1471 American Journal of Economic and Management Business Vol. 4 No. 9 September 2025 The Effect of Endorser Reels Posts on Instagram on Purchase Intention with Advertising Value and Attitude as Mediators: a Study of Batavia Whisky Products Anak Agung Cindy Jayantari, Maria Apsari Sugiat Telkom University, Indonesia Email: cindyj@student.telkomuniversity.ac.id, mariasugiat@telkomuniversity.ac.id Abstract Batavia Whisky faces challenges in marketing alcoholic beverages in Indonesia due to cultural influences, government regulations, and most of the population being Muslim. To overcome these obstacles, the company implements a digital marketing strategy through social media by utilizing the role of endorsers. Endorsers are used as a bridge of communication between the product and consumers, particularly through the Instagram platform. The ads posted contain elements of infotainment, credibility, and irritation, which are believed to influence advertising value, attitude toward advertising, and purchase intention. This study aims to determine the influence between these variables using quantitative methods. The survey was conducted by distributing questionnaires to 384 respondents who were active Instagram users, resided in Bali, consumed alcohol, were familiar with the Batavia Whisky brand, and were aged 21–35 years. Data analysis was conducted using the Structural Equation Modeling (SEM) method through SmartPLS3 software, as well as Multigroup Analysis (MGA) to examine differences in influence based on the strength of the endorser's posts (strong, moderate, and weak). The results of the study indicate that infotainment, credibility, and irritation positively influence advertising value. Advertising value then has a significant effect on attitudes toward advertising and purchase intention. Furthermore, attitudes toward advertising also positively influence purchase intention. Therefore, Batavia Whisky is advised to continue using endorsers as a digital marketing strategy and to conduct periodic evaluations of posted content to ensure it remains relevant and effective. Keywords: Advertising Value, Attitude Toward, Purchase Intention, Instagram, Credibility INTRODUCTION Alcoholic beverages are fermented products that have been known for a long time. The original purpose of fermentation was to enrich diets, preserve food, detoxify food, and reduce cooking time. The people of the Nusantara region are renowned for their ability to manage fermented beverages in harmony with their culture. One of the most famous fermented beverages is tuak. However, by the end of the 13th century, the people of the Nusantara region began consuming distilled spirits (Fathoni, 2023). Although alcoholic beverages have been known for a long time, according to Statista released on October 11, 2024, in 2020 Indonesia became the country with the lowest alcohol consumption (Statista, 2024). In addition, Indonesia's ranking in the Asia-Pacific region in terms of alcohol consumption is directly proportional to the decline in average alcohol consumption in Indonesia. Based on data from BPS (Central Statistics Agency), alcohol The Effect of Endorser Reels Posts on Instagram on Purchase Intention with Advertising Value and Attitude as Mediators: a Study of Batavia Whisky Products 1472 consumption by people aged ≥15 years in the last year has gradually declined from 2018 to 2022. In urban areas, alcohol consumption decreased from 0.28 liters per capita in 2018 to 0.18 liters in 2022. Meanwhile, in rural areas, alcohol consumption also decreased from 0.72 liters per capita in 2018 to 0.53 liters in 2022. When combined, the national average consumption (urban + rural) decreased from 0.48 liters per capita in 2018 to 0.33 liters in 2022. This decrease reflects a downward trend in alcohol consumption across all regions of Indonesia over the past five years. Based on data from the Indonesian Central Statistics Agency updated on March 7, 2023, there has been a decline in alcohol consumption in Indonesia over the past five years. This decline has occurred in both urban and rural areas (BPS, 2023). The Secretary General of the Indonesian Importers and Distributors Association (APIDMI) said that there has been a general slowdown in sales of alcoholic beverages in Indonesia (Kontan, 2024). Indonesia ranks lowest in alcohol consumption, and the decline in alcohol consumption in Indonesia is greatly influenced by cultural factors, regulations, and the majority religion in Indonesia (RRI, 2025). The Indonesian government is serious about addressing the phenomenon of alcohol addiction (Hadi & Kusmawati, 2024) One of them is the Minister of Trade Regulation No. 25 of 2019 concerning the Sixth Amendment to Minister of Trade Regulation No. 20/M-DAG/PER/4/2014 concerning Control and Supervision of the Procurement, Distribution, and Sale of Alcoholic Beverages (Kemendag, 2019). The government has also increased excise tax rates for products in categories B and C based on Minister of Finance Regulation (PMK) No. 160/2023 concerning Excise Tax Rates for Ethyl Alcohol (Kontan, 2024). In terms of health, alcohol can also cause acute and chronic changes to the body due to its addictive properties and proactive substances (Hanifah, 2023). Criminal incidents caused by alcohol are also on the rise, such as bootleg alcoholic beverages. The effects of bootleg alcoholic beverages include blindness, brain damage, and even death (InfoPOM, 2014). The things that have been described not only affect the decline in financial performance (Kontan, 2024) However, this has become a limitation in the marketing of alcoholic beverages in Indonesia, making it difficult to market them. The government does not prohibit alcoholic beverages; rather, it focuses on strict control over their distribution and consumption. This presents an opportunity for alcoholic beverage producers, as their target market and marketing efforts become narrower and more specific. For example, the market segment is targeted at foreign tourists (particularly in Bali) and the local population in those areas, where the majority do not adhere to alcohol consumption restrictions. Based on Batavia Whisky's internal data, the monthly production target for 2023 and 2024 is set at 3,581 cases. However, the average production realization increased from 1,721 cases per month in 2023 to 2,096 cases per month in 2024. Similarly, the average sales volume also showed an increase, rising from 1,408 cases in 2023 to 1,706 cases in 2024. However, both production and sales have not yet reached the set