An Integrated Model for Sustainable Customer Loyalty: Drivers in Thailand’s High-End Home-Building Industry

Abstract

The home-building industry is crucial to Thailand’s economic growth but unfortunately it is plagued by persistent sustainability problems stemming from fierce competition, cost fluctuations, and evolving consumer expectations. Therefore, this study sought to identify the main drivers of sustainable customer loyalty (CL) in the high-end home-building industry by integrating marketing communications strategy (MCS), service quality (SQ), and company management system (CMS) into a unified structural equation model (SEM). Data were collected from 680 customers registered with Thailand’s Home Builders Association through online and in-person surveys. Path analysis of latent variables using LISREL 9.10 showed that the model had strong explanatory power, explaining 72% of the variance of CL (R² = 0.72). MCS had the greatest total effect (β = 0.82), followed by SQ (β = 0.54) and CMS (β = 0.24). The results also showed how transparent communication interventions add to perceived quality while increasing management efficiency which jointly sustains long-term CL. These findings confirm that strategic communication, service excellence, and good corporate management provide the basis for the sustainable competitive advantage and economic resilience of Thailand’s home-building firms. The validated framework provides theoretical and managerial implications for incorporating sustainability into customer relationship strategies.

1. Introduction

The construction industry has always been an important driver for the economic growth of Thailand, whose contribution has often been used as an indicator of Thailand’s overall economic and sustainable growth [1]. Real estate accounts for approximately 10% of national GDP [2] and, as of 2025, the residential sector represents nearly two-thirds of the total market value [3]. Current projections for construction growth stand at 3–5.9%, supported by investments in tourism-related infrastructure, foreigner housing, digital economy modernization, and exports [4].

Despite its economic weight, the residential sector is under strain from multiple pressures. The number of new housing units entering the Bangkok Metropolitan Region (BMR) slumped by −37.8% YoY in 2024, and household debt [5]. Coupled with tighter lending criteria has raised mortgage rejection rates to as high as 60% for properties valued below 3 million ฿ [6].

Although Thailand’s luxury home market (5–20 million ฿) still has demand in the high-end segment, weakening foreigner purchasing power, especially from Chinese, with natural disaster concerns, has dampened market confidence [7]. More broadly, the sector faces structural challenges with macroeconomic headwinds coming from slowing global trade and weakened tourism recovery, rising construction costs from land price inflation and chronic labor shortages, and demographic shifts primarily from Thailand’s transition to an ultra-aged society that is reshaping housing needs [8]. Moreover, there are sustainability imperatives, with developers under pressure to adopt green building practices [9,10], reduce carbon emissions, and comply with the Thailand Taxonomy Phase 2 Environmental, Social, and Governance (ESG) framework [7].

Meanwhile, the concept of “home” is changing. The experience of going through two crises and the trend of remote work from home make sustainable, flexible, and healthy living a requirement, as consumers want homes that have better energy efficiency, are constructed with eco-friendly materials, and are equipped with smart-home devices. Developers cater to these needs and proactively respond to the country’s increasing proportion of older adults, using universal design to build homes for them and introducing new models, including shared centers for seniors to live in together [7,11,12].

In this turbulent ecosystem, customer loyalty building has become the prominent strategy. Instead of retreating from the market, home builders who transitioned into the premium niche market without price competition, needed to stay ahead in trust, sustainability, and relationship quality in the mind of customers [13]. Under these conditions, the present study examines which marketing communications, service quality, and management systems drive sustainable customer loyalty in Thailand’s high-end home building sector.

2. Literature Review and Hypotheses Development

This section presents the conceptual framework for the study’s model development, which includes marketing communication strategy (MCS), service quality, company management system, and customer loyalty, along with five conceptual hypotheses.

2.1. Marketing Communications Strategy (MCS): Generating Credibility and Trust in Sustainability-Conscious Markets

In Thailand’s high-end housing market, MCS transcends its traditional promotional role to become a critical signal of credibility and long-term value for a discerning, sustainability-conscious clientele. The following dimensions are paramount in this context.

While MCS has traditionally focused on persuasion and brand positioning [14,15], the strategy is increasingly used to convey a firm’s sustainability credentials. In this context, developers are responsible for developing tailored messages responding to increased consumer awareness of energy consumption, natural construction materials, and environmental effects [16,17,18,19]. This aligns with recent research demonstrating the sustainability focus of MCS strategies to provide developers with competitive edges in areas of corporate image, exposure to green finance opportunities, and reduced ongoing operating costs [7].

This means that their luxury housing communication methods are no longer just transactional but central to asserting fulfilling ESG certifications [16,18] and new governmental directions, including building energy code (BEC) and floor area ratio (FAR) incentives [7,20]. These sustainability-focused marketing approaches are imperative for high-end consumers who demand visibility and affirmation that their homes bring a positive long-term environmental and social value [17].

The study combined MCS into five key dimensions following research from Mocanu and Szakal [21] and Kotler and Armstrong [14]. These include: (1) advertising, (2) personal selling, (3) sales promotion, (4) public relations, and (5) direct marketing. Specifically, advertising refers to any paid form of nonpersonal presentation and promotion of offerings by an identified sponsor using paid, nonpersonal media [22]. Personal selling relates to communicating between sellers and customers to facilitate mutually satisfying exchange of information, products/services, and/or funds [23]. Sales promotion is associated with short-term incentives that encourage purchasing a service or product [24]. Public relations is concerned with good relations with consumers and the public through favorable publicity, building a good corporate image, and handling or heading off unfavorable rumors, stories, and events [25]. Finally, direct marketing is concerned about connecting directly with carefully targeted consumers to obtain immediate response, and cultivate long-term customer relationships [14,21].

