Analysis of the Influence of Product Quality, Service Quality and Distribution Channels on Customer Satisfaction at PT Suminsurya Mesindoleestari

The aim of this research is to analyze the influence of product quality, service quality and distribution channels on customer satisfaction at PT Suminsurya Machinedolestari. The research method used in this research is quantitative research. The sampling technique used simple random sampling and a sample size of 50 customers was obtained. The research results obtained from this research show that from the results of the t test, the tcount results for all variables are greater than ttable and the significance is smaller than 0.05, so it can be concluded that partially product quality has a significant effect on customer satisfaction, partially service quality has a significant effect on customer satisfaction and partially distribution channels have a significant effect on customer satisfaction. From the results of the F test, it was found that simultaneously the three customer satisfaction factors, namely product quality, service quality and distribution channels, had a significant effect on customer satisfaction with a Fcount value greater than Ftable (14.824 > 2.80) and a significance of 0.000 which was below 0. 05. From the results of the coefficient of determination test, it was found that 45.8% of the customer satisfaction variable was influenced by product quality, service quality and distribution channel variables, while the remaining 54.2% was explained by the influence of other factors or variables outside of the research such as price, brand image. , and others not discussed in this study.


INTRODUCTION
Every company is faced with increasingly tight business competition due to developments in the industrial world.The increasing intensity of competition and the number of competitors requires every company to always pay attention to customer needs and desires, and strive to meet customer expectations (Kindström, 2010).Customer satisfaction is very important in building good relationships with customers (Khan et al., 2022).To meet the needs of customers, a company must know how to place goods and services at the right place, quality, service and time in order to compete in the competitive business world (Dereli, 2015).
Customer satisfaction is a very important element in creating maximum sales as a result of evaluating whether the experience felt is in line with what was expected (Berry et al., 2006).PT Suminsurya Machinedolestari is a distributor company operating in the steel and machinery industry.The products of this company are widely used in the steel industry, palm oil, lathe workshops and the like.During the Covid-19 pandemic, the steel distributor business experienced a drastic decline in sales due to activity restrictions implemented by the government, causing the company's sales turnover to decline.
The problem of declining sales due to the difficulty of growing and maintaining customer demand in the midst of the Covid-19 pandemic has resulted in a decline in customer satisfaction caused by a mismatch in expectations regarding the product, interest in repurchasing and willingness to recommend the product to others.Product incompatibility because there are still customers who are not sure about the quality of the products provided (Dehghani, 2018), the services offered and the goods delivery process which tends to be problematic causes a decrease in customer satisfaction in purchasing products.
To see the development of customer satisfaction with a product, companies can pay attention to the quality of the product produced (Chavez et al., 2016), which includes the overall performance, features, durability, reliability and aesthetics inherent in the product.By paying attention to product quality, companies can carry out a comprehensive customer evaluation of the product standards they have so that they can influence consumers' views of the product.However, without good product quality, consumers and customers will not be interested in buying back the products offered by the company because they will not get good satisfaction with using the product (Syafarudin, 2021).Product quality can be improved by adding product features that suit customer needs and creating the right quality for a product because almost every company uses the quality of a product to attract consumers when marketing its products (Solimun & Fernandes, 2018).
Apart from that, customer satisfaction is also influenced by the quality of service provided by the company.This is because service quality is related to the provision of services to customers.The quality of service provided by the company is still not good as can be seen from the complaints that still arise from the services provided regarding physical evidence such as the lack of additional steel machines and repairs causing reliability in machine operations taking a long time.
The responsiveness of employees who do not properly respond to customer complaints and wishes, the guarantees given to customers regarding the dissatisfaction they experience, and the sense of empathy or attention shown to customers who find it difficult to express their complaints affect customer satisfaction.If the service is provided quickly and responsively by the company, consumers will immediately make a decision to buy the product offered because the satisfaction they get from using the product is achieved.
The development of customer satisfaction can also be seen in the company's distribution channels (Xu et al., 2022).This is because distribution channels are one of the marketing methods carried out by companies to consumers and customers.Distribution channels with slow product delivery due to limited supplies so that steel deliveries are carried out in stages and there are often delays in the delivery of goods so that customers feel disappointed and switch to other companies' steel products (MacCarthy et al., 2016), customers who have ordered but not yet delivered, deliveries via delivery services do not arrive while invoices are combined and the lack of cars in distribution on the grounds that they are currently under repair causes a decrease in customer satisfaction.If customer satisfaction can be achieved with products that are easily available and can be reached by buyers quickly and easily (Shiau & Luo, 2012), then buyers will immediately buy the products offered without thinking again.Based on the background description above, the aim of this research is to analyze the influence of product quality, service quality and distribution channels on customer satisfaction at PT Suminsurya Machinedolestari.

