A Review of Customer Loyalty: An emperical study at CV Bintang

The number of different types of businesses that are starting to emerge in Indonesia shows that business development is now accelerating. As a result of the recent increase in the number of enterprises, a lot of businesses will confront strong rivalry in every aspect of their operations in today's market. As a result, every company must be adept and compete fiercely when dealing with competition. However, the predominant problem that could effects the company sustain is through there product quality, customer perception of price as well as customer satisfaction which could effect the customer loyalty. In aim to learn more about this issue, this research is intended to investigate the effects of product quality, perception of price and customer satisfaction on customer loyalty at CV Bintang Jaya Abadi, Medan. By using a quantitative approach, the sampling method is done by using the Lemeshow formula. Thus, the samples that were taken are 96.04 = 100 respondents. In analyzing all the data, SPSS 25.0 was being conducted in this research and the data was collected by distributing online questionnaires, the data measurement used the 5-point Likert scale to measure all (19) indicators. Besides that, multiple linear regression analysis was utilized in this reseacrh. Based on the results of this research, it shows that the Product quality (X1) and Customer Loyalty (Y) variables have negative and no significant effect. However, Perception of Price (X2), Customer Satisfaction (X3) and Customer Loyalty (Y) variables have positive and significant effect.


INTRODUCTION
As we know, the objective of forming a corporation is to ensure the company's survival through growth and profit maximization. The increase in revenue and profit can be used to assess the company's success in carrying out its operations. If a company's viability to develop and compete is maintained, it will operate more efficiently and to achieve these objectives.
Furthermore, by the result of the recent increase in the number of enterprises, a lot of businesses will confront strong rivalry in every aspect of their operations in today's market. As a result, every company must be adept and compete fiercely when dealing with competition. Hereinafter, there are a variety of challenges that may arise while we operate a business. One of the issues, in today's era, many entrepreneurs have attempted to enter the market in a way that competes with the old enterprise, as they may have reinvented cheaper alternatives that necessitate new innovations in order to meet market needs. However, those things could pose a significant threat to a company's survival, and if they are unable to adjust to these issues, they may have an deficient influence on the company's continuity.
One of the ways is to take advantage of market opportunities and use them to maintain client loyalty, which is the most important and core point in the business. Predominantly, client is considered as one of the company's assets that should not be lost. Customers are the most important aspect in the success of any business since they have the power to make a company prosper or fail. Thus, it's notable for a firm to hold a healthy relationship with customer. Moreover, the most important step that could be followed is to continue improve product quality and maintaing customer good perception towards the price. Customer satisfaction is demonstrated by the importance of having a good assessment of a product from a customer. A competitive selling price is beneficial to the firm in order to create a positive price perception, as it will continue to develop viable in the face of increasing competition, as well as to gain the customer's positive perception, trust, and satisfaction, which could lead to the creation of customer loyalty.
Nonetheless, the company encountered some challenges during the process. For example, many customers did not return to purchase products from CV Bintang Jaya Abadi after several order, and many old customers have forgotten their relationship with the company. Furthermore, there is a situation in which customers consistently complain about the pricing and product quality, indicating customer unhappiness, and the venture also notices that some products are in a state of degradation, indicating the company's poor performance. One of the lists of the problem that are occurring in the firm could be seen from the example above. As a result, it is critical for the company to investigate as well as take action in order to resolve these difficulties and reduce the likelihood of the same problem resurfacing in the future. Thus, based on this description, author is interested in conducting further research with the title about "The Effect of Product Quality, Perception of Price and Customer Satisfaction on Customer Loyalty at CV Bintang Jaya Abadi".

