Volume 6.
Number 1, 2025 https://ijble.
com/index.
php/journal/index The Influence of Brand Ambassador.
User Interface and Free Shipping Program on Purchasing Decisions through Tokopedia Ecommerce with Brand Trust as an Intervening Variable Annisa Anindya1.
Rijal Arslan2.
Kristiurman Jaya Mendrofa3 Ichsan Satya University.
Indonesia1.
Al Azhar University Indonesia2.
Universitas Nias Raya.
Indonesia2 okt@gmail.
com1, rijal.
arslan@uai.
id2, kris@uniraya.
ABSTRACT:
This study aims to evaluate the influence of Brand Ambassador.
User Interface, and Free Shipping Programme on E-Commerce Tokopedia, with Brand Trust acting as an intervening variable.
This study employs quantitative approaches and purposive sampling procedures with a sample of 100 respondents.
Employing path analysis with Smart PLS version 3 for data analysis.
The study's findings demonstrate that brand ambassadors have a favourable and substantial effect on brand trust.
nevertheless, their influence on purchasing decisions through Tokopedia is negligible.
The user interface has a positive and substantial impact on brand trust and purchasing decisions.
The Free Shipping Programme affects brand confidence and purchasing decisions.
Brand trust successfully mediates the impact of Brand Ambassador.
User Interface, and Free Shipping Programme on purchasing decisions.
Keywords:
Brand Ambassador.
User Interface.
Free Shipping Program.
Purchase Decision.
E-Commerce.
Brand
Trust
INTRODUCTION
The progression of information technology has diversified the forms of selling material available on the Internet.
A marketplace is a venue for the exchange of goods, facilitating interactions between vendors and buyers on a platform.
Since its inception in Indonesia in 2009, there are already 88 startups offering marketplace In addition to the corporate aspect, the quantity of markets in Indonesia is consistently rising.
In 2023, the user count had attained 58.
63 million.
According to the 2021 Populix study, the six most popular marketplaces in Indonesia are as follows: Shopee in first position.
Tokopedia in second, and Lazada.
Bukalapak.
JD.
ID, and Blibli in third, fourth, fifth, and sixth places, respectively.
Despite being the first marketplace in Indonesia and a domestic entity.
Tokopedia has only attained a second-place ranking and has been unable to usurp Shopee's dominance in the leading position.
As per the 2022 survey data from the Central Bureau of Statistics, 30 percent of eCommerce firms do online sales through social media platforms, such as Facebook.
Instagram.
Twitter.
TikTok, and YouTube.
Furthermore, only 75 percent of firms manage sales accounts on marketplaces or digital platforms.
The restricted uptake of digital marketplaces poses a barrier for the government in promoting eCommerce businesses in Indonesia to migrate to these platforms, therefore facilitating their access to a wider market.
Furthermore, 7.
percent of firms employ.
The author aims to investigate the influence of Brand Ambassador.
User Interface, and Free Shipping Programme on purchasing decisions through Tokopedia E-Commerce, with Brand Trust serving as an intervening variable, due to the insufficient online sales channel via the marketplace in Indonesia and the inability to dominate Tokopedia.
Literature Review Brand Ambassador Volume 6.
Number 1, 2025 https://ijble.
com/index.
php/journal/index Lea-Greenwood .
asserts that Brand Ambassadors serve as instruments for firms to engage with the public regarding their role in enhancing There are three indicators required by Brand Ambassadors.
Royan .
defines the three characteristics as attractiveness, trustworthiness, and expertise.
Attractiveness extends beyond the physical appeal, encompassing various attributes observable in the supporter, such as intelligence, personality traits, lifestyle, and athletics.
Trustworthiness, where the level of trust, dependability, like someone who can be trusted.
Expertise denotes the information, experience, or abilities possessed by an advocate pertinent to the subject they represent.
User Interface According to Lastiansah .
User Interface is a way for programs and users to interact.
Elements of the User Interface consist of color, writing, layout and shape that are designed to be as attractive and easy as possible so that consumers are interested and easy to use the application.
UI is a link between users and applications where the design of an application is tailored to the needs and desires of consumers.
Sometimes interface design is not so concerned that customers do not get a sense of comfort.
There are indicators used by researchers to be able to describe user interface variables, among others:
Affordance, refers to the clarity and functionality of controls so that users are familiar with and understand how to use them.
Visibility, refers to the availability of control and feedback provided.
