CC BY
5
DOI 10.18551/rjoas.2021-08.18
EFFECT OF E-MARKETING MIX BASED ON E-MARKETPLACE ON MARKETING
PERFORMANCE OF FOOD MSMES
Fachriyan Hilmi Arija*
Faculty of Agriculture, Wahid Hasyim University, Semarang, Indonesia
Jamhari Jamhari, Irham Irham, Waluyati Lestari Rahayu
Faculty of Agriculture, Gadjah Mada University, Yogyakarta, Indonesia
*E-mail: hilmi.arija16@gmail.com
ABSTRACT
Electronic marketplace (E-marketplace) provides a new paradigm for every business organization in managing its business, including food MSMEs. The dominance of the marketing function, which was originally just a marketing mix in traditional marketing, has turned into relationship marketing supported by e-marketing mix activities. This study aims to determine the effect of the e-marketing mix (product, price, place, promotion, people, process) on the marketing performance of food MSMEs through the e-marketplace. E-Marketplaces selected as sample areas were Tokopedia, Shopee, Bukalapak. Respondents who were sampled were MSMEs that sell snack products in the E-Marketplace with a total of 135 respondents. The analysis technique used in this research is Structural Equation Modeling (SEM). The total indicators measured in the model are 23 indicators. The results showed that among the six e-marketing mix variables, place and people variables had a significant effect on marketing performance. Therefore, to produce high marketing performance in e-marketplaces, food MSMEs can prioritize the implementation of "place" strategies, namely maintaining product stock, selecting the right keywords/search words, complete delivery and payment support, including the Cash on Delivery option for consumers, as well as "people" strategies such as responding quickly to customer inquiries, high concern for customer complaints, and personal selling.
KEY WORDS
E-marketplace, e-marketing mix, food MSMEs, marketing performance, SEM.
As a country that is rich in food agriculture, Indonesia has quite a lot of MSMEs (micro, small and medium enterprises) that are engaged in the food sector (food MSMEs). However, the increasingly fierce competition and rapid global development have resulted in food MSMEs having to deal with increasingly complex competitions. One of the efforts to develop and improve the competitiveness of MSMEs can be done through internet-based business programs or else better known as e-commerce (Permana, 2017; Febriantoro, 2018; Zakiah, Ekawijana and Laksana, 2019; Suyanto and Purwanti, 2020). The development of the ecommerce has even become the government's attention with the issuance of the Economic Policy Package 14 on e-commerce (Badan Pusat Statistik, 2019).
E-commerce is defined by Turban et al. (2018) as a concept that describes a process of buying and selling or exchanging products, services, and information via the internet. In line with the development of the internet, a new understanding has emerged regarding the ecommerce paradigm in the form of a modern market-oriented marketing concept or a marketing revolution called the E-Marketplace (Bakos and Brynjolfsson, 1999; Arnott and Bridgewater, 2002; Eid and Trueman, 2002). E-Marketplace provides a new paradigm for every business organization in managing its business, including food MSMEs. The results of the Katadata Insight Center (KIC) survey entitled "MSME Study Report 2021: The Role of Marketplaces for MSMEs" proves that E-Marketplaces are able to help MSMEs survive a pandemic and even penetrate exports (Kurniawan, 2021). Another survey conducted by Visa also states that the new normal condition has encouraged many Indonesian consumers to
consider e-commerce platforms as an ideal means to help buy their daily needs locally, and for MSMEs as the backbone of the Indonesian economy it can continue to survive and grow in this challenging time (Firmansyah, 2021).
In the E-Marketplace, the dominance of the marketing function, which was originally just a marketing mix in traditional marketing, has turned into relationship marketing supported by e-marketing mix (Eid and Trueman, 2002). If the marketing mix consists of several dimensions that can affect the demand for a product, known as 4Ps, namely Product, Price, Place, and Promotion (Kotler and Keller, 2012), then in the e-marketing mix, the elements in the traditional 4P criteria have changed (Tálpáu, 2014). In e-marketing mix, the dimensions of the traditional marketing mix have been modified and expanded from 4P to 7P, by adding three new P's, namely People, Process, and Physical Evidence. But in e-commerce, Physical Evidence turns into Virtual Evidence. The e-marketing mix is a fundamental strategy for the success of any company and is very influential in activities in the e-commerce environment (Pogorelova et al., 2016).
