Analysis of Movie Genre Availability, Service Quality, and Movie Quality on Netflix User Satisfaction in Indonesia

The purpose of this study is to examine and analyze the influence of the availability of movie genre, service quality, and movie quality of Netflix user satisfaction in Indonesia. This type of research is quantitative research. The research variables are availability of movie genre, service quality, movie quality


I. INTRODUCTION
Internet technology is increasingly developing, causing people to increasingly use internet services to meet their needs. One of the needs of using the internet is the need of entertainment by utilizing video-on-demand (video streaming) services, and the trough is increasingly growing along with the Covid-19 pandemic.
The number of video streaming service companies is increasing, and according to Evandio (2020) there are several companies that are popular in Indonesia, and eight popular companies include: Netflix, Amazon Prime, Disney Plus Hotstar, VIU, Iflix, Goplay, Vidio, and Genflix. Based on the number of subscribers, Netflix is the video streaming service company with the largest subscribers in Southeast Asia, reaching 200 million subscribers from all Netflix subscribers, which reached 655 million worldwide as of August 2020 (Handoyo, 2020).
Netflix's position with the largest number of subscribers in Southeast Asia could be replaced by other video streaming service companies if the services provided are considered more attractive. Consumers have the convenience of switching providers if the services provided by Netflix are deemed unable to compete with competitors. Netflix's competitors are trying to provide added value from the services provided so that they can attract consumers' attention. Analysis by Reuters magazine on December 11, 2019, stated that Netflix was in 270 danger of losing four million subscribers if they did not add services at a lower price to compete with new competitors such as Apple TV and Disney+ (Isna, 2019). This condition shows that competition among video streaming service companies is getting tighter, requiring Netflix to make efforts to keep subscribers interested. Based on the results of a survey by Media Partners Asia (MPA) in January 2021, it turns out that the presence of Disney+ is more attractive to Indonesians in terms of the large number of customers. The number of Netflix subscribers in Indonesia as of January 2021 was 850,000, while the number of Disney+ subscribers in Indonesia was more than 2.5 million, even though Disney+ only entered Indonesia in September 2020 while Netflix entered Indonesia in 2016 (Jemadu, 2021).
Netflix customer satisfaction has decreased not only in Indonesia but also in Europe. In March 2020, Netflix took the initiative to reduce the quality of its video streaming in Europe to save internet bandwidth because many European people worked from home during the Covid-19 pandemic so streaming Netflix videos did not return to normal until May 2020 and this has caused complaints from Netflix subscribers in Europe (Yusuf, 2020). Information about the disappointment of Netflix subscribers in Europe can also be easily accessed by the Indonesian public along with the advancement of internet technology, which has led to more Indonesians choosing Disney+ than Netflix.
Efforts to increase Netflix customer satisfaction are worth doing, and to be able to improve it, it is worth identifying the variables that affect customer satisfaction on video streaming services, and according to the movie genre, service quality, and movie quality affecting Netflix user satisfaction, so that customers continue to subscribe to the Netflix video streaming service (Kang et al., 2014). Based on the research problem, the objectives of this study are to examine and analyze the effect of the availability of movie genres on the satisfaction of Netflix users in Indonesia, the effect of service quality on Netflix user satisfaction in Indonesia, and the effect of movie quality on Netflix user satisfaction in Indonesia.

A. Availability of Movie Genre
The theory regarding the availability of movie genre uses the content richness theory stated by Dimmick and Albarran (1994) in Hasan (2017), namely the diversity of interesting programs or content provided by a service provider. Program diversity refers to the number of programs or the number of movie genres that consumers can enjoy. The explanation of content richness stated by Lee and Lehto (2012) is the abundance of knowledge sources that consumers can access to enrich their learning activities. The emphasis of this understanding is "abundance", which is the large amount that consumers can choose to enjoy. The explanation of "abundance" also refers to availability, that is, more availability of movie genres means higher content richness.
The factors that influence the availability of movie genres are based on several opinions. According to Hasan (2017), the factor in the availability of the movie genre on streaming services is in accordance with management policies. Streaming services sold by providers can be grouped into several groups, for example: regular and plus services. Regular service at a regular price is provided for a certain number of movie genres, but for plus services, the number of movie genres can be added because the cost is higher. The number of movie genres for each type of service depends on the provider's management policy.