Consistency, including the arrangement of information, menu arrangement, size and shape of icons, and the order or flow of the system so that users can understand where the menu is located, how the system flows, and what it does at a glance.
Error Handling, a good User Interface design should anticipate common errors and help users to avoid them.
When an error occurs, the system must state the message or information about the error specifically by informing what is wrong and explaining how to fix it.
Free Shipping Program The free shipping program is a promotional program carried out by various ecommerce in collaboration with saller to stimulate consumers to make transactions.
Free shipping promos are another form of sales promotion that provides offers to stimulate consumers to purchase products and increase the number of products purchased by consumers (Amalia and Wibowo, 2.
This free shipping program promo makes consumers not have to spend additional money on shipping costs Maulana and Asra .
So that with this promotion it is expected to influence consumers in purchasing decisions and this free shipping program has many positive impacts on sellers, starting from increasing the number of sales and bringing in new customers.
According to Populix data in 2021, the main reason people choose ecommerce in Indonesia is because of attractive discount promos.
As many as 40% of male respondents and 30% of female respondents agreed with this reason.
Meanwhile, other points of consideration are as follows:
Volume 6.
Number 1, 2025 https://ijble.
com/index.
php/journal/index Shipping promo Habitual use of one particular e-commerce app Product completeness Service interface display Purchase Decision Schiffman .
defines a purchase decision as the selection among two or more alternative purchasing possibilities.
To enable an individual to make a decision, multiple alternative options must be accessible.
The purchasing process encompasses need recognition, pre-purchase research, alternative appraisal, and decision-making.
According to Kotler and Keller .
, the indicators that buyers go through to reach buyer decisions go through five stages, namely:
Problem recognition: The purchasing process commences with problem recognition, wherein the buyer identifies a discrepancy between the actual and the desired state.
Information search: When consumer motivation is robust and the desired product is accessible, the likelihood of purchase increases.
A consumer who is already engaged may pursue additional information.
Evaluation of alternatives: The phase of the Purchase Decision process in which customers assess information or compare other brands among a selection of options.
Purchasing decision: Consumers assess different products, occasionally employing meticulous analysis and rational reasoning, while at other times making impulsive purchases or relying on intuition.
Post-purchase behaviour: Consumers engage in subsequent actions following a purchase, influenced by their level of satisfaction or discontent.
E-commerce E-commerce refers to a set of activities associated with the exchange of commodities or services domestically and internationally, aimed at transferring rights to these items or services in exchange for remuneration, facilitated through various electronic devices and processes.
(Rahman et al.
, 2.
Brand Trust According to Kotler & Armstrong in Parulian .
brand trust or brand trust is a consumer who believes in a brand with all the risks involved.
Due to high expectations for the brand which will provide positive results for consumers so that it will generate loyalty and trust in a brand.
A consumer who has used the brand and then followed by others, it will create evidence that the brand has guaranteed quality, giving rise to a sense of trust in the brand.
(Ranti et al.
Conceptual Framework The conceptual framework will outline the relationship between Brand Ambassador.
User Interface.
Free Shipping Program.
Purchase Decision and Brand Trust.
This conceptual framework acts as a guide for empirical investigation, offering an organized method to identify the complex dynamics at play.
Volume 6.
Number 1, 2025 https://ijble.
com/index.
php/journal/index Figure 1.
Conceptual Framework Based on the above literature, the following hypothesis is proposed:
H01: The Brand Ambassador does not exert a direct, positive, and substantial influence on Purchasing Decisions via Tokopedia e-commerce.
Ha1: The Brand Ambassador exerts a favourable and substantial influence on Purchasing Decisions via Tokopedia e-commerce.
H02: The User Interface does not exert a direct, positive, and substantial influence on Purchasing Decisions via Tokopedia e-commerce.
Ha2: The User Interface has a substantial and favourable impact on Purchasing Decisions within the Tokopedia e-commerce platform.
H03: The Free Shipping Program does not directly have a positive and significant effect on Purchasing Decisions through Tokopedia e-commerce.
Ha3: The Free Shipping Program directly has a positive and significant effect on Purchasing Decisions through Tokopedia e-commerce.
H04: Brand Ambassador directly does not have a positive and significant effect on Brand Trust.
Ha4: Brand Ambassador directly has a positive and significant effect on Brand Trust.
H05: The User Interface does not exert a direct, positive, and substantial influence on Brand Trust.
