Empirical Analysis of the Inﬂuence of Digital Marketing Elements on Service Quality Variables in the Small- and Medium-Sized Enterprises Sector in the Republic of Serbia

: Understanding users’ needs, the continuous growth of the quality of a rendered service, and the differentiation of services is imperative in order to attract new users and retain the existing ones. Numerous authors have conducted different empirical research studies to this day, and those studies have mainly dealt with the inﬂuence of the quality of services on user satisfaction and loyalty. This paper has made a step forward by including in the research an analysis of the inﬂuence of certain elements of digital marketing activities on service quality variables. The research study is aimed at determining how digital marketing elements inﬂuence service quality variables in the small- and medium-sized enterprises sector (SMEs) in the Republic of Serbia. This research study should help perceive the interdependence and the degree of the inﬂuence between the said elements, which may be helpful to managers of SMEs for preparing the plans, strategies and good practices that increase user satisfaction and loyalty.


Introduction
Today, buyers can contact service providers more quickly and easily via new communication channels thanks to the WEB (World Communication Network), which has changed the communication model from one participant to several participants, simultaneously making interpersonal cooperation between participants imperative [1]. Brzaković et al. [2] emphasize the need for integrated marketing, which implies the harmonization of overall marketing activities in order to maximize individual and common benefits. Using contemporary marketing strategies, such as digital marketing, enables organizations separate themselves from the competition and ensure a competitive advantage for themselves in the long run. In order for them to survive on the market, organizations necessitate the continuous improvement of the quality of their services and their relationship with service users. For that reason, marketing can be said to be the profitable management of relationships with consumers [3].
The identification of the most valuable factors that give shape to buyer satisfaction and loyalty has been the subject matter of the analyses made in many scientific papers [4][5][6]. However, it can be noticed that there used to be a gap in research into the influence of digital marketing elements on user service quality in the small-and medium-sized enterprises sector; this topic was not sufficiently studied in the research. The majority of the research studies conducted so far were based on conducting research into service quality elements, such as reliability, responsiveness, assurance, empathy and tangibles [7].
Problem statement: This research study also expands the subject matter of the research to determine the influence of some digital marketing methods on the perception of the quality of a rendered service by the service user. This research study wishes to identify the connections between the different dimensions of the perceived quality of services in the small-and medium-sized enterprises sector using different elements of digital marketing and their contribution to user satisfaction, which can be useful for the development of the strategies and good practices intended to achieve user loyalty. According to the Report on Small-and Medium-Sized Enterprises and Entrepreneurship for the year 2018, which was prepared in May 2020 [8], the small-and medium-sized enterprises sector represents an exceptionally significant segment of Serbia's economy. In 2018, the sector consisted of a total of 99.9% of active enterprises and employed almost two-thirds of the employed in the nonfinancial sector, and contributed 57.5% in the GDV of the nonfinancial sector.
Subject matter of the research: The determination of the relationship between different elements of digital marketing (content marketing, storytelling and eWOM) and the elements of service quality (reliability, responsiveness, assurance, empathy and tangibles).
Aim of the research: Do, and to what extent, some elements of digital marketing (content marketing, storytelling and eWOM) individually exert an influence on the elements of service quality (reliability, responsiveness, assurance, empathy and tangibles)? How do the elements of digital marketing summarily exert an influence on some dimensions of service quality? This research study should help perceive the interdependence and degree of the influence between these aspects, which may help SMEs' managers to prepare their plans, strategies and distribution of resources that would increase user satisfaction and their loyalty.
Limitations of the research: This study is based on the selected elements of digital marketing, on the one hand, and the selected dimensions of service quality, on the other. Due to the fact that there are a large number of the elements of digital marketing and a large number of the elements which service quality depends on, it was impossible to include all of them in the scope of this study. However, this leaves the possibility of making these factors the subject matter of further analyses in the future. Although the existing research study indicates the interdependence of certain elements of digital marketing and the elements of service quality, there is still sufficient scope for expanding this research field by including other questions, especially given the quick changes in methods and strategies in the contemporary marketing practice.

