THE IMPACT OF MARKETING MIX ON CUNSOMER SATISFACTION AND CUSTOMER LOYALTY ON AQUA PRODUCTS Primawati Hayuningtyas

Almost every company in Indonesia carries out a variety of strategies to be competitive. In the Bottled Drinking Water Industry, the result of a research carried out by Mandiri Industry Update (2015) indicated the existence of intense competition, with more than 500 bottled water companies, of which 60% were mostly local players. This should be a concern for Danone Group with the AQUA brand that currently still dominates the market. AQUA must continue to maintain its market share by finding ways to provide satisfying products to costumers in order to encourage loyalty so that AQUA can survive in the competition. This study was chosen based on the researcher's interest in the effect of satisfaction felt by costumers on costumer loyalty to AQUA products. In this case, the Marketing Mix (product, price, place, promotion) forms the factors that lead to costumer satisfaction and loyalty (Wahab et al, 2016). This study was conducted in the campus environment of Airlangga University with as many as 80 students. The results of this study using empirical data indicated that the marketing mix variables consisting of product, price, promotion, place had a positive and significant effect on customer satisfaction, and customer satisfaction has a positive and significant effect on costumer loyalty.

stated that the elements in the marketing mix namely product, price, place, promotion (4P) are tools that can be controlled and reflect customer satisfaction. By examining the elements of the marketing mix that affect customer satisfaction, it can provide assistance for a business to determine the potential of undescribed customers and to retain existing customers.
Costumer satisfaction is the level of one's feelings after comparing (the performance or results) perceived compared to his expectations (Kotler & Keller, 2012). The level of satisfaction is a function of the difference between perceived performance and expectations. According to Kristianto (2011), customer satisfaction or dissatisfaction is the costumer's response to the evaluation of discrepancies (perceived disconfirmation) between previous expectations (or other performance norms) and the actual performance of the product felt after its use. Tjiptono (2012: 301) defines costumer satisfaction as a situation that is shown by costumers when they realize that their needs and desires are as expected and are fulfilled properly. Costumers will be satisfied if the expected quality is at least the same or exceeds the perceived quality. Therefore, if the marketing mix component, which consisted of price, product quality, promotion system, and product availability are in accordance with what is expected by costumers, it is expected that customer satisfaction can be formed which leads to loyalty and intensity of repurchase (buying intention).
According to , consumer satisfaction programs generally include a combination of seven main elements, as follows: a. Quality goods and services Companies that want to implement a customer satisfaction program must have good quality products and services, for example by matching the standards with the main competitors in the industry. b. Relationship marketing Establishing long-term relationships and having a mutually beneficial relationship between companies and consumers, creates customer loyalty to products or services. c. Loyalty promotion program Loyalty promotion program is a form of direct effort that can be applied to establish relationships between companies and consumers. d. Focus on the best customers The best consumers are those who have high frequency of purchases and smooth payments on time, and are relatively insensitive to price. e. Effective complaint handling system An effective complaint handling system requires several aspects such as apologizing to consumers for the inconvenience they experience, empathy for consumers who show disappointment, speed in handling complaints, fairness or fairness in solving complaints, as well as convenience for consumers to contact the company in order to submit comments, critics, suggestions, and questions (Tjiptono, 2011) f. Unconditional guarantee Guarantees show the quality of the product or service where the company does not hesitate to take responsibility for the product for the services offered to consumers. g. Performance payment program Customer satisfaction must be supported by quality rewards which connect the performance appraisal system and employee compensation in order to improve quality and increase customer satisfaction. Costumer loyalty in general can be interpreted as a person's loyalty to a certain product, both goods, and services. According to Kotler and Keller (2016: 153), costumer loyalty is a firm commitment to repurchase a product or service that is preferred in the future despite situational influences and future marketing efforts that have the potential to cause brand switching behavior.
According to Oliver (2010) loyalty is the commitment of costumers to stay in-depth to subscribe or re-purchase selected products or services consistently in the future, even though the influence of the situation and marketing efforts have the potential to cause behavior change. Costumer loyalty is diverse in form and cannot be equalized.
Loyalty is evidence that costumers will stay as costumers, who have the strength and positive attitude towards the company. Costumer loyalty is an important thing that must be maintained by the company for sustainability. Loyal costumers will benefit the company because loyal customers may contribute to high-frequency purchases and also share positive experiences of products or services with the surrounding environment. Loyal costumers will be reluctant to use products from other companies. Oliver (2010) divides costumer loyalty into several levels, namely as follows: a. Cognitive Loyalty: Cognitive loyalty is formed based on information received by costumers. b. Affective Loyalty: Loyalty formed due to an emotional attachment in the minds of costumers.
Affective loyalty arises based on costumers who purchase products or services due to fondness. c. Conative Loyalty: Conative loyalty is formed based on costumers' commitment to consistently repurchase a product and service in the future. d. Action Loyalty: The habits and response behavior of costumers by regularly purchasing a company's products and services.