The e-marketing mix strategy is also directed at producing marketing performance. Marketing performance in business is a measure of performance level which includes sales growth, number of buyers, profits, and market share growth (Anand and Khanna, 2000; Egan, 2011). Marketing performance is a consequence of marketing activities that have been carried out, both internal activities related to the management of internal resources and resources resulting from the consequences of dealing with other parties (Murphy et al., 2005). Measurement of marketing performance is important, because it can be used as input (information) for decision makers on all marketing activities that have been carried out.
E-Marketplace supports all activities related to transactions and interactions or planning for the transformation of goods (Holzmüller and Schlüchter, 2002), thereby enabling the various parties to collaborate with each other in the design, development, production and distribution of the final product in the supply chain. The main goal of the E-Marketplace is to eliminate any inefficiencies in the industry (Barratt and Rosdahl, 2002). This was confirmed in the study by Sin et al. (2016) stated that transactions in e-commerce can increase sales, reduce costs, and explore new business. Sin et al. (2016) continued that E-Marketplace is a form of e-commerce that provides benefits for all parties, especially MSMEs or startup businesses.
The literature on impact on users is still very limited despite the potential for E-Marketplaces and e-commerce which have many advantages. Most of the e-commerce related literature focuses on describing and analyzing the extent of adoption and use by agribusiness (Brush and McIntosh, 2010; Ng, 2013), or adoption by MSMEs (Daniel, Wilson and Myers, 2002; Rahayu and Day, 2015; Yadav and Mahara, 2019). In addition, the majority of studies evaluated only e-commerce sites that focused on assessing user perceptions of web quality (Agarwal and Venkatesh, 2002; Aladwani and Palvia, 2002; Petre, Minocha and Roberts, 2006; Loiacono, Watson and Goodhue, 2007).
Other studies in the e-commerce environment on average still focus on the consumer side, for example the factors that influence online purchasing decisions (Cheung, Lee and Thadani, 2009; Zhu, Zhang and Zhu, 2012; Rudansky-Kloppers, 2017; Khanna and Awal, 2019; Dermawan, Nasution and Sitepu, 2020). Research on E-Marketplaces has been conducted by Jiang & Balasubramanian (2014) to explore differences in market efficiency between traditional and electronic markets. Furthermore, Renna (2010) in her research proposes negotiation policies, customer tactics, and coalition tools, as added value services in the E-Marketplace. Meanwhile, the research on the e-marketing mix is mostly still at the description or introduction stage, it has not yet reached the impact of the e-marketing mix on marketing results achieved (Robins, 2004; Dominici, 2009). Research on the impact of ecommerce on agro-food marketing has been carried out by Baourakis, Kourgiantakis, & Migdalas (2002), but this research is not on the E-Marketplace environment and does not involve the e-marketing mix variable.
This study explains the impact of e-commerce utilization for users (sellers), by developing an empirical model of the Effect of E-Marketing Mix Based on E-Marketplace on Marketing Performance. This empirical model will test several specific hypotheses that
integrate E-Marketplace-based marketing strategies based on the marketing mix theory (Kotler and Keller, 2012), internet marketing theory (Eid and Trueman, 2002), and e-marketing mix theory (Pogorelova et al., 2016), with the marketing performance of relationship marketing activities (Anand and Khanna, 2000; Egan, 2011).
The e-marketing mix model used in this study refers to the model from Pogorelova et al. (2016), namely Product, Price, Place, Promotion, People, Process, and Virtual Evidence. However, because this research is based on E-Marketplace, virtual evidence is removed. Virtual evidence is not included in the SEM model as an exogenous variable because this research is based on e-marketplaces, where virtual evidence in the form of a container for opening an online store on a website or mobile application has been prepared by the e-marketplace, sellers (MSMEs) only need to register, open online stores, and uploading their products in the e-marketplace. Hence empirically, this research will integrate the variables in the e-marketing mix in the form of product, price, place, promotion, people, and process with variables which are positive consequences in the form of marketing performance produced by food MSMEs through e-marketplaces.