In the Thai high-end housing market, such generic dimensions have specific connotations. Advertising is not about awareness in general but about communicating particular ESG certifications (e.g., LEED, TREES) and certain health-conscious architectural details in premium magazines and digital media catering to Thailand’s wealthy. Personal selling is essential as well in establishing a deep, trust-based relationship to sustain the long residential sales cycle, requiring salespersons to also be a consultant to the individual or family about sustainable materials and about aging-in-place design. Public relations are important from the perspective of managing the developer’s reputation, particularly for reliability and social responsibility, a significant concern of buyers making a once-in-a-lifetime investment. Direct marketing targets must be highly individualized, often in the form of exclusive events, demonstrating a personalized understanding of the client’s particular lifestyle and any particular values that pertain to sustainability and to family heritage.

However, a missing knowledge about how Thai homebuilders communicated messages about sustainability orientations focusing on energy efficiencies, eco-materials, and healthy living suggests the trustworthiness and loyalty among discerning and affluent consumers [19]. Accordingly, the study develops the following research hypotheses:

H1: 

Marketing communications strategy (MCS) directly and positively influences customer loyalty (CL) in Thailand’s high-end housing market.

H2: 

Marketing communications strategy (MCS) directly and positively influences service quality (SQ) in Thailand’s high-end housing market.

2.2. Service Quality (SQ): Expanding into Health- and Aging-Oriented Design

Service quality in the Thai residential sector has long been associated with reliability, responsiveness, and assurance [26]. However, recent structural changes in demographics necessitate an expansion of SQ definitions. With Thailand’s transition into a “completed aged society,” universal design principles [27] and health-supportive facilities (e.g., tech-enabled health monitoring, and nursing care integration) are becoming essential elements of service quality [7].

Furthermore, post-crisis consumer preferences highlight sustainable and health-conscious attributes as part of perceived SQ [17]. Developers repurposing condominiums into senior living centers illustrate this expanded role of SQ, where sustainability, health, and inclusivity are as important as traditional service dimensions in generating customer trust and loyalty.

This study employs the five dimensions of the SERVQUAL model [28,29]. These include responsiveness, assurance, tangibles, empathy, and reliability (RATER), to measure service quality [23,29,30]. The key research gap lies in investigating how this broader, more holistic view of SQ mediates customer loyalty, including sustainable and health-focused attributes [17].

The SERVQUAL (RATER) model in this market requires context-driven interpretation. Tangibles go beyond aesthetic finishes to the quality of sustainable materials (reclaimed teak, non-VOC paints, etc.) and integrated health monitoring technology. Reliability is measured in terms of the developer delivering on promises about energy efficiency and project timescales in a market notorious for late delivery. Responsiveness encompasses provision of fast, expert answers to technically complex queries about building science and smart home integration. Assurance comes from the credentials of architecture and engineering professionals and the company’s track record with green building projects. Empathy is shown through understanding of Thai cultural norms, for multi-generational living and for a client wanting a home that reflects their status and values.

Accordingly, the study develops the following research hypotheses:

H3: 

Service quality (SQ) directly and positively influences customer loyalty (CL) in Thailand’s high-end housing market.

H4: 

Service quality (SQ) directly and positively influences company management systems (CMS) in Thailand’s high-end housing market.

2.3. Company Management System (CMS): Operational Foundations for ESG Compliance

Company management systems (CMS) continue to be the core enabler of dependable delivery of housing projects [31,32], but nowadays they are also increasingly mediated by sustainability governance frameworks. The implementation of Thailand Taxonomy Phase 2 compels developers to adhere to more demanding ESG standards, cut carbon footprints throughout a building’s lifecycle, and maximize efficiency of resource utilization [7,33], placing additional demands on such internal management practices as environmental management systems, health-and-safety compliance, and risk control mechanisms.

For high-end developers, an effective CMS secures operational reliability and unlocks access to green finance, which has become a key competitive differentiator [7]. Hence, CMS now embodies an operational and a strategic dimension: a foundation for project execution and a gateway to sustainability-driven competitiveness.

The standard dimensions of a Company Management System are redefined by the demands of the high-end Thai market. Financial Status is not just about solvency but signals the capacity to secure green financing and absorb cost overruns without compromising on premium, sustainable materials. Corporate Organization must be adept at navigating Thailand’s specific building codes and ESG incentives (e.g., Thailand Taxonomy) [7,33]. Human Resources requires retaining specialized talent—architects and project managers skilled in sustainable design and familiar with local regulations and craftsmanship. Management and Control Techniques are paramount for coordinating complex supply chains for imported eco-materials and ensuring meticulous quality control that meets the exacting standards of affluent clients.

Therefore, this research examines the effectiveness of a CMS through six dimensions critical to sustainable operations. These included financial status, corporate organization, work experience, human resources, tools and equipment, and management and control techniques [34,35]. The latter dimension is particularly relevant to sustainability, encompassing systems for environmental management, quality control, and technological integration. The literature reveals a gap in exploring how these internal systems directly influence customer perceptions of reliability and loyalty in the high-end housing market. Accordingly, the study develops the following research hypothesis:

H5: 

A company management system (CMS) directly and positively influences customer loyalty (CL) in Thailand’s high-end housing market.

2.4. Customer Loyalty (CL): Sustainability as a Core Driver

Customer loyalty in high-involvement services reflects a voluntary long-term commitment, which is critical in Thailand’s high-end housing market where purchases represent significant financial and social investments [36]. Therefore, this study measures loyalty through three established behavioral indicators, including willingness to recommend (word-of-mouth), repurchase intention, and price tolerance [36].

However, within the unique context of Thailand’s sustainability-conscious, high-net-worth clientele, these metrics take on distinct changes. Willingness to Recommend is a powerful driver within tight-knit social circles but is contingent on the client’s pride in the home’s distinctive architectural and sustainability features [37]. Repurchase Intention, while rare for a primary residence, translates into future intentions to use the same developer for renovations, commercial projects, or recommendations to family members. Price Tolerance is notably high but is justified by perceived value derived from superior craftsmanship, long-term energy savings [38], and the health benefits of the living environment, moving beyond mere brand prestige.

Thus, in this market, loyalty is intrinsically linked to the developer’s ability to deliver and credibly communicate value that aligns with the client’s sustainability values and lifestyle aspirations [7,16,18].

