Recommend or Not: Is Generation the Key? A Perspective from the SOR Paradigm for Online Stores in Colombia

Abstract

Word-of-mouth behavior is one of the most effective marketing communications. Despite not being able to be controlled by the company, it can be influenced. Its importance lies in its effectiveness in repeat sales and attracting new customers. However, in the study of consumer behavior, it has been shown that the differences between subgroups can be relevant for the business sector, allowing it to generate specific strategies for each segment. In that respect, this research aims to find the determinants of WOM in generations X and Y to know the meanings these consumers give to the elements proposed within the SOR paradigm. A quantitative study was conducted using a sample of 537 e-commerce customers surveyed with a structured questionnaire to test the proposed relationships. Results are analyzed with structural equations, and a multigroup analysis is presented to find the differences between generation X and generation Y. The results indicate that, for millennials, the ease of use of electronic stores is essential to enjoying the shopping experience. Implications are discussed in the paper.

1. Introduction

Word of mouth (WOM) has been established as one of the most effective tools to counteract the perceived risk in electronic transactions and one of the most valuable sources of communication in the virtual environment, especially when the communication comes from a trusted person [1,2]. According to [3], referral communication is more persuasive than commercial recommendations since it is perceived as more credible. Customers referred for a close friend or a family member are 18% more likely to stay with the company, bringing 16% more profits to brands [4]. Subsequently, it is a leading prominent topic in marketing [5,6]. According to [7], WOM is the ability of potential and current consumers to communicate their brand perceptions to others based on their experience with a particular product or service or, in the context of e-commerce, the platform itself [8]. Compared to offline WOM, electronic word-of-mouth (eWOM) can reach a wider audience and last longer [9,10].

Because WOM is not controlled directly by the company, several academics and professionals have been interested in promoting the information circulating between users and consumers towards the sale of their products and customer loyalty. Furthermore, although WOM does not depend directly on the company, it can influence not only the type of information shared by its users but the tone of the communications through non-traditional marketing tools, such as social networks [11], and the improvement of the consumer’s shopping experience.

Regarding sustainability issues, WOM can secure consumers’ trust and loyalty more cost-effectively than advertising [12]. According to [4], it is essential to understand eWOM communication to use social media to develop customer relationships sustainably. According to [13], customers who use a firm’s products can sustainably increase the company’s profits when they feel satisfied and are more likely to spread eWOM continuously. Furthermore, recommendations to others are indicators of measuring customer loyalty. Thus, a company’s understanding of customer loyalty will create a competitive advantage. Lastly, eWOM also makes creating a sustainable competitive advantage effective since it facilitates sharing of innovative ideas and feedback [14].

The importance of understanding the elements that impact the formation of WOM intention has been widely demonstrated in the literature, even understanding WOM as a behavioral consequence of consumer loyalty [15]. Different studies have addressed this issue to provide solutions to industries influencing the much-desired customer behavior. However, most studies focus on developed markets such as the USA and Europe. Emerging economies have historically been ignored [4], leading to a gap in consumer understanding of these regions, which require significant investment to have the development rates they need to be moderately competitive in the world market.

In recent years, Colombia has opened its borders by signing free trade agreements with countries in the region and the European Union, Korea, the United States, and Israel [16]. The competitive advantage of these countries is known and indisputable. Thus, for national companies and companies that enter the Colombian market, it is essential to know the motivations that underlie the behaviors exhibited by the country’s customers.

This research intends to approach the variables influencing the consumer to generate a positive WOM online through the S-O-R model (stimulus, organism, response). The SOR model assumes that external stimuli impact individuals’ evaluation of a given situation. This evaluation, in turn, influences the behavioral response with which the individual faces those stimuli [17]. Thus, WOM is proposed as the behavioral response to be studied. At the same time, the ease of use and the perceived usefulness in online stores will be the elements that configure the external stimuli. Therefore, two components, one from the “affective organism” and the second from the “cognitive organism”, are suggested to analyze an individual’s organic reactions. In the case of the affective organism, the research proposes to study the enjoyment perceived by the customer by browsing the web store, and from the cognitive organism, the self-efficacy with which it assumes the task of navigation and the purchase decision process, thus addressing objective and subjective elements.