Literature Review
Digital marketing is a general term for the targeted, measurable and interactive marketing activities conducted by using infrastructural information networks and a device connected to the network [9]. According to Smith and Chaffey [10], the Internet is an attractive marketing environment, where companies attract buyers to their web locations by optimizing social media browsers. Digital marketing is aimed at promoting consumption by communicating continuous and valuable content to a targeted audience [11]. In digital marketing, the focus is shifted from organizations to individuals, where messages are sent to selected individuals instead of being sent to a mass audience. In the contemporary world, more and more decisions on buying are made before a potential client actually establishes any contact with an organization, and this change in buying behavior has caused marketing to exert a greater influence on the actual sale than ever before [12]. The importance of the integrated marketing communication which appeared at the end of the 20th century has increased day by day [13,14]. Integrated marketing communication helps create coordinated, simultaneous and consistent messages between different communication channels, putting an emphasis on the buyer loyalty that can only be created through strategic relationships [13,15]. A significant part of digital marketing consists of social networks, a space in which buyers can freely express themselves, be kept up to date, learn what other people think, share information, give compliments, submit complaints and ask questions [16]. In contemporary market conditions, the goal is no longer to increase a profit by selling more products or services to the buyer, but to increase sales by making the buyer satisfied. A satisfied buyer is a loyal buyer, who is at the very core of all activities [17]. As the Internet, social media and technology have increasingly entered people's lives, consumer behavior has changed accordingly. The use of social media by buyers leads to different manifestations in behavior, ranging from passive to active [18]. In social media, both positive attitudes and negative content may be shared depending on buyers' attitudes and information processes during interactions [19].
Content marketing implies the creation of content whose primary goal is to make an impression, produce a relevant value and place an idea of a product or service in that manner for a particular target group. In the literature, content relevance is treated as the usefulness of information for the consumer [20]. In order to ensure brand management in a virtual space, there is a constant need for offering new, relevant content and maintaining regular contact with the buyer [21]. Content marketing is identified as a part of digital inbound marketing [22]. Holliman and Rowley [23] understand the term, digital content marketing (DCM), as implying the creation and distribution of relevant and valuable content on digital platforms, so as to develop a relationship of trust and positivity with a brand. Digital content marketing is aimed at identifying, predicting and satisfying users' requests in a profitable manner via digital content distributed through electronic channels [24]. Digital content marketing has become one of the most prominent marketing tools, which has significant benefits and exerts a significant influence in different environments and domains [25]. A study by Berger and Milkman [26] shows that positive content has a stronger effect than negative content.
Human remembrance is based on storytelling [27] and storytelling is efficiently used in social media and traditional media too. While storytelling in traditional media includes the introduction, the body and the conclusion, in social networks stories most frequently only include the beginning and the end [28]. Some authors [29,30] suggest that the most significant elements of a brand are orally communicated through storytelling systems. It seems obvious that brands use storytelling in order to express their image, mainly through symbolism, rhetoric and narrative [31]. According to Twitchell [32], a story should be clearly focused so that it can be summed up in one or two sentences. The plot of a story, its characters and authenticity initiate consumer engagement by activating cognitive, emotional and behavioral responses in a certain sequence with altering intensities [33]. Kemp et al. [34] found that the content of a story was positively connected with emotional content and the personal connection an individual feels that he/she has established with a firm's products. Consumers seek the experiences attracting their emotions and dreams, and stories help create suchlike experiences [35]. The episodic nature of a story increases the probability that it will be conveyed by consumers [36]. Through social media, users have transformed the role in storytelling from nonactive to active users. Lundqvist et al. [37] showed the power of storytelling on consumers' experiences by finding that the consumers who were exposed to a story had described the brand much more positively and had been ready to pay more for the product itself.