2.5. Comparative Context: Loyalty in Global Emerging Markets

Research from other emerging economies provides a valuable backdrop for contextualizing the drivers of loyalty in Thailand’s premium housing sector. In Malaysia, Uzir et al. [39] applied Social Exchange Theory and the Stimulus–Organism–Response (S–O–R) framework, revealing that after-sales service quality, brand image, and social relationships were decisive for long-term engagement. Similarly, in China, Owusu et al. [40] found that strategic post-purchase communication significantly enhances repurchase intentions by reinforcing consumer self-image and satisfaction. In India, Kaur and Mohindru [41] highlighted a transformation in real estate buyer behavior, where decision-making increasingly incorporates post-purchase satisfaction and service quality.

These studies collectively demonstrate that across emerging economies, loyalty in high-end housing stems from the integration of service performance, trust, and meaningful communication, with communication well-supported in marketing theory. The foundational SERVQUAL model [42]. establishes service quality as a direct antecedent to positive behavioral intentions. In high-risk contexts, this relationship is mediated by trust, as posited by Commitment-Trust Theory [43], where trust reduces perceived risk and fosters commitment. Finally, effective marketing communication [44] is recognized as the primary mechanism for building brand knowledge and trust, shaping perceptions of service quality before and during the service encounter. Therefore, this study positions customer loyalty at the nexus of these three interdependent forces, specifically contextualized within the high-end housing market.

The case of Thailand, however, is distinguished by its deeper alignment with national sustainability frameworks (e.g., Thailand Taxonomy) [7,33] and specific demographic aging trends. These factors more fundamentally embed environmental and social responsibility directly into the operational definitions of service quality, management competence, and communication effectiveness, setting the Thai context apart.

2.6. Research Objectives (ROs) and Research Questions (RQs)

The primary objective of this study is to identify and validate the determinants of customer loyalty in Thailand’s high-end home-building sector, with a particular focus on the integration of MCS, SQ, and CMS.

The research objectives (ROs) are as follows:

• RO1: To examine the direct influence of MCS, SQ, and CMS on customer loyalty.
• RO2: To analyze the mediating role of SQ and CMS in strengthening the effect of MCS on customer loyalty.
• RO3: To assess how sustainability-driven marketing and management practices jointly enhance loyalty in the premium housing segment.
• RO4: To develop a validated structural equation model (SEM) demonstrating the explanatory power of MCS, SQ, and CMS in predicting CL (

  Table 1. The study constructs’ theoretical foundations.

  LV	Theoretical Framework	Key Proposition	Supporting Theory
  MCS	Integrated Marketing Communications (IMC)	Effective communication builds credibility and trust, especially when emphasizing sustainability attributes.	[14,21,42]
  SQ	SERVQUAL (RATER dimensions)	Reliability, responsiveness, and empathy—enhanced by sustainability and health-conscious design—increase perceived quality and satisfaction.	[17,23,27,28,29]
  CMS	Resource-Based View (RBV); ESG Compliance Framework	Strong internal systems (financial, organizational, environmental management) underpin operational reliability and sustainable competitiveness.	[7,16,18,33,35]
  CL	Loyalty Theory	CL arises from satisfaction, trust, and perceived sustainability value, reflected in repurchase, recommendation, and price tolerance.	[16,36,39,40,41,42,43,44]

  Note. LV = latent variables.

  ).

The research questions (RQs) are:

• RQ1: To what extent does MCS directly influence CL, and indirectly through SQ and CMS?
• RQ2: How does SQ contribute to CL, and does it mediate the relationship between MCS and CL?
• RQ3: Does CMS strengthen customer loyalty directly, or primarily through its interaction with SQ?
• RQ4: Which construct exerts the most significant total effect on CL in Thailand’s high-end home-building market?

Together, these ROs and RQs guide the investigation toward a comprehensive understanding of loyalty formation in a sector where trust, sustainability, and service excellence are critical for long-term resilience.

3. Materials and Methods

This section outlines the research design, sampling approach, measurement instruments, and analytical procedures employed to test the proposed SEM. Emphasis is placed on methodological rigor, transparency, and reproducibility to ensure that the results on customer loyalty (CL) in Thailand’s high-end residential housing market are valid and reliable.

3.1. Conceptual Framework

The conceptual framework for this study, illustrated in Figure 1, posits that marketing communications strategy (MCS), company management system (CMS), and service quality (SQ) are key independent variables that directly influence the dependent variable, customer loyalty (CL), in Thailand’s high-end home-building business.

3.2. Population, Sampling Frame, and Eligibility

The study’s target population was customers of high-end, custom home builders in the Bangkok Metropolitan Region (BMR). The sampling frame was a list of 13,452 customers registered with the Home Builders Association of Thailand for custom home construction services in 2024 [45]. This official list was obtained directly from the Association after approval for academic use and had previously been validated for doctoral-level research on Thai housing consumers.

To be eligible for the study, individuals had to have direct experience with a custom home building project (e.g., initiating, supervising, or completing a project). This purposive criterion ensured the conceptual relevance of respondents to the study’s focus on customer loyalty in this specific context.

3.3. Sampling Procedure and Sample Size

The required sample size was determined based on recommendations for Structural Equation Modeling (SEM). Following established guidelines, a minimum of 20 respondents per observed variable is necessary for stable model estimation [46]. With 19 observed variables in the measurement model, the minimum sample requirement was 380 respondents. To enhance the robustness and power of the analysis, the target sample size was increased to 680 respondents.

A two-stage quota sampling approach was employed. This non-probability method was selected to ensure the sample represented the geographic distribution of the population while rigorously applying purposive eligibility criteria, which would not be feasible with a purely random sampling technique [47,48].

This approach balanced the need for geographic representation with the critical need for respondent relevance to the research context [48]. This quota-based recruitment ensured that our sample reflected the geographic distribution of the population, a key step for enhancing the external validity and generalizability of our findings across the BMR.

