The Effect E-Wom Website Attractiveness E-Trust and Innovation on Purchase Decision Online Sales

Abstract

Purpose: Technological developments make the business competition even tighter. Digitalization makes it easier for business people in their efforts to encourage the formation of consumer buying behavior because of the availability of effective and efficient means of delivering information to be used to make purchasing decisions. The purpose of this study was to determine the effect of E-Wom and Website Attractiveness on E-Trust and Purchase Decision of online sales on Instagram with innovation acting as a moderating variable. Research design, data and methodology: This research sample is 170 Instagram users who have made online purchases from the local fashion brand. This research is quantitative explanative with data collection using questionnaires and analysis techniques using the Structural Equation Modeling (SEM) approach using AMOS. Results: The results showed that E-wom had a significant positive effect on Purchase decisions. Web Attractiveness has no significant effect on Purchase Decision. Ewom has a significant positive effect on E-trust. Web Attractiveness has a significant positive effect on E-trust. E-Trust has no significant effect on purchase decisions. Conclusions: Innovation can moderate the effect of E-trust on Purchase decisions. Purchase Decisions can be influenced by E-wom and E-trust, but not Web Attractiveness.

1. Introduction

With the rapid development of internet technology and the entry of local brands competing to attract consumers by presenting many innovations, it is very important for a company to pay attention to consumer behavior in making online buying decisions and innovations. Digitalization has changed the business paradigm, from conventional marketing methods to prioritizing internet-based media (Hagberg et al., 2016). Digitalization makes it easier for business people in their efforts to encourage the formation of consumer purchasing behaviour because of the availability of effective and efficient means of delivering information to be used to make purchasing decisions (Wang et al., 2020).

Purchase decisions are the result of the decision-making process that consumers go through, which generally begins with the information search stage, then continues with the information analysis stage, consideration of several existing alternatives, and ends with the determination of one choice that best suits your needs or desires consumer (Virdi et al., 2020). The purpose of this study was to determine the moderating role of innovation on the effect of e-trust on purchase decisions and the mediating role of e-trust on the effect of e-wom and website attractiveness on purchase decisions on online sales on Instagram.

In connection with the ongoing digitization, the risk in a transaction is closely related to trust level. According to Kaur and Arora (2020), the risk that consumers perceive can be mitigated by trust in the seller. This means that the higher the trust that consumers have in the seller, the lower the risk that consumers worry about in making transactions.

2. Literature Review

2.1. The Elaboration Likelihood Model of Persuasion (ELM)

The theory of persuasion communication in research is related to the dissemination of information through advertising and campaigns. Persuasion is communication used to influence and convince people.

The persuasion message processing can be seen from several theories. The theory related to the theory of persuasion in receiving messages is the Elaboration Likelihood Model theory or what we call the elaboration theory.

The Elaboration Likelihood Model of Persuasion (ELM) is a type of persuasion theory developed by Petty and Cacioppo in 1986. When customers receive a lot of messages containing advertisements, each customer certainly has different preferences and abilities in processing advertising messages. be accepted. Therefore, an effective message must grab the customer's attention, and be designed to be easily understood by the customer.

An Elaboration Likelihood Model (ELM) persuasion theory explains how the process of an advertisement, namely when the customer receives a message, the customer starts to pay attention, but it depends on the relevance received by the customer. If the ability of the customer is good, the message will be processed under conditions of high involvement or low involvement. If customer involvement is high, then attitudes can change via a central route to persuasion. If customer involvement is low, attitudes can change through the peripheral route to persuasion. Based on this theory, for messages to be effective, advertising messages must be made to attract attention, be easy to understand and can influence behavior.

2.2. Electronic Word of Mouth (E-Wom)

The internet has provided consumers with the opportunity to increase their choice of finding and sharing information before or after making an online purchase. Meanwhile, consumers have different motivations to search for information online or generate EWOM by posting reviews and experiences about products and services (King et al., 2014; Zhu & Zhang, 2010). This study adopts the conceptualization given by Matute et al. (2016) that EWOM is a positive or negative statement made by a potential customer, actual or former customer about a product or company, made available to many people institutions via the internet.