Prior Research
Customer satisfaction: A key factor of costumer loyalty and buying intention of a PDO food product (Espejel, et al., 2008) This research was conducted to analyze the relationship between satisfaction, loyalty, and buying intentions perceived by costumers in Spain. The object of analysis used in this study was Olive oil from Bajo Aragon which was a commodity from the costumer area.
The data used in this study were obtained through field studies by conducting in-depth interviews with several costumers and then followed by distributing questionnaires to costumers and found that there were 223 valid questionnaires which followed by the Structural Equation Model (SEM) analysis to analyze the relationship between satisfaction, loyalty and intensity purchase.
The results of this study indicated that the higher the customer satisfaction, the higher the level of loyalty and intensity of purchases from PDO "Olive Oil Bajo Aragon". Therefore, it will be very important for a country to provide a feeling of security towards the quality of the region's traditional food so that costumers will be satisfied and loyal and will be willing to make purchases on an ongoing basis so as to be able to bring on local products. The relationship between marketing mix and costumer loyalty in the hijab industry: The mediating effect of customer satisfaction (Wahab, et al., 2016) This research was conducted to describe the relationship between marketing mix and costumer loyalty with customer satisfaction as a mediating variable in the hijab industry. The setting used focuses on female costumers who wear and purchase hijab products from the Shah Alam outlet. The data used in this study were obtained through 234 questionnaires that were successfully collected and can be used in data analysis to explain the hypothesis test that was created.
The results in this study indicated that there was a positive relationship in the research model which explained that the marketing mix was directly related to purchasing loyalty where the strengths and weaknesses were influenced by costumer satisfaction which was the mediating variable in this study.

Hypothesis Formulation
H1: Marketing Mix is positively related and has a significant impact on Customer Satisfaction on Aqua brand bottled water products. H2: Customer Satisfaction is positively related and has a significant impact on Customer Loyalty on Aqua brand bottled water products.

RESEARCH METHODS Research Approach
This study applied a quantitative approach that focused on testing theory. In a quantitative approach, research data are in the form of numbers, and analysis is carried out using statistics (Sugiyono, 2014: 7). According to Deni (2014: 37), quantitative research is a process of finding knowledge that uses data in the form of numbers as a means of finding information.
This study applied a survey as a method of research. According to Sugiyono (2013) survey method is research conducted using a questionnaire as a research tool carried out in large and small populations, but the data studied are data from samples taken from the populations, therefore the relative events, distribution, and relationships between sociological and psychological variable can be obtained.

Variable Identification
Variable according to Sugiyono (2012: 61) is an attribute or nature or value of people, objects, or activities that have certain variations determined by researchers to be studied and then drawn conclusions. Based on the problem formulation and model analysis, the variables in this research consisted of dependent and independent variables. The independent variable in this study was the marketing mix. While the dependent variable in this study consisted of satisfaction and loyalty.

Definition of Variable Operations
The operational definition of a research variable according to Sugiyono (2015) is an attribute or nature or value of an object or activity that has certain variations that have been determined by researchers to be studied and then drawn conclusions.
The variables used in this study consisted of the independent variable (x) and the dependent variable (y) with the following definition:

Marketing Mix (Product, Price, Promotion, Place)
The marketing mix is an independent variable in this study. The marketing mix is a set of marketing tools that companies use to continuously achieve their marketing goals in the target market (Kotler and Keller, 2012: 119). According to Kotler and Keller (2016: 47) four variables in marketing mix activities have components, namely product, price, promotion, and place. The four elements of the marketing mix are interconnected and influencing each other, therefore efforts must be made to produce a marketing policy that leads to effective service and customer satisfaction. In this research, several things were being used as measurements for the marketing mix.
Product quality (Product), is the relative size of an item or service, which indicates the level of excellence of a product to fulfill the desires of costumers. This variable was measured based on costumers' perceptions of the quality of AQUA which reflected in four dimensions, namely the impression of quality, taste, packaging, and production processes.
Price is the amount of money needed to obtain a combination of a product and the accompanying services, in this case, is the amount of money to obtain AQUA products. This variable is measured by indicators: costumer perception of price quotes, AQUA price levels with competing brands, price conformance with quality, perception of drinking water price standards.
Promotion is the company's effort to inform, influence and persuade, and remind costumers about the company's products. This variable was measured using four dimensions, namely prizes, sponsorships, impressions of advertisements in printed and electronic media.
Product availability is a condition to easily find and to purchase the product. This variable was measured based on the ease and availability of the product in several product sales locations