targets, despite showing a positive upward trend year over year. As shown in Table 1, although the sales target has been consistent and relatively high, the average production in 2023 and 2024 fell significantly short of the production target. In 2023, the average production was only 1,721 boxes, which is approximately 48% of the target. Although there was an increase in average production in 2024 to 2,096 boxes, this figure is still only around 58% of the target. This indicates that the targets 1473 have not been met. Production is often driven by inaccurate or miscalculated market demand forecasts. In addition, sales (online and offline) that never exceeded production volumes led to stockpiling. In 2023, average sales of 1,408 boxes were lower than the average production of 1,721 boxes, creating a surplus of 313 boxes per period. This also occurred in 2024, where the average sales volume of 1,706 boxes was lower than the average production volume of 2,096 boxes, resulting in a surplus of 390 boxes per period. This product surplus has the potential to cause issues such as increased storage costs and the risk of damage to packaging materials due to excessive accumulation. From general indications and internal indications such as the failure to achieve average production targets and the failure to achieve average sales targets, it can be seen that Batavia Whisky is experiencing difficulties in sales. This poses a challenge in marketing Batavia Whisky products. Batavia Whisky is one of Indonesia's alcoholic beverage brands with potential for growth. Batavia Whisky has adopted various approaches to market its products amid the rapid advancement of technology. One such approach is the implementation of digital marketing, which focuses on leveraging technology. According to (Kotler, 2021) In digital marketing, digital marketing tools such as websites, social media, advertisements, mobile applications, online videos, emails, blogs, and other digital platforms will be utilized. This also involves consumer engagement anywhere, anytime through computers, smartphones, tablets, internet- connected TVs, or other digital devices. One example of digital marketing conducted by Batavia Whisky is the use of social media (Instagram, TikTok, Facebook, and YouTube) as its medium. Social media, according to (Kotler, 2021) is a medium that allows consumers to share text, images, sound, and video information with each other and with companies, and vice versa. In addition, (Kotler, 2021) also mentions that social media is an important component of digital marketing. Batavia Whisky actively utilizes various social media platforms as a means of promotion and communication with consumers. These include Instagram and YouTube, which feature visual content such as reels, product campaigns, and consumer testimonials. In addition, Batavia Whisky is also present on TikTok with short, entertaining videos that attract the attention of young audiences, as well as on Facebook, which is used to build interaction and reach a wider community. These four platforms demonstrate the brand's consistency in building its image and enhancing engagement in the digital realm. In marketing its products on social media, Batavia Whisky employs subtle branding by using endorsers as part of its marketing strategy. This is to attract the attention of alcohol consumers who tend to prefer things that are entertaining, healing, and appealing. It is hoped that this will facilitate sales. Additionally, based on previous research, due to the success of this type of endorsement and its high impact, advertisers often sponsor influencers or collaborate with them to promote their brands (Herrando & Hoyos, 2022). The use of endorsers is prioritized on the Instagram social media platform because endorsers have more freedom to showcase products, although there are still limitations (regarding the platform's posting rules). The Effect of Endorser Reels Posts on Instagram on Purchase Intention with Advertising Value and Attitude as Mediators: a Study of Batavia Whisky Products 1474 Figure 1. Example of Batavia Whisky Endorser Usage (Source: Batavia Whisky Instagram) Figure 1 shows an example of the use of endorsers in the marketing of Batavia Whisky on Instagram. Batavia Whisky endorsers play a role as individuals who bridge the gap in the process of advertising products to consumers through social media. Endorsers have a connection with social cognitive theory, which helps to understand the dynamics between social media users and endorsers (Volkmer & Meißner, 2025). At Batavia Whisky, endorsers must make Batavia Whisky a part of their social life or lifestyle. To make Batavia Whisky a part of their lives, endorsers must post images (feeds) and videos (reels). The selected endorsers must be alcohol consumers (specifically whisky) and be 21 years of age or older. Batavia Whisky has a brand image of “Lifestyle and Nightlife Friendly,” so the selected endorsers must align with or adapt to this brand image. Batavia Whisky has several endorsers, including Roy CDC, Coki Pardede, DJ Avril, DJ Putri December, and Laurensia Ineke. The purpose of this study is to determine the “Influence of Endorser Reels Posts on Instagram on Purchase Intention with Advertising Value and Attitude as Mediators: A Study of Batavia Whisky Products.” It is hoped that this study can delve deeper into the subject and contribute to the literature. Thus, even though alcoholic beverages have restrictions in their marketing, their marketing strategies can still be developed. However, it is important to remember that consuming alcoholic beverages should be done responsibly and in accordance with the rules set by the government. RESEARCH METHOD This study aimed to determine the influence between variables by testing existing hypotheses using a quantitative method with an explanatory research approach. Quantitative methods generally use surveys to collect data from large populations, typically by distributing questionnaires to respondents. The unit of analysis in this study was individuals who use Instagram, focusing on data derived from these individuals to address the research problem. No data intervention or manipulation was performed, indicating that the study took place in a non-contrived setting. The research adopted a cross-sectional design, where data collection and analysis were conducted within a specific time frame, and conclusions were drawn based on the collected information. 