Ever since the Internet took the lead in communication, companies have increasingly been turning to electronic WOM communication [38]. Electronic word-of-mouth (eWOM) is a dynamic and continuous process of information exchange between former or potential consumers of products and services, which is available via the Internet [39]. eWOM is considered an important source of information that influences human behavior [40] and significantly influences the manner in which consumers make their purchase decisions [41]. eWOM has an influence on the intention to buy through influencing user trust [42]. The important aspects of eWOM imply that it is simple, personal and includes the truth and the organization's reliable independent message [43]. eWOM is a powerful market force, since consumers consider it to be a fast, impartial and reliable source of information about products or services [44]. According to Hennig-Thurau et al. [45], based on eWOM, the consumer shares his/her positive or negative opinions of products, services or brands with many institutions and people in a virtual environment.
Service quality is a concept which has drawn a fairly large interest and provoked discussion in the research literature, and there is no consensus on how it is defined and measured [46]. There are many different definitions for what service quality implies.
Most frequently, these definitions pertain to the extent to which services satisfy buyers' needs or expectations [47,48]. Service quality is also defined as a difference between buyers' expectations and a perception of the rendered service; so, if expectations are greater than performances, the perceived quality is less than satisfactory, which leads to buyer dissatisfaction [49]. Parasuraman et al. [50] defined service quality through ten dimensions (the SERVQUAL model), and the same group of authors [51] reduced those ten dimensions to five dimensions several years later: reliability, assurance, tangibles, empathy, and responsiveness. Researching the influence of service quality on customer satisfaction and loyalty by the SERVQUAL measurement scale was the matter dealt with by many other scientists as well [52][53][54][55][56][57][58]. Although the SERVQUAL model was initially designed to measure the quality of the service of trading companies, it was only later that its use spread into other fields, such as bank services, the field of informatics, telecommunications services, air companies' services, among others [59]. The SERVQUAL model can serve as a useful indicator for information system managers who are trying to identify the required service improvements, and for the researchers who are seeking a measure of the success of information system services [60].
Researchers have studied the topic of the influence of digital marketing on the business operations and success of small-and medium-sized enterprises. Authors have dealt with how digital marketing influences an enterprise's business performances and how it helps to develop a more effective business system [61]. Ritz et al. [62] examine the application of digital marketing in small enterprises and conclude that technological benefits are not the only drivers for the application of this form of marketing. In order to define the relationship that consumers create with digital marketing, an appropriate research study was conducted in Bangladesh [63]. Njau and Karugu [64] studied the influence of e-marketing on the performances of small-and medium-sized enterprises in Kenia. The possibilities of applying social networks in communication with potential and existing clients are discussed in the papers by Papa et al. [65] and Durkin et al. [66]. Research was also conducted into the level to which e-marketing was adopted in small-and mediumsized enterprises in Nairobi and Kenia [67]. The existence of barriers in the application of the e-marketing strategy was also the subject matter of research [68]. All the abovementioned strategies indicate that there is still scope for the examination of the influence that the application of digital marketing has on service quality performances in small-and medium-sized enterprises. Bearing in mind the fact that digital marketing influences the performances of quality, Karabašević et al. [69] also highlight the fact that digital marketing is one of the key elements for defining modern organizations' e-business strategies.