The electronic of mouth (EWOM) variable is a variable that has been widely studied in relation to the increasingly widespread use of internet media in marketing activities. In fact, it can be said to have replaced the significance of traditional marketing which relies on processes that require marketers to meet directly with consumers. In order to create ewom, marketers, or in a digital context commonly called influencers, must explore the digital environment in order to attract consumers' attention and encourage the formation of purchasing decisions (Silva et al., 2019).

2.3. Website attractiveness

Website attractiveness shows whether web pages are pleasant to read and subjectively pleasing (Cao et al., 2005). In the digital era, it is very important to pay attention to the attractiveness of a website as a marketing strategy. The results of Zui Chih's (2011) study show that consumer perceptions of web attractiveness positively affect consumer perceptions of trust and trust is positively related to their intention to buy.

2.4. E-Trust

Trust in online transactions, also known as electronic trust (e-trust), can be interpreted as the trust that underlies online transactions, which can be formed due to positive perceptions of the characteristics of the parties involved in the transaction. On the consumer side, e-trust is formed when there is a belief in the positive characteristics of the seller; for example, consumers believe that the seller is honest, committed, and consistently sells quality products. Thus, e-trust is the foundation for creating healthy interactions and strong long-term relationships between sellers and buyers in the digital market (Al-dweeri et al., 2019).

Online transactions are generally carried out in the form of buying and selling between consumers and online stores that join an electronic marketplace (e-marketplace). For sellers, the existence of an e-marketplace provides many advantages, such as access to consumers in a larger scope and allows sellers to focus on the customer service side without having to think too much about marketing problems. For buyers, e-marketplaces are a second guarantee or other important factor considered in trusting sellers and can encourage buyers to make purchasing decisions (Gcora et al., 2019).

The understanding of trust is explained by Trivedi and Yadav (2020) as the willingness of one party to accept the influence or consequences of actions by another party, which is based on the expectation that the other party can actually act as expected. Trust is formed in a situation that is free from control. That is, a party can become a trusted party when it is able to meet the expectations of other parties without being based on coercion in an effort to fulfill those expectations.

According to Liu and Tang (2018), consumers who shop online cannot see the products sold directly, so they cannot check the physical condition or utility of the product. This is prone to make consumers feel that the risk of conducting online transactions is very large. If the product purchased is physically defective or cannot be used according to its function, then consumers are worried about getting difficulties in exchanging or not getting after-sales service that can overcome the problems they are facing. Therefore, in order to build consumer trust so that they are willing to transact online, an online trust building mechanism is formed, which is manifested in the form of presenting detailed information about the seller and the products being sold, presenting testimonials from other consumers, to providing guarantees that consumers' money can be paid. returned if it turns out that there is an error or discrepancy between the product displayed in the online shop window and the product received by the consumer (Khobzi et al., 2019).

In the study of Chou et al. (2015) researchers examined the effect of e-satisfaction and e-trust on e-loyalty. This type of quantitative research using a sample of 482 respondents. Research data were collected using a questionnaire and analyzed by the SEM program. The results of the study state that e-satisfaction and e-trust affect e-loyalty.

2.5. Innovation

Product innovation is one of the main keys to product success in attracting consumer interest (Wahyono, 2020). This is based on the basic concept of innovation, namely as the application of various new technologies to increase the utility and appearance of products so as to meet the needs and desires of consumers who tend to change from time to time (Nørskov et al., 2015). According to Bharadwaj (2018), innovation is the main thing that determines any company's financial success. Innovation refers to a process of translating new ideas into a value proposition for commercial consumers. The results of an innovation are in the form of new products or old products with new features and greater usability, which generally become the main source of new income for the company.

Choshaly (2019) aims to examine the effect of innovation on purchase intention. This type of quantitative research using a sample of 180 respondents. Data were collected using a questionnaire and analyzed by SEM-PLS. The results of the study state that the three dimensions of innovation consisting of relative advantage, trialability and observability have a positive relationship with purchase intention, while the complexity dimension has a negative relationship with purchase intention. The fifth dimension, namely compatibility, is not proven to have a positive relationship with purchase intention.