In addition, this research intends to profile consumers according to their willingness to recommend. Therefore, we assume that the demographic variable “age” can generate different perceptions of the stimuli and, thus, different responses. As per [18], the profile of generation Y is heterogeneous in different countries, indicating a growing need for targeted marketing for specific countries and highlighting the importance of identifying and understanding Gen Y’s profile before implementing their marketing strategy. The sample has been divided between generation X and generation Y (millennials) consumers. Defining generational groups is challenging [19] since there has yet to be a consensus on the year a generational group begins and ends. According to [20,21], the generational cohort can be extended by 20–25 years or more, depending on how long it would take for the group to be born, grow up, and have their own family. According to [22], millennials were born between 1980 and 1995, while generation X was born between 1965 and 1979.

To our knowledge, the literature on WOM and generational cohorts is scarce. This evidence emerged as a preliminary result using the Scopus database with the following queries, as shown in

Table 1. Literature search on word of mouth.

Criteria	Concept	Scopus Algorithm	Results
Criterion 1	Word-of-mouth	(TITLE-ABS-KEY (“word of mouth”))	10,359
Criterion 2	Word-of-mouth and generation X	(TITLE-ABS-KEY (“word of mouth”) AND TITLE-ABS-KEY (“generation x”))	7
Criterion 3	Word-of-mouth and millennial	(TITLE-ABS-KEY (“word of mouth”) AND TITLE-ABS-KEY (“millennial *”))	33
Criterion 4	Word-of-mouth and generation	(TITLE-ABS-KEY (“word of mouth”) AND TITLE-ABS-KEY (“generation Y”))	53

* Note: The asterisk is used in Scopus as a “wild card” to include plural and singular results.

.

The literature states that consumers of generation Y or millennials, called “the internet generation”, will always use the Internet to obtain official information [23]. Due to their inherent confidence in using new technologies, such as the Internet, this cohort perceives technology as vital. Furthermore, they are a broadly segmented audience who are more challenging to reach through traditional marketing efforts [24]. Specifically, millennials were born with the rise of information and communications technology (ICT) and dominated technological devices such as tablets and smartphones. Consequently, they naturally adopt online commercial transactions due to their convenience [10].

In contrast, generation X is considered a “digital immigrant” since they did not grow up immersed in digital media. Still, they try to understand and get the skills to use it [21,25]. Since online users’ attitudes differ according to the age group to which they belong, it is expected that trust towards the online store will vary according to the generational cohort [26]. Moreover, the generation X segment forms a significant group of consumers, so it is crucial to extend research on this topic, given their attitudes and behaviors from younger age groups [27]. Subsequently, WOM behavior is expected to be higher for millennials than for generation X. Thus, we propose the research questions:

RQ1. What external and internal factors affect WOM intentions in Gen X customers?

RQ2. What external and internal factors affect WOM intentions in Gen Y customers?

RQ3. Are there differences between factors affecting WOM intentions in customers, according to their generation?

With these research questions, we aim to find the determining factors of WOM intention in generations X and Y in an emerging economy.

The paper is organized as follows: Section 2 provides an in-depth literature review of the S-O-R paradigm. In Section 3, the hypotheses are proposed, and the methodological strategy is presented. Section 4 presents the results. Finally, Section 5 presents the discussion, and Section 6 shows conclusions, limitations, etc.

2. The SOR Paradigm

The SOR model or paradigm refers to a sequential model, where it is assumed that external stimuli (S) condition the organism or individual’s evaluations (O) and that those evaluations, in turn, influence a behavioral response (R) [28]. The SOR paradigm was developed under the Environmental Psychology Theory by [17], who contributed to the knowledge of consumer behavior. Their theory explains the psychological elements underlying consumer decision-making. In this theory, the authors note that environmental stimuli can influence consumer emotions and impact consumer behavior. Based on this theory, stimuli have been studied from different perspectives in brick-and-mortar and online stores.