Methodological Approach
Data source: In the empirical section of the paper, the survey method was used, based on the questionnaire technique especially created for the needs of this research study. The research study was conducted on a sample of 360 respondents, who each answered 59 questions on a satisfaction scale from 1 to 5 (1: least satisfied, 5: most satisfied). The basic statistical indicators for structure of the observed 360 samples, the frequency and the percentual share of the respondents as per different levels: as per the basic activity of the enterprise (manufacturing and services), the legal form of the enterprise (a private independent company and a part of some bigger enterprise/shop), and the respondents' position in the enterprise (an owner, a director, a manager, and an adviser), are shown in Table 1 below. Theoretical model of research: For the needs of the research study presented in this paper, a theoretical research model was formed. The initial theoretical research model ( Figure 1) consists of the "Service quality" dependent variable and the "Digital marketing" independent variable.  Based on the new systemic model, the following sub-systemic research models were made, with the relation connections between the dependent and the independent variables, which simultaneously propose the following hypotheses: The reliability sub-systemic models ( Figure 3). The responsiveness sub-systemic models ( Figure 4). The assurance sub-systemic models ( Figure 5). The empathy sub-systemic models ( Figure 6).  Research methodology: The following mathematical and statistical methods were used to process the results obtained through empirical research: (1) the Cronbach alpha coefficient, in order to determine the reliability of the variables of the formed model; (2) descriptive statistics, with the aim of summing up the collected data in a clear and understandable manner; (3) correlation analysis, in order to determine the interconnectedness between the phenomena; (4) regression analysis, aimed at assessing the connections between the independent and the dependent variables; (5) linear regression, which modelled the connections between two variables by forming a linear equation; (6) the ANOVA test, in order to compare the two groups of the variables; (7) multiple linear correlation and regression analysis, which shows the influence of several independent variables. The collected data were processed and presented textually, tabularly and graphically. The statistical software, the SAS JMP v.14 software was used.

Research Results and Discussion
At the beginning of the research, the reliability of the research variables was determined based on the Cronbach alpha coefficient, which should be greater than the recommended coefficient of ≥0.6 [70]. This research has proven that all the Cronbach coefficients are remarkably reliable; the formed questionnaire is well constructed for this research study and the sample is remarkably reliable, which is accounted for in Table 2 below. In further research, the descriptive statistics of the respondents' answers to the set assertions are performed for all the variables separately: the mean value and the standard deviations of the answers. The research has shown that, when speaking about the assertions of the dependent variables, the greatest mean value is for the answers to the assertion EMP 1 (4.8416666667), and the least is for REL 4 (3.8666666667). The greatest standard deviation value is that for the assertion EMP 8 (0.8829868279), and the least value is that for the assertion EMP 1 (0.3655613457). The research has shown that, when speaking about the assertions of the independent variables, the greatest mean value is for the answers to the assertion CMA 8 (4.72), and the least is for the assertion CMA 2 (4.03). The greatest standard deviation value is for the assertion CMA 5 (0.76), and the least value is for the assertion CMA 6 (0.46). Table 3 below allows us to conclude for all the groups of the variables that the mean value of the answers is the greatest for the groups of the ASS variables (4.4986111111), and the least for the assertion REL (4.3579861111). The greatest standard deviation value is for the assertion eWOM (0.5219256117), and the least value is for the assertion EMP (0.2798800199). In Figure 8 below, the Pearson correlation values are given. The directions of all possible connections between the variables are positive, which means that there is a positive correlation between those variables. The greatest correlation coefficient, 0.9536, is between the Tangibles variable and the Storytelling variable, and it is strong, which means that the Storytelling variable can be explained by the Tangibles variable at 0.90935296 (or 90.93%). The least correlation coefficient, 0.1757, is between the Reliability variable and the Assurance variable, which means that the Assurance variable can be explained by the Reliability variable at 0.03087049 (3.08%). Based on the reliability sub-systemic model (Figure 3), Table 4 presents the grades for the statistical significance of the influence of the independent variables on the Reliability dependent variable; the magnitude of the contributions of the independent variables in the prediction of the Reliability dependent variable are also defined. Based on the obtained results, the set hypotheses of the reliability sub-systemic model can be either confirmed or rejected.