2.6. Purchase Decision

Each transaction can be realized if the consumer makes a decision to make a purchase of a product (purchase decision), both in the form of goods and services. Therefore, the sellers or producers try hard to be able to encourage consumers to the point of making a decision to buy after going through a long predecessor process. According to Hanaysha (2018), purchasing decisions are a stage that consumers reach after exerting their various resources to obtain information about products, processing them to obtain points of consideration, and evaluating several other alternative products.

According to Salem (2018), purchasing decisions are a process that involves the cognitive aspects of consumers in identifying needs or wants, generating certain options or considerations, and determining specific products or brands to buy. Purchasing decisions include various aspects, including aspects of the decision on which seller to purchase their product, product brand, purchase model, time of purchase, the amount of money to be spent, and the payment method to be used. In addition to these aspects, there are also other aspects that are part of the purchase decision, which may vary from one consumer to another. It is based on a diverse purchase decision-making cycle in providing experiences to consumers.

Purchase decisions between consumers can vary due to the speed of the process that is different. Kaufmann and Gaeckler (2015) state that the speed of consumers in going through various activities in the process of achieving purchasing decisions is very influential on the decisions taken by consumers. Literally, this speed refers to the length of time it takes for consumers to make an initial identification of the things that become their needs or desires to the stage of determining an alternative that is considered the most appropriate among several alternatives that have been found.

The purchase decision is the end point of a cycle that consumers go through. According to Song and Yoo (2016), there is a pre-purchase phase that precedes the purchase decision, where consumers begin to search for information based on their awareness of certain needs or desires. Furthermore, when consumers manage to get enough information, they will be involved in evaluating various alternatives that have been found. The results obtained from the evaluation of these alternatives are in the form of a decision to determine one alternative that is considered the most appropriate and able to meet consumer expectations.

Purchasing decision is a phase in consumer buying behavior that lies between consumer needs and product purchase actions. When consumers are in the purchase decision phase, it means that consumers have gone through several evaluation processes, namely evaluating the needs and alternative products to be purchased. The results obtained from the evaluation serve as the basis for consumers to make decisions to purchase a product. Furthermore, these decisions can be realized in the form of buying products that are actually carried out by consumers (Voyer & Ranaweera, 2015).

According to Sharma and Faropon (2019), purchase intention is a factor that drives a purchase action. Thus, purchase intention has a close relationship with purchasing decisions. Consumers who have the intention to make a purchase can make a decision to realize this intention, namely by making a purchase, or vice versa, canceling their intention to buy a product. In a digital environment, this decision-making process becomes increasingly critical due to limitations in direct interaction between consumers and sellers, causing consumers to feel that they are facing a big risk in making their purchasing decisions.

Purchasing decision-making is a cycle that can vary in the process when it is passed by consumers. The experience that one consumer gets is sometimes not the same as that of other consumers. Therefore, those who are sellers or marketers need to specifically emphasize the entire process that consumers go through, not just at the point of the purchase decision. Thus, it can be seen that achieving the purchasing decision phase requires hard effort not only from sellers but also from consumers (Hanaysha, 2018).

The puchase decision is the endpoint of the consumer decision-making process, which can be influenced by various factors: e-Wom. In online retailing, WOM can often have a significant impact, both positively and negatively, on acquiring new customers. Given the importance of e-Wom, online retailers must implement a specific strategy to foster e-Wom. e-Wom is much stronger than offline Wom because it affects many people in a short time (Chung & Shin, 2010). According to the research results of Matute et al. (2016) the quality of e-wom has a positive direct effect on consumer purchase intention. The perceived usefulness of the product mediates the influence of all e-wom characteristics on online purchase intentions. E-wom’s credibility and quality also indirectly influence repurchase intention through e-trust.