Customer Satisfaction
Costumer satisfaction is the feeling of pleasure or disappointment that arises after comparing the performance (results) of products thought to the performance (or results) expected (Kotler, 2014). This variable was measured in this study through the suitability between expected product quality, costumer confidence in the product consumed, interest in regularly purchasing the product, and a feeling of satisfaction with the product consumed

Costumer Loyalty
Costumer loyalty is a dependent variable in this study. Costumer loyalty is a commitment that is firmly held to repurchasing products or services that are preferred in the future regardless of situational influences and marketing efforts that have the potential to cause behavioral shifts (Kotler and Keller, 2016). This variable was measured based on the satisfaction felt by costumers of the product, as well as the desire to continue using the product even though new products emerge.

Data Collection Procedure Determination of Population and Sample
The population is a collection of individuals with quality and characteristics that have been determined (Nazir, 2011: 271). The population in this study were students majoring in Management at Airlangga University who had consumed AQUA. The sample in this study was determined by the non-probability sampling technique which was an accidental sampling technique. This method took samples by chance, easily accessible subjects by the researcher can be used as a sample. (Sugiyono, 2012: 122).
The sample size of 30 to 500 people can be regarded to be effective depending on the sampling design and the research statement. In addition, a minimum number of 30 people was taken because the selected respondents must have specific characteristics to be studied. Respondent characteristics that were relevant to the research were respondents who had made purchases and consumed the AQUA brand bottled water products.

Method of collecting data
This research required data collection which needed in the discussion of the problems of research. Data collection was carried out by means of library research and field research. In library study, researchers collecting secondary data. Secondary data was defined as data gathered and then compiled by other individuals and being used by the author as research data. Data was in the form of theories obtained from reference books, journals, and reports relating to the object of research.
In the field study, the instrument used to collect data from each respondent was a questionnaire. The questions used were closed questions, with possible answers were provided to the respondents.

Measurement Scale and Data Interpretation
The measurement scale used in this research was the Likert scale which also called the summated-rating scale. Likert scale is one of the most used techniques in costumer research and marketing. According to Sugiyono (2012: 93), the Likert scale is used to measure the attitudes, opinions, and perceptions of a person or group of people about social phenomena. Respondents are being asked to provide their response to an issue or object then respondents are required to indicate their level of agreement with each statement and various degrees of the agreement would be given a value. In this research, the scale used was in the range of one to five, with number one as the indication of strong disagreement and five for strong agreement (Sugiyono, 2012: 94).

Data analysis technique
This research applied the frequency and Likert Scale or the method of summated ratings in measuring data. The rating value of each answer or response was added to obtain the total score. The data then processed through the editing, coding, and presented in the data tabulation table. A descriptive statistical analysis was carried out to describe the variables that in turn, could identify the characteristics of the respondents as well as the attributes that influence costumer behavior. Each question in the questionnaire included in the SPSS (Statistical Package for the Social Science) software variable.

Validity Test and Reliability Test
Validity is a measure that shows the validity of an instrument. Arikunto (2010: 211) stated that the validity test is a measure that shows the levels of validity of an instrument. A valid instrument has high validity. Conversely, a less valid instrument means it has low validity.  (Sugiyono, 2013).

Multiple Linear Regression Test
Regression analysis is used to measure the effect of one independent variable and also to indicate the direction of the relationship between the independent variable and the dependent variable. Regression analysis in this study was used to see whether there was an effect of marketing mix on customer satisfaction and customer loyalty.