By applying the ANOVA test, the grade for the statistical significance of the individual and group independent variables Content marketing, Storytelling and eWOM is confirmed. The magnitude of the contribution of the Content marketing independent variable to the prediction of the Reliability dependent variable is 0.439709, which means that it has a relatively weak influence (19.33%). Based on the obtained results, the Hypothesis H 11 , which reads as follows can be rejected: "The level of Content marketing has an influence on the level of Reliability". The magnitude of the contribution of the Storytelling independent variable to the prediction of the Reliability dependent variable is 0.266298, which means that its influence is relatively weak (7.09%). Based on the obtained results, the Hypothesis H 12 , which reads as follows can be rejected: "The level of Storytelling has an influence on the level of Reliability". The magnitude of the contribution of the eWOM independent variable to the prediction of the Reliability dependent variable is 0.421446, which means that it exerts a relatively weak influence (17.76%). Based on the obtained results, the Hypothesis H 13 , which reads as follows can be rejected: "The level of eWOM has an influence on the level of Reliability". The total multiple determination coefficient of the independent variables Content marketing, Storytelling and eWOM in the prediction of the Reliability dependent variable is 0.368277 (or 36.82%), which means that the independent variables exert a medium-strong influence on the dependent variable. Based on the obtained results, the auxiliary Hypothesis H 01 , which reads as follows can be accepted: "The levels of Content marketing, Storytelling and eWOM have an influence on the level of Reliability". Based on Table 4, it is possible to generate the diagrams of the regression equations for the Reliability dependent variable (Figure 9). Based on the responsiveness sub-systemic model in Figure 4, Table 5 presents the grades for the statistical significances of the influence of the independent variables on the Responsiveness dependent variable; the magnitude of the contributions of the independent variables to the prediction of the Responsiveness dependent variable are also defined. Based on the obtained results, the set hypotheses of the Responsiveness sub-systemic model can be either confirmed or rejected. By applying the ANOVA test, the grade for the statistical significance of the individual and group independent variables Content marketing, Storytelling and eWOM is confirmed. The magnitude of the contribution of the Content marketing independent variable to the prediction of the Responsiveness dependent variable is 0.711372, which means that it has a medium-strong influence (50.60%). Based on the obtained results, the Hypothesis H 21 , which reads as follows can be accepted: "The level of Content marketing has an influence on the level of Responsiveness". The magnitude of the contribution of the Storytelling independent variable to the prediction of the Responsiveness dependent variable is 0.823451, which means that it exerts a strong influence (67.80%). Based on the obtained results, the Hypothesis H 22 , which reads as follows can be accepted: "The level of Storytelling has an influence on the level of Responsiveness". The magnitude of the contribution of the eWOM independent variable to the prediction of the Responsiveness dependent variable is 0.787636, which means that it has a strong influence (62.03%). Based on the obtained results, the Hypothesis H 23 , which reads as follows can be accepted: "The level of eWOM has an influence on the level of Responsiveness". The magnitude of the contributions of the independent variables Content marketing, Storytelling, eWOM (the total multiple determination coefficient) to the prediction of the Responsiveness dependent variable is 0.696143 (or 69.61%), which means that the independent variables strongly influence the dependent variable. Based on the obtained results, the auxiliary Hypothesis H 02 , which reads as follows can be accepted: "The levels of Content marketing, Storytelling and eWOM have an influence on the Responsiveness level".