Product Quality
Product quality, according to Kotler and Keller (2016: 164) is an item's ability to deliver results or performance that meet or exceed consumer expectations. Product quality refers to a product's ability to perform its functions, such as durability, dependability, accuracy, ease of use, and repair, among other things. (Tazkiyah, 2016)

Perception of Price
Consumers' price perception, according to Firmansyah (2018), relates to how pricing information is fully absorbed and given meaningful meaning to them. Understanding pricing perceptions can be done in several ways. (Rohmat, 2019)

Customer Satisfaction
Customer satisfaction, according to Schisffman and Kanuk, is defined as a person's feelings about the performance of a product that is felt and expected. So, based on the definitions above, a person can be said to be content if their sentiments meet or even exceed their expectations. (Indrasari & Press, 2019)

Customer Loyalty
Customer loyalty is defined as a customer's commitment to a brand, company, or supplier based on a favorable nature in long-term purchases, according to Tjiptono in Putri and Santoso (2018). Customer satisfaction is determined by how much the company's performance is to produce satisfaction by limiting numerous complaints, whereas loyalty is determined by a mix of satisfaction and complaints.(A. C. M. Sari & Lestariningsih, 2021)

Relationship between variables Product Quality and Customer Loyalty
According to stefanus maximus lamere's (2017) research, the higher the quality of food firms' goods, the higher the degree of customer loyalty; conversely, the lower the level of consumer loyalty, the lower the quality of the items supplied, and vice versa. (M. Sari, 2019) H 1 : There is effect of Product Quality will increase Customer Loyalty.

Perception of Price and Customer Loyalty
Marketers must analyze pricing tactics because price plays a crucial role in determining the effectiveness of a marketer's activity. According to Stefanus maximus lamere (2017), the higher the company-determined price, the greater the degree of customer loyalty; conversely, the lower the company-determined price, the lower the amount of consumer loyalty. (M. Sari, 2019) H 2 : There is effect of Perception of Price will increase Customer Loyalty

Customer Satisfaction and Customer Loyalty
Customer satisfaction is a key aim of all business organizations, since a corporation must be able to please its consumers in order to generate a profit (Ganiyu, 2017). (Wiradarma & Suasana, 2019) H 3 : There is effect of Customer Satisfaction will increase Customer Loyalty Based on the above discussions, the research model below is as follows:

METHODOLOGY
This research is descriptive based study, with the aim of describing the types of data collected and analyzed. Moreover, this study uses quantitative methods. Quantitative research methods, according to Sugiyono (2017:8), are research methods that study populations or samples based on the positivist ideology.(D. P. Sari & Sutapa, 2020)

Population and Sample Population
According to Silaen (2018: 87), the term "population" refers to the total collection of things or people being studied for specific attributes (traits). The population is frequently referred to as the universe (universe), which encompasses all living and non-living things. The population from this company is unknown. (Nurjanah et al., 2018) Sample The sample is a portion of the population that has been taken in certain ways in order to measure or observe its features, and subsequently conclusions about these qualities that are deemed typical of the population have been formed. 2018:87 (Silaen). (Mardiyaningsih & Andhitiyara, 2020) Because the entire population is unknown or infinite, the number of samples taken in this study was calculated using the Lemeshow formula. (Nur, n.d.) Here's the Lemeshow formula: Note: = number of samples = z score at 95% confidence = 1.96 = maximum estimate = 0,5 d = alpha (0,01) or sampling error = 10% As the result of this method, the obtained is 96.04 = 100 people, implying that the author must collect data from a sample of at least 100 people for this study.

Sampling Method
The sample approach utilized in this research is non-probability sampling with purposive sampling.
Purposive sampling, according to Sugiyono (2017: 218), is a data sample technique or based on specific assumptions. (Deriyanto & Qorib, 2019). Purposive sampling is when a sample is taken for a specific reason rather than based on strata, chance, or geography. This technique is sampling method based on certain features, criteria, and qualities that are the population's key characteristics.
The characteristics or prerequisites of the purposive sampling used in this study for customers that use CV Bintang Jaya Abadi's product are as follows: a. It is permissible to sample both genders. b. This age range is chosen because the researcher believes that beyond the age of 18, people will be able to think critically. c. Only Medan, Indonesia, is the site of research.
Customers that bought the goods from CV Bintang Jaya Abadi.

Data Collection Method a. Primary Data
Data obtained directly from data collectors is referred to as primary data, according to Sugiyono (2018: 213). Respondents will answer questions in a methodical manner utilizing data gathered through questionnaires presented to them. (Imron, 2019).
The writer will use primary data to obtain data for this research by doing observations, distributing online questionnaires, and conducting personal interviews. b. Secondary Data Secondary data, according to Sugiyono (2018:213), is information that is not directly submitted to data collectors, usually in the form of document files or through other people. Researchers get additional data from a variety of sources as supporting data and extra data. (Imron, 2019) The writer uses secondary data in this research to support the primary data and supporting theories by referring to journal articles, books, and the internet.
The four factors studied in this study are product quality, price perception, customer satisfaction, and customer loyalty. There are 19 indicators in total from those four variables, with 2 questions in each variable, for a total of 38 questions. In the same way, the total number of persons who will be polled is likely to be 100.