According to [29,30], in e-commerce, external stimuli could be due to the order and complexity of the web page, graphics, colors, links, menus, the aesthetic appearance of the page, background music, availability of products, ease of use of the site, m-payment, design, information, and effectiveness, among others. Therefore, for this research, consumers’ perceptions of the web store’s ease of use and usefulness will be analyzed as external stimuli. Likewise, [29] identified that the most frequently used emotional evaluations, categorized as “organism”, are excitement, pleasure, impulse, enjoyment, and flow. Furthermore, the same study points out the most studied behavioral responses in e-commerce: repurchase intention, impulse buying behavior, purchase, consultation, and search for other shopping websites. Therefore, in this research, starting from the gap in the literature, we propose to study the WOM behavior of consumers as a response within the SOR model, identifying the variables enjoyment and self-efficacy as organism elements.

2.1. Stimulus—Ease of Use

Perceived ease of use refers to a person’s perceived lack of effort in learning to use and use new technology [31,32]. Ref. [33] defined ease of use as “the degree to which a person believes that using a particular system will be free from effort.” In other words, based on the TAM theory, a consumer will consider that a system is helpful if they perceive it is easy to use [34]. Therefore, as [35] mention, users’ perceived ease of using technology will provide more benefits.

In e-commerce, [36] explain that ease of use is the degree to which a user expects the online store to be effortless. Ease of use is linked to consumers’ perceptions during the experience of using the store [37]. Thus, perceived ease of use will be understood as the consumer’s belief that online shopping will require minimal mental and physical effort, including ease of learning. Ref. [38] found that easy-to-understand websites allow consumers greater control over navigation and purchasing and, consequently, a positive attitude towards online stores. In addition, according to [39], consumers exposed to digital ads that are easy to see and understand will have a similar positive attitude towards advertising presented on social networks.

The literature supports the existence of a significant relationship between ease of use and consumer attitudes [40]. By its nature, ease of use will be particularly significant in the early stages of the experience of using new technology [33]. As the experience of use increases, consumers will adapt their perspective on the system’s ease of use [41].

2.2. Stimulus—Perceived Usefulness

Conceived in the TAM technology acceptance model, perceived usefulness refers to a person’s belief about improved performance and productivity that they will achieve by using new technology [31,37]. Specifically, [33] defines perceived usefulness as the “degree to which a person believes that using a particular system would enhance his or her job performance”. Ref. [32] refer to perceived usefulness as a person’s belief that using a particular system will improve their job performance and, in general, effectiveness, productivity in terms of time, and the importance of the system for personal work [42].

Likewise, perceived usefulness includes a reward system that dominates the user’s attitude and intention to use technology and, generally, their behavior [43]. Consequently, when a user is in a computer domain, this knowledge will allow them to feel confident in achieving their objectives, contributing to perceived usefulness [44].

Although perceived usefulness was developed in the context of workplace systems, due to its importance, several studies applied the concept in e-commerce and social commerce [45]. For this study, in e-commerce, perceived usefulness is understood as the consumer’s belief that online stores will improve their purchasing performance. Ref. [46] explained that consumers’ commitment to an online store increases when the website offers valuable information to help them make purchasing decisions and achieve their goals. Thus, the online store’s perceived usefulness is expected to positively affect word-of-mouth behavior through the organic reactions of the consumer.

2.3. Cognitive Organism—Self-Efficacy

First studied in Social Cognitive Theory (SCT), self-efficacy refers to people’s judgment about their abilities to organize or execute specific actions to achieve a goal [47]. Additionally, self-efficacy influences behavior, effort expenditure, behavioral patterns, and emotional reactions [48]. The proposed relationship between cognitive factors and the individual’s behavior is vital among the different dimensions proposed in SCT. Thus, a person’s likelihood of engaging in a specific activity is directly related to the expectations inferred from the behavior and the individual’s perceptions of their ability to carry out the behavior [49].

Because Internet shopping is not intuitive [50] and requires specific skills, consumers’ perception of their self-efficacy in online shopping would significantly affect the acceptance of online activities. Therefore, users with a high level of self-efficacy towards a particular task, such as surfing the Internet, searching for information through corporate websites, as well as opinion websites (forums, blogs, video blogs), and electronic money transfers, will see themselves as capable of performing this task [51].

