The Digital Influence: Exploring the Impact of E-Marketing Techniques on Tiktok Users' Purchase Decisions

This research aims to examine the influence of E-Marketing, namely Influencer Marketing, Content Marketing, and User-Generated Content on TikTok users in Daerah Istimewa Yogyakarta regarding their purchase decisions. The criteria for this research include TikTok users aged 18-34 residing in Daerah Istimewa Yogyakarta. The technique used to gather the sample is purposive sampling, and data is collected through an online questionnaire from 100 respondents. The research follows a descriptive analysis with a quantitative approach, and data analysis is conducted using Partial Least Square (PLS) with the data management program Smart PLS 3.0. The results of this study show that influencer marketing has a positive and significant effect on purchase decisions, content marketing similarly has a positive and significant effect on purchase decisions, and user-generated content also positively and significantly influences purchase decisions.


INTRODUCTION
In the current era of globalization, technology has been rapidly advancing.One notable example of this technological progress is the internet (Pratama & Muchlis, 2020).The development of social media technology has led to various innovations, particularly in the field of marketing (Kotler et al., 2015).This is done to enhance consumer retention and increase a company's market value (Ali & Allan, 2017).One approach that companies can use to respond to the evolution of social media is by enhancing the effectiveness of marketing activities through social media platforms.Social media is one thing that is very effective and efficient in helping the economy to develop and introduce products to many people (Crystrie & Sartika, 2022).TikTok is one such social media platform widely used by the public today.Ginee reported in 2021 that 76% of TikTok users in Indonesia are adults.The demographic age group of TikTok users indicates that young adults fall within the target audience for a brand.This is why researchers have focused their study on TikTok users aged between 18 and 34 years.
The term "purchase decision" can be defined as what and whether to buy, be it products or services, whether to buy or not, where to buy, and how to make payments for the purchased items (Kotler & Amstrong, 2018).In the digital era, consumers need factual information that can be useful for them in making decisions.Content marketing is an art that can provide information to the audience (Fadhilah & Saputra, 2021).Socialinsider released data indicating that the engagement rate on the TikTok application decreased by 28% in 2022 compared to the previous year.This decline in engagement can impact customer purchase decisions in various ways.For example, reduced engagement can decrease brand exposure to users.
Currently, one of the factors suspected to influence purchasing decisions on the TikTok application is influencer marketing, content marketing, and user generated content.TikTok employs influencers to review a brand using a communication style that is easily understood and resonates with consumers (Mahardini et al., 2023).Influencer marketing is both an art and a science aiming to engage influential individuals online in sharing brand messages with their followers and audiences through sponsored content (Sammis et al., 2015).TikTok also provides features allowing consumers to interact directly with influencers during live streaming sessions through the comment section.
Content marketing focuses on creating and distributing valuable, relevant, and consistent content in the form of videos, blogs, and social media posts to attract and retain consumers, ultimately driving profitable consumer actions for the company (Pulizzi, 2012).Content marketing falls under both direct and indirect marketing that can be used to help build awareness, recall, recognition, and actions towards the company, brand, or products (Gunelius, 2011).
User Generated Content (UGC) is voluntary content uploaded by third parties or consumers with content related to products or services (Wafiyah & Wusko, 2023).The TikTok application represents User Generated Content, which is characterized as free and involves users in creating and exchanging content (Omar & Dequan, 2020).According to Wyrwoll (2014), user generated content is content created by users and published on online platforms.Social media consists of platforms containing user generated content.Social media users do not need programming skills to publish content on these platforms.Therefore, users can use social media according to their respective purposes.
Based on research conducted by Mahardini et al. (2023) it is stated that influencer marketing has a positive and significant impact on purchasing decisions.This proves that influencer marketing can reach the market potential and build consumer trust on social media.Research by Mahardini et al. (2023) indicates that content marketing has a positive and significant impact on purchasing decisions.Then, based on the study conducted by Mahardini et al. (2023) discussing influencer marketing, content marketing, and purchasing decisions, it is suggested that further research should add internal and external variables that can be linked to purchasing decision variables.
Based on the aforementioned background phenomena, the author is interested in conducting a study titled "The Influence Of E-Marketing on Purchase Decision"