B. Service Quality
Parasuraman et al. (1985) in Naik, Gantasala, and Prabhakar (2010) define service quality as the customer's perception of service based on a comparison between expectations and the actual service perceived by the customer. Mmutle and Shonhe (2017) explain that service quality is the impression or consumers' perception of the superiority and inferiority of an organization and its services. Companies that are able to provide good service quality will increase customer satisfaction (Bernarto et al., 2019). Some factors that affect the quality of service of a provider, as explained by Mmutle and Shonhe (2017), namely the advantages and disadvantages of competitors because the assessment of quality is based on superiority compared to competitors' services. Based on the strengths and weaknesses of these competitors, a provider tries to improve the quality of service at least commensurate with competitors' services.

C. Movie Quality
The quality of movies in this study is more focused on analyzing the quality of video streaming because the quality of a movie is not the responsibility of the video streaming provider on demand but is the responsibility of 271 the director, producer, and production house (Welianto, 2020), while the provider is on-demand with transmission services movies videos to subscribers. The quality of streaming video stated by Krishnan and Sitaraman (2012) is based on the convenience of video to be enjoyed by customers, and a streaming video is considered quality if the streaming video is easy to watch, the loading process is fast, there is no rebuffering which can cause interruptions.
Austerbery (2013) in Diwi, Rumani, and Wahidah (2014) explain the factors that affect the quality of streaming movie video, including bandwidth, jitter, and loss rate. The explanation of each factor is as follows; (1) Bandwidth is the maximum size of the transfer that can be done at one time in data exchange. The sender sending data faster than the available bandwidth causes congestion on the network, resulting in packet loss which causes poor video quality. The opposite condition is that if the sender sends a data packet slower than the available bandwidth, the video quality that reaches the receiver is less than optimal. The suitability of the bandwidth size and data transmission speed determines the quality of the streaming video (including video movies). (2) Jitter is a variation of the arrival time of data packets. Jitter requires the receiver to receive a decode (display frame) at a constant rate, and any late frame will make it difficult to reconstruct the received video so that the video quality becomes poor for customers to enjoy. (3) Error rate control, the loss rate on the wireless network can be caused by bit error and burst error. This loss rate causes damage to the reconstructed video quality so that the error rate control can design the video streaming system so that the quality of the video streaming is good.

D. Customer Satisfaction
Vasic, Kilibarda, and Kaurin (2019) explain that satisfaction is a description of the emotional condition of consumers who are feeling happy or glad about company services. Feelings of happiness and pleasure are caused because consumers feel all their desires are fulfilled by the company's services. According to Levy and Weitz (2001, p. 152), satisfaction or dissatisfaction is the result of an evaluation of the consumer's experience in using a product. Evaluation of satisfaction or dissatisfaction is assessed as an activity commonly carried out by consumers after experiencing the purchase or use of a product. Kotler and Keller (2012, p. 128) explain that customer satisfaction depends on the expectations and performance of the product or service. Consumer expectations are influenced by promises or information from marketers and competitors, thus raising a certain level of expectation on consumers. The better service performance causes consumers to feel more satisfied.
Based on the above explanation in this conceptual framework, the following hypotheses and research framework is as follow: H1 = Availability of movie genre affects consumer satisfaction H2 = Service quality of the movie has an effect on customer satisfaction H3 = The quality of the movie has an effect on customer satisfaction Figure 1. Research framework (Source: Kang et al., 2014;Berman & Evans, 2006, p. 469;Jain & Madan, 2017;Dobrian et al., 2011;Vasic et al., 2019) 272

A. Research Design
According to Kuncoro (2003, p. 8), research can be classified into several groups based on its method. There are four types of research namely historical research (research activities, understanding, and explanation of circumstances), descriptive research (collecting data to answer questions about final status of research subjects), correlational research (research to get the relationship between two or more variables), and causal research (research that shows a causal relationship).

B. Population and Sampling
The population is the entire group to be studied (Walliman, 2011, p. 94). The population of this research is all Netflix video-on-demand subscribers in Indonesia. The exact population size of this study is not known because Netflix management does not publish it to the public. Therefore, according to Hair, Black, and Babin (2010, p. 711), the number of research samples can be determined as much as 5-10 times the number of research indicators. The number of indicators for this research is 17 indicators, so the number of research samples is 5 × 17 indicators = 85 samples. For this reason, the number of research samples that were determined was 85 samples. The sampling technique used is purposive sampling because there are criteria for being a sample, namely being over 17 years old and subscribing to Netflix.