Ha5: User Interface directly has a positive and significant effect on Brand Trust.
H06: Free Shipping Program does not directly have a positive and significant effect on Brand Trust.
Ha6: Free Shipping Program directly has a positive and significant effect on Brand Trust.
H07: Brand Trust does not directly have a positive and significant effect on Purchasing Decisions through Tokopedia e-commerce.
Ha7: Brand Trust directly has a positive and significant effect on Purchasing Decisions through Tokopedia e-commerce.
H08: The Brand Ambassador exerts no positive or substantial influence on Purchasing Decisions via Tokopedia e-commerce, with Brand Trust serving as an intervening variable.
Ha8: The Brand Ambassador exerts a favourable and substantial influence on Purchasing Decisions via Tokopedia e-commerce, with Brand Trust serving as an intervening variable.
H09: User Interface indirectly has no positive and significant effect on Purchasing Decisions through Tokopedia e-commerce with Brand Trust as an intervening variable.
Volume 6.
Number 1, 2025 https://ijble.
com/index.
php/journal/index Ha9: User Interface indirectly has a positive and significant effect on Purchasing Decisions through Tokopedia e-commerce with Brand Trust as an intervening H010: The Free Shipping Programme does not exert a positive and meaningful influence on Purchasing Decisions via Tokopedia e-commerce, with Brand Trust serving as an intervening variable.
Ha10: The Free Shipping Programme exerts a favourable and significant influence on Purchasing Decisions via Tokopedia e-commerce, with Brand Trust serving as an intervening variable.
METHODS
Place and Time of Research This study utilises quantitative methods to examine the impact of brand ambassadors, user interface, and promotional programmes .
, free shippin.
on purchasing decisions on the Tokopedia e-commerce platform, with brand trust acting as an intervening variable.
The research was conducted from October 2024 to December 2024 in the Greater Jakarta area, targeting persons aged 17 to 60 who have made purchases through Tokopedia e-commerce in the preceding month.
Data Source and Type This study collected primary data on brand ambassador, user interface, and Free Shipping Programme variables affecting purchasing decisions via Tokopedia e-commerce, utilising brand trust as a mediating variable.
Primary data was obtained by an online survey utilising distributed questionnaires, as this method is seen as more effective and efficient.
This study utilised secondary material sourced from books, journals, magazines, statistical publications, and the Internet.
Population and Sample Sugiyono .
3, p.
Population refers to a generalised group of entities or individuals possessing specific attributes and characteristics selected by researchers for study and analysis.
The demographic of this study comprises individuals from Jabodetabek who engage in purchases via the Tokopedia ecommerce platform.
The sample is part of the population that is used as a source of data in research, as stated by Sugiyono .
3, p.
In this study, the sample consists of people who have made purchases through Tokopedia e-commerce for at least the last 1 month (October-November 2.
, with an age range between 17 and 60 years.
This sample includes demographics that are relevant for the study being conducted.
Non-probability sampling method used in this study.
Non-probability sampling is a methodology that does not afford equal chances for each element or individual within the population to be chosen as a sample Sugiyono .
3, p.
Therefore, the procedure used is Purposive Testing.
Sugiyono .
3, p.
says that purposive sampling is a way of taking samples by considering certain things.
The selection of this research sample is based on two certain factors, namely having made purchases through Tokopedia e-commerce and purchases made in at least the last 1 month.
This study use the Volume 6.
Number 1, 2025 https://ijble.
com/index.
php/journal/index Lemeshow formula due to the population being unknown or infinite, it is determined that the number of respondents to be studied is 96.
04 or 97 people.
Data Collection and Processing Techniques Sampling Technique According to Sujarweni .
, research that uses data collection techniques to reveal and collect quantitative data from respondents in accordance with the scope of a study is called data collection techniques or samples.
This study utilises secondary and primary data as the foundation for analysis to address research Because the selection of appropriate data collection methods will produce relevant, accurate, and reliable data, the significance of a study that profoundly influences research outcomes.
This research uses the following approaches:
Primary Data Researchers ask several lists of questions that can generally be closed or open to selected respondents.
Sugiyono's opinion .
3, p.
The attitudes, views, and perceptions of an individual or group regarding social phenomena can be quantified using a Likert scale in research.
Secondary Data Sugiyono .
defines data sources as entities that do not directly furnish data to collectors, including documents or other individuals.