The technology and functions of social media have significantly altered interactions on the internet. This change affects the perceived attractiveness of website. In online sales, website attractiveness is one of the online marketing strategies to attract consumers. The results of research by Mandal et al. (2017) website attractiveness was found to positively affect the intention of revisiting the visitor’s website. The research results of Lee et al. (2011) show that consumer perceptions of web attractiveness have a positive effect on consumer perceptions of trust, and trust has a positive effect on purchase intention.

Website attractiveness that can be accepted by users can increase user trust. Without trust, the buyer will not be willing to exchange money for products or services offered by the seller. The trust factor is becoming increasingly important in the era of digital technology, where many buying and selling transactions occur online. Consumers are increasingly getting alternative products and services as well as sellers that can be chosen according to the criteria (Liu & Tang, 2018).

The trust factor is an important issue in digital transactions. The term that is widely used for this variable is electronic trust, which is interpreted by Al-dweeri et al. (2019) as the trust that underlies online transactions, which can be formed due to positive perceptions of the characteristics of the parties involved in the transaction.

Innovation is the process of applying ideas and technology to produce new products that can provide greater benefits to the company (Bharadwaj, 2018). The innovations made make products more in line with consumer wants and needs (Qi et al., 2020). Product innovation is one of the main keys to product success in attracting consumer interest (Nørskov et al., 2015). If consumers receive product and service comments from reliable sources, clear enough information with strong reasons for up-to-date innovation will be able to improve purchasing decisions.

Based on the overall explanation above, this study was conducted to examine the influence of the role of the innovation and e-trust variables on the influence of e-commerce and website attractiveness on purchase decisions. The relationship between the five variables is an important concept to discuss in a business environment that is increasingly digitalized as it is today. Besides being able to present information that can complement the findings of previous research, the results of this study can also enrich the literature regarding attractiveness in a digital context represented by the attractiveness website variables. This is based on the rarity of research that specifically examines these variables in relation to purchasing decisions.

3. Research Methods

This type of research is explanative quantitative. The object of this research is a local fashion brand startup that sells online through their respective Instagram accounts. The sample in this study was selected using the purposive sampling technique. The determination of the sample in this study refers to the opinion of Hair et al. (2019), which states that the number of samples must be met if using the Structural Equation Model (SEM) analysis, the sample size ranges from 100-200. Based on this opinion, it was determined that the sample in this study was 170 respondents.

The types and sources of data used in this study are primary data. The primary data obtained in this study were the results of a questionnaire distributed to respondents who became the research sampleignations provided.

The data analysis used is SEM with the help of IBM SPSS and AMOS.

Figure 1 : Research Model

4. Results and Discussion

4.1. Validity Test

Before being analysed, it first tested the validity and reliability of the questionnaire distribution. The results of the validity test show that one questionnaire item is invalid (WA4) because it has a Loading Factor value <0.5, so that the item is excluded from the analysed model.

Table 1: Validity Test

4.2. Reliability Test

The results of reliability testing on the research questionnaire showed reliable results because the total value of construct reliability was ≥ 0.70 and AVE ≥ 0.50, so that the data obtained could be further processed for model testing.

Table 2: Reliability Test

Continued multicollinearity and singularity testing where the results obtained show the value of the Determinant of sample covariance matrix = 0.000, so it can be concluded that there are no multicollinearity and singularity problems in the analysed data.

Figure 2: Full Model Standardized SEM

4.3. Testing the Reseach Model Path Analysis

Decision-making to determine whether the hypothesis is accepted or rejected is based on the resulting significance probability; the significance level is α = 5% (0.05). The findings are presented in the following path coefficient table:

Table 3: Path Model Significance Test Results

The results of the significance test in table show that there is a significant direct effect of E-wom on Purchase Decision (0.009 < 0.05), so H1 is accepted, Web Attractiveness has no significant effect on Purchase Decision (0.821 > 0.05), so H2 is not accepted, E wom has a significant effect on E-trust (0.014 < 0.05), then H3 is accepted, Web Attractiveness has a significant effect on E-trust (0.021 < 0.05), then H4 is accepted, and E-Trust has no significant effect on Purchase Decision (0.111 > 0.05) then H5 is not accepted.