Data Analysis Method
In analyzing all the data, SPSS 25.0 was being conducted in this research, the list of the test is as follows: According to Ghozali (2018: 51) a validity test is used to determine whether or not a questionnaire is valid. If the questions on the instrument or questionnaire are able to expose anything that will be assessed by the questionnaire, it is considered to be valid (Mufida et al., 2021).
The determination for validity test to be valid or not is by using the Pearson correlation, the test result that dictated as valid if the have to be higher than the with the significant level of 0.5% b. Reliability test Ghozali (2018:45) defines reliability as "a way for analyzing the validity of a questionnaire that serves as an indication of a variable or construct." The questionnaire is regarded reliable if a person's response to a statement is consistent or stable throughout time. (Fauzan, 2021).
The reliability test is being tested by using the Cronbach's Alpha method, and the determination is the value have to higher > 0.60 to be declared as reliable or consistent.

Classical Assumption Test a. Normality Test
The normality test determines whether the independent and dependent variables, or both, in a regression model have a normal distribution (Ghozali, 2018:161). (Fauzan, 2021) This research will be using Kolmogorov-Smirnov test and graphic analysis. In Kolmogorov-Smirnov test, the significance value (Sig.) must be > 0.05 to be considered as the research data is normally distributed, as well as for the histogram of the normality test the result must be performing a bell shape curve to be indicated that the residual data is normally distributed. Moreover, to evaluate whether the P-P Plot data have a normal distribution is, the points or data are near or close to the diagonal line. On the other hand, the residual value is not normally distributed if the points or data are far away or scattered and do not follow the diagonal line.

b. Multicollinearity Test
The multicollinearity test is performed to see if the regression model finds a link between the independent variables, according to Ghozali (2018, p. 105). (Manulang et al., 2021) The decision guidelines based on tolerance value is, if the tolerance value is > 0.10 then it means that there is no multicollinearity in the regression model. However, if the tolerance value is < 0.10 then it means that there is multicollinearity in the regression model. There are also the decision-making criteria based on the value of VIF (Variance Inflation Factor), if the VIF value is < 10.00, the regression model does not have multicollinearity. However, if the VIF value is > 10.00, the regression model has multicollinearity.

c. Linearity Test
According to Ghozali (2018:167), the linearity test is used to determine if the model's specifications are valid. Whether empirical research should utilize a linear, quadratic, or cubic function. The dependent variable and the independent variable should have a linear relationship with good data. (Fauzan, 2021) The way to determine the linearity test is, if the Sig. value of deviation from linearity is > 0.05, the independent variable and the dependent variable have a significant association. However, if it's < 0.05, the independent variable and the dependent variable do not have a meaningful relationship.

d. Heteroscedasticity Test
The heteroscedasticity test is used to assess if there is an inequality in variance between the residuals of one observation and the residuals of another observation in a regression model, according to Ghozali (2018, p. 135). (Imron, 2019) The statistical test will be done by using the Glejser test, and the heteroscedasticity symptom will be determined by looking at the scatterplot pattern.
Heteroscedasticity should not be present in a good regression model, the determination is when the points or data gather slightly above or below and the spread of data points forms a wavy pattern that widens, narrows, and widens again, as well as whether the spread of data points is patterned, this indicates that, there are no signs of heteroscedasticity in the regression model. However, the determination for Glejser test is, if the significance value (Sig.) is > 0.05, then the regression model has no signs of heteroscedasticity. On the other hand, if the significance value (Sig.) is < 0.05, the conclusion is that the regression model has a symptom of heteroscedasticity.