C. Data Analysis
The data analysis technique of this research includes three analysis techniques: validity and reliability analysis, data normality analysis, multicollinearity analysis, and multiple linear regression analysis. The validity and reliability analysis test was conducted to measure the ability of the statements in the questionnaire to be used in research. Validity testing uses the rcount value obtained from the corrected item-total correlation value. Validity testing is done by comparing the rcount and rtable values. The rtable value is obtained from the statistical table according to the number of research samples (n). The statement is declared valid if the value rcount > rtable (Garson, 2012, p.31). The reliability test was conducted to measure the consistency of each statement in the questionnaire. The reliability test uses the Cronbach alpha statistic, and the questionnaire statement is said to be realistic if the Cronbach alpha value is above 0.60 (Garson, 2012, p.16).
A classic assumption test is a test in multiple linear regression analysis. The classical assumption is a requirement to be able to use multiple linear regression statistics. According to Widarjono (2010, p.75), the classic assumption test in multiple linear regression include: the normality test, multicollinearity test, autocorrelation test, and heterogeneity test.
Descriptive statistics to explain respondents' responses to each statement in the questionnaire. The analysis is focused on the mean value and standard deviation. The average value indicates the level of the respondent's answer for each statement, while the standard deviation indicates the degree of diversity in the respondent's answer. The average scores are grouped into three groups by categories: low, medium, and high. The basis for grouping the average values is as follows; (1) determine the range and (2) determine the scale.
Multiple linear regression analysis was carried out to examine the effect of variable movie genre availability, service quality, and movie quality on satisfaction (Widarjono, 2010, p. 15). The effect of this hypothesis is testing the effect of each independent variable on the bound variable, testing the effect of the variables: the availability of the movie genre, service quality, quality of the movie on satisfaction, testing using the ttest (Widarjono, 2010, p. 26), that is, if tcount is higher than ttable, so the tested variables have an effect on satisfaction.

A. Validity and Reliability Test of Movie Genre Availability Variables
The validity analysis in this research uses the corrected to total correlation (rcount) which is compared with the rtable value. The rtable value of the respondents (n) = 85 is 0.21 (from the statistical table), so if the rcount value is higher than the rtable value, then the indicator is valid. Based on Table 1, the overall corrected to total correlation (rcount) value of each indicator of the research variable was higher than the rtable value. These findings ensure that each indicator of the research variables is declared valid, means that the statement on each research indicator can be understood by all research respondents so that it does not cause misunderstanding. The Cronbach alpha value of the four variables tested ranged from 0.73 to 0.87 and based on the test conditions on Table 2, it was declared reliable because it was above 0.60. The results of this test indicate that each research variable has consistency in providing measurements so that it is declared reliable.

B. Testing Data
Based on Table 3, the overall corrected to total correlation (rcount) value of each indicator of the research variable was higher than the rtable value. These findings ensure that each indicator of the research variables is declared valid. The Cronbach alpha value of the four variables tested ranged from 0.73 to 0.87 and based on the test conditions it was declared reliable because it was above 0.60. The Kolmogorov-Smirnov Z value is 0.55 and Asymp. Sig. (2-tailed) of 0.93. The test provision is if the Asymp. Sig. (2-tailed) above 0.05, the data is declared to be normally distributed. The VIF value for each independent variable of the study is below 10, and based on the multicollinearity testing provisions it can be stated that the independent variables are not interrelated so that the multicollinearity test is fulfilled. Sig value. t the effect of the availability of film genres on abs_res is 0.67. The effect of service quality's sig tvalue is on abs_res is 0.06. The effect of movie quality's sig t value is on abs_res is 0.70. Based on the sig tvalue each variable for abs_res is above 0.05 so that the heteroscedasticity test is fulfilled.