1475 Based on this approach, the study was expected to provide explanations of the relationships and influences between the independent and dependent variables outlined in the hypotheses. Table 1. Types of Research Data Type Description Research Method Quantitative Research Objective Explanatory Research Research Strategy Survey/Questionnaire Research Analysis Unit Individual Level of Researcher Involvement No Data Intervention Research Background Unforced Environment. Time of Implementation Cross-sectional Research (Source: Data Processing) RESULT AND DISCUSSION Partial Least Square - Structural Equation Modelling (PLS-SEM) Construct Validity Test – External Model Construct validity testing is conducted to reveal the relationships between constructs and their variables. To determine the relationships between constructs, convergent validity testing, discriminant validity testing, and reliability testing are conducted. Convergent validity tests are used to see how valid the measurement indicators of latent variables are or to show the extent to which a measurement is positively correlated. Convergent validity can be measured by looking at the factor loading values for each variable, which should be greater than 0.7 (Joseph F. Hair, 2022). Table 2. Outer Loading – Convergent Validity Test Variable Item Code Outer Loading Description Infotainment INF1 0,837 Valid INF2 0,874 Valid INF3 0,877 Valid Credibility CRE1 0,870 Valid CRE2 0,877 Valid CRE3 0,869 Valid Irritation IRR1 0.881 Valid IRR2 0,908 Valid Advertising Value ADV1 0,906 Valid ADV2 0,923 Valid ADV3 0,901 Valid Attitude Toward the Advertising ATT1 0,881 Valid ATT2 0,908 Valid ATT3 0,906 Valid Purchase Intention PI1 0,886 Valid PI2 0,896 Valid PI3 0,875 Valid (Source: Data Analysis) Based on Table 2, the correlation between item scores and construct scores or loading factors shows that all values for each indicator are above 0.7. This indicates that all indicators are valid. Furthermore, after examining the loading factor values, the convergent validity test was continued by measuring or examining the Average Variance Extracted (AVE) values. The The Effect of Endorser Reels Posts on Instagram on Purchase Intention with Advertising Value and Attitude as Mediators: a Study of Batavia Whisky Products 1476 AVE values for each variable are greater than 0.5. Therefore, when the value is greater than 0.5, the indicators are deemed usable together and can represent the variable (Joseph F. Hair, 2022). Table 3. AVE – Convergent Validity Test Variable AVE Description Infotainment 0,744 Valid Credibility 0,761 Valid Irritation 0,800 Valid Advertising Value 0,828 Valid Attitude Toward the Advertising 0,807 Valid Purchase Intention 0,785 Valid (Source: Data Processing) Based on Table 3, the AVE value for each variable is greater than 0.5. This indicates that all variables are valid. The results of the convergent validity test can also be shown in the PLS – Algorithm outer model graph. Figure 2. Outer Model PLS Algorithm (Source: Data Processing) Figure 2 shows the loading factors depicted in the PLS – Algorithm output graph. All loading factor values above 0.7 significantly indicate that each indicator can reflect the latent construct being measured. For example, items in the infotainment variable have loading factor values ranging from 0.837 to 0.877, indicating that this variable can significantly reflect its latent construct. This also applies to items in other variables. In general, items in each variable or model used in this study have good convergent validity. In addition to convergent validity testing, discriminant validity testing was also conducted to determine the extent to which a construct is truly different. Discriminant validity testing can be measured by observing several indicators such as cross loading, Fornell-Larcker Criterion, and Heterotrait-Monotrait Ratio (HTMT) (Joseph F. Hair, 2022). Table 4. Cross Loading – Discriminant Validity Test ADV ATT CRE INF IRR PI ADV1 0.906 0.724 0.705 0.644 0.582 0.589 ADV2 0.923 0.698 0.672 0.609 0.550 0.598 ADV3 0.901 0.701 0.694 0.599 0.554 0.605 ATT1 0.704 0.881 0.644 0.540 0.506 0.650 1477 ATT2 0.710 0.908 0.633 0.574 0.540 0.631 ATT3 0.681 0.906 0.644 0.568 0.521 0.659 CRE1 0.666 0.646 0.870 0.547 0.572 0.544 CRE2 0.663 0.588 0.877 0.564 0.588 0.512 CRE3 0.657 0.632 0.869 0.543 0.576 0.596 INF1 0.580 0.557 0.533 0.837 0.420 0.410 INF2 0.577 0.511 0.546 0.874 0.394 0.399 INF3 0.599 0.546 0.555 0.877 0.452 0.439 IRR1 0.517 0.460 0.581 0.401 0.881 0.442 IRR2 0.584 0.575 0.605 0.471 0.908 0.482 PI1 0.635 0.703 0.612 0.463 0.502 0.886 PI2 0.596 0.626 0.532 0.419 0.445 0.896 PI3 0.499 0.571 0.524 0.394 0.420 0.875 (Source: Data Processing) Based on Table 4, where the external load of the indicator on the related construct is greater than its cross-load on other constructs, or in other words, the correlation value of the construct with its indicator is greater than the correlation value of the construct with other constructs. For example, ADV1, ADV2, and ADV3 have values of 0.906, 0.923, and 0.901, respectively, when compared to ADV. These values are larger than their cross-loadings on other constructs, such as ATT1, ATT2, and ATT3 when compared to ADV, and so on. This also applies to other related constructs. Thus, the items in each variable have differences and can be considered valid. The next step involves conducting a discriminant analysis by examining the Fornell-Larcker Criterion. This involves comparing the square root of the AVE value with the correlation of the latent variables. The square root of the AVE for each construct must be greater than its highest correlation with other constructs (Joseph F. Hair, 2022). Table 5. Fornell-Larcker Criterion – Discriminant Validity Test ADV ATT CRE INF IRR PI ADV 0.910 ATT 0.778 0.898 CRE 0.759 0.713 0.872 INF 0.679 0.624 0.632 0.863 IRR 0.618 0.582 0.663 0.489 0.895 PI 0.656 0.720 0.631 0.483 0.517 0.886 (Source: Data Processing) Based on Table 5, it can be seen that the results show that the AVE square value of each construct exceeds the AVE square value of other constructs. For example, the AVE square root of ADV is 0.910. Next, consider the correlation between ADV and other constructs: ATT 0.778, CRE 0.759, INF 0.679, IRR 0.618, and PI 0.656, where the values are smaller. This indicates an empirical difference and signifies good discriminant validity. The final step in this discriminant validity test is to examine the Heterotrait-Monotrait Ratio (HTMT) value. The HTMT threshold is 0.9. If the value exceeds 0.9, it indicates a lack of discriminant validity (Joseph F. Hair, 2022). Table 6. Heterotrait-Monotrait Ratio (HTMT) ADV