METHODS OF RESEARCH
This research begins with an activity of exploring the theories and concepts that will be used. This explanatory research activity is used to find and limit research problems so that they are applicable and researchable. The yardsticks of the sample area in this study include two things, i.e. the E-Marketplace and the location of MSMEs. The E-Marketplaces chosen as sample areas are Tokopedia, Shopee, and Bukalapak, because theoretically and empirically they have characteristics that are in accordance with the topic and research objectives. The sample taken is MSMEs that sell snack products in one of the E-Marketplaces (Tokopedia / Shopee / Bukalapak). Furthermore, the location or area of origin of MSMEs which was determined as the sample area was Central Java Province because it had the highest distribution of snacks than other provinces.
The sampling method uses accidental sampling where the sample is selected and determined based on time and cost limitations, without considering the population as a sample frame, because it is very heterogeneous in terms of numbers, and the selected respondents are considered to be able to provide a picture that is closer to the truth. Thus to control for the suitability of the study with the research objectives, the indicators of selecting respondents as a sample include: 1) MSMEs sell more than one (> 1) snack food product; 2) MSMEs have been selling snack products in one of the E-Marketplaces for at least 2 years and are still active until now; 3) MSMEs have already completed sales transactions at one of the E-Marketplaces (products are sold and reach consumers). So, the indicators of the sampling technique are more on the aspect of the relevance to the research topic than on the population representation. These stipulated requirements are needed to make it easier for researchers to explore research data input.
The sample size in this study is based on various considerations. First, the population size is dynamic, although the number of population members in this study is counted, but to decide whether this number will decrease or increase in a certain time is not certain. Second, this study uses the Maximum Likelihood (ML) estimation model, which requires a minimum size of 100 to 200 samples. The sample size in this study refers to Ghozali (2014) which states that the number of samples can be calculated from the size of the parameter multiplied by 5 to 10. Because this study uses 7 construct variables with a total indicator number of 23, the sample required is at least 23 x 5 = 115, and a maximum of 23 x 10 = 230. The number of samples in this study is 135 which already exceeds the minimum sample size. , so that is sufficient for analysis.
Data was collected through a questionnaire in the form of Google Form which was distributed to respondents in the form of a link via e-mail or Whatsapp number that was previously obtained. The type of question in this questionnaire is a closed question using an interval measurement scale in the form of a biporal adjective with a score range of 1 to 5 with two extreme points (agree-disagree scale), namely strongly agree (score 5) and strongly
disagree (score 1). Furthermore, the data obtained from the Google Form is downloaded and entered in a database in the Microsoft Excel program. This database is then processed with the AMOS Graph 22.0 program according to the needs and interests to achieve the objectives of the analysis in this study.
The analysis technique used in this research is Structural Equation Modeling (SEM) using the AMOS Graph Version 22.0 analysis tool. The advantage of this analysis technique is its ability to test together the structural models and the measurement models. SEM can confirm various indicators / dimensions of a concept / construct and measure the theoretical relationship between variables. This study will analyze the influence between variables, where there are several dependent variables and this dependent variable can be an independent variable for other dependent variables. So, the reason for using this technique is because methodologically the design of this study is relatively complicated, and using SEM is predicted to be able to test what is wanted to be achieved in this study.
SEM analysis stages are as follows: 1) building an inner model; 2) building an outer model; 3) constructing a path diagram (Figure 1); 4) conversion of the path diagram into the equation (Table 1.); 5) select the input matrix and model estimation; 6) assessing the possibility of identification problems; 7) evaluation of goodness of fit (Table 2.); 8) model interpretation and modification; 9) hypothesis testing (hypothesis is rejected if the critical ratio value is < 2.0 and P-value > 0.05, the hypothesis is accepted if the critical ratio value is > 2.0 and P-value < 0.05) (Ghozali, 2014).