Multiple Linear Regression Analysis
Researchers use multiple linear regression analysis to analyze the relationship between independent variable (Product Quality, Perception of Price and Customer Satisfaction) on the dependent variable (Customer Loyalty).
According to Sugiyono (2017: 275), to forecast how the dependent variable (criteria) will rise and fall in value if two or more independent variables as predictor factors grow and fall in value (manipulated). Multiple regression analysis will be used if there are at least two independent variables. (Casmadi, 2018). Moreoever, the following is the formula:

Hypothesis test a. Determination Test
The coefficient of determination (R2), according to Ghozali (2016: 95), effectively assesses the model's capacity to explain fluctuations in the dependent variable. The capacity of the independent factors to explain the dependent variables is extremely restricted if the modified R2 value is minimal or near to zero. The more the independent variable can explain the variance in the dependent variable, the higher the modified R2 value. (Wahyuni & Suryakusuma, 2018) b. Partial T-Test The goal of the t-test, according to Ghozali (2018: 98), is to see how independent factors influence the dependent variable. A t-test, often known as a t-test, is a statistical test that compares t-count and t-table data. (Maulana et al., 2021) The T test is carried out using a significance level of 0.05 (a = 5%). Thus, if the significant value of T < 0.05, then H 0 is rejected and H a is accepted, meaning that there is a significant effect between independent variable on the dependent variable. However, if the significant value of T > 0.05 then H 0 is accepted, and H a is rejected, meaning that there is no significant effect between independent variable on the dependent variable.

c. F-Test
The goal of this hypothesis testing is to determine an estimated parameter, or how much impact the independent variables have on the dependent variable collectively (Ghozali, 2018: 98). (Maulana et al., 2021) The F test is carried out using a significance level of 0.05 (a = 5%). In addition to that if the significance value is ≥ the real level (0.05), then H 0 is accepted and H a rejected, meaning that there is no significant effect between independent variable on the dependent variable. However, if the significance value is < the real level (0.05), then H 0 is rejected and H a accepted, meaning that there is a significant effect between independent variable on the dependent variable.

RESULTS
In analyzing and calculating all the data, SPSS 25.0 was being conducted in this research, the list of the test results is as follows:       Based on the table above, it shows all the reliability test from the Independent variabel (Product Quality, Perception of Price and Customer Satisfaction) and dependent variable (Customer Loyalty) are dictated as reliable, because the results is greater than 0.6 which is the average outcome of Cronbach's Alpha method.  The normality test histogram is performing a bell shape curve, or the data slope is symmetrical curve. The residual data has a normal distribution as a result. Furthermore, the normality test data P-P plot in the data figure above is near or on the diagonal line. As a consequence, the residual value is said to be dispersed frequently. The tolerance value for variable X1 (Product Quality) is 0.277, for variable X2 (Price Perception) is 0.293 and for variable X3 is 0.340. As previously stated, the regression model does not exhibit multicollinearity if the tolerance value is greater than 0.10.

Multicollinearity Test
Variable X1 (Product Quality) has a value of 3.613, variable X2 (Price Perception) has a value of 3.412, and variable X3 has a value of 2.944, according to the VIF value. The regression model does not exhibit multicollinearity if the VIF value is less than 10.00. The significant value for deviation from linearity is 0.022, which indicates that there is no significant relationship between the independent variable X1 (Product Quality) and the dependent variable (Customer Loyalty) because the Sig. < 0.05. The significant value for deviation from linearity is 0.374, which indicates that there is significant relationship between independent variable X2 (Perception of Price) and the dependent variable Y (Customer Loyalty) because the Sig. > 0.05. The significant value for deviation from linearity is 0.163, which indicates that there is significant relationship between independent variable X3 (Customer Satisfaction) and the dependent variable Y (Customer Loyalty) because the Sig. > 0.05.

Figure 4 Scatterplot of Heteroscedasticity Test
Source: Data Processing Result (SPSS 25, 2021) The data points or data do not gather just above or below, the spread of data points does not form a wavy pattern that widens then narrows then widens again, and the spread of data points is not patterned, as can be seen in the graph above. This means that in this regression model, there are no signs of heteroscedasticity.
The statistical analysis of heteroscedasticity test by using Glejser test:  If the product quality, pricing perception, and customer satisfaction are all 0. As a consequence, CV Bintang Jaya Abadi's customer loyalty intention will be 7.249. Consumer loyalty will suffer if the firm does not handle the independent variables of product quality, pricing perception, and customer satisfaction.