Based on Table 5, it is possible to generate the diagrams of the regression equations for the Responsiveness dependent variable ( Figure 10). Based on the assurance sub-systemic model ( Figure 5), Table 6 gives the grades for the statistical significances of the influence of the independent variables on the Assurance dependent variable; the magnitude of the contributions of the independent variables to the prediction of the Assurance dependent variable are also defined. Based on the obtained results, the set hypotheses of the Assurance sub-systemic model can be either confirmed or rejected. By applying the ANOVA test, the grade for the statistical significance of the individual and group independent variables Content marketing, Storytelling and eWOM was confirmed. The magnitude of the contribution of the Content marketing independent variable to the prediction of the Assurance dependent variable is 0.591122, which means that it exerts a medium-strong influence (34.94%). Based on the obtained results, the Hypothesis H 31 , which reads as follows can be accepted: "The level of Content marketing has an influence on the level of Assurance". The magnitude of the contribution of the Storytelling independent variable to the prediction of the Assurance dependent variable is 0.812698, which means it has a strong influence (66.04%). Based on the obtained data, the Hypothesis H 32 , which reads as follows can be accepted: "The level of Storytelling has an influence on the level of Assurance". The magnitude of the contribution of the eWOM independent variable to the prediction of the Assurance dependent variable is, in this case, 0.774963, which means that its influence is strong (60.05%). Based on the obtained results, the Hypothesis H 33 , which reads as follows can be accepted: "The level of eWOM has an influence on the level of Assurance". The magnitude of the contributions of the independent variables Content marketing, Storytelling and eWOM to the prediction of the Assurance dependent variable (the total multiple determination coefficient) is 0.673914 (or 67.39%), which means that the independent variables strongly influence the dependent variable. Based on the obtained results, the auxiliary Hypothesis H 03 , which reads as follows can be confirmed: "The levels of Content marketing, Storytelling and eWOM have an influence on the level of Assurance".
Based on Table 6, it is possible to generate the diagrams of the regression equations for the Assurance dependent variable (Figure 11).
Based on the empathy sub-systemic model in Figure 6, Table 7 shows the grade for the statistical significance of the influence of the independent variables on the Empathy dependent variable; the magnitude of the contributions of the independent variables to the prediction of the Empathy dependent variable are also defined. Based on the obtained results, the set hypotheses of the Empathy sub-systemic model can be either confirmed or rejected.  By applying the ANOVA test, the grade for the statistical significance of the individual and group independent variables Content marketing, Storytelling and eWOM was confirmed. The magnitude of the contribution of the Content marketing independent variable to the prediction of the Empathy dependent variable is 0.679602, which means that it has a medium-strong influence (46.18%). Based on the obtained results, the Hypothesis H 41 , which reads as follows can be accepted: "The level of Content marketing has an influence on the level of Empathy". The magnitude of the contribution of the Storytelling independent variable to the prediction of the Empathy dependent variable is, in this case, 0.616554, which means that it exerts a medium-strong influence (38.01%). Based on the obtained results, the Hypothesis H 42 , which reads as follows can be accepted: "The level of Storytelling has an influence on the level of Empathy". The magnitude of the contribution of the eWOM independent variable to the prediction of the Empathy dependent variable is, in this case, 0.61834, which means that it has a medium-strong influence (37.31%). The VIF level is 1.000. Based on the obtained results, the Hypothesis H 43 , which reads as follows can be accepted: "The level of eWOM has an influence on the level of Empathy". The magnitude of the contributions of the independent variables Content marketing, Storytelling and eWOM to the prediction of the Empathy dependent variable (the total multiple determination coefficient) is 0.698706 (or 48.81%), which means that the independent variables exert a medium-strong influence on the dependent variable. The obtained results tell us that the auxiliary Hypothesis H 04 , which reads as follows: "The levels of Content marketing and eWOM have an influence on the level of Empathy." can be confirmed. Table 7, it is possible to generate the diagrams of the regression equations for the Empathy dependent variable (Figure 12). Based on the Tangibles sub-systemic model in Figure 7, Table 8 presents the grades for the statistical significances of the influence of the independent variables on the Tangibles dependent variable; the magnitude of the contributions of the independent variables to the prediction of the Tangibles dependent variable are also defined. Based on the obtained results, the set hypotheses of the Tangibles sub-systemic model can be either confirmed or rejected. By applying the ANOVA test, the grade for the statistical significance of the individual and group independent variables Content marketing, Storytelling and eWOM is confirmed. The magnitude of the contribution of the Content marketing independent variable to the prediction of the Tangibles dependent variable is, in this case, 0.81957, which means that it has a strong influence (67.16%). Based on the obtained results, the Hypothesis H 51 , which reads as follows can be accepted: "The level of Content marketing has an influence on the level of Tangibles." The magnitude of the contribution of the Storytelling independent variable to the prediction of the Tangibles dependent variable is, in this case, 0.953563, which means that it exerts a strong influence (90.92%). Based on the obtained results, the Hypothesis H 52 , which reads as follows: "The level of Storytelling has an influence on the level of Tangibles." can be accepted. The magnitude of the contribution of the eWOM independent variable to the prediction of the Tangibles dependent variable is in this case 0.932974, which means that its influence is strong (87.04%). Based on the obtained results, the Hypothesis H 53 , which reads as follows can be accepted: "The level of eWOM has an influence on the level of Tangibles". The magnitude of the contributions of the independent variables Content marketing, Storytelling and eWOM to the prediction of the dependent variable (the total multiple determination coefficient) is 0.944133 (or 94.41%), which means that the independent variables strongly influence the dependent variable. Based on the obtained results, the auxiliary Hypothesis H 05 , which reads as follows can be confirmed: "The levels of Content marketing, Storytelling and eWOM have an influence on the level of Tangibles".