3.3.1. Stage 1: Quota Setting by Geographic Strata

To ensure geographic representativeness across the Bangkok Metropolitan Region (BMR), the population was stratified into three zones (Inner, Middle, and Outer) based on official designations [49]. The proportion of respondents to be drawn from each zone was calculated based on the actual distribution of registered customers in the sampling frame, establishing proportional quotas as detailed in

Table 2. Sample Distribution by Geographic Area.

Survey Areas	Population	Sample	%
Inner Area	4863	246	36.17
Middle Area	4352	220	32.35
Outer Area	4237	214	31.48
Total	13,452	680	100.00

. While similar to stratified random sampling in its initial division of the population into subgroups, quota sampling differs by using non-random selection methods to fill these subgroups [48].

3.3.2. Stage 2: Purposive Sampling to Fill Quotas

Within each geographic stratum, a controlled quota sampling method was implemented. This involved applying specific purposive criteria to screen all potential participants. The sole method for selecting individuals was their fulfillment of two conditions: (1) being part of the sampling frame and (2) meeting the eligibility criteria of having direct experience with a custom home building project. Recruitment continued until the pre-defined quotas for each geographic zone were filled. This approach balanced the need for geographic representation with the critical need for respondent relevance to the research context [48].

3.3.3. Test for Geographic Invariance

To directly address the potential for geographic differences in customer preferences, as raised in the literature regarding varying sustainability demands and urban development patterns, the authors performed a Multi-Group Analysis (MGA) using our structural model. This analysis tested whether the path coefficients (e.g., MCS→CL, SQ→CMS) were statistically equivalent across the three geographic strata (Inner, Middle, and Outer BMR). A lack of significant difference in these paths would indicate that the model’s mechanisms hold consistently across the region, thereby strengthening the generalizability of our results.

Stage 2: Purposive Sampling to Fill Quotas: Data collection proceeded by inviting potential respondents from the sampling frame via email, SMS, and in-person at Association events. The purposive eligibility criteria (direct custom home building experience) were applied to screen all potential participants. Recruitment continued within each geographic stratum until the pre-defined quotas shown in Table 2 were filled.

3.4. Data Collection, Screening, and Nonresponse

Invitations were distributed to 2150 individuals from the sampling frame during the survey period (April–May 2025). A total of 812 initial responses were received (a 37.7% response rate). Following data collection, 132 cases were removed due to incomplete responses (>20% missing data), patterned answering (e.g., straight-lining), or failure to meet the eligibility criteria upon closer inspection. The final dataset comprised 680 valid responses, matching the target sample size and quotas.

To assess potential nonresponse bias, early respondents (first 25%) were compared with late respondents (last 25%) on key demographic and model variables [50]. No statistically significant differences were found, suggesting that nonresponse bias was not a major concern in this dataset.

Prior to analysis, the data were screened for normality, outliers, and multicollinearity [51]. The amount of missing data was minimal (<2% per variable) and was handled using mean substitution. Skewness and kurtosis for all variables were within acceptable limits (±2), and Variance Inflation Factor (VIF) values were all below 5, indicating no critical multicollinearity issues [52].

3.5. Research Instrument

Data were collected using a structured questionnaire divided into two main parts. The first part gathered demographic information, including gender, age, education, income, and occupation.

The second part measured the study’s four primary constructs using five-point Likert scales. All scales demonstrated strong psychometric properties, with discrimination indices and reliability scores (Cronbach’s α) for each construct.

3.6. Data Analysis

The data analysis employed a two-step approach. First, descriptive statistics (frequencies, percentages, means, and standard deviations) were used to summarize the sample characteristics. Second, a path analysis with latent variables was conducted using LISREL 9.10 to test the hypothesized structural model of factors influencing CL [53]. The model’s validity was assessed using standard goodness-of-fit indices.

3.7. Measures

The questionnaire was structured into four constructs. These included MCS, SQ, CMS, and CL. Each construct was operationalized through multiple items adapted from previously validated international measurement scales in marketing and service management research. All items were measured on a five-point Likert scale (1 = strongly disagree, 5 = strongly agree).

An expert panel of five housing and marketing scholars reviewed the translated items for content validity and clarity, with an Index of Item-Objective Congruence (IOC) score ranging between 0.82 and 0.96, indicating satisfactory content validity.

A pilot test with 30 respondents with experience with custom home-building was conducted to evaluate reliability and comprehension. Feedback from the pilot led to minor adjustments in item wording for clarity, particularly with technical terms relating to construction quality and after-sales service. The pilot demonstrated strong internal consistency, with Cronbach’s alpha values for the four constructs all exceeding the recommended threshold of 0.70.

3.8. Ethics Statement

This study involved adult consumers purchasing luxury housing and did not engage vulnerable groups or involve sensitive personal information. Participation was voluntary, with informed consent obtained electronically or in writing. All respondents were assured of the confidentiality of their data and informed of their right to withdraw at any time. As such, the study was approved by the Kasetsart University Research Ethics Committee, obtaining the approval number COA No. COA68/031. The study also adhered to the Guidelines for Conducting Human Subjects Research in Behavioral and Social Sciences issued by the National Research Council of Thailand, which technically exempts social science research in Thailand [30].

3.9. Data Collection

Data were collected between April and May 2025 through a mixed approach: (i) online surveys distributed via Google Forms to emails and mobile numbers registered with the Home Builders Association, and (ii) in-person collection at home-building exhibitions and Association-organized events.

To ensure high response quality, respondents were provided with explicit written instructions, confidentiality assurances, and information on the academic purpose of the research. For in-person surveys, trained enumerators assisted participants where necessary and checked the completeness of responses before submission. As a result, missing data were rare, and the overall dataset met high standards of accuracy and completeness.

4. Results

The results of the SEM are presented in this section. Analyses include the sample’s descriptive characteristics, latent construct correlations, and tests of the hypothesized relationships. Model adequacy and validity are first established, followed by detailed reporting of the DE, IE, and TE of MCS, SQ, and CME on CL.

4.1. Sample Group Characteristics

The demographic profile of the 680 respondents is summarized in

Table 3. Demographic profile of the sample (n = 680).