2.4. Affective Organism—Enjoyment

Cognitive evaluation involves a mental process, while affective evaluation requires an emotional process that is subjective to the individual and their interaction with Internet stimuli. Thus, perceived enjoyment is one of the consumer’s intrinsic emotions when surfing the Internet [52]. Furthermore, enjoyment is essential in predicting Internet behaviors, such as social networking behavior [53]. However, its definition and conceptualization are under construction. The literature provides some definitions, such as [54], who proposes that enjoyment has traditionally been understood as “something joyful that creates fun”; however, he expounds that the construct has a deeper meaning that equates to psychology or education of the psyche.

As it is perceived during the purchase, a vital enjoyment factor is the experience. The study of experiential shopping has been motivated by the positive and negative feelings associated with the moment of purchase [55]. As a result, it has been defined as the purchase of goods and services prompted by the combination of the store and the environment, which can elicit an emotional response in shoppers [56]. The emotional response is proposed as the consumer’s evaluation of affective organism in this study and subsequent work [57]. Its association with positive WOM behavior will be examined.

2.5. Behavioral Response—WOM

The importance and reach of WOM have been recognized since 1950, appearing in different disciplines in popular literature, academic journals, and research centers. Furthermore, its importance is seen in academic google searches where WOM yields 1,400,000 results [58]. Consumers generally search for information related to the brand or product to be purchased in the purchase decision-making process. This data is gathered not only from companies but also from other product users. WOM refers to the informal communication between users of a product or service. That is the behavior to be studied as a response within the S-O-R paradigm in this study.

WOM has been categorized because of service quality in the context of the financial impact of retention [59]. The authors argue that customers who remain loyal to a store longer are more likely to issue positive WOM communications. Consumer emotions drive WOM as post-purchase behavior from purchasing or using a product or service [60]. The concept of WOM originates in the theory of social comparison, which assumes that individuals need to confirm their perceptions and feelings with others, especially when their emotions are ambiguous and confusing [61]. On the other hand, the theory of social sharing indicates that emotions generate sharing-oriented behaviors, which would explain why consumers want to communicate their emotions openly with others to obtain help, and attention or strengthen relationships [62].

2.6. Research Model

The research hypotheses are presented based on the literature reviewed and the study variables to be examined in stimulus, organism, and response. In this work, the sample was divided between generation X and generation Y to find differences between consumers in these ages and, thus, make recommendations according to the target to influence WOM behavior.

Perceived ease of use is defined in the context of consumer e-shopping behavior research as an online store’s effortless and easy-to-learn shopping experience. As a result, it is expected that if a consumer perceives online shopping as simple, they will have a positive attitude toward shopping. This attitude will be reflected in the perceived enjoyment of e-commerce. Likewise, it is expected that there is a significant difference between the generations discussed in the study, as shown by the first hypothesis of the research. The research hypotheses are presented based on the literature review and the study variables to be examined in stimulus, organism, and response.

Hypothesis 1.

Ease of use determines enjoyment in (a) generation X and (b) generation Y.

Additionally, in the context of online consumer behavior research, perceived usefulness is understood as a consumer’s belief about improving their shopping performance and product information search by using an online store [36]. Therefore, following the literature review, it is expected that perceived usefulness in e-shops directly and positively impacts perceived shopping enjoyment. The results are different generations X and Y of the selected sample.

Hypothesis 2.

Perceived usefulness determines enjoyment in (a) generation X and (b) generation Y.

On the other hand, self-efficacy is presented in the research as the consumer’s evaluation of their capabilities to carry out commercial transactions over the Internet [50]. Therefore, this characteristic has been understood as the consumer’s cognitive evaluation, and external stimuli are expected to affect self-efficacy, as stated in Hypotheses 3 and 4 of the research.

Hypothesis 3.

Ease of use is a determinant of self-efficacy in (a) generation X and (b) generation Y.

Hypothesis 4.

Perceived usefulness is a determinant of self-efficacy in (a) generation X and (b) generation Y.

Perceived enjoyment is a hedonic element of value [63] and has been proposed as one of the determinants of consumer loyalty [64]. Consumer loyalty can be reflected in certain types of behavior. The most frequent are repeat purchases and positive WOM about the brand. As a result, consumers who enjoy the online shopping process will develop positive emotions that will influence the tendency for the consumer to give positive product recommendations [52]. This research aims to identify the elements that lead to a positive WOM. Therefore, Hypothesis 5 of the research is proposed:

Hypothesis 5.