C. Multiple Linear Regression Analysis
The value of the equation constant is positive, which is 1.63, so that when the three research variables are zero (0), the respondent's satisfaction is stated statistically at 1.63. The coefficient value of each research variable: availability of movie genre, service quality, and movie quality is positive. The positive coefficient value of the independent research variable can mean that when the respondent's assessment of the availability of the movie genre, service quality, and movie quality increases, it causes increased satisfaction, and when the assessment of the availability of movie genres, service quality, movie quality decreases, satisfaction also decreases.
Multiple Correlation Coefficient shows the strength and weakness of the two-way relation between the independent research variable and the dependent variable. Multiple (simultaneous) correlation is shown from the R-value of 0.618. This correlation value can be interpreted that there is a reciprocal relationship between the availability of the movie genre, the quality of the movie, and the quality of the movie and the satisfaction. The maximum value of the correlation is one (1) so the value of 0.618 can be categorized as moderate, meaning that the relationship between the three research variables and satisfaction is moderate.
Based on the answers of respondents from the questionnaires distributed, the data were inputted and processed using SPSS. The statistic chosen in the SPSS is multiple linear regression. The result of data processing according to the attachment turns out that the coefficient of determination (RSquare) is 38%. This value is in accordance with the respondent's answer, meaning that according to the respondent it turns out that the effect of X1, X2, and X3 on satisfaction is 38%.

D. Model Feasibility Testing
Testing the feasibility of the model is testing whether a research model is suitable for use or not (Ghozali, 2011, p. 112). Testing uses the Ftest by comparing the values of Fcount and Ftable. The Fvalue of the multiple linear regression results is 16.73 and the Ftable value (from the statistical table) is 2.71. The value of Fcount (16.71) is greater than Ftable (2.71) so it can be said that the model is able to explain the effect of the independent variables on the dependent variable.

E. Hypothesis Testing
The effect of movie genre availability on satisfaction explains the t-count for the availability of the film genre variable is 0.97 while the ttable value is 1.99. Based on the comparison of the two, it can be stated that the t value (0.97) is lower than the ttable value (1.99). The result of the test decision, namely the availability of the film genre, was stated to have an insignificant effect on satisfaction. The effect of Service Quality on Satisfaction can be read from the tcount value of the Netflix service quality variable is 2.84, while the ttable value is 1.99. Based on the comparison of the two, it can be stated that the tcount (2.84) is higher than the ttable value (1.99). The test result decision is that Netflix service quality is stated to have a significant effect on satisfaction. The effect of movie quality on satisfaction can be explained that the tcount value of the Netflix movie quality variable is 2.27 while the ttable value is 1.99. Based on the comparison of the two, it can be stated that the tcount (2.27) is higher than the ttable (1.99). The decision on the test results is the quality of the Netflix movie which was stated to have a significant effect on satisfaction.

V. DISCUSSION
The statistical results of the test show that the t-count value of the film genre availability variable is lower than the ttable value. The satisfaction that is taken from this test is that the availability of the film genre is declared to have no significant effect on satisfaction. The insignificant influence can be interpreted, namely, the respondent's assessment of the availability of the film genre has an effect but is relatively small (insignificant) on the level of satisfaction, so the high and low satisfaction level is not caused by the availability of film genres provided by Netflix.
The findings of this study are different from those of Kang et al. (2014) who found that the availability of movie genres has an influence on satisfaction. This difference in findings is due to the fact that the respondents in the previous study were TV streaming, while the current research was video streaming. If TV is streaming, respondents are used to changing channels containing different themes so that media richness has an effect on satisfaction. However, the current research respondents are video streaming subscribers. Usually, the tendency to choose a video or movie is based on the attractiveness of the story content in the movie so that satisfaction is not caused by the availability of the movie genre.
Other studies with different objects but related to product variations were carried out by Maichael (2018) with the finding that product variation had no effect on satisfaction. Product variation does not affect satisfaction because for consumers who are already segmented, consumers no longer think about various product choices but focus more on the products they like. In general, video streaming subscribers are interested in various serial movies because they are already dissolved in the storyline in the film. This condition causes Netflix users to no longer to think about the availability of various film genres but rather think about the continuation of the movie series that are of interest to customers. Consumers who feel they can enjoy a movie; consumers tend to emotionally follow the plot in the movie.