Moreover, data collection methods may include interviews, questionnaires, observations, or a mix Literature concerning brand ambassadors, user interfaces, promotional initiatives, purchasing decisions, and brand trust constitutes secondary data sources.
These books can be printed books or electronic books.
Theories that can be used as guidelines in this research.
Data Analysis and Methods Data Analysis The previously distributed questionnaires were tabulated into data that was ready to be processed.
The data that is ready to be processed is then processed again with the help of the Path Analysis model.
According to Sani and Maharani .
, this model can determine whether a set of independent .
variables has an endogenous or direct influence on the dependent variable.
Standardized regression coefficients, also known as path analysis, are regression coefficients calculated from a database that has been set in standard numbers (Zscor.
SmartPLS (Partial Least Squar.
software assists in this analysis.
Data Analysis Method Path analysis using the partial square method and the PLS program was used for data processing in this study.
The PLS analysis procedure comprises three stages: initial outer model analysis, subsequent inner model analysis, and final testing of established hypotheses.
Outer Model (Measurment Mode.
The outer model delineates the relationships between each indicator and other factors.
This model establishes the correlation between latent variables and their indicators.
The following are some analyses about the outer model check:
Convergent validity Volume 6.
Number 1, 2025 https://ijble.
com/index.
php/journal/index The factor loading value of the latent variable with its indication represents the degree of convergent validity.
A factor cross loading value beyond 0.
7 is deemed optimal, although a number surpassing 0.
5 remains acceptable.
Discriminant Validity This value is a cross-loading factor loading, and can be used to compare the loading value on the construct in question to see whether or not the construct has sufficient discriminant with other constructs, this value must be greater than the loading value.
Composite Reliability.
High reliability data has a composite reliability greater than 0.
Inner Model Ghozali .
says that structural model analysis or inner model is carried out to ensure the structure is strong and accurate.
There are a number of indicators that can be used to evaluate the inner model, including:
The Coefficient of Determination (RA) measures the extent to which the independent variable affects the dependent variable.
The coefficient of determination suggests that a RA value of 0.
5 is classified as good, a value 25 and 0.
5 is viewed as moderate, and a value below 0.
25 is perceived as poor.
Size Effect An endeavour to acquire an appropriate model.
The exogenous variable lengths exhibit effect sizes of 0.
02, 0.
15, and 0.
35, all of which are little however significant concerning the structural components.
Hypothesis Testing The significance level of a hypothesis can be used statistically to determine whether the hypothesis should be accepted or rejected.
In this study, the significance level used is 5%.
To reject a hypothesis, the significance level or confidence level is 05 if the chosen significance level is 5%.
In this study, there is a 95% chance of making the correct choice and a 5% chance of making the wrong choice.
The following factors are the basis for decision making:
If the P-value is less than 0.
H0 is rejected and Ha is accepted, indicating that the endogenous variables significantly affect the exogenous variables.
If the P-value is more than 0.
05, it indicates that the endogenous variable has no significant effect on the exogenous variable, which indicates that H0 is accepted and Ha is rejected.
RESULTS AND DISCUSSION
Research Results Outer Model (Measurment Mode.
During the data analysis phase, we will evaluate three criteria employed in Data Analysis Techniques within the smart PLS data analysis process.
Convergent validity, discriminant validity, and composite reliability.
Convergent Validity Table 1.
Outer Loading (Measurement Mode.
BA (X.
X1_1 X1_2 X1_3 UI (X.
X2_1 X2_2 X2_3 SDP (X.
X3_1
X3_2
X3_3
BT (Z)
X4_1
X4_2
X4_3
BD (Y)
X5_1
X5_2
X5_3
Volume 6.
Number 1, 2025 https://ijble.
com/index.
php/journal/index BA (X.
X1_4 X1_5 UI (X.
X2_4 X2_5 SDP (X.
X3_4 X3_5 BT (Z)
X4_4
X4_5
BD (Y)
X5_4
X5_5
Based on the outer loading table, it can be seen that the correlation value between the construct and the variable in all variable indicators has an outer loading value> 0.
70, so the data can be used for the next test.
Figure 2.
SmartPLS Algorithm Results Discriminant Validity and Composite Reliability The reliability coefficient of a construct and the Average Variance Extracted (AVE) for each construct can be employed to evaluate the validity and reliability criteria in the Discriminant Validity Test.
The established requirements include a Cronbach Alpha value beyond 0.