The mediation test of the effect of e-wom on purchase decisions via e-trust shows that the direct effect of EW to PD is path c ‘, which is 0.225 and p <0.05 (significant). Meanwhile, the indirect effect is the multiplication between axb or the output above. It can be immediately known that it is -0.025 and p> 0.05 (insignificant). Because the indirect effect decreases from 0.225 to -0.025, it means that e-trust fully mediates the effect of e-wom on purchase decisions, so H6 is not accepted.

The mediation test of the effect of Web Attractiveness on purchase decisions through e-trust shows that the direct effect of WA to PD is path c ‘, which is -0.014 and p> 0.05 (not significant). Meanwhile, the indirect effect is the multiplication of axb or the output above, and it can be immediately known that it is -0.024 and p> 0.05 (insignificant). Because the direct and indirect effects are insignificant, meaning that e-trust cannot fully mediate the influence of Web Attractiveness on purchase decisions, H7 is not accepted.

4.4. Significance Test of Innovation Moderation Variables

The moderation test in this study uses a moderation interaction model. The formula for calculating the loading factor value for the latent interaction indicator is as follows:

Manual calculation of loading values for interaction variables:

= 0,625 0,961 0,579 x 0,754 0,845 0,745

= 2,165 x 2,344

= 5,0748

So, the loading value for the interaction variable is equal to 5,0748. The formula for calculating the variance error value of the interaction latent indicator:

Manual calculation of the error variance value for the interaction variable:

= 0,097119 + 0,157248 + 0,06993

= 0,3243

The error variance value for the interaction variable is equal to 0,3243.

After obtaining the loading factor and error variance values, the interaction variable is then constrained to the research model so that we can see whether or not the moderating variable is influencing this study.

Table 4 explains that the value of the interaction variable between E-trust and innovation on Purchase Decision is 0.21 and p-value: 0.000 < 0.05. It can be concluded that innovation can moderate the effect of E-trust on Purchase Decision, H8 is accepted.

Table 4: Interaction Moderation Test Results

4.5. Discussion

The results showed that hypothesis 1 is accepted, e-wom has a positive and significant effect on Purchase Decisions. This is because consumers positively value the publication of other user reviews on the seller’s own website, as this social space provides added value and usability to the site. Consumers are more likely to revisit online stores to make future purchases if they feel that another user’s comments provide qualifying information. This hypothesis is in line with the research results of Matute et al. (2016) who stated that e-wom has a positive direct effect on consumer purchase intention and is supported by Chung and Shin (2010), which e-Wom is an important marketing tool for retailers today. e-Wom is much more powerful than offline Wom as it affects a large number of people in a short amount of time.

The results showed that hypothesis 2 was not accepted, Web Attractiveness had no significant effect on Purchase Decision (0.821 > 0.05). This explains that the attractiveness of Instagram which is fun, entertaining, has not been able to encourage users to make purchase decisions; this is possible because users tend to only look at displays on Instagram but have not been able to stimulate to decide to buy products offered on Instagram. These results are according to research by Lee et al. (2011) stated that consumer perceptions of web attractiveness positively affect purchasing decisions, trust must be followed/controlled in advance of the sellers on the website, so that it cannot directly influence consumers to decide to buy.

The results show that hypothesis 3 is accepted, that is, E-wom has a significant positive effect on E-trust. This is because it is possible that users will not base their purchases on the correctness of publications on Instagram if they do not trust the seller. In line with the research results of Matute et al. (2016) which state that e-wom has a positive direct effect on e-trust. In this context, e-wom has been identified as a predictor of user’s tendency to trust the Instagram site before making a purchase.

Web Attractiveness has a significant positive effect on E-trust. The results showed that hypothesis 4 is accepted, namely Web Attractiveness, Web Attractiveness has a significant positive effect on E-trust. This explains that Website attractiveness is the attractiveness of the web to attract website users more intently. Website attractiveness that can be accepted by users can increase user trust. Without trust, the buyer will not be willing to exchange money for products or services offered by the seller. The trust factor is becoming increasingly important in the era of digital technology, where many buying and selling transactions occur online. Consumers are increasingly getting alternative products and services as well as sellers that can be chosen according to the criteria (Liu & Tang, 2018). The results of this study are in line with the research of Lee et al. (2011) show that consumer perceptions of web attractiveness have a positive effect on website trust.