276
The effect of Netflix service quality on satisfaction is shown from the results of statistical testing conducted. The findings from this test indicate that service quality is proven to have a significant effect on satisfaction, meaning that the Netflix service that is considered quality by respondents determines the level of respondent's satisfaction. The findings of this study support the research results of Kang et al. (2014) and Maichael (2018) also show the same findings, where service quality has an effect on satisfaction. Quality service means that the service provides comfort and convenience for consumers to be able to enjoy the movies that are offered by the provider. Indicators to measure the quality of Netflix service include: the availability of movies in various formats, Netflix has its own original content, parental controls, availability of multiple languages for each video, Netflix is able to assure customers of the quality of service and the high concern for the desires of the subscribers shown by Netflix. These indicators of service quality will pamper customers more, so that customers can enjoy movies more comfortably.
The effect of movie quality on satisfaction was also tested statistically with the result that the movie quality was statistically proven to have a significant effect on satisfaction. The higher quality of the movies provided by Netflix will significantly affect customer satisfaction. However, if the customer assessment of the quality of the movie decreases, it will cause a decrease in customer satisfaction. The findings of this study support the research of Kang et al. (2014) stated that the quality of the movie affects satisfaction so that the higher the quality of the movie presented by the provider causes the customer to be more satisfied. The ability of providers to present movies that have quality affects consumer satisfaction because movie quality, both seen from the storyline and the quality of the media, means that consumers do not experience interference when enjoying them. Putro et al. (2014) also found that quality products affect satisfaction. A product is something that is consumed or enjoyed by a buyer, as well as a movie, which is something that is enjoyed by the viewer. The ability of a movie to meet the expectations and desires of its viewers will be able to make viewers feel happy so a movie that is getting better quality will also make viewers more satisfied because it can meet the expectations of the viewers.

VI. CONCLUSION AND RECOMMENDATION
The availability variable of movie genres does not have a significant effect on satisfaction, so the assessment of the availability of movie genres does not have an impact on Netflix customer satisfaction. Netflix service quality is proven to have an effect on customer satisfaction, so a higher rating of service quality leads to increased satisfaction and a lower rating of service quality causes a decrease in satisfaction. The quality of the movies is proven to have a significant effect on satisfaction, the more the quality of the movies provided by Netflix significantly affects customer satisfaction. However, if the customer assessment of the quality of the movie decreases, it will cause a decrease in customer satisfaction.
Based on the results of the research, the suggestions are (1) Netflix's concern for the customer needs continue to be improved. One way that can be done is by providing customer complaint services so that Netflix management understands what customers want. As far as possible, all complaints are accepted, but if the complaint is not a weakness of Netflix, then communication must be made with the customer to explain the customer's complaint problem so that the customer can understand the problem that occurred. (2) The quality of the movies provided by Netflix has an influence on customer satisfaction. Based on the average value of customer ratings of film quality, it is known that the lowest rating is on the indicator which stated that the average bitrate of the Netflix service is short. Average bitrate describes the average bit rate per second when a customer is enjoying a video, so the shorter the average bitrate, the better the movie quality. All digital data-based media is, in principle, a collection of bits (the smallest unit of all data in a computer). This collection of bits is what causes images, sound, text, and video on a PC, internet, smartphone, or other digital media. The advice put forward regarding the problem of average bitrate is Netflix management can increase the bitrate, but the consequence is that the file size becomes larger and increasing the bitrate must also pay attention to the player, capable or not. For this reason, management must look for a midpoint (adjustment) where the increase in the bitrate is still able to keep the film running on the customer's cellphone or PC. (3) Based on the availability of the number of movies in each genre, Netflix is still below Amazon Prime, so Netflix needs to increase its collection of movies in each genre. (4) According to a report by research firm Media Partners Asia in 2020, the number of Netflix subscribers is still inferior to Disney Plus Hotstar which controls the subscription video service market in Indonesia with total paid subscribers of 2.5 million while the number of Netflix subscribers according to this source is 850,000 subscribers. According to a 2020 report by research firm Media Partners Asia, Netflix is still behind the Hong Kong platform Viu, which has 1.5 million paid users. The reason why competitors are able to beat Netflix is due to the factor of price, the number of local movies that they have, and the partnership pattern with Telkomsel (Pertiwi, 2021).

277
For those reasons, the suggestion for Netflix to become the best provider for consumer choice in Indonesia must pay attention to these three factors: (1) we recommend Netflix to conduct a review of the price of the service, pricing that resonates with competitors can be a choice for consumers to prefer to subscribe to Netflix. (2) It is better if Netflix continues to increase the amount of local film content because local films are also attractive to paid video consumers in Indonesia. The selected local films should also pay attention to the level of local film ratings so that local films that are added become the attraction of consumers to subscribe to Netflix. (3) Netflix should do some partnerships with local companies such as Telkomsel so that Telkomsel customers can also easily subscribe to Netflix.