Perceived purchase enjoyment directly and positively affects WOM behavior in (a) generation X and (b) generation Y.

Similarly, self-efficacy is an antecedent of consumer behavior in online stores. For example, suppose the consumer knows the person’s values, objectives, and personal interests to whom they will give a recommendation. In that case, the WOM can be adjusted and related to the recipient’s goals [65]. Therefore, in this research, we propose to explore the effect of self-efficacy on WOM behavior, as presented in the last hypothesis of the research:

Hypothesis 6.

Self-efficacy in the online store directly affects WOM behavior in (a) generation X and (b) generation Y.

The proposed hypotheses are presented in the research model in Figure 1.

3. Materials and Methods

3.1. Study Context

According to the 2018 census carried out by DANE (National Administrative Department of Statistics) [66], Colombia has a total population of 44,164,417 people, of whom 51.2% are women. There is a literacy rate of 92.1% for men and 93% for women (over 5 years old). In total, 77.1% of the population resides in municipal capitals, while 15.8% lives in rural areas and 7.1% in populated centers.

The structure of the population has changed since the last census carried out in 2005. The population pyramid has shrunk at its base, with fewer births and a smaller young population in 2018 compared to 2005. According to the information provided, people between 0 and 10 years of age represent 15% of the population, adolescents and young people between 10–19 years old represent 17%, while adults between 20 and 60 years old represent 55%, being the majority of the population and the current objective of many companies since they represent the population with the highest purchasing power. Contrastingly, people over 60 represent 13% of the population, thus growing from 6.3% to 9.1% between 2005 and 2018 the population over 65 years old. This data may be interesting for future research since they have just been considered an attractive target. Colombia is an emerging market and needs many studies to understand the value of its market and create competitive strategies that allow eCommerce to grow.

According to census data, in 2018, only 43.4% of the population had access to the Internet. In 2019, DANE conducted a study to determine the primary indicators of ICT use in homes. The results indicate that in 95.2% of households, at least one person has a cell phone. In comparison, only 37.3% of households have a desktop computer, and 39.6% of people over 5 years of age use a computer (belonging to the household, educational institutions, offices, and libraries). According to the DANE report [67], 89.5% of the population over 5 years old used the cell phone as a means to connect to the Internet, being the means that then facilitates the Colombian client’s entry into electronic commerce.

3.2. Study Design

A quantitative cross-sectional study was carried out to contrast the proposed hypotheses. In addition, all the ethical guidelines for data collection, informed consent, and appropriate disclaimers were reviewed and approved by the ethics committee of CESA. Five hundred thirty-seven (537) e-commerce users from generations X and Y answered a questionnaire. In addition, a market research firm collected data in November 2015. The sample was obtained from undergraduate and graduate university students from the Faculty of Administration to obtain different age segments.

A previous exploratory study in Colombia [66] showed the main factors in online purchase intention and provided information on the composition of online buyers in the country.

3.3. Sample and Scales

A quantitative cross-sectional study was carried out to contrast the proposed hypotheses. In addition, all the ethical guidelines for data collection, informed consent, and appropriate disclaimers were reviewed and approved by the ethics committee of CESA. The questionnaire is presented in Appendix A. Five hundred thirty-seven (537) e-commerce users belonging to generations X and Y answered a questionnaire created from scales tested in the literature: ease of use and perceived usefulness were measured based on the work of [68] and [69]. Similarly, [70] and [71] were used to adapt the scale measuring self-efficacy in online stores. Furthermore, the work of [72] was utilized to calculate user satisfaction. Finally, WOM was measured from the work of [73]. The information collected was processed with SPSS, EQS, and Smart-PLS programs.

4. Results

4.1. Demographics

The sample is made up of 51% women and 49% men. Most of the sample reported themselves as single (80%), while only 19% classified themselves as married. The age of the participants divided by generations shows us that millennials represent 64% of the sample, while generation X represents the remaining 36%. Finally, 56% of those surveyed affirm that their main occupation is studying, followed by 34% of employees and 9% of entrepreneurs.

In total, 46% of the sample belongs to socioeconomic stratum 3, which, according to the Colombian classification, belongs to a lower middle class; 29% live in stratum 4, representing the middle class, while 12% belong to low-income households—the lower class. There is a representative 10% of residents in upper-middle-class households.