Char.	Category	n	Percent
Gender	Male	362	53.24%
	Female	318	46.76%
	Total	680	100.00%
Age	18–20 years	31	4.56%
	21–30 years	145	21.32%
	31–40 years	146	21.47%
	41–50 years	149	21.91%
	51–60 years	142	20.88%
	Over 61 years	67	9.85%
	Total	680	100.00%
Education	Below Bachelor’s	89	13.09%
	Bachelor’s Degree	417	61.32%
	Above Bachelor’s	174	25.59%
	Total	680	100.00%
Monthly income ฿	<25,000 ฿ ≈ <$694	11	1.62%
	25,001–50,000 ฿ ≈ $694–$1389	188	27.65%
	50,001–75,000 ฿ ≈ $1389–$2083	96	14.12%
	75,001–100,000 ฿ ≈ $2083–$2778	91	13.38%
	100,001–125,000 ฿ ≈ $2778–$3472	67	9.85%
	125,001–150,000 ฿ ≈ $3472–$4167	45	6.62%
	150,001–175,000 ฿ ≈ $4167–$4861	44	6.47%
	175,001–200,000 ฿ ≈ $4861–$5556	69	10.15%
	>200,001 ฿ ≈ >$5556	69	10.15%
	Total	680	100.00%
Occupation	Freelance/Self-employed	106	15.59%
	Government/State Enterprise	176	25.88%
	Private Company Employee	175	25.74%
	Business Owner	223	32.79%
	Total	680	100.00%

Note. The approximate Thai baht to USD conversion in April and May 2025 was one USD ≈ 36 ฿.

. The sample was predominantly male (53.24%), with the largest age group being 41–50 (21.91%). The majority held a bachelor’s degree (61.32%) and reported a monthly income between 25,001 and 50,000 ฿ (27.65%). Most respondents were business owners (32.79%).

4.2. Assessment of Measurement Model and Descriptive Statistics

The measurement model was evaluated using descriptive statistics of the measurement constructs and their psychometric properties based on Confirmatory Factor Analysis (CFA). The results, reported in

Table 4. Descriptive statistics and CFA results for the measurement model.

Latent (4) and Observed Variables (20)	M	SD	β	t-Value	R2
Marketing Communications Strategy (MCS)	3.97	0.66
Advertising (MCS1)	3.98	0.65	0.80 **	12.68	0.47
Personal Selling (MCS2)	3.95	0.68	0.85 **	13.60	0.55
Sales Promotion (MCS3)	4.00	0.65	0.86 **	13.74	0.57
Public Relations (MCS4)	3.96	0.68	0.83 **	12.97	0.51
Direct Marketing (MCS5)	3.99	0.65	0.84 **	13.02	0.53
Service Quality (SQ)	4.00	0.65
Responsiveness (SQ1)	3.91	0.67	0.85 **	-	0.54
Assurance (SQ2)	4.06	0.64	0.86 **	13.86	0.56
Tangibles (SQ3)	3.99	0.68	0.83 **	12.89	0.49
Reliability (SQ4)	4.03	0.69	0.87 **	14.02	0.65
Empathy (SQ5)	4.04	0.69	0.88 **	14.28	0.66
Company Management System (CMS)	3.99	0.63
Financial Status (CMS1)	4.06	0.68	0.83 **	-	0.50
Corporate Organization (CMS2)	4.02	0.65	0.82 **	12.80	0.49
Work Experience (CMS3)	4.02	0.67	0.84 **	13.12	0.52
Human Resources (CMS4)	3.96	0.65	0.86 **	13.85	0.57
Tools & Equipment (CMS5)	3.92	0.68	0.85 **	13.75	0.56
Management Techniques (CMS6)	3.93	0.66	0.84 **	13.17	0.53
Customer Loyalty (LO)	4.04	0.65
Willingness to Recommend (LO1)	4.01	0.67	0.89 **	-	0.65
Price Tolerance (LO2)	4.02	0.69	0.90 **	15.24	0.67
Repurchase Intention (LO3)	4.10	0.63	0.92 **	15.47	0.68

Note: For the latent constructs, M and SD represent the mean and standard deviation of the average score of their respective items. For individual items, M and SD are the raw scores. All mean scores are interpreted on a 5-point Likert scale (1 = Very Low to 5 = Very High) and fall within the ‘High’ range. The standardized factor loadings (β), t-values, and R² are only reported for the observed indicators (items). ** p < 0.01; A dash (-) indicates a fixed parameter in the CFA model.

, summarize key features of the dataset and provide evidence of the validity of the measurement scales.

The mean scores for all primary constructs are all similarly high (M MCS = 3.97, SD = 0.66; SQ: M = 4.00, SD = 0.65; CMS: M = 3.99, SD = 0.63; CL: M = 4.04, SD = 0.65), demonstrating that customer perception is very positive across all four constructs. At the dimension level, “Repurchase Intention” (CL3: M = 4.10, SD = 0.63) is the highest scoring item, which suggests that it is a key behavioral indicator of loyalty to the brand/store. Similarly, the high scores of dimensions such as “Assurance” (SQ2) and “Empathy” (SQ5) confirm the importance of trust and personalized service within this high involvement environment.

The CFA outcomes, also shown in Table 4, provide strong evidence of construct validity. All standardized factor loadings (β) were statistically significant (p < 0.01), ranging from 0.80 to 0.92, which were significantly greater than the recommended cutoff value of 0.70. This confirmed that the indicators represented their corresponding constructs, indicating convergent validity [54].

It should also be noted that the strength of the factor loadings with ‘Repurchase Intention’ (β = 0.92) is determined to be the strongest component of CL. Additionally, the importance of a customer-centric mode of service delivery is emphasized by the high loadings of Service Quality indicators, for instance, SQ5: “The service and employees’ attitude are consistently reliable” (β = 0.88). In addition, the R² values that indicate the extent to which the constructs explain the indicator variables were high (ranging from 0.47 to 0.68), indicating the acceptable explanatory power of the measurement model.