RESULTS AND DISCUSSION
This research used a sample of 80 students from the Management of Airlangga University Surabaya. The data collection used 100 questionnaires and as many as 80 questionnaires filled with responses being returned. It was expected that all the answers given by respondents would indicate the true picture of respondents' expectations. At the beginning of the analysis process in the results of this research, an analysis of the characteristics of respondents was grouped by gender, age, last education, the amount of drinking water consumption per day, and the length of time-consuming bottled drinking water products.
The initial step in analyzing the data of this research was to do a descriptive analysis of each indicator of the research variable using the index value. A descriptive analysis was being applied to describe the condition of respondents' answers on each research variable. Furthermore, the results of these answers were used to obtain the tendency of respondents' answers regarding the condition of each research variable.
After the descriptive analysis was carried out, it was followed by an analysis of the data obtained through data quality testing with validity and reliability testing. Classic Assumption Test was also carried out so that the resulting regression model was in Best Linear Estimate (BLUE) mode. Multiple regression analysis test was then performed. Statistical tests used in this research consisted of the F test and t-test.
This study used a sample of 80 students from the Management of Airlangga University Surabaya. The data collection was carried out using 100 questionnaires and as many as 80 questionnaires were returned. From all the answers given by respondents, it was expected that the true picture of respondents' expectations could be obtained.
At the beginning of the analysis process, the analysis of the characteristics of respondents was classified according to gender, age, last education, the amount of drinking water consumption per day, and the length of time-consuming bottled drinking water products.  Table 1 showed that as many as 48 respondents or 60% of respondents were male and the remaining 32 respondents or 40% of respondents were female. Most of the respondents that undergoing studies at the Management Department of Airlangga University were male.  Table 2 showed that as many as 31 respondents or 39% of respondents were aged less than 20 years and the remaining 49 respondents or 61% of respondents aged between 20 to 30 years. From table 2 it can be seen that most respondents in this study were aged between 20 to 30 years. Table 3 shows that as many as 80 respondents or 100% of all respondents' last education were high schools because the questionnaire was distributed to students who were undergoing a bachelor's degree at Management Major of Study program at Airlangga University.  Table 4 shows that as many as 8 respondents or 10% of respondents consumed less than 1 liter of mineral water a day; 33 respondents or 41% of respondents consumed 1 to 2 liters of mineral water a day; 31 respondents or 39% of respondents consumed 2 to 3 liters of mineral water a day and 8 respondents or 10% of respondents consumed more than 3 liters of mineral water a day.  Table 5 shows that as many as 4 respondents or 5% of respondents had been consuming mineral water for 1 to 5 years; 21 respondents or 26% of respondents had been consuming mineral water for 6 to 10 years; 36 respondents or 45% of respondents had been consuming mineral water for 11 to 15 years and 19 respondents or 24% of respondents had been consuming mineral water for more than 20 years.

Descriptive statistic
Descriptive statistics are descriptions of respondents' responses to each research variable. In the first part, the description showed the frequency value and the percentage of respondents' answers, while the second part showed the mean or average value which processed using descriptive statistics, SPSS version 22.00. The average value was then categorized to find out the respondent's responses. To determine the average value of the respondents included in certain categories, the following are the categorization rules.

Class interval formula
Notes: K = Number of classesbanyaknya kelas Xn = Biggest observation score X1 = Smallest observation score From the formula above, a class interval value of 0.8 was obtained, this number was the class intervals in each category with the following results: To illustrate the respondents' perceptions of the questions raised, the average value was used. The results of the calculations and analysis of the average value of each research variable were as follows: 1. Product Variable Product variables were measured through two indicators namely product quality (X1.1) and form of packaging (X1.2). From table 7, after carrying out the calculation of the average value it was understood that the average value of the frequency of respondents' responses to product variables was 3.86 which included in the high category. was understood that the average value of the frequency of respondents' responses to the promotion variable was 3.64 which included in the high category.

Place Variable
Place variables were measured through two indicators namely easiness to find the product (X4.1) and product availability (X4.2). From table 10, after carrying out the calculation of the average value it was understood that the average value of the frequency of respondents' response to the place variable was 4.30 which included in the very high category.

Costumer Satisfaction Variable
The Costumer Satisfaction variable was measured through two indicators namely complaints against the product (Y1) and product expectations (Y.2). From the results of table 11, after carrying out the calculation of the average value it was understood that the average value of the frequency of respondents' response to the variable customer satisfaction was 3.58 which included in the high category.