The results showed that hypothesis 5 was not accepted, E-Trust did not have a significant effect on Purchase Decision (0.111 > 0.05). This explains that respondents feel that Instagram is not yet committed to providing the needs and what users want. Instagram is only a seller mediator to offer its products so that the level of user trust lies with the seller who uses Instagram, not on the Instagram site itself. According to Al-dweeri et al. (2019) the trust underlies online transactions, which can be formed due to positive perceptions of the characteristics of the parties involved in the transaction. This result is not in line with the research of Lee et al. (2011) that examining the antecedents of purchase intention from clothing retail website, shows that the level of website trust has a positive effect on purchase intention.

E-wom affects Purchase Decision through E-trust. The results show that hypothesis 6 is accepted. That is, e-trust fully mediates the effect of e-wom on purchase decisions. This explains that e-wom is not able to significantly influence purchase decisions without going through e-trust. This explains that in the online realm, trust refers to a customer’s belief that a retail website is legal, ethical, and credible and is able to protect consumer privacy. The importance of trust in online shopping cannot be overstated. The effect of e-wom on trust indicates that consumers who have relationships with online or multi-channel sellers are more willing to trust the seller’s website. As a result, these consumers will be more willing to buy from websites because consumers will see less risk associated with purchasing decisions. The results of this study are in line with Matute et al. (2016) stated that EWOM’s credibility and quality indirectly affect purchases through e-trust.

The results show that hypothesis 7 is not accepted by e-trust cannot fully mediate the influence of Web Attractiveness on purchase decisions. This explains that the high attractiveness of the web cannot guarantee increased user trust. The high attractiveness of the web also cannot guarantee an increase in users’ purchasing decisions. It is possible that the website’s attractiveness Instagram, the seller’s account, is still perceived as less attractive to consumers. In other words, users don’t care how good the content is or how reliable and easy it is to browse the web, but if users don’t find the site attractive, they won’t spend much time paying attention to the web. The results of this study are not in line with the research of Lee et al. (2011) stated that consumer perceptions of web attractiveness positively affect consumer perceptions of trust, and trust has a positive effect on purchase intention.

Innovation as a Moderating Variable on The Effect of e Trust on Purchase Decisions Innovation can moderate the effect of e-trust on Purchase Decisions. This explains that if consumers receive comments on products and services from reliable sources, clear enough information with strong reasons for up-to-date innovation will be able to improve purchasing decisions. Product innovation is one of the main keys to product success in attracting consumer interest (Nørskov et al., 2015). If consumers receive product and service comments from reliable sources, clear enough information with strong reasons for up-to-date innovation will be able to improve purchasing decisions.

The innovations made are not only intended to make products look better with the various new features and benefits that are offered. But also as a tool to explore potential new market shares, encourage the accelerated increase in income, boost market competition, and primarily as a guardian of business continuity (Marín-García et al., 2020).

In this study, researchers only measure purchasing decisions based on E-wom, Web Attractiveness and e-trust variables, so that they are still unable to describe the effect on Purchase Decision as a whole because measuring Purchase Decision can still be measured by various variables and other aspects. In this study, researchers only examined one social media, namely Instagram, maybe for further research, they could use other social media or e-commerce in Indonesia, so that they could get even better results.

5. Conclusions

Based on the research results, it can be concluded that there is a significant direct effect of E-wom on Purchase Decisions. Web Attractiveness has no significant effect on Purchase Decision. E-wom has a significant effect on E-trust. Web Attractiveness has a significant effect on E-trust. E-Trust has no significant effect on Purchase Decisions. E-trust can play a full mediation role E-wom influence on Purchase Decision but not on the influence of Web Attractiveness has no significant effect on Purchase Decision. Innovation is able to moderate E-trust towards Purchase Decision.