4.3. Structural Model and Goodness-of-Fit (GoF) Appraisal

The hypothesized structural model was evaluated for its congruence with the empirical data using a suite of goodness-of-fit indices, as recommended by established methodological literature. The assessment followed a multi-faceted approach to ensure an effective evaluation of model fit.

The absolute fit of the model was excellent. The non-significant Chi-square statistic (χ² = 0.05, p = 0.84) indicated that the model was not significantly different from the perfect-fitting model [55]. This was supported by a Relative Chi-square (χ²/df) of 0.66, well below the conservative threshold of 2.00, suggesting a strong fit without being overly sensitive to sample size [56]. Further evidence of absolute fit was provided by a Root Mean Square Error of Approximation (RMSEA) of 0.00 and a Standardized Root Mean Square Residual (SRMR) of 0.00, both of which fell well within the recommended limit of 0.05 for a close-fitting model [57].

In terms of incremental fit, which compares the hypothesized model to a baseline null model, the results were equally strong. The Normed Fit Index (NFI) of 1.00 and the Comparative Fit Index (CFI) of 0.99 both substantially exceeded the benchmark of 0.90, indicating a high level of improvement over the independence model [58]. The parsimony-adjusted indices, which evaluate fit while penalizing for model complexity, also demonstrated excellent fit. The Goodness-of-Fit Index (GFI) was 0.99 and the Adjusted Goodness-of-Fit Index (AGFI) was 0.99, both surpassing the 0.90 threshold [59].

Finally, the reliability of the measurement scales was confirmed, with Cronbach’s alpha coefficients for all constructs ranging from 0.87 to 0.89, indicating a high degree of internal consistency and exceeding the accepted standard of 0.70 [60].

Collectively, these indices provide comprehensive evidence that the proposed structural equation model demonstrates an excellent fit to the observed data, thereby allowing for a valid interpretation of the structural path coefficients.

4.4. Questionnaire Reliability

The data collection instrument for this study was a questionnaire consisting of five sections as follows:

Section 1 contained a multiple-choice (checklist) questionnaire that examined the respondent’s personal information. It included questions regarding gender, age, education, average income, and occupation.

Section 2 contained questions regarding MCS, including questions on advertising, employee sales, sales promotion, news dissemination, and direct marketing. The questionnaire was administered on a five-point rating scale with a discrimination power of 0.45–0.62 and a reliability of 0.87.

Section 3 contained questions regarding SQ, including responsiveness, assurance, concreteness, and understanding and recognition of needs. The questionnaire was administered on a five-point rating scale with a discrimination power of 0.44–0.52 and a reliability of 0.88.

Section 4 contained questions regarding the home construction company’s CMS. The questionnaire included questions regarding the company’s financial status, organization within the unit, work experience, human resources, tools, machinery, equipment, and management and control techniques. The questionnaire is a 5-point rating scale with a discrimination power of 0.35–0.74 and a reliability of 0.89.

Section 5 examines CL in the home building industry in Thailand. It includes questions on referrals, willingness to purchase, and repeat purchases. The questionnaire is a 5-point rating scale with a discrimination power of 0.52–0.76 and a reliability of 0.89.

The reliability of the measurement scales was confirmed, with Cronbach’s alpha coefficients for all constructs ranging from 0.87 to 0.89, indicating a high degree of internal consistency and exceeding the accepted standard of 0.70 [60]. These indices provide comprehensive evidence that the proposed SEM demonstrated an excellent fit to the observed data, thereby allowing for a valid interpretation of the structural path coefficients.

4.5. Hypothesis Testing and Path Analysis

Before testing the structural model, the correlations between the latent constructs were examined (

Table 5. Descriptive statistics and correlation matrix of latent variables.

LV	M	SD	Skew	Kurt	MCS	SQ	CMS	CL
MCS	3.97	0.66	−0.94	−5.32	1
SQ	4.00	0.65	−0.05	−3.65	0.64 **	1
CMS	3.99	0.63	−0.69	−4.74	0.52 **	0.67 **	1
CL	4.04	0.65	−0.47	−5.00	0.68 **	0.47 **	0.53 **	1

Note: LV = latent variable, ** p < 0.01.

). All constructs were significantly and positively correlated at p < 0.01, providing initial support for the proposed relationships. The strongest correlation was observed between MCS and CL (r = 0.68), while the correlation between SQ and CMS was also notably strong (r = 0.67). The data showed no issues with multicollinearity, and the skewness and kurtosis values for all constructs were within acceptable ranges, indicating univariate normality.

4.6. Assessment of Geographic Invariance

To ensure the strength of our findings, a multi-group analysis (MGA) was conducted to test for significant differences in the structural path coefficients [61] across the three geographic zones (Inner, Middle, and Outer BMR). The results of the chi-square difference tests revealed no significant moderation effect by geographic zone (p > 0.05 for all constrained paths), indicating that the relationships between MCS, SQ, CMS, and CL are invariant across the BMR. This finding suggests that the drivers of CL identified in this study are generalizable across the different geographic areas of the BMR.

4.7. Path Analysis Results

The results of the path analysis, detailing the direct, indirect, and total effects, are presented in

Table 6. Direct, indirect, and total effects on the endogenous variables.

DV	R2	Effect	CMS	SQ	MCS
SQ	0.49	DE	-	-	0.38 **
		IE	-	-	-
		TE	-	-	0.38 **
CMS	0.64	DE	-	0.40 **	-
		IE	-	-	-
		TE	-	0.40 **	-
CL	0.72	DE	0.54	0.46	0.58
		IE	-	0.21	0.24
		TE	0.54	0.67	0.82

Note: DV = dependent variable, ** p < 0.01; DE = Direct Effect, IE = Indirect Effect, TE = Total Effect; (-) indicates a path not specified in the hypothesis.

. The overall model demonstrated exceptional explanatory power, accounting for 72% of the variance in CL (R² = 0.72). This high R² value indicates that the combination of MCS, SQ, and CMS strongly explains what drives CL in the high-end Thai home-building market.

4.8. Bootstrapped Mediation Analysis

The bootstrapped mediation analysis (

Table 7. Bootstrapped Indirect Effects (IE) of Mediation Paths.

