specifically, prior studies reporting direct effects may have omitted one or both of the key psychological pathways examined here. When both affective (service experience) and cognitive (customer satisfaction) mediators are simultaneously modeled, they fully account for the influence of servicescape on behavioral outcomes, suggesting that environmental cues are processed internally before manifesting as loyalty or e-WOM.

This finding aligns with the core premise of the S-O-R framework, which posits that environmental stimuli (S) must first affect internal organismic states (O) before eliciting behavioral responses (R). The non-significant direct paths in our model indicate that in experience-centric café settings, servicescape operates primarily through psychological mechanisms rather than exerting unmediated behavioral influence. However, we acknowledge that contextual factors—such as industry type, cultural setting, or service involvement—may moderate this relationship. For instance, in low-involvement or utilitarian services, environmental cues might directly shape behavior without intensive affective or cognitive processing. Thus, while our dual-mediation model helps clarify inconsistencies in extant literature, further research is needed to establish its boundary conditions across different service domains and cultural contexts.

5.2. Dual Pathways: Affective Versus Cognitive Mechanisms

The strong effects of servicescape on both service experience (H1c: β = 0.805, p < 0.001) and customer satisfaction (H1d: β = 0.816, p < 0.001) underscore the environment’s fundamental role in shaping customer perceptions, consistent with contemporary research recognizing servicescape as a holistic socio-physical construct (Dedeoglu et al., 2018). More importantly, the differential influence patterns lend support to the distinct theoretical nature of each pathway, as conceptualized in our integrated S-O-R, EDT, and Affective-Cognitive framework.

The significant but comparatively modest effects of service experience on loyalty (H2b) and e-WOM (H2a) are consistent with affective response theory’s emphasis on immediate emotional and sensory engagement (Mehrabian & Russell, 1974), rather than with post-consumption cognitive evaluations. This affective pathway appears to create memorable experiences that customers are motivated to share through digital channels, particularly in visually driven industries like cafés where social media sharing is prevalent (X. Xu, 2020). The finding that service experience directly drives e-WOM aligns with previous research demonstrating that positive experiences create “stories worth sharing” (Pai et al., 2013; Juliana et al., 2025).

In contrast, the stronger effects of customer satisfaction on both loyalty (H3b) and e-WOM (H3a) support EDT’s proposition that satisfaction constitutes a rational, post-consumption evaluation that provides stronger justification for behavioral commitments (Oliver, 1980). This cognitive pathway reflects a more deliberative assessment of whether the servicescape meets or exceeds expectations. The observed satisfaction–loyalty relationship aligns with established literature in marketing, while its link to e-WOM suggests satisfaction can also motivate evaluative advocacy (Kuo & Nakhata, 2019). In competitive café markets where switching costs are often low, satisfaction may provide a cognitive basis for repeat visitation.

The dual-pathway model extends experiential marketing frameworks by examining both affective and cognitive processes concurrently. While previous studies have often focused on one pathway, our findings suggest both can operate in parallel with distinct behavioral influences. This encourages a more integrated approach to understanding how environmental cues are processed in experience-rich service settings.

5.3. Clarification of Behavioral Sequence and Mediation Patterns

The significant parallel mediation effects through both service experience (H6a, H6b) and customer satisfaction (H7a, H7b), coupled with the non-significant mediation through customer loyalty (H5), offer insights into the behavioral sequence. This pattern suggests that loyalty and e-WOM, while interrelated (H4: β = 0.349, p < 0.001), represent distinct behavioral responses that are independently stimulated by the same underlying psychological states rather than operating in a sequential chain where loyalty mediates the environment-eWOM relationship.

The non-significant mediation through loyalty (H5: β = 0.016, p = 0.614) indicates that, within our model, loyalty does not act as a transmission mechanism for servicescape effects on e-WOM. This finding invites further theoretical reflection: while loyalty is often positioned as an antecedent to advocacy in sequential behavioral models (Thakur, 2019; Walean et al., 2025), our results suggest that in experiential café settings, service experience and satisfaction may simultaneously foster both loyalty intentions and e-WOM behaviors independently. This implies that customers might engage in digital sharing as an immediate emotional or evaluative response, without requiring the prior development of loyalty as a necessary step. Such a pattern aligns with emerging perspectives in digital consumption, where sharing can be an intrinsic part of the experience rather than a post-loyalty outcome (Koçan & Yıldız, 2025).

The observed full mediation patterns are consistent with recent studies emphasizing the role of organismic states in transmitting environmental influences. Our findings corroborate recent evidence that servicescape affects e-WOM intention through customers’ overall service experience (Suanchimplee et al., 2023) and that customer satisfaction serves as a mediating mechanism connecting service facilities to word-of-mouth (Kazungu & Kubenea, 2023). By testing both affective and cognitive pathways simultaneously, this study provides a more comprehensive understanding of the psychological processes through which designed environments translate into digital advocacy and loyalty behaviors.