60, an Average Variance Extracted (AVE) value 50, and a composite reliability value greater than 0.
70 for a novel The subsequent results pertain to the Construct Reliability and Validity assessments conducted by researchers:
Table 2.
Average Variance Extracted (AVE).
Composite Reliability and Cronbach's Alpha Brand Ambassador (X.
User Interface (X.
Free Shipping Program (X.
Brand Trust (Z) Purchase Decision (Y) Cronbach`s 0,914 0,939 0,865 0,912 0,918 Composite reliability .
0,926 0,945 0,865 0,917 0,925 Composite Average variance reliability .
extracted (AVE) 0,936 0,747 0,954 0,804 0,903 0,650 0,934 0,739 0,939 0,755 The following are the results of the researcher's outer model in Table 2:
Brand Ambassador (X.
exceeds the critical limit of 0.
60 with a Cronbach's Alpha value of 0.
while the AVE value is 0.
747 above 0.
50 and the composite reliability value is 0.
926 above 0.
The reliability of the Brand Ambassador variable (X.
has been proven by these results.
User Interface (X.
has a Cronbach's Alpha value of 0.
939 which is greater than 60, its AVE value is 0.
804 which is greater than 0.
50 and its composite reliability value is 0.
945 which is greater than 0.
The reliability of the User Interface variable (X.
has been confirmed by these results.
The Postage Discount Program (X.
has a Cronbach's Alpha value of 865>0.
60, the AVE value is 0.
650>0.
50, and the composite reliability is Volume 6.
Number 1, 2025 https://ijble.
com/index.
php/journal/index 865>0.
The reliability of the Postage Discount Program (X.
variable has been proven by these results.
Brand Trust (Z) has a Cronbach's Alpha with a value of 0.
912 greater than 0.
then the AVE value is 0.
739>0.
50, while the composite reliability is 0.
917>0.
From these results, the reliability of the Brand Trust (Z) variable has been .
Purchase Decision (Y) has a Cronbach's Alpha result of 0.
918 which is greater 60 then the AVE value is 0.
755>0.
50 while the composite reliability value 925>0.
From these results, the reliability of the Purchase Decision (Y) variable has been fulfilled.
Inner Model Coefficient of Determination (R)2 The analytical phase termed the inner mode, or structural model, aims to analyse the links among constructs, evaluate their importance, and calculate the Rsquare value within this research framework.
Model evaluation involves employing R-square to determine the dependent variable of the t-test.
Path analysis using the Partial Least Squares (PLS) method initially assesses R-square to evaluate the degree of influence some variables have on others inside the model.
Furthermore, it is essential to assess the significance of the impact of each of these variables.
This stage aids in assessing the model's efficacy and elucidating the correlation between variables.
The criteria stipulate that an R-square value of 0.
75 to less than 0.
50 indicates a substantial effect .
arge/stron.
, an R-square value of 0.
to less than 0.
25 signifies a moderate effect, and an R-square value of less than 0.
denotes a weak effect.
Table 3.
R-Square Brand Trust (Z) Purchase Decision (Y) R-square 0,673 0,782 Adjusted R-square 0,663 0,773 The R-square value for the Brand Trust variable (Z) is 0.
67, indicating a significant influence.
The Purchasing Decision variable (Y) similarly possesses an Rsquare value of 0.
78, indicating a significant influence.
Size Effect .
Direct Effect Table 4.
Direct Effect Brand Ambassador (X.
-> Brand Trust (Z) Brand Ambassador (X.
-> Purchase Decision (Y) User Interface (X.
-> Brand Trust (Z) User Interface (X.
-> Purchase Decision (Y) Free Shipping Program (X.
-> Brand Trust (Z) Free Shipping Program (X.
-> Purchase Decision (Y) Brand Trust (Z) -> Purchase Decision (Y) Original Sample (O) Sample mean (M) Standard
(STDEV)
0,196
0,123
0,199
0,111
statistics values (|O/STDE V|) 0,069 2,825 0,006 0,085 1,458 0,148 0,500 0,416 0,507 0,442 0,085 0,102 5,851 4,095 0,000 0,000 0,304 0,294 0,093 3,270 0,001 0,466 0,445 0,108 4,304 0,000 0,454 0,449
0,102
4,441
0,000
Based on the table above, it can be concluded as follows:
Volume 6.