Table 1 - Construct Variables, Indicators, and Measurement Models
Construct Variables Indicators Measurement Models
Product (X1) Photos or Product Images (X1.1) Product Variations(X1.2) Product Description (X1.3) Product Update Information (X1.4) ÀX1.1.X1 + e1 ÀX1.2.X1 + e2 ÀX1.3.X1 + e3 ÀX1.4.X1 + e4
Price (X2) Price Comparative Analysis (X2.1) Discount or Wholesale Price (X2.2) Price Changes Based on Temporal Dynamics (X2.3) ÀX2.1.X2 + e5 ÀX2.2.X2 + e6 ÀX2.3.X2 + e7
Place (X3) Stock / Product Availability (X3.1) Keywords Selection / Product Search Words (X3.2) Shipping and Payment Support (X3.3) ÀX3.1.X3 + e8 ÀX3.2.X3 + e9 ÀX3.3.X3 + e10
Promotion (X4) Customer Reviews about Products (X4.1) Promotion Intensity (X4.2) Promotion Variations (X4.3) ÀX4.1.X4 + e11 ÀX4.2.X4 + e12 ÀX4.3.X4 + e13
People (X5) Speed of Responding to Questions (X5.1) Concern for Customer Complaints (X5.2) Personal Selling (X5.3) ÀX5.1.X5 + e14 ÀX5.2.X5 + e15 ÀX5.3.X5 + e16
Process (X6) Standard Operating Procedure to Customers (X6.1) Order Processing Speed (X6.2) Order Packaging (X6.3) ÀX6.1.X6 + e17 ÀX6.1.X6 + e18 ÀX6.1.X6 + e19
Marketing Performance (Y1) Sales growth (Y1.1) Profit growth (Y1.2) Customer growth (Y1.3) Market share growth (Y1.4) ÀY1.1.Y1 + e20 ÀY1.2.Y1 + e21 ÀY1.3.Y1 + e22 ÀY1.4.Y1 + e23
Table 2 - Goodness of Fit Criteria
Goodness of Fit Cut-Off Value
X2 Chi Square Probability RMSEA GFI AGFI CMIN/DF TLI CFI Small value is expected > 0.05 < 0.08 > 0.90 > 0.90 < 2.00 > 0.95 > 0.95
Source: (Ghozali, 2014).
Figure 1 - Path Diagram of Structural Equation Model RESULTS AND DISCUSSION
SEM Assumption Test. The SEM assumption test in this study includes data normality evaluation (Table 3), outliers evaluation, multicollinearity and singularity evaluation, and residual testing. The data normality test is important and a requirement for processing data using the Maximum Likelihood (ML) estimation technique. The compliance of data normality can avoid bias and inefficient results. The results of outliers testing in this study did not indicate the existence of univariate and multivariate outliers in the observed variables. The assumption of multicollinearity and singularity can be detected from the determinant value of the covariance matrix. However, the AMOS 22.0 program has provided a "Warning" facility if there are indications of multicollinearity and singularity. If there really is multicollinearity and singularity, the treatment data that can be taken is to remove the variables that cause multicollinearity and singularity and then create a "composite variable" and use it for further analysis. The criteria for the value of residual covariances accepted in SEM are < 2.58 (Hair Jr, Joseph et al., 2009). Based on the results of data processing using AMOS version 22.0, there is no standardized residual covariance data greater than 2.58.
Validity and Reliability Test. The validity in this study was tested using convergent validity by taking into account the loading factor value obtained from the standardized regression weight compared to the rule of the thumb value used in this study, which is 0.6. Meanwhile, reliability is a measure of the internal consistency of the indicators of a construct which shows the degree to which each indicator indicates a common latent factor. The minimum reliability value of the dimensions of the latent variable forming dimensions that can be accepted is 0.7, but in explanatory research, the reliability of 0.5 - 0.6 has been accepted (Nunnally and Bernstein, 1994). The results of testing the validity and reliability of the indicators on each of the research variables are presented in Table 3.
Structural Model and Evaluation of Goodness of Fit. Structural models are used to describe models of research causality with tiered relationships. Proposed Model that has been made in this study is analyzed with a structural equation model with the help of AMOS 22 software.
Based on Figure 2, it can be seen that there are several goodness of fit criteria that have not been fulfilled. To get a Goodness of Fit value according to the recommended criteria, modifications must be made to the structural equation model which allows a relationship between error measurement variables.