ATT CRE INF IRR PI ADV ATT 0.875 CRE 0.873 0.827 INF 0.788 0.731 0.756 The Effect of Endorser Reels Posts on Instagram on Purchase Intention with Advertising Value and Attitude as Mediators: a Study of Batavia Whisky Products 1478 IRR 0.750 0.711 0.833 0.617 PI 0.739 0.819 0.735 0.567 0.637 (Source: Data Processing) Based on Table 6, the values of each construct are below the threshold of 0.9. This indicates that each variable has good discriminant validity. The final stage after conducting convergent and discriminant validity tests is the reliability test. The reliability test is conducted by looking at Cronbach's alpha and composite reliability (ρC) values. It is considered reliable if Cronbach's alpha and composite reliability (ρC) values are greater than 0.6 (Joseph F. Hair, 2022). Table 7. Reliability Test Cronbach's alpha Composite reliability (rho_c) Description ADV 0.896 0.935 Reliable ATT 0.880 0.926 Reliable CRE 0.843 0.905 Reliable INF 0.828 0.897 Reliable IRR 0.751 0.889 Reliable PI 0.864 0.916 Reliable (Source: Data Processing) Based on Table 7, it can be seen that the results show that the Cronbach's alpha and composite reliability values for each variable are greater than 0.6. This indicates that all variables are reliable. Structural Model Test – Inner Model Structural model testing is conducted to evaluate the relationships between variables in the designed model, to test the proposed hypotheses, and to assess how well the data supports the proposed relationships between constructs. In this structural model testing, R-squared, Q- squared, path coefficient, and hypothesis testing are performed. The initial stage involves conducting an R-squared test to determine the combined effect of exogenous latent variables on endogenous latent variables. A value of 0.75 indicates a strong relationship, 0.5 indicates a moderate relationship, and 0.25 indicates a weak relationship (Joseph F. Hair, 2022). However, previous studies also mentioned R-Square values of 0.67, 0.33, and 0.19, where the structural model identified the model as “good,” “moderate,” and “weak” (Sugiat, Saabira, & Witarsyah, 2023). Table 8. R-Squared Test R-square ADV 0.657 ATT 0.605 PI 0.541 (Source: Data Processing) Based on Table 8, where the R-Squared value of the combined effect on ADV (Advertising Value) is 0.657, it can be explained that all exogenous variables simultaneously affect ADV (Advertising Value) by 65.7%, which indicates a strong relationship. Next, the R- 1479 Squared value of the combined effect on ATT (Attitude Toward the Advertising) is 0.605, indicating that all exogenous variables collectively influence ATT (Attitude Toward the Advertising) by 60.5%, which signifies a strong relationship. Finally, the R-Squared value for the combined influence on PI (Purchase Intention) is 0.541, indicating that all exogenous variables collectively influence PI (Purchase Intention) by 54.1%, which signifies a moderate relationship. Next, a Q-square test is conducted to assess the predictive power outside their sample model, also known as predictive relevance. A value greater than 0 is considered good (Joseph F. Hair, 2022). Table 9. Q-Squared Test SSO SSE Q² (=1-SSE/SSO) ADV 1152.000 533.649 0.537 ATT 1152.000 593.668 0.485 CRE 1152.000 1152.000 INF 1152.000 1152.000 IRR 768.000 768.000 PI 1152.000 677.307 0.412 (Source: Data Processing) Based on Table 9, the predictions for ADV (Advertising Value), ATT (Attitude Toward Advertising), and PI (Purchase Intention) by all exogenous latent variables are relevant and accurate because their Q-squared values are > 0. Specifically, 0.537 > 0, 0.485 > 0, and 0.412 > 0. After the Q-squared test, the path coefficient test and hypothesis test were conducted. In path coefficient analysis, assuming a significance level of 5%, the p-value must be less than 0.05 to conclude that the relationship under consideration is significant. For hypothesis testing, the hypothesis is accepted when the t-statistic is greater than 1.645 or when the p-value is less than 0.05, and vice versa (Joseph F. Hair, 2022). Figure 3. Bootstrapping T-Value (Source: Data Processing) Figure 3 shows a diagram of the calculated T value model from the loading factor and the calculated t value from the path coefficient direct effects. The Effect of Endorser Reels Posts on Instagram on Purchase Intention with Advertising Value and Attitude as Mediators: a Study of Batavia Whisky Products 1480 Figure 4. P Value Path Coefficient (Source: Data Processing) Figure 4 shows the P Value Path Coefficient model diagram. Figures 3 and 4 are direct effect model diagrams. The two figures above can be explained in detail as follows. Table 10. Path Coefficient – Direct Effect Hypothesis Variable Original sample (O) Sample mean (M) Standard deviation (STDEV) T statistics (|O/STDEV|) P values Descripti on H1 INF -> ADV 0.312 0.310 0.056 5.611 0.000 Accepted H2 CRE -> ADV 0.453 0.452 0.062 7.246 0.000 Accepted H3 IRR -> ADV 0.164 0.168 0.052 3.188 0.001 Rejected H4 ADV -> ATT 0.778 0.778 0.043 18.046 0.000 Accepted H5 ADV -> PI 0.244 0.254 0.085 2.864 0.004 Accepted H6 ATT -> PI 0.530 0.522 0.085 6.217 0.000 Accepted (Source: Data Processing) Table 10 shows the results of the direct effect path coefficient to see the magnitude of the direct influence of each variable. For example, the magnitude of the parameter coefficient INF (Infotainment) on ADV (Advertising Value) is 0.312, which indicates the influence of INF on ADV. The higher the INF (Infotainment), the higher the ADV (Advertising Value). A path coefficient or estimated coefficient of 0.312 was also obtained, with a calculated t-value of 5.611 > 1.645, resulting in a p-value of 0.000 < 0.05, thus accepting H1. The same applies to CRE on ADV, ADV on ATT, ADV on PI, and ATT on PI. Only the hypothesis for IRR on ADV was rejected. The conclusions of the hypotheses are as follows: a. The effect of INF (Infotainment) on ADV (Advertising Value). H1: Infotainment has a positive effect on advertising value with a path coefficient of 0.312, with a t-value of 5.611 > 1.645, so the p-value is 0.000 < 0.05, therefore H1 is accepted. This means there is a significant effect of INF (Infotainment) on ADV (Advertising Value). H0: Infotainment does not affect advertising value. b. The Effect of CRE (Credibility) on ADV (Advertising Value). H2: Credibility has a positive effect on advertising value with a path coefficient of 0.453. With a t-value of 7.246 > 1.645, the p-value is 0.000 < 0.05, so H2 is accepted. This means that there is a significant effect of CRE (Credibility) on ADV (Advertising Value). H0: Credibility does not affect advertising value. c. The effect of IRR (Irritation) on ADV (Advertising Value). H3: Irritation has a negative effect on advertising value. 