THEORETICAL REVIEW The Influence of Product Quality on Customer Satisfaction
According to Chandra & Megawati (2022), product quality is a difference in the amount of quality and a difference in quantity in the materials or attributes used, where the better the product quality, the more satisfied customers will be and make repeat purchases.According to Aisyah et al. (2022), product quality is closely related to product performance, where products that have good performance will always be in demand by users and will generate interest in buying goods.The good benefits that arise from using the product will tend to increase customer satisfaction.

The Influence of Service Quality on Customer Satisfaction
According to Kusmadeni (2022), service quality shows how far the difference is between reality and customer expectations regarding the service received/obtained.Service quality really depends on the employee's ability to provide service.The better the quality of service provided, the higher the satisfaction felt by customers.
According to Priyowibowo et al. (2022), service quality is the overall characteristic of a product that influences its ability to satisfy consumer needs.Service quality is considered good if the service provided is in accordance with what consumers expect, so that the higher the quality of service provided, the higher the customer or consumer satisfaction with the service.

The Influence of Distribution Channels on Customer Satisfaction
According to Nasruddin & Nurchayati (2019), a distribution channel is a route or series of intermediaries in conveying goods from producers to consumers.Having a good distribution channel can guarantee the availability of products needed by customers easily and quickly so that customers will feel satisfied and continue to make repeat purchases.
According to Pandeirot et al. (2021), a good distribution channel will make it easier for customers to transact with the company so that customers are satisfied and continue to make purchases at the company.Choosing the right distribution channel will influence the smooth flow of products or goods to customers quickly.Meanwhile, for the Conceptual Framework that the author has implemented, this can be seen in the following conceptual framework image:

METHODOLOGY
The research was carried out at PT Suminsurya Machinedowisata located at Jl. Irian Barat 320, Kongsi VI, Percut Sei Tuan District, Medan City, North Sumatra Province.This research uses quantitative research methods.Quantitative research is a research approach that focuses on collecting and analyzing data in the form of numbers or numerical data.Quantitative research methods are used to test hypotheses, identify patterns, and make generalizations that can be applied to a broader population.One of the main goals of quantitative research is to test hypotheses or predictive statements about relationships between variables.
This method allows researchers to collect empirical data that can be used to prove or disprove the hypothesis.Quantitative methods are used to describe certain phenomena or variables and measure them objectively.This allows researchers to present information about how often or how large an event or characteristic is.Through data analysis, quantitative research helps authors to make predictions about future behavior or events based on patterns or relationships identified in the data related to Analysis of the Influence of Product Quality, Service Quality and Distribution Channels on Customer Satisfaction at PT Suminsurya Machinedolestari.Quantitative research is directed at producing empirical evidence that is reliable and scientifically accepted.The population in this research is 64 customers of PT Suminsurya Machinedolestari in 2022.Meanwhile, the sample is as follows, with a population of 64 customers and an error rate of 5%, the sample for this research is: So the sample in this study was 50 respondents, and 30 respondents were chosen randomly for testing validity and reliability.

Validity and Reliability Test Validity Test
According to Sugiyono (2017), a validity test is a test used to measure whether a questionnaire is valid or not.To determine whether an item is suitable for use or not, namely by carrying out a correlation coefficient significance test with a significance level of 0.05 which is considered valid if it correlates significantly with the correlation coefficient.If r count is greater than r table then a questionnaire is said to be valid, conversely if r count is smaller than r table then a questionnaire is said to be invalid.1.If r count > r table then the question item is said to be valid 2. If r count < r table then the question item is said to be invalid

Reliability Test
According to Sugiyono (2017), reliability testing was carried out with the aim of determining the consistency of measuring instruments in questionnaires with the Cronbach Alpha scale, where the items included in the test were valid items with a limit of 0.6 to determine their reliability.

Classical Assumptions Test
The classical assumption test is a statistical requirement that must be carried out in multiple linear regression analysis based on ordinary least squares.In OLS there is only one dependent variable, while there are more than one independent variable.According to Ghozali (2018: 159), to determine the accuracy of the model, it is necessary to test several classical assumptions, namely, normality test, multicollinearity test, heteroscedasticity test and autocorrelation test.