Table 3 - Validity and Reliability Test
Variable Indicators Loading Factor Standard Error Construct Reliability Variance Extracted Validity Testing Interpretation Reliability Testing Interpretation
X1.1 0.854 0.271 Val d
Product X1.2 X1.3 0.808 0.838 0.347 0.298 0.903 0.700 Val Val d d Reliable
X1.4 0.846 0.284 Val d
X2.1 0.877 0.231 Val d
Price X2.2 0.898 0.194 0.894 0.739 Val d Reliable
X2.3 0.800 0.360 Val d
X3.1 0.723 0.477 Val d
Place X3.2 0.814 0.337 0.850 0.655 Val d Reliable
X3.3 0.883 0.220 Val d
X4.1 0.868 0.247 Val d
Promotion X4.2 0.841 0.293 0.892 0.734 Val d Reliable
X4.3 0.861 0.259 Val d
X5.1 0.838 0.298 Val d
People X5.2 0.821 0.326 0.876 0.702 Val d Reliable
X5.3 0.855 0.269 Val d
X6.1 0.791 0.374 Val d
Process X6.2 0.780 0.392 0.838 0.632 Val d Reliable
X6.3 0.814 0.337 Val d
Marketing Performance Y1.1 Y1.2 Y1.3 0.814 0.762 0.778 0.337 0.419 0.395 0.868 0.623 Val Val Val d d d Reliable
Y1.4 0.801 0.358 Val d
Figure 2 - Proposed Model
Modifications were made by taking into account the suggestions of the AMOS 22.0 program. in the form of modification indices by taking into account the estimated value and the significance of the relationship between error measurement. After modifying the index of the covariance between the error variables in the Proposed Model, the Final Model looks like in Figure 3.
TLI=.988 CFI=.991
Figure 3 - Final Model
Table 4 - Goodness of Fit Results in the Final Model
Goodness of Fit Indices Cut-Off Value Final Model Test Results Information
X2 Chi Square Small value is expected 224.425 Good
Probability > 0.05 0.114 Good
CMIN/df < 2.00 1.122 Good
RMSEA < 0.08 0.030 Good
GFI > 0.90 0.881 Marginal
AGFI > 0.90 0.835 Marginal
TLI > 0.95 0.988 Good
CFI > 0.95 0.991 Good
The Final Model is a fit model with an acceptable range of values (Table 4.). This shows that there is no significant difference between the covariance matrices of the data from the observed variables and the covariance matrices of the specified models, and it can be said to have conformity with the empirical data used in this study.
Testing the Relationship between Variables and Indicators. Final model in Figure 3 is a model that describes the causal relationship between the variables in the e-marketing mix and its influence on the variable marketing performance achieved by food MSMEs in the E-Marketplace. The structural model is then analyzed with the help of AMOS 22.0 software to produce numbers that show the relationship between indicators as part of the construct and the strength of the relationship between constructs. The results of estimating the relationship between variables and these indicators can be seen from output regression weight and standardized regression weight which are presented in Table 5.
Table 5. shows the strength and weakness of the causal relationship between variables, as well as between variables and their indicators. Based on Table 5., it can be seen that the indicators that form the constructs or latent variables in the model have a C.R. > 2.00 which means that all indicators are acceptable. The loading factor value on all indicators shows a value > 0.7 and a significant P value, so it can be said that all indicators have a strong relationship and can explain the existence of the construct.
The results of the goodness of fit test that have been carried out previously show that the final model as a whole is a fit structural model. The next process to see here is if there is a significant and close relationship between the independent variable (exogenous) and the dependent variable (endogenous). This relationship can be known based on the value of the Critical Ratio (C.R.) and P value shown in Table 5. If the value of C.R. > 2 or P value < 0.05, then H0 is rejected and H1 is accepted.
Table 5 - Results of Estimation of Relationship Between Variables and Indicators
Indicators dan Variables Estimate Standardized Estimate (Loading Factor) S.E. C.R. P
Y1 <— X1 -.114 -.125 .241 -.471 .638
Y1 <— X2 .079 .102 .247 .319 .750
Y1 <— X3 .597 .587 .191 3.125 .002
Y1 <— X4 .075 .087 .195 .385 .700
Y1 <— X5 .414 .435 .206 2.011 .044'
Y1 <— X6 -.082 -.074 .261 -.315 .753
X3.3 <— X3 1.165 .883 .133 8.751 ***
X3.2 <— X3 1.090 .814 .123 8.852 ***
X3.1 <— X3 1.000 .723
X2.2 <— X2 .987 .898 .066 14.917 ***
X2.1 <— X2 1.000 .877
X1.4 <— X1 1.113 .846 .101 10.969 ***
X1.3 <— X1 1.110 .838 .090 12.285 ***
X1.2 <— X1 .999 .808 .086 11.557 ***
X1.1 <— X1 1.000 .854
X4.3 <— X4 .946 .861 .072 13.120 ***
X4.2 <— X4 .980 .841 .078 12.528 ***
X4.1 <— X4 1.000 .868
X5.2 <— X5 .891 .821 .082 10.907 ***
X5.1 <— X5 1.000 .838
X6.2 <— X6 .980 .780 .103 9.533 ***
X6.1 <— X6 1.000 .791
Y1.1 <— Y1 1.000 .814
Y1.2 <— Y1 .899 .762 .090 10.018 ***
Y1.3 <— Y1 .957 .778 .093 10.291 ***
Y1.4 <— Y1 1.053 .801 .099 10.605 ***
X6.3 <— X6 1.151 .814 .114 10.083 ***
X5.3 <— X5 .932 .855 .092 10.086 ***
X2.3 <-- X2 .881 .800 .074 11.894 ***
The hypothesis in this study examines how the e-marketing mix variables influence marketing performance.The proposed hypothesis is:
• H0: e-marketing mix has no effect on marketing performance;
• H1: e-marketing mix affects marketing performance.