Number 1, 2025 https://ijble.
com/index.
php/journal/index The path analysis results indicate several key findings regarding the relationships between variables in the study.
First, the Brand Ambassador Ie Brand Trust relationship shows a path coefficient of 0.
with a P-value of 0.
which is smaller than 0.
This indicates a significant relationship, meaning that an increase in the value of the Brand Ambassador can enhance Tokopedia's Brand Trust.
However, the Brand Ambassador Ie Purchase Decisionrelationship has a path coefficient of 0.
with a P-value of 0.
148, which is greater than 0.
indicating that although positive, the influence is not significant.
This suggests that increasing the value of the Brand Ambassador does not significantly impact purchasing decisions through Tokopedia.
Next, the User Interface Ie Brand Trust relationship has a path coefficient 500 .
and a P-value of 0.
000, which is less than 0.
This result confirms a significant positive relationship, indicating that improvements in the User Interface can significantly increase Brand Trust in Tokopedia.
Similarly, the User Interface Ie Purchase Decisionrelationship shows a path coefficient 416 .
with a P-value of 0.
000, suggesting that enhancing the User Interface directly contributes to increasing purchasing decisions through Tokopedia.
Additionally, the Free Shipping Program Ie Brand Trust relationship reports a path coefficient of 0.
with a P-value of 0.
001, which is also significant.
This indicates that a higher value in the Free Shipping Program can effectively increase Brand Trust.
Furthermore, the Free Shipping Program Ie Purchase Decision relationship has a path coefficient of 0.
with a P-value of 000, showing that an increase in the Free Shipping Program has a significant positive effect on purchasing decisions through Tokopedia.
Lastly, the relationship between Brand Trust Ie Purchase Decision reveals a path coefficient of 0.
with a P-value of 0.
000, which is smaller than This indicates a significant and positive relationship, meaning that an increase in Brand Trust will significantly enhance purchasing decisions through Tokopedia.
Indirect Effect Original Sample Sample (O) mean (M) Brand Ambassador (X.
-> Brand Trust (Z) -> Purchase Decision (Y) User Interface (X.
-> Brand Trust (Z) -> Purchase Decision (Y) Free Shipping Program (X.
-> Brand Trust (Z) -> Purchase Decision (Y) 0,089 0,091 0,227 0,226 0,138 0,131 Standard T statistics (|O/STDEV|) values (STDEV) 0,039 2,281 0,02 0,063 3,604 0,00 0,048 2,846 0,00 Based on the table above, it can be concluded as follows:
Path Coeficient Brand AmbassadorIe Brand trustIe Purchase Decision = 0.
while the P-Value is 0.
025 less than 0.
, it can be concluded that the higher the value of the Brand Ambassador can increase the value of purchasing decisions through Tokopedia which is mediated by brand .
Path Coeficient User InterfaceIe Brand trustIe Purchase Decision = 0.
while the P-Value of 0.
000 is less than 0.
, it can be concluded that increasing the value of User Interface can increase the value of purchasing decisions through Tokopedia which is mediated by brand trust.
Volume 6.
Number 1, 2025 https://ijble.
com/index.
php/journal/index .
Path Coeficient Free Shipping ProgramIe Brand trustIe Purchase Decision = 138 (Positiv.
while the P-Value is 0.
005 less than 0.
, it can be concluded that the increasing value of the Free Shipping Program can increase the value of purchasing decisions through Tokopedia which is mediated by brand Hypothesis Testing Brand Ambassador (X.
-> Brand Trust (Z) Brand Ambassador (X.
-> Purchase Decision (Y) User Interface (X.
-> Brand Trust (Z) User Interface (X.
-> Purchase Decision (Y) Free Shipping Program (X.
-> Brand Trust
(Z)
Free Shipping Program (X.
-> Purchase Decision (Y) Brand Trust (Z) -> Purchase Decision (Y) Rand Ambassador (X.
-> Brand Trust (Z) > Purchase Decision (Y) User Interface (X.
-> Brand Trust (Z) -> Purchase Decision (Y) Free Shipping Program (X.
-> Brand Trust
(Z) -> Purchase Decision (Y) Original Sample Sample mean (M) (O) 0,196 0,199 0,123 0,111 Standard
T statistics (|O/STDEV| values (STDEV) 0,069 2,825 0,006 0,085 1,458 0,148 0,500 0,416 0,507 0,442 0,085 0,102 5,851 4,095 0,000 0,000 0,304 0,294 0,093 3,270 0,001 0,466 0,445 0,108 4,304 0,000 0,454 0,089 0,449 0,091 0,102
0,039
4,441
2,281
0,000
0,025
0,227
0,226
0,063
3,604
0,000
0,138
0,131
0,048
2,846
0,005
First, the Brand Ambassador Ie Brand Trust hypothesis shows a P-value of 006, which is less than 0.