Normality Test
The normality test is used to test whether the regression model in this study has residuals that are normally distributed or not.An indicator of a good regression model is having normally distributed data.The way to detect whether the residuals are normally distributed or not can be done using the non-parametric Kolmogorov-Smirnov (K-S) statistical test contained in the SPSS program.Data distribution can be said to be normal if the significance value is > 0.05 (Ghozali, 2021).Detecting whether the data is normally distributed or not can also be done using a more reliable method, namely by looking at the Normal Probability Plot.A good regression model is normally distributed data, namely by detecting and seeing the distribution of data (points) on the diagonal axis of the graph.2.

Multicollinearity Test
The multicollinearity test is used to test whether a research regression model has a correlation between independent (free) variables.A good regression model is one where there is no correlation between the independent variables and is free from symptoms of multicollinearity.To find out whether or not there are symptoms of multicollinearity is by looking at the magnitude of the VIF (Variance Inflation Factor) value and also the Tolerance value.Tolerance measures the variability of a selected variable that is not explained by other independent variables.The values used to indicate the presence of symptoms of multicollinearity are the VIF value < 10.00 and the Tolerance value > 0.10 (Imam Ghozali, 2018).

Heteroscedasticity Test
The heteroscedasticity test aims to test whether in the regression model there is an inequality of variance from the residuals of one observation to another (Imam Ghozali, 2018).Testing heteroscedasticity can be done by looking at the scatterplot graph between SRESID and ZPRED, namely whether there is a certain pattern or not.The basis for decision making is as follows: 1.If there is a certain pattern, such as the points forming a certain regular pattern (wavy, widening then narrowing, then this indicates heteroscedasticity has occurred.2. If there is no clear pattern, the points points spread above and below the number 0 on the Y axis, then heteroscedasticity does not occur (Imam Ghozali, 2018).As a way to strengthen the scatterplot test, there is another way, namely by testing the park test.Namely if the independent variable has a significance level value exceeding 0.05 so it can be concluded that there are no symptoms of heteroscedasticity in the regression model of this research.

Multiple Linear Regression Analysis
In this research, the data analysis technique uses multiple linear regression, namely an analysis technique to determine the effect of independent variables on the dependent variable.According to Ghozali (2021), the t test is used to test the hypothesis of two samples in the form of interval or ratio data.If the significant value of t <0.05, then partially the independent variable has a significant influence on the dependent variable and vice versa.Simultaneous Hypothesis Testing (F Test) According to Ghozali (2021), the F test generally shows whether all the variables in the regression model have a joint influence on the dependent variable and dependent variable.Coefficient of Determination (R2) According to Ghozali (2021), the coefficient of determination (R²) is used with the aim of measuring the ability of the regression model to explain variations in the dependent variable.

Descriptive Statistical Analysis
This research was conducted at PT Suminsurya Machinedolestari with independent variables, namely product quality, service quality and distribution channels and the dependent variable, namely customer satisfaction.Descriptive statistics are used in analyzing data to describe or describe the data that has been collected so that it is easier to understand.The results of descriptive statistical analysis research can be seen in table 1.1 with a total of 50 data.
1. Product quality has a minimum value of 30 and a maximum value of 50 with an average value of 40.18 and a standard deviation of 7.523.This shows that the standard deviation value tends to be quite varied or it can be said that the data collected is good.2. Service quality has a minimum value of 30 and a maximum value of 49 with an average value of 44.08 and a standard deviation of 3.109.This shows that the standard deviation value tends to be quite varied or it can be said that the data collected is good.3. The distribution channel has a minimum value of 24 and a maximum value of 40 with an average value of 38.12 and a standard deviation of 2.980.This shows that the standard deviation value tends to be quite varied or it can be said that the data collected is good.4. Customer satisfaction has a minimum value of 15 and a maximum value of 29 with an average value of 24.18 and a standard deviation of 3.874.This shows that the standard deviation value tends to be quite varied or it can be said that the data collected is good.

Normality Test Results
The function of the normality test is to see how the regression results are by measuring whether the data is normal or not.To test whether the data is normally distributed or not, there are two ways to detect it, namely: 1. Chart analysis Test results using graphic analysis can be done using histogram graphs and P-Plot normality.
a. Histogram graph In accordance with Figure 1.2, it shows that the curve is curved and forms a bell shape, so it can be concluded that the data is normally distributed.In accordance with Figure 1.3, it can be concluded that the data is normally distributed.This is indicated by the presence of data points that spread along and approach the axis line.