Regarding experience on the Internet, the majority of the sample (57%) claim to have more than 7 years of experience, 14% say they have between 4–6 years of experience, and only 2% say they are beginning to learn about the tool less than 6 months ago.

To connect to the Internet, most of the sample use their cell phones more frequently (59%), followed by a laptop (18%) and a tablet (16%). Nearly 40% of the sample claimed they were connected for 6 to 15 h a week, while 45% reported spending more than 15 h per week accessing the Internet.

4.2. Reliability and Validity of the Measurement Instrument

A Confirmatory Factor Analysis CFA was applied to confirm the measurement instrument’s reliability and validity. The results are presented in

Table 2. Reliability and Validity of the Measurement Instrument.

Factor	Item	Indicator	β	T	Cr. a	CR	AVE
Self-Efficacy	SE1	I would be able to use the Web myself to find online stores.	0.709	19.562	0.812	0.876	0.639
	SE2	If I wanted to, I would be able to buy from an online store in the next 30 days.	0.903
	SE3	If I wanted to, I’m sure I could buy from an online store in the next 30 days.	0.897
Ease of use	EOU1	My interaction with online stores is clear and understandable.	0.800	27.324	0.882	0.912	0.674
	EOU2	Interacting with an online store does not require much mental effort.	0.834
	EOU3	I find online stores easy to use.	0.797
	EOU4	It is easy to become skilled in the use of online stores.	0.820
	EOU5	Learning to manage online stores is easy	0.852
Perceived usefulness	PU1	Online stores improve my ability to search for and purchase products/services.	0.861	41.710	0.889	0.923	0.751
	PU2	Online stores allow me to search for and buy products/services faster	0.878
	PU3	Online stores improve my effectiveness when buying	0.862
	PU4	Online stores increase my productivity in searching for and purchasing products/services.	0.865
Enjoyment	ENJ1	I have fun interacting with the store’s website.	0.929	86.210	0.951	0.965	0.873
	ENJ2	Using the store’s website provides me with a lot of entertainment.	0.949
	ENJ3	I enjoy using the store’s website	0.952
	ENJ4	Using the store’s website is interesting to me.	0.906
WOM	WOM1	I say positive things about this online store to other people.	0.936	61.825	0.921	0.950	0.863
	WOM2	I recommend this online store to anyone who asks for my advice.	0.952
	WOM3	I encourage my friends to buy from this online store.	0.899

NFI = 0.800; SRMR = 0.07. Cr. a = Cronbach’s alpha; CR = composite reliability.

.

The values obtained exceed the expected minimum: AVE > 0.5, item loadings on the factors above 0.6, composite reliability values, and Cronbach’s alpha above 0.7. The Fornell–Larcker criterion was used to find the discriminant validity, as shown in

Table 3. Discriminant Validity.

	ENJ	EOU	SE	PB	WOM
ENJ	0.934
EOU	0.415	0.823
SE	0.387	0.658	0.759
PB	0.521	0.624	0.562	0.867
WOM	0.399	0.395	0.357	0.430	0.929

.

The diagonal in the table represents the AVE squared, and the values below the diagonal represent the correlations between the constructs. According to the Fornell–Larcker criterion, the value of the AVE squared must be higher than the correlation between the factors, thus testing the discriminant validity of the measurement instrument.

4.3. Hypotheses Testing

The next step consisted of a PLS-SEM through a Bootstrapping analysis. Finally, we tested the model using the entire sample to establish the statistical significance of the proposed relationships. The results are shown in

Table 4. Bootstrapping.

H	Relationship	Result	Beta	T	p-Value
H1	Ease of use → Enjoyment	Accepted	0.147	2.803	0.000
H2	Perceived usefulness → Enjoyment	Accepted	0.430	8.298	0.000
H3	Ease of use → Self-Efficacy	Accepted	0.504	11.489	0.000
H4	Perceived usefulness → Self-Efficacy	Accepted	0.247	6.001	0.000
H5	Enjoyment → WOM	Accepted	0.307	6.554	0.000
H6	Self-Efficacy → WOM	Accepted	0.238	5.500	0.000

The relationships in the total sample have been accepted with high significance, p < 0.005.

.