1481 H0: Irritation does not have a negative effect on advertising value with a path coefficient of 0.164, with a t-value of 3.188 > 1.645, so the p-value is 0.001 < 0.05, so H3 is rejected and H0 is accepted. This means that there is a significant positive effect of IRR (Irritation) on ADV (Advertising Value). d. The effect of ADV (Advertising Value) on ATT (Attitude Toward the Advertising) H4: Advertising Value has a positive effect on attitude toward the ad, with a path coefficient of 0.778. The calculated t-value is 18.046 > 1.645, so the p-value is 0.000 < 0.05, thus H4 is accepted. This means that there is a significant influence of ADV (Advertising Value) on ATT (Attitude Toward the Advertising). H0: Advertising Value does not influence attitude toward the ad. e. The influence of ADV (Advertising Value) on PI (Purchase Intention) H5: Advertising Value has a positive influence on purchase intention, with a path coefficient of 0.244 and a t-value of 2.864 > 1.645, resulting in a p-value of 0.004 < 0.05, thus accepting H5. This means that there is a significant influence of ADV (Advertising Value) on PI (Purchase Intention). H0: Advertising Value does not influence purchase intention. f. The influence of ATT (Attitude Toward the Advertising) on PI (Purchase Intention) H6: Attitude toward the ad has a positive effect on purchase intention, with a path coefficient of 0.530. The t-value is 2.864 > 1.645, so the p-value is 0.000 < 0.05, thus H5 is accepted. This means that there is a significant influence of ATT (Attitude Toward Advertising) on PI (Purchase Intention). H0: Attitude toward advertising does not influence purchase intention. In addition to the direct effect, an indirect effect test was also conducted. The purpose was to determine the effect of exogenous variables on endogenous variables through mediating variables. The mediating variables were ADV (Advertising Value) and ATT (Attitude Toward Advertising). The results are as follows: Table 11. Path Coefficient – Indirect Effect Original sample (O) Sample mean (M) Standard deviation (STDEV) T statistics (|O/STDEV|) P values ADV -> ATT -> PI 0.412 0.404 0.058 7.146 0.000 CRE -> ADV -> ATT -> PI 0.187 0.182 0.033 5.579 0.000 CRE -> ADV -> ATT 0.352 0.348 0.055 6.458 0.000 CRE -> ADV -> PI 0.110 0.116 0.045 2.459 0.014 INF -> ADV -> ATT -> PI 0.129 0.126 0.032 4.039 0.000 INF -> ADV -> ATT 0.243 0.241 0.046 5.292 0.000 INF -> ADV -> PI 0.076 0.078 0.028 2.749 0.006 IRR -> ADV -> ATT -> PI 0.068 0.068 0.023 2.902 0.004 IRR -> ADV -> ATT 0.128 0.132 0.040 3.183 0.002 IRR -> ADV -> PI 0.040 0.042 0.019 2.068 0.039 (Source: Data Processing) The Effect of Endorser Reels Posts on Instagram on Purchase Intention with Advertising Value and Attitude as Mediators: a Study of Batavia Whisky Products 1482 Based on Table 11, it can be seen that the overall t-value is > 1.645 and the overall p- value is < 0.05. This indicates that there is a significant effect. This shows how the relationship between variables is indirectly influenced. For example, the indirect effect of ADV (advertising value) on PI (purchase intention), mediated by ATT (attitude toward the advertising), is positive and significant. The higher the perception of ADV (advertising value), the more positive the ATT (attitude toward the advertising) will be, and this will also influence PI (purchase intention). The same applies to other indirect relationships. Partial Least Square – Multigroup Analysis (PLS-MGA) PLS-MGA was conducted to determine whether there were significant and meaningful differences. This was done by calculating the model separately (beta coefficient) and comparing the exposure level (p-value) or the endorser's posting level (high, medium, low) from the selected endorser's reels posts. Significance is determined at a 5% probability of error level. If the p-value is less than 0.05, there is a significant difference; otherwise, there is no significant difference. Table 12. Selected MGA Reels Endorsers Variable Coefficient Beta P - Value Reels 1 - Strong Reels 2 - Medium Reels 3 - Low Reels 1- 2 Reels 1- 3 Reels 2- 3 ADV -> ATT 0.800 0.720 0.833 0.484 0.567 0.280 ADV -> PI 0.272 0.215 0.239 0.749 0.885 0.881 ATT -> PI 0.465 0.587 0.545 0.538 0.646 0.795 CRE -> ADV 0.291 0.453 0.488 0.322 0.284 0.816 INF -> ADV 0.344 0.370 0.214 0.855 0.382 0.176 IRR -> ADV 0.314 0.076 0.242 0.094 0.662 0.150 (Source: Data Processing) Based on Table 12, the beta coefficients for each level of posts all have a positive relationship. For example, in reels 1 – strong, the relationship between ADV (advertising value) and ATT (attitude toward advertising) has a value of 0.800, which means it has a stronger positive relationship than reels 2 – moderate and a weaker positive relationship than reels 3 – low. Similarly, in Reels 1 – Strong, the relationship between ADV (advertising value) and PI (purchase intention) has a value of 0.272, indicating a stronger positive relationship than in Reels 2 – Moderate and a weaker positive relationship than in Reels 3 – Low. In Reels 1-Strong, ATT (attitude toward) with PI (purchase intention) has a value of 0.465, indicating a smaller positive relationship than Reels 2-Moderate and smaller than Reels 3-Low. On reels 1-strong, CRE (credibility) with ADV (advertising value) has a value of 0.291, indicating a positive relationship that is smaller than reels 2-moderate and smaller than reels 3-low. In reel 1-strong, INF (infotainment) with ADV (advertising value) has a value of 0.344, which has a positive relationship smaller than reel 2-medium and larger than reel 3-low. And on reels 1-strong, IRR (irritation) with ADV (advertising value) has a value of 0.314, which has a positive relationship greater than reels 2-medium and greater than reels 3-low. 1483 It can also be seen that the p-values for the comparisons between reels 1 (strong) and reels 2 (moderate), reels 1 (strong) and reels 3 (low), and reels 2 (moderate) and reels 3 (low) all have p-values > 0.05, indicating no significant differences. For example, for reels 1–2 with a value of 0.484 > 0.05, this indicates that the effect of advertising value on attitude toward is not significantly different between reels 1 and reels 2. This is also the case for other effects and at other posting levels. It can therefore be concluded that there is a positive correlation between each level