Statistical Analysis
Analytical data normality testing can be done using the Kolmogorov-Smirnov test.

Source: SPSS Output Results
Based on table 1.4, it can be seen that all variables have a significance value (Sig.)> 0.05, namely product quality with a significance value of 0.079, service quality with a significance value of 0.245 and distribution channels with a significance value of 0.343 so that the data concludes that there are no symptoms.heteroscedasticity.

Results of Multiple Linear Regression Analysis
This analysis will explain the equation of multiple linear regression of the three independent variables, namely product quality, service quality and distribution channels.in purchasing the product.The good benefits that arise from using the product will tend to increase satisfaction so that consumers will only use products from the company because they are considered to have provided a guarantee of good quality, thereby creating a sense of satisfaction for their customers.

Effect of Service Quality (X2) on Customer Satisfaction (Y)
The results of the research hypothesis H2 were accepted, which means that service quality has a significant effect on customer satisfaction.The results of this research are in line with research conducted by Priyowibowo et al (2022) which states that service quality has a positive and significant effect on customer satisfaction.With good quality service in a company, it will create satisfaction for its customers.Employees who provide comfort in providing services to customers such as maintaining cleanliness and tidiness, providing immediate service if there are difficulties experienced by customers, providing options for products needed by customers, responsiveness provided by employees in providing explanations and guarantees of good service such as informing good promotions and product types will create satisfaction for customers in purchasing company products.

Influence of Distribution Channels (X3) on Customer Satisfaction (Y)
The results of the research hypothesis H3 were accepted, which means that distribution channels have a significant effect on customer satisfaction.The results of this research are in line with research conducted by Pandeirot et al (2021) which states that distribution channels have a positive and significant effect on customer satisfaction.A good distribution channel will make it easier for customers to transact with the company so that customers feel satisfied and always decide to purchase from the company.The majority of customers assess that the distribution of steel products by the company is timely and easy so that customers can easily obtain the desired products using the company's distribution channel system.Customers also assess that the number of product transportation vehicles provided by the company is sufficient so that product availability and ease of product distribution are fast so that customers feel satisfied.

Influence of Product Quality, Service Quality and Distribution Channels on Customer Satisfaction
The results of the research hypothesis H4 are accepted with a Fcount value greater than Ftable with a value of 14.824 > 2.80 and a significance of 0.000 which is smaller than 0.05, which means that product quality, service quality and distribution channels have a significant effect on customer satisfaction by 45.8%.

CONCLUSIONS AND RECOMMENDATIONS
This research aims to examine the influence of product quality, service quality and distribution channels on customer satisfaction.The research results show that: 1) The results of the product quality hypothesis test show a significance value of 0.008 <0.05 so that partially product quality has a significant effect on customer satisfaction; 2) The results of the service quality hypothesis test show a significance value of 0.000 <0.05 so that partially service quality has a significant effect on customer satisfaction; 3) The results of the distribution channel hypothesis test show a significance value of 0.046 <0.05 so that partially the distribution channel has a significant effect on customer satisfaction; 4) The results of the F test hypothesis test show a significance value of 0.000 < 0.05 so that simultaneously product quality, service quality and distribution channels have a significant effect on customer satisfaction by 45.8%.
Based on the research results, suggestions that can be given are as follows: For companies, companies should continue to improve the quality of their products so that they are of higher quality, such as increasing sales of steel products that have good materials, strong durability and thick layers as well as providing good information and service as well as increasing product distribution times and ease of obtaining products to increase customer satisfaction.

FURTHER STUDY
For future researchers, it is hoped that this research can be used as a reference in similar research and it is recommended for future researchers to observe other factors that influence customer satisfaction such as brand image, price and promotion so that the results obtained can be developed; 3) For Prima Indonesia University, it is hoped that this research can help as an example for future research.

Figure
Figure 1.1 Histogram graphic results Source: Research Results (2023) Figure 1.2 P-Plot Normality Test Results Source: Research Results (2023)

Table 1 .
Operational Research Variables The model in this research is:

Table 1 .2
Normality Test ResultsSource: Research Results (2023)Based on table 1.2, the Asymp value is obtained.A Sig of 0.2 which has a value greater than 0.05 indicates that the regression model is normally distributed.

Table 1 .5
Multiple Linear Regression Results