These findings challenge strictly hierarchical models of customer behavior and support a more parallel processing view, where affective and cognitive responses jointly—and independently—shape multiple forms of behavioral engagement. This encourages future research to consider loyalty and e-WOM not only as sequential outcomes but as potentially co-occurring manifestations of underlying psychological states.

5.4. Theoretical Focus on the Service Environment

This study was theoretically delimited to examine the effects of the servicescape, independent of core product attributes. Our findings suggest that the physical and social environment, through its distinct affective and cognitive pathways, can meaningfully contribute to positive customer outcomes. This points to the potential value of investments in the servicescape as a strategic lever for managers in experience-centric venues, even when product-related factors are held constant.

The significant effects observed through both service experience and satisfaction underscore that environmental elements—ranging from ambient design to social interactions—are not merely peripheral, but central to customer perception and behavior in services where the experience itself is a key component of the offering. This supports the view that in certain service contexts, especially those oriented toward leisure, socialization, or aesthetic consumption, the environment can act as a primary driver of customer responses.

It should be noted, however, that this focus does not imply that product quality is unimportant. Rather, it highlights that in experience-dominant settings such as nature-themed cafés, a well-crafted servicescape can independently foster positive psychological and behavioral outcomes. This perspective encourages a more nuanced approach to service design, where environmental and social elements are strategically developed alongside product excellence to create a holistic value proposition.

6. Conclusions

6.1. Theoretical Contributions

This study contributes to servicescape research. First, it resolves the theoretical ambiguity in the servicescape–behavior relationship by demonstrating that service experience and customer satisfaction function as parallel yet distinct mediating mechanisms. The finding that both pathways fully mediate the relationship explains why previous studies reported inconsistent results, with some finding direct effects (Line et al., 2018), while others found mediated relationships (Inthasang et al., 2022). By specifying the precise nature of the “Organism” in the S-O-R framework, this study addresses the problem of unclear internal mechanisms that has long limited theoretical advancement in servicescape research.

Second, this research provides empirical validation for an integrated theoretical framework that combines the S-O-R paradigm with Expectancy-Disconfirmation Theory (EDT) and affective response theory. The differential effects observed—with service experience capturing immediate affective responses and customer satisfaction reflecting post-consumption cognitive evaluations—demonstrate the value of distinguishing between these psychological processes. This approach offers a more nuanced understanding of how environmental cues are processed through both emotional and rational pathways.

Third, this study extends servicescape theory into the digital realm by establishing clear psychological pathways from physical environment to electronic word-of-mouth (e-WOM). As digital advocacy increasingly influences consumer behavior (Kandampully et al., 2023), understanding how designed environments stimulate online sharing through distinct affective and cognitive mechanisms represents a meaningful theoretical contribution to services marketing literature.

6.2. Practical Implications

The findings offer several actionable insights for café managers and service designers:

• Dual-Pathway Design: Servicescape elements should address both affective and cognitive pathways. Atmospheric features (e.g., nature-themed designs, Instagrammable spots) primarily enhance service experience and stimulate digital sharing, while functional elements (e.g., layout efficiency, employee professionalism) more directly influence satisfaction evaluations. This distinction enables targeted resource allocation based on strategic objectives.
• Employee Development: Service staff should be trained to support both pathways simultaneously. Beyond functional competence, employees should develop interpersonal skills that create positive emotional experiences, as the social dimension of servicescape significantly influences both affective and cognitive responses.
• Strategic Monitoring: Service experience and customer satisfaction should be monitored as leading indicators of behavioral outcomes. Given their full mediation role, changes in these psychological states will manifest in loyalty and e-WOM, providing early signals for managerial intervention.
• Experience Integration: Environments should balance visual appeal with functional comfort to encourage both immediate sharing and sustained patronage. The finding that both pathways influence e-WOM suggests that aesthetically pleasing yet practically comfortable spaces optimize organic digital marketing through user-generated content.

6.3. Limitations and Future Research Directions

Several limitations should be considered when interpreting these findings. First, the cross-sectional design limits causal inference, though it aligns with the theoretical directionality proposed in the S-O-R framework. Future research could employ longitudinal designs or experimental manipulations to strengthen causal claims.

Second, while our dual-mediation model helps reconcile previous inconsistencies, other untested variables (e.g., brand attachment, cultural factors) or alternative model specifications might offer complementary explanations. Future studies should explore additional moderators and mediators to provide a more comprehensive understanding.

Third, the gender classification used in this study, while inclusive, groups diverse identities within broad categories. Future research should employ more granular measures to examine potential moderating effects of gender identity on the psychological pathways identified here.

Fourth, the context-specific nature of this study—conducted in nature-themed Thai cafés—may limit generalizability. Replications in different service settings (e.g., utilitarian services) and cultural contexts would help establish boundary conditions for the proposed model.

Finally, reliance on self-reported behavioral intentions represents a common limitation in servicescape research. Future studies would benefit from incorporating objective behavioral metrics, such as actual revisit frequency or content analysis of online reviews, to validate the intention–behavior relationship within this dual-mediation framework.