05, and a positive original sample (O) value.
This indicates that Ha1 is accepted while H01 is rejected, confirming a significant positive relationship between Brand Ambassador and Brand Trust.
In contrast, the Brand Ambassador Ie Purchase Decision hypothesis has a P-value of 0.
148, greater than 05, despite a positive original sample value.
This result leads to the acceptance of H02 and the rejection of Ha2, indicating no significant influence of Brand Ambassador on Purchase Decision.
Next, the User Interface Ie Brand Trust hypothesis reports a P-value of 000, which is less than 0.
05, alongside a positive original sample value.
This confirms that Ha3 is accepted and H03 is rejected, indicating that User Interface significantly impacts Brand Trust.
Similarly, the User Interface Ie Purchase Decision hypothesis shows a P-value of 0.
000, less than 0.
05, with a positive original sample value.
This supports the acceptance of Ha4 and the rejection of H04, proving that User Interface significantly influences Purchase Decision.
The Free Shipping Program Ie Brand Trust hypothesis reveals a P-value of 001, less than 0.
05, with a positive original sample value, leading to the acceptance of Ha5 and rejection of H05, confirming a significant positive relationship between the Free Shipping Program and Brand Trust.
Moreover, the Free Shipping Program Ie Purchase Decision hypothesis has a P-value of 0.
000, less than 0.
with a positive original sample value, supporting the acceptance of Ha6 and rejection of H06, demonstrating that the Free Shipping Program significantly affects Purchase Decision.
Volume 6.
Number 1, 2025 https://ijble.
com/index.
php/journal/index For the Brand Trust Ie Purchase Decision hypothesis, the P-value is 0.
less than 0.
05, with a positive original sample value, indicating that Ha7 is accepted and H07 is rejected, showing a significant positive influence of Brand Trust on Purchase Decision.
Furthermore, the Brand Ambassador Ie Brand Trust Ie Purchase Decision hypothesis shows a P-value of 0.
025, less than 0.
05, with a positive original sample value.
This results in the acceptance of Ha8 and the rejection of H08, confirming the mediating role of Brand Trust in the relationship between Brand Ambassador and Purchase Decision.
Additionally, the User Interface Ie Brand Trust Ie Purchase Decision hypothesis reports a P-value of 0.
000, less than 0.
05, with a positive original sample value, supporting the acceptance of Ha9 and the rejection of H09, proving that Brand Trust mediates the relationship between User Interface and Purchase Decision.
Finally, the Free Shipping Program Ie Brand Trust Ie Purchase Decision hypothesis shows a P-value of 0.
005, less than 0.
05, with a positive original sample value, indicating the acceptance of Ha10 and the rejection of H010, demonstrating that Brand Trust mediates the influence of the Free Shipping Program on Purchase Decision.
CONCLUSION
The path analysis of the examined variables reveals that most have a substantial influence on purchasing decisions, except for brand ambassadors, which show no direct significant impact on purchasing decisions via Tokopedia.
This suggests that Tokopedia's brand ambassador strategy primarily enhances brand awareness without directly driving consumers to make purchases.
The role of brand ambassadors appears more focused on increasing visibility and recognition of the platform rather than directly influencing buyer behavior.
As a company within the GoTo Group.
Tokopedia already possesses substantial brand awareness.
However, to further enhance consumer purchasing decisions, it is critical to strengthen other key variables identified in this study.
These include improving the user interface to ensure a seamless shopping experience and introducing strategic initiatives such as free shipping programs.
These factors can directly contribute to increasing consumer convenience and satisfaction, which are crucial for driving purchasing decisions.
Furthermore, the analysis indicates that brand ambassadors can still play an indirect role in enhancing purchasing decisions via Brand Trust as an intervening variable.
This means that brand ambassadors influence purchasing behavior primarily among buyers who have prior purchasing experience and have already established trust in the Tokopedia brand.
Therefore, building and maintaining brand trust is essential to amplify the effectiveness of brand ambassadors in influencing purchase decisions on the platform.
Reference