RESEARCH METHOD
This research is of a quantitative nature.Based on the data analysis, the research design used in this study is descriptive statistical analysis.By employing descriptive statistical analysis, a clear overview of the collected data can be provided, offering a deep understanding of the characteristics of the studied population.In this research, the descriptive analysis utilized describes various variables, including Influencer Marketing, Content Marketing, User Generated Content, and Purchase Decisions.The object of this study is the TikTok application used by users residing in the Special Region of Yogyakarta.This research was conducted between January 2022 and June 2022.
The population for this study consisted of TikTok application users in the Special Region of Yogyakarta.The sample was selected using non-probability sampling techniques, specifically purposive sampling, which involved TikTok application users residing in the Special Region of Yogyakarta selected based on specific criteria.The sample size consisted of 100 respondents, calculated using the Rao Purba (1996) formula.Data sources were divided into two types: primary data (questionnaires) and secondary data (Ginee, 2021;Ali, N. N. H., & Allan, M., 2017).
The data used in this study were obtained through the use of a questionnaire method with a closed questionnaire type.The questionnaire in this study contained questions and statements related to Influencer Marketing, Content Marketing, User Generated Content, and Purchase Decisions on the TikTok application.Variables used in this study included independent variables (Influencer Marketing (X1), Content Marketing (X2), and User Generated Content (X3)) and dependent variables (Purchase Decisions (Y)).Likert scale was employed as the measurement scale.After data collection, the data were tested and analyzed using validity tests, reliability tests, descriptive analysis, and quantitative analysis (outer model, inner model, and model scheme).

Descriptive Analysis of Respondents
From the data gathered through the distributed questionnaires, we obtained information regarding the gender-based characteristics of the respondents, as depicted in Table 1.Understanding the characteristics of the participants in this study is a crucial preliminary step before delving into a detailed analysis.
Table 1 offers an overview of the gender distribution among the 100 respondents involved in our research.This data serves as a valuable factor to consider in understanding how consumer behavior, influenced by gender, can impact their purchasing decisions.In the Table 1, it is noted that the number of questionnaires filled out by 100 respondents resulted in two types of respondents: males, constituting 20%, and females, constituting 80%.This data indicates that the percentage of female respondents significantly dominates, totaling 80 individuals.
In addition to analyzing gender-based data, this study categorized respondents based on their age, using the information collected from the distributed questionnaires.The characteristics of respondents concerning their age are presented in Table 2, representing data from the 100 participants in our research.This data serves as a fundamental component for understanding the potential impact of various age categories on purchasing decisions.Based on the Table 2, out of 100 respondents who completed the questionnaires, it can be concluded that the majority of users in this study are dominated by respondents aged 20-25 years, comprising 80% or 80 respondents.
From the data gathered through the distributed questionnaires, the study derived the characteristics of the respondents based on their occupation or status, which are presented in Table 3.This table, displayed below, provides an overview of how respondents are distributed across various occupational categories.It offers valuable insights into how differing professional and life status backgrounds might influence their preferences and decision-making, particularly in the context of online purchases.This data forms the basis for comprehending the diverse representation of occupations among the study's participants.According to the Table 3, out of 100 respondents who completed the questionnaires, it can be inferred that the majority of users in this study are students, totaling 71 (71%) respondents.
The study drew characteristics of the respondents based on their average monthly income from the data obtained through the distributed questionnaires.These findings are displayed in Table 4, providing a breakdown of the income distribution among 100 respondents.This financial demographic data is crucial for our research as it offers insights into the participants' financial backgrounds, which play a significant role in understanding consumer behavior and purchase decisions.Based on the Table 4, 39 (39%) respondents dominate the average monthly income with earnings less than Rp1,000,000 per month.

Partial Least Squares Model Scheme
The analysis employing Partial Least Square (PLS) through SmartPLS provides essential metrics, including Cronbach's alpha, composite reliability, average variance extracted (AVE), R-square, T-statistics, and P-values.These metrics serve to elucidate the research results systematically, allowing the researcher to offer comprehensive and precise explanations of the final research findings.In this study, the relationships between variables among the 100 respondents are illustrated in the model schema, as depicted in both Figure 1 and Figure 2.

Results of Outer Model Testing
Before conducting further research, it is important to know the level of reliability of the variables used.In Table 5, the Cronbach's Alpha value is presented, this value is a reference that the variables used are reliable.From the Table 5, it can be seen that the Cronbach's alpha values for each variable are above 0.7.Therefore, it can be stated that all variables in this study have good reliability.

Results of Inner Model Testing
The R-Square test is used to determine how much the endogenous variable is influenced by other variables.Below we present the results of internal model testing in table 6 which reveal the extent to which the Purchase Decision variable is influenced by the variables influencer marketing, content marketing, and user-generated content.The Table 6 indicates that the purchase decision variable is influenced by influencer marketing, content marketing, and user generated content variables by 71.1%, with the remaining 28.9% being influenced by other factors not included in the research model.The influence of influencer marketing, content marketing, and user generated content variables on purchase decisions, totaling 0.711, falls into the strong or good category.
The Q-Square test is utilized to measure how well the observations generated by the model and parameter estimates align.In table 7, the results of the Q-Square calculation are displayed which measure the level of relevance of model predictions for the variables used in this research.According to the Table 7, the purchase decision variable obtains a Q-Square value of 0.438.From this result, it can be concluded that this research model possesses predictive relevance, meaning that the observations generated by the model and parameter estimates are strong or good.
In Table 8, there are path coefficient test results that measure the relationship between the variables influencer marketing, content marketing, and user-generated content with the Purchase Decision variable.In the path coefficient test presented in Table 8, if the path coefficient value > 0, it means the variable has a positive relationship.If the path value < 0, the variable has a negative relationship with another variable.The higher the value of the independent variable, the lower the dependent variable, with the t-statistic value having to be greater than the significance t-table value, which is 1.66 at 5% significance level (Ghozali & Latan, 2015).

Hypothesis Testing
H1: There is an influence of influencer marketing on purchasing decisions.Based on Table 8, it is known that the t-statistic result is 3.912 > t-table (1.96) and the P-Value is 0.000 < 0.05.Therefore, it is concluded that influencer marketing has a significant impact on purchasing decisions.Thus, hypothesis 1 is accepted because the t-statistic is greater than 1.96 and the P-Value is less than 0.05.It can be interpreted that if influencer marketing increases, the purchasing decisions of TikTok application users in the Special Region of Yogyakarta will also increase.
H2: There is an influence of content marketing on purchasing decisions.Based on Table 8, it is known that the t-statistic result is 2.033 > t-table (1.96) and the P-Value is 0.043 < 0.05.Therefore, it is concluded that content marketing has a significant impact on purchasing decisions.Thus, hypothesis 2 is accepted because the t-statistic is greater than 1.96 and the P-Value is less than 0.05.It can be interpreted that if content marketing increases, the purchasing decisions of TikTok application users in the Special Region of Yogyakarta will also increase.
H3: There is an influence of user generated content on purchasing decisions.Based on Table 8, it is known that the t-statistic result is 2.657 > t-table (1.96) and the P-Value is 0.008 < 0.05.Therefore, it is concluded that user generated content has a significant impact on purchasing decisions.Thus, hypothesis 3 is accepted because the t-statistic is greater than 1.96 and the P-Value is less than 0.05.It can be interpreted that if user generated content increases, the purchasing decisions of TikTok application users in the Special Region of Yogyakarta will also increase.

The influence of influencer marketing on purchasing decisions
From the results of this study, it can be inferred that if companies utilize influencer marketing strategies by increasing collaborations with influencers tailored to consumers' characteristics and interests, it is expected to enhance purchasing decisions.
The presence of influencers can assist companies in their marketing activities.Through influencer marketing, companies can engage in effective marketing in today's digital era.The significance of influencer marketing on purchasing decisions lies in the fact that influencers can create and share content related to users' interests.TikTok users can follow influencers they like and resonate with their interests.In this study, it was found that the resonance indicator emerged as the strongest indicator within the influencer marketing variable.This was discerned based on respondents' answers, indicating that the level of engagement, encompassing views, likes, shares, and comments on an influencer's content, can increase awareness and interest in the content they view.Therefore, an influencer's content engagement on the TikTok application tends to enhance purchasing decisions within the app.
This research finding is supported by a previous study by Mahardini et al. (2023) indicating that influencer marketing has a positive and significant impact on purchasing decisions.

The influence of content marketing on purchasing decisions
From the results of this study, it can be inferred that if companies employ content marketing strategies by creating and distributing valuable, relevant, and consistent content, it is expected to enhance purchasing decisions.
Content marketing can assist companies in their marketing efforts.Through content marketing, companies can engage in effective marketing in today's digital era.The importance of content marketing on purchasing decisions lies in its ability to attract and retain consumers, ultimately driving favorable consumer actions for the company.In this study, the reader cognition indicator emerged as the strongest indicator within the content marketing variable.This was determined based on respondents' answers, indicating that the content created and shared on the TikTok application is interactive and easy to understand.Therefore, interactive and easily understandable content tends to enhance purchasing decisions within the TikTok application.
This research finding is supported by a previous study by Mahardini et al. (2023) indicating that content marketing has a positive and significant impact on purchasing decisions.

The influence of user generated content on purchasing decisions
From the results of this study, it can be inferred that if companies employ user generated content strategies by allowing users to express themselves by uploading content and giving opinions, indirectly, user generated content can provide exposure or benefits that can increase awareness of the TikTok application itself.
User generated content can also assist companies in their marketing activities.The significance of user generated content on purchasing decisions lies in the fact that user generated content is created by users voluntarily, instilling greater trust in the content or reviews uploaded by other users, thus influencing the decisions made by consumers.In this study, the attitude on user generated content indicator emerged as the strongest indicator within the user generated content variable.This was determined based on respondents' answers, indicating that user generated content is more trustworthy than advertisements within the TikTok application and can aid consumers before making purchasing decisions.Therefore, voluntarily created content by users can enhance purchasing decisions within the TikTok application.
This research finding is supported by a previous study by Scholz et al. (2013) indicating that user generated content directly influences purchase decisions.This research has yielded insights into the impact of various marketing strategies on purchasing decisions.Three key components were meticulously examined in this research: Influencer Marketing, Content Marketing, and User Generated Content.In the realm of Influencer Marketing, the results clearly indicate a positive and significant influence on purchasing decisions.
The effect size, indicating the magnitude of this influence, was found to be significant.Similarly, Content Marketing emerges as a crucial force shaping consumer choices.The analysis shows a positive and significant influence on purchasing decisions attributed to effective content strategies.This study also delves deeper into the impact of User Generated Content on consumer behavior.
The analysis illuminates the positive influence evident in purchasing decisions originating from content created by users.This suggests that consumers not only respond to professionally curated content but are also profoundly influenced by opinions and experiences shared by fellow consumers.
The results of this research show a strong correlation between Influencer Marketing, Content Marketing, User Generated Content, and purchasing decisions.It can be concluded that influencer marketing has a positive and significant effect through purchase decision, also content marketing has a positive and significant effect through purchase decision, and under generated content has a positive content and significant through buyer decision.

RECOMMENDATIONS
Based on the research results, discussions, and conclusions obtained, this study identifies several limitations that provide room for suggestions and recommendations.First, Business people are expected to be able to utilize Influencer Marketing, Content Marketing, User Generated Content to attract more purchasing decisions.Especially influencer marketing which has the highest value, of course this is a huge opportunity for business people to take advantage of technological developments.
Also, for TikTok as a company, the research indicates that the influencer marketing variable has the highest impact on purchasing decisions.However, it was also found that dissatisfaction with influencer behavior can affect consumer trust.Therefore, it is recommended for TikTok to carefully select influencers, considering aspects that reflect the brand image of the application.This action is expected to improve purchasing decisions and prevent users from seeking alternatives on other platforms.
Second, in the context of future research, it is expected to add other variables besides influencer marketing, content marketing, user-generated content, and purchasing decisions.Are there other variables that are more significant in purchasing decisions compared to the influencer marketing variable.
Variables such as viral marketing, price, product quality, and other factors can enrich this study and provide a more comprehensive scientific insight.Additionally, it is important to expand the age range of the sample, which is currently limited to TikTok users aged 18-34.
Research could investigate how technology adoption influences consumer responses to digital content, especially in the context of generational differences.This study will provide insights into how technology and age differences affect consumer preferences in digital marketing.
Figure 1.Outer Model Figure 2. Inner Model