Mediation Path	Indirect Effect (β)	95% CI (Lower–Upper)	p-Value	Results
MCS→SQ→CL	0.18	0.11–0.29	<0.001	Significant
MCS→SQ→CMS→CL	0.06	0.02–0.12	0.004	Significant
MCS→CMS→CL	0.24	0.14–0.36	<0.001	Significant
SQ→CMS→CL	0.21	0.12–0.32	<0.001	Significant

Note: Bootstrapped with 5000 samples; all effects standardized. CL = Customer Loyalty; SQ = Service Quality; CMS = Company Management System; MCS = Marketing Communication Strategy.

) confirmed a network of significant indirect pathways [62] through which marketing communications strategy (MCS) influences customer loyalty (CL). The results reveal that MCS fosters loyalty not only directly but also by sequentially enhancing perceived service quality (SQ) and strengthening company management systems (CMS). Specifically, the significant indirect effects demonstrate that: (1) SQ is a primary mediator between MCS and CL; (2) CMS acts as both a direct mediator and part of a longer chain (MCS→SQ→CMS→CL); and (3) SQ itself influences CL through its positive effect on CMS. This underscores that communication, service, and management systems are deeply interconnected in cultivating loyalty.

4.9. Hypotheses Testing and Theoretical Implications

As confirmed by the path analysis in

Table 8. Bootstrapped the IE of mediation paths.

Hypothesis	Mediation Path	Std. β	t-Value	Results
H1	MCS→CL	0.58	10.64 **	Supported
H2	MCS→SQ	0.38	4.28 **	Supported
H3	SQ→CL	0.46	6.56 **	Supported
H4	SQ→CMS	0.40	5.43 **	Supported
H5	CMS→CL	0.54	8.74 **	Supported

Note: ** p < 0.01.

, all five research hypotheses were supported. Marketing communications strategy (MCS) exerted the strongest total effect on CL (β = 0.82, p < 0.01). This substantial value represents the sum of its strong direct effect (β = 0.58) and its multiple significant indirect effects operating through service quality (SQ) and company management systems (CMS), underscoring its paramount importance as the key driver in our model.

This finding aligns with Integrated Marketing Communications (IMC) theory [63], which posits that a consistent and multi-channel communication strategy is critical for building strong brand relationships and, ultimately, CL [64].

The significant mediation effects revealed in Table 7 can be interpreted through several theoretical lenses. The finding that MCS enhances perceived SQ (H2), which in turn drives loyalty, resonates with the Service Profit Chain [65], where internal efforts drive external service value. Furthermore, the link from SQ to CMS (H4) suggests that a reputation for high-quality service builds a company’s internal capabilities and resources, a key tenet of the Resource-Based View (RBV) theory [66].

Finally, the substantial direct effects of MCS (H1) and CMS (H5) on CL indicate that clients of high-end home builders respond to communication and management as direct signals of value. This is consistent with Signaling Theory [67], where a company’s communication and robust management systems act as credible signals of its ability to deliver on a complex, high-stakes project, thereby fostering customer allegiance directly.

In conclusion, the analysis confirms a complex web of relationships where marketing communications are the primary driver of loyalty (Figure 2). However, its power is amplified through its ability to shape perceptions of service quality, which in turn strengthens the company’s operational foundation. The combined direct and indirect effects of these three factors provide a comprehensive model for understanding and fostering customer loyalty in this niche market.

5. Discussion

This study aimed to explore the antecedents of CL in the Thai luxury home-building industry using an integrated structural equation model. Research questions (RQs) were formulated to understand the role of an MCS and identify the relevant SQ and CMS constructs affecting CL. Findings are discussed below in relation to the four RQs. Finally, the theoretical and practical implications of the findings close this section.

5.1. RQ1: To What Extent Does Marketing Communications Strategy (MCS) Enhance Service Quality (SQ) and Customer Loyalty (CL) in Thailand’s High-End Housing Market?

The analysis confirms that MCS exerts the strongest total effect on CL (β = 0.82) and significantly influences SQ (β = 0.38). This demonstrates that transparent, sustainability-oriented communication not only builds trust but also shapes customers’ perceptions of quality. Consistent with Kotler and Armstrong [14] and Mocanu and Szakal [21], strategic communication emphasizing environmental responsibility and corporate integrity fosters emotional connection and credibility among affluent consumers. In the Thai context, this finding aligns with Pimpa [16], who notes that green marketing narratives increasingly underpin brand differentiation in the property sector. Hence, MCS acts as both a direct and an indirect driver of CL through enhanced SQ.

5.2. RQ2: How Does Service Quality (SQ) Influence Customer Loyalty (CL)?

SQ demonstrates a strong direct (β = 0.54) and indirect influence on CL via CMS (β = 0.21). These results affirm that perceptions of reliability, empathy, and responsiveness remain central to loyalty formation, but now extend into sustainability and wellness attributes. This supports the expanded SERVQUAL model proposed by Lieophairot and Rojniruttikul [28,29], in which sustainability and health considerations complement traditional quality dimensions. The results also echo Tangyuyang and Chartrungruang [17], who found that environmental quality and aging-friendly design enhance satisfaction and trust. Within the luxury housing niche, superior SQ thus strengthens both relational loyalty and brand advocacy.

5.3. RQ3: To What Extent Does the Company Management System (CMS) Contribute to Customer Loyalty (CL)?

CMS shows a positive direct effect (β = 0.24) and a mediated effect through SQ. This highlights that internal management quality—spanning organizational structure, human resources, and environmental compliance—translates into customer-perceived reliability. The result mirrors Klokar et al. [34] and Zayed et al. [35], who emphasized that robust management systems underpin project efficiency and stakeholder confidence. Moreover, alignment with Thailand’s ESG taxonomy provides developers access to green finance and strengthens reputational capital [7]. CMS therefore acts as an operational backbone reinforcing both sustainability compliance and customer trust, two cornerstones of durable loyalty.