of endorser posts and no significant differences, indicating that the model (whether the product or endorser being highlighted) has the same or stable value across all levels of posts (strong, moderate, low). The Influence of Infotainment on Advertising Value Based on the results of the tests conducted, the infotainment variable has a positive influence on advertising value. This supports the statement H1 or, in other words, H1 is accepted. The results of this study are consistent with previous studies that concluded that infotainment has a positive influence on advertising value. These results also indicate that the Batavia Whisky endorser's Reels posts contain good and interesting information and entertainment. When users perceive influencer posts as a source of interesting information or infotainment, the advertising value they attribute to them also increases (Herrando & Hoyos, 2022). The more informative and interesting the Batavia Whisky endorser's reels posts are, the more useful, important, and valuable the advertising value will be, and vice versa. The influence of credibility on advertising value Based on the results of the tests conducted, the credibility variable has a positive influence on advertising value. This supports statement H2, or it can be said that H2 is accepted. The results of this study are directly proportional or in line with the results of previous studies which concluded that credibility has a positive influence on advertising value (Mustafi & Hosain, 2020) said in his research that credibility has a positive and significant effect on consumers' intention to purchase smartphones in Bangladesh. This shows that consumers feel confident, trust, and rely on Batavia Whisky endorsers' reels posts or reels posts that can give a convincing, trustworthy, and reliable impression. To maintain this, companies must carefully select suitable endorsers (TrianasariA, FitrianiB, & Rachmawati, 2023). This positive effect also shows that effective online advertising can significantly increase the likelihood of consumers making a purchase (Mustafi & Hosain, 2020). This also shows that the credibility contained in Batavia Whisky endorser reels posts can increase consumer interest in purchasing Batavia Whisky products. The more convincing, trustworthy, and reliable Batavia Whisky endorser reels posts are the more useful, important, and valuable their advertising value will be. The more convincing, trustworthy, and reliable the Batavia Whisky endorser's Reels posts are, the more beneficial, important, and valuable their advertising value will be, and vice versa. The effect of irritation on advertising value The Effect of Endorser Reels Posts on Instagram on Purchase Intention with Advertising Value and Attitude as Mediators: a Study of Batavia Whisky Products 1484 Based on the results of the tests conducted, the irritation variable has a positive effect on advertising value. This does not support statement H3, or it can be said that H3 is rejected. The statement is that irritation has a negative effect on advertising value. Although irritation is often associated with negative things, this does not apply in this study. These results are consistent with previous studies, which indicate that irritation does not only have a negative effect but can also have a positive effect. In the study, it was stated that mobile ads are not considered to cause irritation (Goh, Tan, & Tan, 2023) This shows that Batavia Whisky endorser reels posts do not cause annoyance or disturbance to those watching the ads. Despite the product being advertised having a high level of sensitivity, this is also supported by infotainment that has been tailored to existing guidelines and the use of endorsers as part of a subtle marketing strategy. Marketing on Instagram also supports these results, as Instagram users tend to seek entertainment to pass the time, with the understanding that product endorsements are more common on Instagram compared to other social media platforms (Herrando & Hoyos, 2022). This is also supported by the fact that Batavia Whisky prefers to use Instagram as a promotional medium because of its broader posting restrictions. The less disruptive and annoying the Batavia Whisky endorser's reels posts are, the more useful, important, and valuable they will be in terms of advertising value, and vice versa. The influence of advertising value on attitude toward Based on the results of the test conducted, the advertising value variable has a positive influence on attitude toward. This supports statement H4 or, in other words, H4 is accepted. The results of this study are directly proportional or in line with the results of previous studies. This also shows that Batavia Whisky endorser reels posts are useful, important, and valuable to attitude toward advertising. The more beneficial, important, and valuable an advertisement is, the more its advertising value will drive positive outcomes toward attitude toward. Higher advertising values can lead to better consumer attitudes, which in turn increase the likelihood of purchase intent (Sharma A. , Dwivedi, Arya, & Siddiqui, 2021). (Nguyen T. P., 2023) also states that Attitude Toward Advertising has the greatest impact on purchase intent. Therefore, it is important to understand how the positive impact on attitude toward advertising can influence consumers or audiences to increase their purchase intent. The more useful, important, and valuable the Batavia Whisky endorser's reels posts are perceived to be, the better the attitudes generated will be, and vice versa. The influence of advertising value on purchase intention Based on the test results, the advertising value variable has a positive influence on purchase intention. This supports statement H5, or in other words, H5 is accepted. The results of this study are directly proportional or in line with the results of previous studies. This also indicates that the Batavia Whisky endorser's Reels posts are beneficial, important, and valuable in terms of purchase intention. The more beneficial, important, and valuable an advertisement is, the higher its advertising value will be, thereby driving positive outcomes and influencing attitudes toward the product, which in turn will impact purchase intention for Batavia Whisky. Attitudes toward advertising serve as a significant mediator between advertising value and purchase intention. This means that good advertising value can increase positive attitudes 1485 toward advertising, which in turn encourages purchase intention (Sharma A. , Dwivedi, Mariani, & Islam, 