Data Quality Test
This test was conducted to determine the quality of research data obtained using research instruments (questionnaires). The validity test was used to find out whether the questions on the questionnaire were able to indicate the subjects under research, while the reliability test was used to determine the consistency of respondents' responses to the questions asked.
A validity test was applied to determine the feasibility of each variable in a list of questions. This test was carried out to measure the data obtained after the research. Validity is related to the determination of the measuring instrument and whether the data obtained through a questionnaire fulfill research objectives. In order to obtain the results, the following criteria were being applied: 1. If r arithmetic> r table then the statement declared valid (the value of r table with 80 respondents was 0.1829 with a significance of 0.05) 2. If r arithmetic <r table then the statement declared invalid The validity test of the questionnaire was carried out using SPSS software applications to obtain targeted results. Based on the results of running it showed that the corrected item total correlation value of each indicator was greater than 0.1829, therefore the indicator for each variable was valid.
The reliability test is applied to indicate the consistency of the respondents' responses to the questions contained in the questionnaire when validity has been tested. Decision making can be done with the following conditions: 1. If r alpha is positive> r table then declared reliable 2. If r alpha is positive <r table then declared unreliable The reliability test results based on data that researchers processed with SPSS software is presented in the following table:  Ekspektra : Jurnal Bisnis dan Manajemen, Volume 4, Nomor 2, Hal. 83 -103 ISSN 2549-3604 (Online), ISSN 2549 Based on table 13 it can be seen that the Cronbach alpha of all variables above was 0.4 therefore the value of the instrument was declared reliable.

Hypothesis Testing (Test F)
The F test was carried out to determine whether simultaneously the independent variable had a significant positive effect on the dependent variable. The table below shows the F test conducted for product, price, promotion, place for customer satisfaction.

Figure 3. Results of the SPSS F test regarding the simultaneous effect of marketing mix on customer satisfaction
Simultaneous testing (X1, X2, X3, X4) against Y that is based on the table obtained the calculated F value of 21,626 with a probability value (sig) of 0,000. Calculated F value (21,626)> F table, from the significance value smaller than the probability value of 0.05 or 0,000 <0.05, then H1 is accepted, meaning that together the marketing mix (product, price, promotion, place) has a significant effect on costumer satisfaction.
Then for testing the F Test hypothesis the effect of customer satisfaction on loyalty is obtained as follows:

Influence Between Variables
The influence of marketing mix on costumer satisfaction showed that the marketing mix (product, price, promotion, place) influenced costumer satisfaction on AQUA products. This showed that all the variables that existed in the marketing mix could be positively related to customer satisfaction. The combination of the whole concept of the marketing mix needs to be considered by the company because it has a significant impact on the strengths or weaknesses of customer satisfaction.
The Influence of Costumer Satisfaction on Costumer Loyalty showed that customer satisfaction affected customer loyalty. The higher level of customer satisfaction, the higher costumer loyalty in using products. This is supported by research carried out by Fandy Tjiptono (2011;169), that states costumer satisfaction has a positive impact on costumer loyalty. Lovelock and Wright (2002) state that there were several benefits to the company if it could satisfy the customer, among others, that a high level of customer satisfaction would lead to customer loyalty.

CONCLUSIONS, IMPLICATIONS, SUGGESTIONS, AND LIMITATIONS OF RESEARCH Conclusion
1. The results of research using empirical data showed that marketing mix variables consisting of products, price, promotion, location had a positive and significant influence on customer satisfaction. 2. The results of research using empirical data indicated that costumer satisfaction had a positive and significant impact on costumer loyalty.

Research Limitations
In conducting research there were various weaknesses and shortcomings, even though the author had made various efforts to have near-perfect results.
• Researchers only discussed the effect of the overall marketing mix on customer satisfaction. Objectively there are still many other factors that support costumer satisfaction such as brand equity. • The population or sample is only taken from Airlangga University management students.
• Before conducting this research, a series of trials were conducted to obtain a valid and reliable instrument so that the research instrument was feasible to measure the marketing mix of customer loyalty through customer satisfaction, however, through the questionnaire had some weaknesses such as inaccurate answers, the possibility of respondents providing incorrect responses and there were some less understood statements by the respondents.

Further Research Suggestions
For further research, there are suggestions in terms of the number of respondents and the use of other variables in the context of research development. The author in subsequent studies can use populations and other samples or more broadly in the environment or outside the environment of the author to adjust the needs. In addition, the variables used in subsequent research can be buying intention as the dependent variable and marketing mix as the independent variable. It is expected that it can be investigated more broadly whether AQUA products can fulfill the buying intention of its costumer.