5.4. RQ4: How Do MCS, SQ, and CMS Interact to Drive Sustainable Customer Loyalty (CL)?

The integrated SEM explains 72% of CL variance (R² = 0.72), indicating a powerful synergy among the three constructs. MCS initiates trust and expectations, SQ translates these into tangible experience, and CMS ensures consistency and reliability. This triadic mechanism creates a virtuous cycle in which communication authenticity enhances perceived quality, and quality, in turn, reinforces managerial credibility. Such integration supports Valero [18] and Yalçın & Çatlı [19], who contend that firms embedding sustainability throughout communication, service, and governance achieve superior resilience. In Thailand’s volatile housing market [68,69,70], this alignment between business processes and sustainability objectives constitutes a strategic path to long-term competitive advantage.

5.5. Summary Interpretation

Collectively, these findings confirm that sustainable customer loyalty emerges not from isolated marketing efforts but from an ecosystem of coordinated strategies encompassing transparent communication, exceptional service, and strong management governance [71,72]. The study extends loyalty theory by demonstrating how sustainability principles can be operationalized through the marketing–operations interface, thereby linking firm resilience with societal value creation.

5.6. Theoretical and Practical Implications for Mid-Market and Non-BMR Provinces

While this study focused on the high-end segment within the BMR, the validated model of loyalty drivers offers valuable theoretical and practical insights for the mid-market and regional provinces in Thailand. The core relationships between MCS, SQ, CMS, and CL are likely to hold, but their relative importance and manifestation may shift due to differing economic constraints, consumer priorities, and competitive landscapes.

5.6.1. Implications for the Mid-Market Segment

For cost-conscious mid-market buyers, the value proposition must be recalibrated. The findings suggest that:

Marketing Communications (MCS) should pivot from signaling prestige to demonstrating tangible value and risk reduction [73]. Communication should emphasize clarity, transparency, and the practical benefits of a reliable CMS (e.g., on-time delivery, adherence to budget) rather than high-end sustainability credentials [74]. Trust remains paramount, but it is built on reliability and financial prudence [75].

Service quality (SQ) dimensions of reliability [30] and responsiveness [76], become the primary drivers, potentially outweighing “empathy” or “tangibles” related to luxury. Post-purchase service and warranty fulfillment are critical for loyalty in a segment where a home purchase represents a lifetime’s savings.

Company management systems (CMS) remain crucial but are valued for different reasons. A strong CMS is a mechanism for cost efficiency and quality assurance at an accessible price point. The ability to deliver project fundamentals on budget and to a reliable standard becomes the key signal of competence, rather than the ability to manage complex sustainability features.

Theoretically, this implies a shift in the salience of the constructs. While the integrated model persists, the direct path from CMS to CL may strengthen relative to the more experiential path through SQ, as mid-market buyers prioritize operational proof over aspirational marketing.

5.6.2. Implications for Non-BMR Provinces

In regional provinces, where the market is often less saturated and relationships are more localized, the model adapts further:

Marketing communications can leverage hyper-local public relations and word-of-mouth [36]. A developer’s reputation is built deeply within a community. ‘Direct marketing’ and ‘personal selling’ become fused with community engagement. Sustainability messaging might focus on localized issues like water management or durability in the face of regional weather patterns [77].

Service quality (SQ) is intensely personal. Assurance and empathy, demonstrated through understanding local customs, family networks, and long-term presence in the community, are likely to be significantly more influential than in the more anonymous BMR market.

Company management systems must be agile enough to handle regional supply chains and labor pools. A CMS that proves efficient in navigating local regulations and sourcing materials reliably is a powerful competitive advantage. The ‘management and control techniques’ dimension is critical for maintaining quality over dispersed projects.

5.6.3. Theoretical Contribution

Exploring these contextual shifts provides a fertile ground for future research. A multi-group analysis comparing BMR vs. non-BMR [78], or high-end vs. mid-market, could quantitatively test these proposed differences in structural paths, further refining the generalizability of the integrated loyalty model.

5.6.4. Practical Recommendations

Developers expanding into these segments should not simply dilute their high-end strategies. Instead, they must re-calibrate their value chain: reinforcing MCS that builds local trust, focusing SQ on core reliability, and adapting CMS for regional operational excellence and cost-effectiveness. This demonstrates the model’s utility as a strategic framework, adaptable to different segments within the Thai housing ecosystem.

6. Conclusions

Our study found that customer loyalty (CL) in the high-end home-building sector in Thailand is governed by an integrated framework where marketing communications strategy (MCS), service quality (SQ), and company management systems (CMSs) act as interdependent drivers. We found that MCS has the greatest total effect on CL, with additional indirect effects through SQ and CMS. It was also noted that SQ and CMS mutually reinforce each other in transforming positive service experiences into sustainable customer commitment.

The findings demonstrate that the path to competitive success in Thailand’s luxury housing market is also the path to sustainability. By building trust through transparent communication [43,79], delivering value through high-quality and environmentally responsible service, and ensuring long-term reliability through sound management systems, developers can achieve the customer loyalty essential for economic resilience and sustainable business performance.

The validated structural model (R² = 0.72) contributes both theoretically and practically. Theoretically, it integrates fragmented perspectives on CL into a holistic system that embeds sustainability principles within marketing and operational strategies. Practically, it provides a framework for Thai housing developers to align business practices with environmental, social, and governance (ESG) goals, meeting growing consumer expectations for green, safe, and inclusive housing.

7. Limitations and Future Research

While the study offers novel insights, its findings are derived from data collected solely within Thailand’s high-end housing sector, limiting the generalizability of results. The socio-economic and cultural context of Thai consumers—particularly the influence of sustainability awareness and status-oriented consumption—may differ from other markets. Future research should validate the model through cross-national comparisons and longitudinal designs to test stability over time. A multi-group analysis comparing Bangkok Metropolitan Region (BMR) and non-BMR provinces, or high-end and mid-market segments, could further refine understanding of contextual variation in loyalty drivers. Additionally, future studies may incorporate constructs such as digital innovation, customer satisfaction, or perceived value to enhance the explanatory power and adaptability of the integrated loyalty model.