4.4. PLS-MGA Multigroup Analysis

Finally, a multigroup analysis was applied to determine the differences between generation X and generation Y consumers. The results are presented in

Table 5. Multigroup Analysis.

H	Gen X		Gen Y		Β (X vs. Y)	p-Value (X vs. Y)
	β	T	β	T
H1. Ease of use → Enjoyment	0.061 ns	0.721	0.165 *	3.108	0.103 ns	0.156
H2. Perceived usefulness → Enjoyment	0.592 *	8.077	0.331 *	5.240	0.261 *	0.997
H3. Ease of use → Self-Efficacy	0.356 *	3.948	0.506 *	10.123	0.151 ns	0.064
H4. Perceived usefulness → Self-Efficacy	0.400 *	4.981	0.171 *	3.058	0.228 *	0.990
H5. Enjoyment → WOM	0.282 *	3.904	0.264 *	4.210	0.017 ns	0.569
H6. Self-Efficacy → WOM	0.303 *	3.985	0.221 *	4.293	0.083 ns	0.820

* p < 0,05; ns = no significant.

.

The total sample was divided into two subsamples according to the age of the participants, resulting in 194 generation X consumers and 343 generation Y consumers. The data show that ease of use is not a determining factor in perceived purchase enjoyment for generation X consumers but for generation Y consumers. Although the path difference is not significant, the multigroup study indicates a difference in constructing the affective organic reaction of the consumers of both subsamples. Generation Y consumers are susceptible to perceived purchase enjoyment through ease of use, while for generation X, ease is not a key determinant.

Hypothesis 2 tested the relationship between perceived usefulness and perceived purchase enjoyment. Again, the total sample and the subsamples were positively contrasted; however, the betas indicate that for generation X, it has a higher impact than for generation Y, demonstrating the moderating effect of age on this relationship.

Hypothesis 3 states that ease of use directly impacts perceived self-efficacy; this relationship has been positively contrasted in the total sample and the subsamples, with a slight difference, although not significant (p < 0.1), between generations, showing that for millennials, the relationship is more robust.

Hypothesis 4 explores the relationship between perceived usefulness and self-efficacy. The bootstrapping analysis shows that the relationship is positive in the total sample. In the multigroup analysis, the relationship is positively contrasted in both subsamples. However, there is a significant difference between generations: for generation X, perceived usefulness in the online store is more important than it is for millennials.

Finally, Hypotheses 5 and 6 explore the relationship between perceived shopping enjoyment and self-efficacy and WOM behavior. In this case, both relationships were positively contrasted in the total and subsamples.

5. Discussion

Using the S-O-R model, the study aimed to find the antecedents of WOM behavior in the online store’s context and the differences between two groups of customers: generation X and Y. The results indicated that, for the total sample, external elements such as ease of use and perceived utility directly impact affective and cognitive organism variables. Likewise, the affective and cognitive elements proposed in this research positively and directly impact WOM behavior. However, the relations between the proposed variables change when the sample is segmented into two groups according to customer age. The PLS-MGA analysis showed that generation X is more critical of utility than ease of use to perceive more enjoyment and self-efficacy. Additionally, even when the significance was p < 0.1, it can be said that self-efficacy is more critical to generation X than generation Y to generate positive WOM about their online stores. Generation X is looking for utility value. Meanwhile, generation Y is looking for hedonic value.

Although previous research has shown that millennials are more resistant to the traditional communication efforts of companies [54], their purchase decisions depend primarily on the opinion of their peers and WOM recommendations [55]. This research found that studying the elements influencing WOM behavior is still necessary. Thus, this paper contributes to the literature from the S-O-R model to understand the consumer behavior of generations X and Y. Following [74], generation Y is significantly different from previous generations. Millennials are also known as digital natives. They demonstrate higher levels of trust, are more tolerant, support social causes, and are socially responsible companies. According to [75], millennials are group-think-oriented with a strong sense of identity, so this research expected this subsample to exhibit higher values than generation X consumers. However, although relationships are significant and positive, generation X also obtained significant results in terms of WOM behavior.