Based on
Based on Table 8, it is possible to generate the diagrams of the regression equations for the Tangibles dependent variable (Figure 13).

Conclusions
In order for organizations to survive on the market, it is necessary that they should constantly improve the quality of their services and their relationships with their service users. Service quality is a concept which has drawn a large interest and provoked discussion in the research literature, although there is no consensus on defining this term. In researching service quality, the extent to which services satisfy buyers' needs or their expectations is most frequently assessed, i.e., the difference between buyers' expectations and the perception of a rendered service is determined. The majority of the research studies on service quality conducted so far have been based on conducting research into the elements of service quality, such as reliability, responsiveness, assurance, empathy and tangibles. In this paper, the subject matter of the research is expanded to determine the influence that some digital marketing methods, such as content marketing, storytelling and eWOM, exert on the perception of the quality (reliability, responsiveness, assurance, empathy and tangibles) of a rendered service by the service user. The importance of the research and empirical analyses of the influence of some digital marketing elements on service quality variables in SMEs is also demonstrated and highlighted by some researchers, such as: Du plesis [71,72], Mei et al. [73] and Liu et al. [74].
The research study has fulfilled its goal, and has provided an answer to the question: To what extent, do some elements of digital marketing individually exert an influence on the elements of service quality? This research study is very helpful for perceiving the interdependence and the degree of the influence between the aforementioned aspects, which ultimately may help SMEs' managers to prepare their strategies, plans, actions and distribution of resources which increase user satisfaction and their loyalty.
Therefore, the research study has revealed that Content marketing, Storytelling and eWOM, observed individually, have a relatively weak influence on the level of Reliability, whereas, when observed as a group, they exert a medium-strong influence on the level of Reliability. The magnitude of the contribution of the Content marketing independent variable to the prediction of the Responsiveness dependent variable has a medium-strong influence, whereas storytelling and eWOM strongly influence Responsiveness. Observed as a group, Content marketing, Storytelling and eWOM strongly influence the level of Responsiveness. The magnitude of the contribution of the Content marketing independent variable to the prediction of the Assurance dependent variable has a medium-strong influence, whereas Storytelling and eWOM strongly influence Assurance. Observed as a group, Content marketing, Storytelling and eWOM strongly influence the level of Assurance. The magnitude of the contributions of the independent variables Content marketing, Storytelling and eWOM, observed both individually and as a group, exert a medium-strong influence on the level of Empathy. Observed both individually and as a group, Content marketing, Storytelling and eWOM strongly influence the level of Tangibles.
As previously is stated regarding the paper limitations, one of the main limitations of this paper is reflected in the fact that not all the elements of digital marketing and service quality were included in this study. As a direction for future research, there is the possibility of applying the same methodology and including factors and elements that are not included in this study; therefore, this can be the subject matter of further analysis in the future. Additionally, there is still ample space for extending this research field by involving other questions too, particularly given the rapid changes in methods and strategies in contemporary marketing practice.