Coefficient of Regression of Product Quality
The variable has a positive coefficient of 0.013, as indicated by this number. As a result, if product quality increases by 1%, consumer loyalty rises by 0.013 percent. As a consequence, improving X1 (Product Quality) on CV Bintang Jaya Abadi would enhance customer loyalty. 3. Coefficient of Regression of X2 (Perception of Price) / b2 = 0.459 This number shows that the variable have a positive coefficient of 0.459 that effects customer loyalty; for example, if price perception increase by 1%, consumer loyalty will improves by 0.459 percent, As a consequence, improving X2 (Price Perception) on CV Bintang Jaya Abadi would enhance customer loyalty. 4. Coefficient of Regression of X3 (Customer Satisfaction) / b3 = 0.433 This number denotes a variable with a positive coefficient of 0.433 that effects consumer loyalty, implying that if customer satisfaction grows by 1%, consumer loyalty will rise by 0.433 percent. As a consequence, improving X3 (Customer Satisfaction) on CV Bintang Jaya Abadi would enhance customer loyalty. As indicated in the table above, the coefficient of adjusted R square is 0.651, suggesting that there is an independent variable with a substantial effects to the dependent variable. Where the data is used to count 65.1 percent of the elements that effects the dependent variable Y (Customer Loyalty) from the independent variable X3 (Customer Satisfaction). However, other factors that are not studied in this study, have an impact on the remaining 34.9 percent. The t test of X1 (Product Quality) has a significance value of 0.817, as 0.000 for X2 (Price Perception) and 0.001. from X3 (Customer Satisfaction). As a result, because the X1 (Product Quality) significance value is greater than 0.05, the H 0 is accepted and the H a is rejected, indicating that there is no significant effect of one independent variable on the dependent variable. However, the results of significance values for X2 (Price Perception) and X3 (Customer Satisfaction) are both less than 0.05, then it indicates that H 0 is rejected and H a is accepted, implying that one independent variable has a substantial effect on the dependent variables. The significance value for F test is 0.000, as can be seen in the table above. As a result, because the significance value is less than 0.05, it indicates that H 0 is rejected and H a is accepted, indicating that there is a significant effect between one independent variable on the dependent variable.

DISCUSSION
The research found on the t test significance value of Product Quality was 0.817, indicating that the H 0 is accepted and H a is rejected, implying that there is no significant effect between independent variable and the dependent variable because the significance value is > 0.05. However, the t test significance value of X2 (Price Perception) is 0.000 and it's 0.001 for X3 (Customer Satisfaction). Because the significance value is less than 0.05, then it indicates the H 0 is rejected and H a is accepted, implying that independent variable has a significant effect on the dependent variable.
Nonetheless, multiple linear regression is used in this study, with the equation Y=7.249+0.013X1+0.459X2+0.433X3 indicating that for every 1% increase in product quality, consumer loyalty will increase by 0.013 percent and if the perception of price increases by 1%, consumer loyalty will increase by 0.459 percent. Even yet, a 1% increase in customer satisfaction will result in a 0.433 percent increase in consumer loyalty.
To summarize, all elements play an essential part in the establishment of the firm in terms of increasing firm growth, sales, and continuity. However, the dependent variable Y (Customer Loyalty) is not directly affected by the independent variable X1 (Product Quality). But the X2 (Price Perception) and X3 (Customer Satisfaction) on the other hand, have direct effects on the dependent variable Y (Customer Loyalty).

CONCLUSION
The independent variable X1 (Product Quality) indicates there is no significant effect between the dependent variable (Customer Loyalty). Moreover, the independent variable X2 (Perception of Price) indicates there is significant effect between the dependent variable (Customer Loyalty) the same goes with the independent variable X3 (Customer Satisfaction) indicates there is significant effect between the dependent variable (Customer Loyalty).

RECOMMENDATION
The author advised the company to continue looking for a method to improve and maintain product quality, as well as to look for better alternatives that deliver a high-quality product at an inexpensive price, that could entice more customers to buy or form a relationship with the company. Hereinafter, the company could also try to control their pricing policies as it really effects the customer purchasing decision and it have substantial effects towards the firm sales and performance. Furthermore, it is crucial for company to ensure everything is under control and