THE INFLUENCE OF PROMOTION, BRAND IMAGE, AND ADVERTISING ATTRACTIVENESS THROUGH INSTAGRAM MEDIA SOCIAL ON BUYING DECISIONS OF NAMASTE RESTO INDONESIA

: This study aims to find out how much influence the variables are researched and studied by researchers, and also to prove that purchasing decisions that occur at namaste restaurants are influenced by some of the variables above, such as Advertising Attractiveness, Promotio, and Brand Image through instagram media social. The type of research used in this study is quantitative, where later there will be the use of important data that can be used as evidence or validation of data that will later be used for this thesis. This research will later use as many as 100 correspondents to be surveyed at the location of the Indonesia restaurant namaste, and the data obtained will later be processed into new data that is accurate and can be accounted for for its validity.


INTRODUCTION
. This is due to the presence of information technology Instagram Media Social On Buying Decisions Of Namaste Resto Indonesia products that are getting better and can meet the needs of the general audience (Firmawati, 2019).
Over time, the revolution in the field of information, technology and communication continues to run until it reaches great developments for the world community.This can be seen more and more people who use mobile phones with advanced features or referred to as smartphones.In Indonesia itself, the phenomenon of the mobile phone revolution has also had a considerable influence on various fields.
This influence certainly has a beneficial impact on many parties.In addition, there are applications that can be used to interact between internet users with each other, this is known as social media (Fitri & Erdiansyah, 2022).
Social media is widely used and very popular almost everyone uses social media.In addition to social media is used as a medium to interact, but now social media is also widely used by companies as marketing media.Based on Nabila et al. (2020), social media is an online media that operates with the help of web-based technology that makes changes in terms of communication that used to only be one-way and turned into two directions or can be referred to as interactive dialogue.Social media is a place, service, and tool that allows everyone to connect so that they can express and share with other individuals with the help of the internet.
Social media is a tool or means to exchange information using internet  From the data that has been presented above, it can be concluded that using Instagram social media as a promotional medium is one way that can be used to be able to promote products sold by the company.With the promotion carried out by the company, it will later be able to increase the brand and can also influence the decision process to make purchases with attractive promotions and can be done by the company.

Populasi
The unit of analysis is something related to the purpose to be studied.

Data Analysis Techniques
In Because of the large amount of data that can be processed, and not necessarily all of the data is needed in the analysis, the longest process in this analysis method is the collection of data and validation of each data that has been obtained.In addition to being a filter for data that may not be needed or significantly affect the results of the analysis, the data validation process can also function as a way to ensure that the data collection carried out is in accordance with the procedure.
After the data has been collected and validated precisely and accurately, the analysis process can proceed to the next stage, which is looking for relationships between the data that has been obtained.Because the processed data source is in the form of numbers, automatically searching for relationships from each of these data needs to be done through a variety of precise calculation formulas.
In short, quantitative data analysis is an activity that has a lot to do with numbers, statistics, and calculations.
Without relevant skills or abilities, of course, the results of the analysis carried out become less accurate.In fact, the results of the analysis can later be used to explain the relationship between one business activity and another business activity in the entire business management process.In the table above, it can be seen that the results of outer loadings displayed are > 0.70, and with that it is declared that all variables are valid in this test.

Reliability Testing (Composite
Reliability and Cronbach Alpha) The number 0.50 or above becomes the standard value of variable AVE (Haryono, 2017).The results of research data processing show that all variables have a good validity construct.

Uji Discriminant Validity
Whether or not the construct is sufficient as a function of the value of the cross loading factor.The stages to determine Discriminant validity are:

View Cross loading values
Loading is determined if the correlation coefficient is above each construct value juxtaposed with the coefficient number of indicator relationships in other construct columns, so that a conclusion can be drawn that each indicator in the block is a constructor in that column.(Haryono, 2017).

Comparing AVE root values
In this test is the result of comparing the root of the ave with other model constructs, if the number is greater than the correlation value between constructs with others, it can be concluded to have good discriminant validity.From the results of the data processing above, it is obtained that all indicators have a higher correlation coefficient number for each variable itself compared to the correlation coefficient number of indicators with other variables, so that in the end it is summarized that each indicator in the block is a form of variables or constructs in the column above >0.70 (Ghozali, 2021).

Uji Inner Model
Testing this model was taken to see the value of R Square, Q2, GoF, and test the influence between variables.

R Square value
This analysis is to determine the percentage of endogenous construct variability that may be explained by exogenous construct variability.This analysis is also to find out the goodness of structural equation models.The higher the R-square value indicates the larger the exogenous variable can explain the endogenous variable so that the better the structural equation.
Attached output value R Square value: The final conclusion of the influence is high.

Effect Size F 2
This formula is to explore whether or not endogenous latent variables are strongly influenced from exogenous latent variables.The following can be processed: (Ghozali and Latan, 2015).If the number f 2 produces 0.02 then the effect is small, the value of 0.15 is medium and the value of 0.35 then the influence of exogenous latent variables is stated to be large (Ghozali and Latan, 2015).
From the output above can be seen as follows: 1. Promotion Variable on Purchase Decision f square value of 0.011, including small effect.

Brand Image Variable on Purchase
Decision f square value of 0.101, including medium influence.

Variable Ad Attractiveness to
Purchase Decision The f square number is 0.436, which has a big impact.
Uji Signifikasi (Two Tailled) Development of communication technology in the modern era as it is today, humans can carry out various marketing activities at home or in the office without having to go directly into the field.With the existence of internet technology, humans can do many things in promoting their products such as making advertisements online, making promotions that are done online without having to spend the cost of printing services charged, and obtain various information.Technological developments have changed the way of interaction in marketing communications from face to face (conventional) to screen to face (internet marketing).This has led to an increase in internet users and social media users in Indonesia which has an impact on increasing to ten years which can change the human lifestyle which is very much different when compared to now

Figure 3 .Figure 4 .
Figure 3. Example of Promotion on Instagram Account Source : Instagram account @namasterestoindonesia According to Abdullah in the Yam& Taufik (2021) study, a hypothesis is a temporary answer that will be tested for truth through a study.In hypotheses, there are several important elements, namely temporary guesses, relationships between variables and truth tests.Based on the formulation of the problem and research objectives described above, the hypotheses in this study are: Groups & Individuals according to what the goals and problems are, the unit of analysis is in the form of people who use social media to find or find information about Namaste Resto Indonesia.According to Sugiyono (2017), the research population is a generalized area consisting of objects or subjects that have certain quantities and characteristics that researchers apply to study them and draw conclusions.In conducting research, it is necessary to determine the population so that the research carried out gets the appropriate and expected data.The population used in this study is followers from Instagram accounts @namasterestoindonesia Indonesia which have as many as 1600 followers.

Figure 5 .
Figure 5. Number of Followers @namasterestoindonesia Source: (@namasterestoindonesia, number of Instagram social media followers, 2023) by looking at the item reliability (validity indicator) which is shown by the loading factor value.Loading factor is a number that shows the correlation between the score of a question item and the score of the construct indicator that measures that construct.A loading factor value greater than 0.7 is said to be valid.However, according to Hair et al. (2019) for the initial examination of the loading factor matrix, approximately 0.3 is considered to have met the minimum level, and for loading factors approximately 0.4 is considered better, and for loading factors greater than 0.5 is generally considered significant.In this research, the loading factor limit used was 0.7.After processing the data using SmartPLS 4.0, the loading factor results can be shown as in the following table: