Analysis of Promotion and Product Differentiation of Jukajo on Consumer Purchase Decision

The purpose of this research was to measure the relationship and the impact of promotion and product differentiation on Jukajo consumer purchase decision. This research was a quantitative research with a questionnaire as the data collection method. The sampling technique in this research is purposive sampling technique. Then, the questionnaire was measured by Likert-scale and multiple linear regression. The result shows that promotion and product differentiation variable have positive impact on purchase decision partially and simultaneously. Both variables have 50,5% of impact towards consumer purchase decisions of Jukajo. Moreover, product differentiation becomes the most dominant variable which partially affects consumer purchase decision.


INTRODUCTION
Usaha Mikro, Kecil, dan Menengah (UMKM -Micro, Small, and Medium Enterprise) has a vital role in the economic development of Indonesia.UMKM plays a vital part in the absorption of manual labors and distribution of the development results.After the economic crisis of 1997-1998, UMKM has increased magnificently and can absorb about 85-107 million of manual labors until 2012 (Bank Indonesia, 2015).
UKKM also has a proportion of 99,99% of the total of Indonesian business.Around 56,54 million unit was in 2012, and the 0,01% or 4.968 units consisted of Badan Usaha Milik Negara (BUMN -Huge Governmental Business) or Private business (Bank Indonesia, 2015).The role of UMKM had already been explained by Joko Widodo, Indonesian President, when he was attending a meeting with the UMKM units in Istana Merdeka on 25 November 2016.He explained that UMKM was very important in Indonesian economic development (Romadoni, 2016).According to Badan Pusat Statistik (2016), during 2011-2013, the amount of UMKM in Indonesia has increased from 55.206.444 unit to 57.895.721units.
One of the well developed UMKM industries is foods and drinks.The Ministry of Industry of the Republic of Indonesia stated that in 2016, national food and drink industry had shown a positive performance of 9,82% from 6,9% in 2015, or equivalent to Rp192,6 trillion.This significant development was assumed because of a tendency in the community, especially middle and higher class in consuming hygienic foods and drinks (Putra, 2016).
An example of developing UMKM in foods and drinks in Indonesia is Jukajo.Jukajo is UMKM established since 2013 based on SPP-IRT 2133603010101-20.Sertifikat Perizinan Pangan Industri Rumah Tangga (SPP-IRT -Licensing Certificate of Household Food Industry) is a certificate of guarantee from local mayor through Public Health Department to the production of household industries.It has met certain safety requirements and standards in the framework of production and distribution of food product (BPOM RI, 2012).Jukajo is a beverage producer of a bean juice.The beginning of Jukajo is unique.It is because Jukajo is only sold in Presidential Plane of Republic Indonesia from Aero Catering Service, a part of Garuda Enterprise specialized on catering (SWA, 2017).Since introduced in the Presidential Plane of Republic Indonesia, Jukajo proclaims itself as a producer of a premium beverage.Because of the "premium" proclamation, Jukajo needs to do a product differentiation to differentiate their product from the competitors.In addition, Jukajo vision is to be a world-class company with a safe and high-quality product, also having the best service.Meanwhile, their missions are to apply a good production and good services.
Product differentiation that Jukajo has done is by showing the products that are claimed to be better than others, or as a premium brand.Thus, even though the production cost is high, consumers still look at Jukajo positively and uniquely and purchase it.
Product differentiation is an important strategy to differentiate a certain product from the rest, including attributes, design, service, location, except for price (Brue, McConnell, & Flynn, 2013).Furthermore, Allen et al. (2007) stated if companies used the differentiation strategy, companies would have focused their efforts on providing a unique product or service.Thus, it set their offerings apart from other competitors.This strategy allows organizations to charge a premium price to capture market share.The differentiation strategy is effectively implemented when the business provides unique or superior value to the consumers through product quality, features, or after-sale support and service.Based on this theory, to strengthen the Jukajo's appeal, it launches the product with a plus point that makes them different from the competitors' products.
From the interview with Raja Muhammad Fauzi, CEO of Jukajo by the researchers, for now, the product differentiation that Jukajo provides is divided into several types.Those can be (1) innovations and new variants of flavors, (2) the density of the drink, (3) hexagonal water, (4) created without coconut milk, sweetener, and preservatives.Moreover, the strategy applied affects Jukajo purchase rate directly.Even with a high price, the sales on Jukajo keeps rising each month.The graphic of Jukajo sales from June 2016 to April 2017 can be seen in Figure 1.
Jukajo product is a premium product with several product differentiations.Because of that, Jukajo cost is scaled to their targets (middle class and upper-class consumers).According to Datta (2014), upper-class people have higher income and use a lot of luxuries and comforts as their basic needs.Meanwhile, middleclass people are between upper and lower class (poor people) in many fields such as education, income, occupation, and others.Furthermore, Banerjee and Duflo (2008) with absolute approach stated that middle class was people with per capita expenditure per day about US$2-4 and also the people with per capita expenditure per day about US$6-10.
Then, to achieve optimal sales, Jukajo needs a good promotion strategy.Promotion is the last substance from marketing.A company has to do more than just creating a consumer value with a clear and persuasive promotion to communicate the consumer value.Promotion is not a single tool for marketing, but it is multiple tools for marketing (Kotler & Armstrong, 2014).In addition, Familmaleki et al. (2015) stated that promotion referred to the motivational methods of getting consumer to purchase the product.Promotion strategy is necessary because attracting new consumers has become so important in modern retailing.The companies need to communicate to the consumers about what they have to offer.Besides that, they also need to rethink the relationship between the attitude and behavior of their consumers.In this case, the promotion that has been done by Jukajo are exhibitions, word of mouth, social media, and news.Uniquely, it never uses advertisements to promote.Based on the information of the graphic of Jukajo sales from June 2016 -April 2017 in Figure 1, Jukajo sale seems to be relatively raising.However based on the interview with Anton Haryatmo, Jukajo marketing manager, in term of endorsements, Jukajo never does a direct contact with the final step for the consumers.Those activities are done by the resellers.Therefore, Jukajo knows its consumer characteristics indirectly from reseller.This phenomenon makes the final step of consumer behavior towards Jukajo to be interesting to be researched further.One of them is to know how big the impact that promotion and differentiation have done towards purchase decision of Jukajo final consumer.
The process of purchase decisions is all experienced in learning, choosing, using, and disposing a certain product by a consumer.A smart company will always try to understand how the process of purchase decision works for their consumer (Kotler & Keller, 2012).Therefore Jukajo needs to learn the process of purchase decisions by their consumer so that it can put the strategy into their business and marketing consideration.
This research has some research problems.Those are (1) are there any partial effect on promotional activities towards Jukajo consumer purchase decision?(2) Is there any partial effect on product differentiation towards Jukajo consumer purchase decision?(3) Is there any simultaneous effect of product differentiation and promotional activity towards Jukajo consumer purchase decision?(4) How are the managerial implications of relationships between promotional variable and product differentiation towards Jukajo consumer purchase decision?
Then, based on the research problems, the objectives of this research are explained.First, it is to analyze and to know the partial effect of promotional activities towards Jukajo consumer purchase decision.Second, it is to analyze and to know the partial effect of product differentiation towards Jukajo consumer purchase decision.Third, it is to analyze and to know if there is simultaneous effect of product differentiation and promotional activity Jukajo consumer purchase decision.Fourth, it is to provide appropriate managerial implications regarding the relationship between promotional variables and price differentiation towards Jukajo consumer purchase decision.
This research is expected to be helpful and has a benefit for Jukajo.First, the result can be used as a consideration and evaluation of performance towards promotional activities and product differentiation done to decide the right step to raise the sales and widen the marketplace.It can be conducted by paying attention to the factors which affect purchase decision of their products.Second, for the readers, it can help to give knowledge towards the process of purchase decision that can be used as reference of a new marketing research.

METHODS
The analytical method in this research uses quantitative analysis with multiple linear regression tests.Multiple linear regression analysis is used as a tool to find out how far the value of dependent variable changed, if the value of independent variables are manipulated or changed (Sugiyono, 2013).The mathematical formula of multiple linear regression in general is as follows.
Y is dependent variable.In this research, it is consumer purchase decision.Then, a is constants; and β... is coefficient independent variable.Next, X1 is the first independent variable.In this research, it is promotion.Meanwhile, X2 is the second independent variable which is product differentiation.Last, e is error level.
The data used in the research are primary data from questionnaires, and the secondary data are from literature study.The questionnaires are distributed through meeting the end consumers directly door to the door.First, the researchers contact the resellers that have been listed by Jukajo.After meeting with resellers, each reseller directs and helps the researchers to meet with their end consumers.
Moreover, population used is all Jukajo consumers located in the region of Tangerang (Tangerang Kota), Banten.Tangerang Kota is chosen as research location because it is based on data from Jukajo.Tangerang area (Tangerang Kota, Tangerang Selatan, and Kabupaten Tangerang) has the most number of resellers in Banten Province, which is 715 people.346 of them are in Tangerang Kota.Even though the exact number of end consumers is not recorded by Jukajo or the population of Tangerang Kota end consumers is not known for certain, it can be estimated that the end consumer can reach more than 1700 people in Tangerang Kota.It is assumed that each reseller has at least five regular consumers.
The sampling technique in this research is purposive sampling technique.It means that the sample in this research is taken with certain considerations (Sugiyono, 2013).Thus, the sample is a part of the population consisting of selected members of the whole population.Since the population number is not known for certain, this research uses the sampling determination technique for the infinite population.According to Widiyanto (2008), the formula of this sampling determination technique for infinite population can be seen in equation ( 1).This method is also used by Abdurrahman (2013) in his research.
(2) n is sample size.Then, Z is Z value at desired error level (5% in this research).Meanwhile, MoE is the maximum of error level (10% in this research).
With the intended error of 5% or the confidence level of 95% by following the consideration of using a lower confidence level (90% or 10% error), it can reduce the interval while estimating the population in total.Because of this, the interval becomes smaller, so the chance of mistake also raises.In contrary, if the researchers use 1% of error rate, the result will be more accurate.However, it affects the width of the sample and increases the research cost.Because of that, using 1% error rate will be meaningless if the result does not differ that much from 5% error rate (Argyrous, 1997).Based on those considerations, the sample determination formula uses 5% of error rate.It is known that the total of sample on this research are 96 respondents.Meanwhile, the expected results or the hypothesis on this research are as follows.
H 1 : There is partial influence of promotion variable to consumer purchase decision.H 2 : There is partial influence of product differentiation variable to consumer purchase decision.H 3 : There is simultaneously influence of promotion and product differentiation variable to consumer purchase decision.

RESULTS AND DISCUSSIONS
Product differentiation of Jukajo is innovation of flavor variation like Azuki red bean, which is rarely found in Indonesia.Then, the density of Jukajo reaches 27% for all flavor.This is because, in the process, the waste of the beans is not filtered, unlike the other competitor which only takes the flavored water.Jukajo also uses hexagonal water.According to Dr. Walter Kim in Nugraha (2013), hexagonal water is very good for the body because it is easier to be absorbed during the metabolism process.Jukajo ensures a healthier product to be consumed.For example, Jukajo in the production process does not use coconut milk, preservation, or synthetic sweetener.
Moreover, price of Jukajo product as premium product is rather high.It can be Rp15.000,00-Rp17.000,00per bottle with normal price, and Rp22.000,00-Rp24.000,00per bottle for supermall area like AEON, LOTTE, Premium Mall, and PAPAYA Supermarket.The high price shows that it is intended to be bought by middle or upper class, and needs a good marketing so it can be optimal.One of the strategies is promotion such as exhibitions, bazaar, and others.The marketing to modern retailers is more like a reputation and premium brand for image building.Similarly, the marketing in some corporate canteens and cooperatives such as the National Palace, Garuda Indonesia, Bank Indonesia, and the Ministry of Agriculture has the same function.However, marketing with resellers is Jukajo ultimate weapon.
For resellers in Tangerang, there are 715 in total.346 of them are within the city of Tangerang, and the distribution center is in Karawaci.47 resellers are located in Perumnas Tangerang Kota.If it is assumed one reseller has at least five consumers, the amount of Jukajo final consumer is 235 in Karawaci.
Correspondent's characteristics in this research will be classified according to gender and occupation.Based on the obtained data, the characteristics of consumers in Perumnas Tangerang Kota are dominated by female as shown in Table 1.
This result is just like Raja Muhammad Fauzi said, that Jukajo's targets are female around 23-40 years old, which to be specified are pregnant lady and housewives.Then based on correspondent's working status as could be seen in Table 2.
The validity and reliability of the result of responses obtained are tested.The validity test is done by taking 30 respondents who are Jukajo consumer in Perumnas Tangerang Kota.The validity test and reliability will be conducted five times because there is an invalid and unreliable item.There is also a valid yet unreliable question, so it has to be removed and needs to be retested.The invalidity happens to items is caused by the unsuitable response with the expected results.At first, the researchers are confident that every questions used provide an appropriate response for each indicator in each variable tested as it has followed the rules of the theory used in this research.However, the possibility of the invalidity of the questions can still happen.
Based on the total, there are 17 questions which cannot be used and 24 questions which can be used.The questions which do not pass the test have r arithmetic value or factor correlation less than 0,361 (r table).The results can be seen in Table 3, 4, and 5.
The, the reliability test has the purpose to know whether the questionnaire used by the researcher is reliable or not as a research instrument.If the alpha Cronbach value is bigger than 0,6, it can be said that the questionnaire is reliable and produces a consistent result (Ghozali, 2013).Reliability test results of each variable in the research instrument are contained in Table 6, and overall reliability test results of variables are shown in Table 7.
After validity and reliability test, the questionnaire distribution will be done to 96 samples.The result is tested with classical assumption.It is a test to know whether the regression model used will give a representative results or not (Ghozali, 2013).
Classic assumption test used in this research includes normality, multicollinearity, and heteroscedasticity.This test is helped by IBM SPSS 2.2.
Normality test is done by looking at the data spread at the diagonal graph.There are two ways to know whether the data distribution is normal or not.First, it uses graphic analysis.Second, it use Kolmogorov-smirnov statistic test (Ghozali, 2013).The normality result in form of graphs is seen in Figure 2, and the Kolmogorov-smirnov is in Table 8.The result of Kolmogrov-smirnov is bigger than 0,05, which is 0,200.It can be said that data error is distributed normally and fulfils normality assumption.
Next, it is multicollinearity.It is done to know if there is any relation between the variables.Those variables have tolerance bigger value than 0,10.It means there is not any relation between them.This indicated by the value.If it is bigger than 0,10 and has VIF less than 10, there is no multicollinearity between them (Ghozali, 2013).The multicollinearity test result is in Table 9.
Then, the classical assumption test is heteroscedasticity which serves to avoid biased regression model.It results in the variance of the residuals an observation that is inconstant (Ghozali, 2013).This can be done in two ways.Those are by Spearman's Rho and scatter plot.The results are in Table 10 and Figure 3.
Based on the results, it shows that all residual variances of each independent variable in this research are free from heteroscedasticity.It is because it has bigger value than 0,05.Based on scatter plot, the observation also does not indicate the formation of a particular pattern.
Furthermore, because there is no heteroscedasticity, the data is processed with a model feasibility test.It aims to prove whether the independent variables (X) simultaneously or together influence the dependent variable (Y).If the probability value of F is <0,05 with significance level = 0,05, then there is the influence of independent variables simultaneously to the dependent variable, and vice versa (Ghozali, 2013).The result can be seen in Table 11.
Then, it can be concluded that all independent variables in this research simultaneously influence the consumer purchasing decisions by the final consumer Jukajo in Perumnas Tangerang Kota.H 3 is accepted.The result of this calculation is also consistent with research conducted Riyaningrum et al. (2015).They agreed that the variable of product differentiation and promotion had positive and significant impact on consumer purchasing decisions.It means if the product differentiation and promotion have been increased, it will also increase consumer purchasing decisions to purchase a product.Next, there is a multiple linear regression analysis.It is used as a tool to determine how far the changes in the value of the dependent variable if the value of the independent variables is manipulated or changed (Sugiyono, 2013).The results are obtained from calculations with IBM SPSS 2.2 related to regression coefficients.It is shown in Table 12.
The results show that the multiple linear regression constants obtained are 5,545.This constant value indicates if X 1 (Promotion) and X 2 (Product Differentiation) are zero, then consumer purchase decision on Jukajo product is 5.545 units.With regression coefficient value of each independent variable is 0,101 and 0,314 with error level of 0,05, the regression model is formed as follows.
Y=5,545 +0,101X1+0,314X2+0,05 (3) The regression coefficient of 0,101, it means that if Jukajo improves or enhances its promotion activity, the consumer purchase decision will increase by 0,101 units.It is assumed that the other variables in this research are fixed.These results are consistent with the results by Purba (2016) that the promotion  variables significantly influence consumer purchasing decisions.
For product differentiation, it has regression coefficient value about 0,314.It means if Jukajo improves its product differentiation, the possibility of purchasing by consumers in Perumnas Tangerang Kota will increase by 0,314 unit by assuming the other independent variable is fixed value.Based on the results of this research is in line with the research by Ong (2013).They said that product differentiation affected the purchasing decisions of consumers, so the increase or decrease in consumer ratings of product differentiation significantly influencing consumer purchasing decisions was true.
The next test is a partial test.It aims to measure how far the influence of an individual variable independently in explaining the dependent variable, namely the purchase decision.If t arithmetic is greater than t-table, then the independent variables affect the dependent variable (Ghozali, 2013).The value of t-table in this research is 1,98580 obtained with degree of freedom (df) value of 93.It is from reducing the number of samples taken (96) with a total of variables used, which is 3. Partial test results can be seen in Table 13.
Based on the results obtained, it can be stated promotion variable has partial effect on Jukajo consumer purchasing decisions in Perumnas Tangerang Kota.It means H 1 is accepted.The promotion has value of t arithmetic about 2,552.It is bigger than t-table which is 0,198580.Thus, this result supports the results of Warne and Drake-Brooks (2016).It is shown by several dimensions of promotional activities of social media, mass media coverage, and blog, or website that have a positive effect on consumer purchasing decisions.The results of this research are also similar Riyaningrum et al. (2015), and Nawawi and Ikhaz (2015).Promotion variables have a partial positive effect on consumer purchasing decisions.Because of that, the researchers suggest that promotional activities should be intensified further, in addition to promotional activities conducted by personal selling or reseller as Jukajo marketing strategy.For example, the promotion can be through exhibitions and events.There are 30-40 respondents who provide 2 and 3 value to four items to the questions about dimensional events and experience.They have never seen exhibitions or fairs organized by Jukajo.The number is quite significant because almost 50% of the total respondents are encountered randomly.
Although it has a smaller regression coefficient value of 0,101, the promotion still has a positive effect on consumer purchasing decisions.Thus, it still needs to be improved to expand the range of market.
Likewise, with product differentiation variables, with a value of t arithmetic of 5,752 is greater than t-table, which is about 0,198589.It can be stated that product differentiation also affects the Jukajo consumer purchase decision of in Perumnas Tangerang Kota positively and partially.H 2 is accepted.The results of this research are indirectly in accordance with research conducted by Nishino, Akai, and Tamura (2014).They stated that product differentiation activities conducted on a product would affect the buying desire and consumer purchasing decisions.Furthermore, according to Nawaz et al. (2012), packaging as one dimension of product differentiation and questionnaire indicators in this research affects consumer purchase decision positively.The product differentiation becomes the most dominant independent variable with positive influence on consumer purchasing decisions with regression coefficient about 0,314.Therefore, the researchers suggest that the strategy related to Jukajo product differentiation should be a major concern.
Then, quality conformance is one dimension of product differentiation variables.Consumers always expect quality conformity.Quality conformity is the degree to which all units produced are identical and meet the promised specifications (Kotler & Keller, 2012).Researchers suggest this because when collecting questionnaires, some respondents as the consumers complain that Jukajo products are sour when they buy it once.This may happen because Jukajo products do not use artificial preservatives, which makes it unable to last long.Based on interviews with Anton Haryatmo by the researchers, Jukajo Marketing Manager, currently in its distribution system, it uses six modes of car transportation and three motorcycles.For delivery motors, it is done by using a homemade cool box and for close-up delivery.Meanwhile, for long-distance delivery, it uses car transport mode, and the packaging for distribution uses styrofoam boxes added with ice or also called as cold styrofoam.Although Mr. Anton Haryatmo states that the period of the cold temperature resistant styrofoam is 2 x 24 hours in a sealed condition, in fact, there has been a change in the taste of Jukajo products.It becomes acidic (perhaps the product are going stale) when it reaches those consumers.The researchers suggest that Jukajo should increase its supervision on the process of distributing the products, such as cold temperature checks of cold styrofoam in transportation mode of Jukajo distribution.It also should make sure the cold styrofoam is always tightly closed.Then, it can speed up the required time during the process of moving Jukajo products when they arrive at their destination.Cool storage can also be done on the car, but the investment cost is too expensive and is not advised to do.
In addition to these problems, it is also necessary to improve the quality control of Jukajo products related to product viscosity.There are at least 8 out of 96 samples complaining about the Jukajo products that it is too thin and not thick enough.With the improvement of quality control on Jukajo product consistency, it is also expected to contribute to the improvement of Jukajo consumer purchasing decision.
Based on the explanation given, the results of this research are in line with Riyaningrum et al. (2015).It is true that promotion and product differentiation have partial positive effects on consumer purchasing decisions.
The last test is to find the coefficient of determination.Based on calculations with IBM SPSS 2.2, the coefficient of determination of multiple linear regression models is 0,505.It indicates variable ability of promotion and product differentiation of Jukajo in explaining its relation to Jukajo consumer purchasing decision at Perumnas Tangerang Kota.It is equal to 50,5%.Meanwhile, the remaining 49,5% is influenced by other factors outside this research.