2022). This was also conveyed (Goh, Tan, & Tan, 2023) When consumers experience a positive flow and perceive high advertising value, they tend to be inclined to purchase the advertised product or service. The more beneficial, important, and valuable the Batavia Whisky endorser's Reels posts are perceived to be, the more interest, desire to try, and willingness to purchase the product will be generated, and vice versa. The effect of attitudes toward advertisements on purchase intentions Based on the results of the tests conducted, the attitude toward variable has a positive influence on purchase intention. This supports statement H6 or, in other words, H6 is accepted. The results of this study are consistent with the results of previous studies. Purchase intention is positively influenced by attitude toward advertising, product evaluation, and incentives provided (e.g., discount coupons). This confirms that incentives such as special offers can encourage consumers to be more inclined to make a purchase (Chen, Ling, & Chen, 2023). This also shows that the attitudes created by the positive effects of advertising value such as the attitude of being able to know trends, being able to know the best products and knowing the appropriate products, can cause a sense of interest, wanting to try to buy, and finally the decision to buy Batavia Whisky products. Purchase intent is influenced by a range of factors, providing businesses with crucial information to comprehend and cater to consumer preferences (Iskamto & Rahmalia, 2023). From the results of his research (Nguyen T. P., 2023) said that users' attitudes towards advertising can influence how they respond to these factors in the context of making purchasing decisions. The more positive the attitude towards the advertising, the higher the chance of purchase intention, namely interest, interest in trying to buy, to buy the product. And vice versa. CONCLUSION This study shows that endorser reels posts on Instagram have a positive influence on attitude toward the advertising and purchase intention for Batavia Whisky products. Infotainment and credibility were shown to have a positive effect on perceived advertising value, meaning that the more informative and credible a post is, the higher the perceived advertising value perceived by the audience. Meanwhile, irritation did not show a significant negative effect, because Batavia Whisky's strategy of selecting the right endorsers - such as DJs and actors who fit the product image - was able to minimize the audience's sense of annoyance. High advertising value significantly encourages a positive attitude towards advertising, which in turn increases consumer purchase intention for the product. Moreover, a positive attitude towards the ad was shown to increase consumers' intention to purchase the product. Through Multigroup Analysis, it was found that the effectiveness of endorser reels remained consistent across different levels of message strength (strong, medium, and low), with no significant differences. Therefore, companies are advised to continue utilizing endorser reels consistently by selecting relevant figures, creating informative and entertaining content, and engaging audiences interactively to build sustainable positive attitudes towards advertisements. Future research can expand the study to other platforms such as TikTok and explore other social media activity variables to produce a more thorough understanding of consumer behavior in the digital realm. The Effect of Endorser Reels Posts on Instagram on Purchase Intention with Advertising Value and Attitude as Mediators: a Study of Batavia Whisky Products 1486 REFERENCES Abbasi, A. Z., Rehman, U., Hussain, A., Ting, D. H., & Islam, J. U. (2021). The impact of advertising value of in-game pop-up ads in online gaming on gamers’ inspiration: An empirical investigation. Telematics and Informatics, 1-15. Bermúdez-González, G., Sánchez-Teba, E. M., Benítez-Márquez, M. D., & Montiel-Chamizo, A. (2021). Generation Z Young People’s Perception of Sexist Female Stereotypes about the Product Advertising in the Food Industry: Influence on Their Purchase Intention. Foods, 1-16. Borges-Tiago, M. T., Santiago, J., & Tiago, F. (2023). Mega or macro social media influencers: Who endorses brands better? Journal of Business Research. BPS. (2023, Maret 7). Konsumsi Alkohol Oleh Penduduk Umur ≥ 15 Tahun Dalam Satu Tahun Terakhir (Liter Per Kapita), 2018-2022. Retrieved from Badan Pusat Statistika: https://www.bps.go.id/id/statistics-table/2/MTQ3NSMy/konsumsi-alkohol-oleh- penduduk-umur--15-tahun-dalam-satu-tahun-terakhir.html Chaffey, D., & ElliS-ChadwiCk, F. (2016). Digital Marketing Sixth Edition . United Kingdom: Pearson Education Limited. Chen, W.-K., Ling, C.-J., & Chen, C.-W. (2023). What affects users to click social media ads and purchase intention? The roles of advertising value, emotional appeal and credibility. Asia Pacific Journal of Marketing and Logistics. Chetioui, Y., Butt, I., & Lebdaoui, H. (2021). Facebook advertising, eWOM and consumer purchase intention-Evidence from a collectivistic emerging market. Journal of Global Marketing, 1-18. Clow, K. E., & Baack, D. (2022). Integrated Advertising, Promotion, and Marketing Communications. United Kingdom: Pearson Education Limited. Cuesta-Valiño, P., Rodríguez, P. G., & Núñez-Barriopedro, E. (2020). Perception of Advertisements for Healthy Food on Social Media: Effect of Attitude on Consumers’ Response. International Journal of Environmental Research and Public Health, 1-19. Damayanti, C. P., Ninggar, D., & Sugiarto, C. (2024). The effect of influencer advertising on the intention to buy Wardah cosmetics in Indonesia. Innovative Marketing, 88-100. Dessy Kurnia Sari, S. S. (2020). An evaluation of social media advertising for Muslim millennial parents . Journal of Islamic Marketing, 1-19. Fathoni, R. S. (2023, Juli 21). Jejak Sejarah Minuman Beralkohol di Nusantara. Retrieved from Wawasan Sejarah: https://wawasansejarah.com/sejarah-minuman-beralkohol-di- nusantara/ Goh, M. L., Tan, S. H., & Tan, V. D. (2023). How Flow Experience and Advertising Value Influence Purchase Intention towards the Mobile Advertisement? International Conference on Digital Applications, Transformation & Economy (ICDATE). Hadi, M. I., & Kusmawati, A. (2024). Perubahan Perilaku Sosial Remaja Mantan Pecandu Alkohol Terhadap Interaksi Sosial. Jurnal Integrasi Ilmu Social dan Politik, 34-42. Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2019). Multivariate Data Analysis. United Kingdom: Annabel Ainscow. Halim, E., Tanadjaja, E. P., Hartono, H., & Hebrard, M. (2022). Video-Based Advertisement Value Impact on Brand Awareness and Purchase Intention in Social Media. International Seminar on Intelligent Technology and Its Applications (ISITIA). Hanifah, L. N. (2023). Kalian Literatur : Faktor yang Mempengaruhi Konsumsi Alkohol dan Dampak Alkohol terhadap Kesehatan Berdasarkan Teori Perilaku. Media Gizi Kesmas, 453-462. 1487 Herrando, C., & Hoyos, M. J.-D. (2022). Influencer endorsement posts and their effects on advertising attitudes and purchase intentions. International Journal of Consumer Studies, 2288-2299. Indrawati. (2015). Metode Penelitian Manajemen dan Bisnis. Bandung: Rafika Aditama. InfoPOM. (2014, Mei-Juni). InfoPOM - Vol 15 No. 3 Mei-Juni 2014. Retrieved from Perpustakaan POM: https://perpustakaan.pom.go.id/slims/repository/0314.pdf Iskamto, D., & Rahmalia, K. F. (2023). Customer Review, Influencer Endorsement, and Purchase Intention: The Moderating Role of Brand Image. Shirkah: Journal of Economics and Business, 234-251. Jamil, R. A., Qayyum, U., Hassan, S. R., & Khan, T. I. (2023). Impact of social media influencers on consumers’ well-being and purchase intention: a TikTok perspective. European Journal of Management and Business Economics, 366-385. Joseph F. Hair, J. (2022). A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM). Los Angeles: SAGE. Karamchandani, S., Karani, A., & Jayswal, &. M. (2024). Linkages Between Advertising Value Perception, Context Awareness Value, Brand Attitude and Purchase Intention of Hygiene Products During COVID-19: A Two Wave Study. Journal Sage, 1-14. Kemendag. (2019, Maret). Peraturan Mentri Perdagangan. Retrieved from kemendag: https://jdih.kemendag.go.id/peraturan/peraturan-menteri-perdagangan-nomor-25-tahun- 2019-tentang-perubahan-keenam-atas-permendag-nomor-20m-dagper42014-tentang- pengendalian-dan-pengawasan-terhadap-pengadaan-peredaran-dan-penjualan- minuman-beralkohol Kontan. (2024, September 25). Banyak Tantangan, Permintaan Minuman Beralkohol Cenderung melemah. Retrieved from Kontan.co.id: https://industri.kontan.co.id/news/banyak-tantangan-permintaan-minuman-beralkohol- cenderung- melemah#:~:text=Direktur%20Delta%20Djakarta%20Ronny%20Titiheruw,sejalan%20 dengan%20adanya%20libur%20Nataru. Kotler, P. (2021). Marketing Management. United Kingdom: Pearson Education Limited. Kumar, S., Nair, P., & Sridharan, A. (2023). The rise of new age social media influencers and their impact on the consumers’ reaction and purchase intention. International Journal of Business and Emerging Markets , 89-106. Li, K., & Shen, F. (2023). Irritating or Enjoyable? Exploring the Effects of Soft- Text Native Advertising and Social Media Engagement Metrics. Journal of Promotion Management. Luhgiatno et al, S. M. (2024). Metode Penelitian Manajemen. Jawa Tengah: Eureka Media Aksara. Lustono, S. M. (2024). Metode Penelitian Manajemen. Jawa Tengah: Eureka Media Aksara. Murillo-Zegarra, M., Ruiz-Mafe, C., & Sanz-Blas, S. (2020). The Effects of Mobile Advertising Alerts and Perceived Value on Continuance Intention for Branded Mobile Apps. Sustainability. Mustafi, M. A., & Hosain, M. S. (2020). The role of online advertising on purchase intention of smartphones: mediating effects of flow experience and advertising value. Journal of Contemporary Marketing Science, 2516-7480. Nguyen, H. H., Nguyen-Viet, B., Nguyen, Y. T., & Le, T. H. (2022). Understanding online purchase intention: the mediating role of attitude towards advertising. Cogent Business & Management, 1-23. Nguyen, T. P. (2023). THE INFLUENCE OF YOUTUBE ADVERTISING VALUE ON BRAND AWARENESS AND PURCHASE INTENTIONS OF VIETNAMESE CUSTOMERS. International Journal of Business and Society,. The Effect of Endorser Reels Posts on Instagram on Purchase Intention with Advertising Value and Attitude as Mediators: a Study of Batavia Whisky Products 1488 Obadă, D.-R., & Țugulea, O. (2024). What drives online impulse buying among Zoomers on TikTok in an emerging market? A new model based on flow theory . Frontiers in Communication, 1-15. Pelsmacker, P. D., Geuens, M., & Bergh, J. V. (2013). Marketing Communications A European Perspective. United Kingdom: Pearson Education. PLS, S. (2025, Juni 13). Multigroup Analysis (MGA). Retrieved from smartps.com: https://smartpls.com/documentation/algorithms-and-techniques/multigroup-analysis/ RRI. (2025, Januari 06). Indonesia Menjadi Negara dengan Konsumsi Alkohol Terendah. Retrieved from Radio Republik Indonesia: https://www.rri.co.id/lain- lain/1239249/indonesia-menjadi-negara-dengan-konsumsi-alkohol-terendah Sari, D. K., Suziana, S., & Games, D. (2020). An evaluation of social media advertising for Muslim millennial parents. Journal of Islamic Marketing, 1-19. Sharma, A., Dwivedi, R., Mariani, M. M., & Islam, T. (2022). Investigating the effect of advertising irritation on digital advertising effectiveness: A moderated mediation model. Technological Forecasting & Social Change, 0040-1625. Sharma, A., Dwivedi, Y. K., Arya, V., & Siddiqui, M. Q. (2021). Does SMS advertising still have relevance to increase consumer purchase intention? A hybrid PLS-SEM-neural network modelling approach. Computers in Human Behavior, 0747-5632. Shimp, T. A., & Andrews, J. C. (2018). Advertising, Promotion, and other aspects of Integrated Marketing Communications . United States America: South-Western. Statista, D. R. (2024, Oktober 11). Alcohol consumption per capita in the Asia-Pacific region 2020, by country. Retrieved from Statista: https://www.statista.com/statistics/1418676/apac-alcohol-consumption-per-capita-by- country/ Sugiat, M., Saabira, N., & Witarsyah, D. (2023). Omni-Channel Service Analysis of Purchase Intention. International Journal On Informatics Visualization, 2543-2549. Sugiyono. (2021). Metode Penelitian Kuantitatif dan Kualitatif. Bandung: Alfabeta. TrianasariA, N., FitrianiB, N., & Rachmawati, I. (2023). The Influence Of Social Media Marketing And Influencer Endorsement Through Brand Image And Trust, And Their Impact On The Purchase Intention Of The Ms Glow Brand Through The Tiktok Application. International Journal of Professional Business Review, 01-23. Volkmer, S. A., & Meißner, M. (2025). Delight my brain and my eyes: Credibility and aesthetic judgments of endorsers. Journal of Retailing and Consumer Services. Wardhana, A. (2024). Perilaku Konsumen di Era Digital. Jawa Tengah: Eureka Media Aksara. Copyright holders: Anak Agung Cindy Jayantari, Maria Apsari Sugiat (2025) First publication right: AJEMB – American Journal of Economic and Management Business