One of the critical differences in the research is that ease of use is typical for generation X individuals to perceive enjoyment during online shopping. In contrast, perceived usefulness does have an impact on enjoyment. Millennials are a generation that expects processes to be effortless, which is why making online stores easy to use and intuitive will be a good place to start influencing WOM behavior. On the other hand, self-efficacy in online stores represents an essential element for generation X, who demonstrated that perceived usefulness in the web store has a direct effect on self-efficacy and, subsequently, on WOM behavior. This relation is congruent with [76], finding that generation X need to feel confident about their skills in using the website to engage in word of mouth. Therefore, it is concluded that for millennials, affective elements are more important in influencing WOM, while for generation X, cognitive and functional elements are fundamental in this process.

The literature shows that as consumers reach an older age, they become more independent and able to make decisions without waiting for the opinion of their peers (Bellman et al., 2009). Hence, the results of this research may only last for a while, as when millennials reach an older age, they may exhibit different consumption behaviors than they do today. As a result, centennials may then become the focus of market attention.

6. Conclusions, Implications, Limitations and Future Research

A natural question for any marketing manager is understanding customers’ word of mouth. This article takes notable differences between generations X and millennials as a case. Using the SOR model, the study aimed to find the antecedents of WOM behavior in the online store context and the differences between two groups of customers: generations X and Y. In addition, the current study can relate to other approaches developed from complex sciences using other metrics that might be informative of customers’ WOM.

Our study found that generation X is more critical of utility than ease of use and self-efficacy. While the generation X cohort is looking for utility value, millennials are looking for hedonic value. Based on this finding, some practical implications should be clear. First, commercial advertisements should be adapted accordingly. Managers can exploit this insight in different concrete scenarios. For generation X, in contrast, advertisements should be consistently written and designed around the utility of products and services. These differences in addressing the message content and the morale could be summarized as “one size does not fit all”.

The implications of this study lie in effectively managing customer segments. This study proposes that segmenting by generational cohort is an efficient alternative [77]. Considering the differences between the two generations, as per [22], managers should not treat their customers alike and should properly consider customer value to satisfy better and positively impact WOM. Decision-makers must implement more robust marketing strategies, maintain customer relationships, and adjust to market changes. Moreover, considering millennials, online content should be visually appealing and pleasant to improve hedonic value. However, these strategies imply that investments should be made in more than just product/service offerings and digital channels but also personnel. Such training endeavors can ensure conducive social atmospheres in brick-and-mortar stores to boost positive perceptions by customers and, in turn, increase intentions to return and recommend [78].

Further implications for managers include the importance of online reviews concerning their customers’ generational cohort since millennials and generation X react differently to online reviews [79]. Respectively, generation X has been identified as being more skeptical, practical, and insecure, leading them to seek alternative information to reassure themselves of their choices. Thus, increased involvement in the decision-making process positively affects the credibility of eWOM [80]. Furthermore, regarding the user-generated content created, managers can obtain insights into trends and evolving market needs through data mining and analytics [81]. These insights are crucial for younger generations that use social media as a primary source of information [82].

This study, like any other study, has several limitations. First, the data was collected from customers belonging to generations X and Y and neglected generation Z, which has gained relevance in the last couple of years given their strong entry to the market, being the first electronic native generation. We recommend that future studies use more complex measurements to capture the impact of WOM on social networking. According to [83], eWOM among generation Z is more prone to be carried out from various aspects, perspectives, and levels. Second, the data was collected before the pandemic when e-commerce was less robust than now. Third, many customers have had to change their purchasing habits due to COVID-19 restrictions, forcing them to switch from traditional brick-on-mortar store purchases to online purchases. Regarding these two limitations, we expect future WOM research to capture more negative comments due to product scarcities, a decrease in service quality, and pricing surges during and after the pandemic, markedly in developing countries.

Finally, as a fourth limitation, we use the SOR paradigm to understand the motivations that underlie customer behavior, considering external and internal elements that influence the behavioral response; nevertheless, more approximations available in the literature can be explored to understand the target phenomenon. It should also be considered that the SOR paradigm allows the integration of different variables that can explain the external elements that organizations can control, such as the data presented, images, branding, and designs made, so the customers perceive flow and “teleport” to the website, the use of technological tools such as AI or “phygital”. The experience perceived by the customer across all the channels offered by the companies could also be considered a relevant topic that could be addressed in future research. Future research should include another theoretical approach and more significant samples.