CONCLUSIONS
Based on the findings of this research, there are several conclusions.Firstly, Jukajo is a producer of premium beverage which has done at least four major differentiation on its products.Those are innovation of new flavor variants; beverage consistency level; using hexagonal water; and excluding the use of coconut milk, preservatives, and artificial sweeteners.Jukajo is also known for its promotional strategy by focusing more on using personalized mix selling through their resellers of more than 2.000 people.They are spread in various cities such as Jakarta, Tangerang Kota, Tangerang Selatan, Pandeglang, and Bandung.
Second, promotion and product differentiation affect the decision of Jukajo end consumer purchases at Perumnas Tangerang Kota respectively and partially.H 1 and H 2 are accepted.Product differentiation becomes the most dominant independent variable with positive influence on consumer purchasing decisions with regression coefficient of 0,314.Because of that, the researchers suggest that strategy related to product differentiation should be a major concern for Jukajo.It can be done by keeping the quality of Jukajo product conformity.
Third, the multiple linear regression models are generated with coefficient of determination about 0,505.It indicates that Jukajo promotion and product differentiation influence 50,5% of consumer purchasing decision in Perumnas Tangerang Kota.This regression model can be considered feasible because it has a value of F arithmetic of 47,442 which is bigger than F table.Based on these results, it can be concluded that all independent variables simultaneously affect the Jukajo consumer purchase decision in Perumnas Tangerang Kota (H 3 is accepted).

Table 1
Characteristics of Respondents by Gender (Source: Primary Data, 2017)

Table 5
Validity Test Results of Consumer Purchasing Decision Variable

Table 8
Normality Test Result

Table 9
Multicollinearity Test Result

Table 11 Model
Feasibility Test Result

Table 12 Multiple
Linear Regression Coeffiicient