The hypothesis above can be described as follows:
• H0a: Product has no effect on marketing performance;
• H1a: Product affects marketing performance;
• H0b: Price has no effect on marketing performance;
• H1b: Price affects marketing performance;
• H0c: Place has no effect on marketing performance;
• H1c: Place affects marketing performance;
• H0d: Promotion has no effect on marketing performance;
• H1d: Promotion affects marketing performance;
• H0e: People has no effect on marketing performance;
• H1e: People affects marketing performance;
• H0f: Process has no effect on marketing performance;
• H1f: Process affects marketing performance.
If you look at Table 5, the thing that is of concern is that the product (X1) shows a negative loading factor on marketing performance (Y1). Interestingly, this result also occurs in research Fatonah (2009) which examines the effect of the marketing mix (product, price, place, promotion) on marketing performance in batik companies, where the product variable also shows a negative influence on marketing performance. Fatonah (2009) argues that this is most likely due to the very varied tastes of consumers. Researchers agree with this opinion, in addition to consumer tastes which are very varied, batik products and snacks are indeed very varied. The negative loading factor value on the product means that the more
variety and information on snack products displayed on the e-marketplace, it will reduce the marketing performance achieved by food MSMEs.
In addition to the product variable, the process variable (X6) also has a negative influence on marketing performance. This means that improving process strategies such as provisions/SOPs for prospective customers, speed of processing orders, and packaging can reduce marketing performance. For example, the provisions of the SOP for consumers who want to place an order. Some consumers or potential customers may cancel the decision to buy the product, if they consider the SOP provisions provided by the seller (MSMEs) to be too troublesome. A fast order process is what consumers or customers expect, but large orders can also cause errors in processing the order due to lack of accuracy. As a result, the product sent may be wrong or not in accordance with what was ordered by the consumer/customer.
Price and promotion variables have a positive influence (loading factor) on marketing performance, but not significant. This means that although the price and promotion variables have a positive influence on marketing performance, these variables do not have a major impact on the marketing performance of food MSMEs. While the results of the study found that the place and people variables had a positive and significant effect on marketing performance. The results in this study are different from the results of Guisi research (2018) which examines the effect of the 7P marketing mix (product, price, place, promotion, people, process, physical evidence) on marketing performance at shopping malls. Guisi research (2018) finds that the 7P marketing mix has a significant effect on marketing performance. Guisi (2018) uses the 7P marketing mix as a single construct variable, while in this study the e-marketing mix is broken down into several construct variables. There are different concepts between the traditional market and the electronic market, so the construction of the model will of course also be different.
Furthermore, from the results of the SEM analysis in Table 5., this study resulted in the finding that the place variable has a positive and significant influence on marketing performance, as indicated by the loading factor value of 0.587, the C.R. value of 3.125 (> 2.00), and the P value of 0.002. (< 0.05), so the research hypothesis (H1c) is accepted. This means that matters relating to the place strategy such as product availability/stock, keyword selection/product search terms, as well as delivery and payment support including the option to pay on the spot (COD/Cash on Delivery), has a significant influence on the marketing performance of food MSMEs in the e-marketplace. Allen and Fjermestad (2001) explain that the traditional marketing mix (4Ps) can be the basis of an e-commerce strategy, identifying the necessary changes to the need to create an appropriate model for e-marketing. Allen and Fjermestad (2001) added that there was a big change in the 4Ps in e-commerce, one of which was that place turned into reach. This large change from place to reach means that the better the place strategy, the higher the market reach that can be achieved. This means that the place strategy plays a very important role in e-commerce, which has also been proven from the results of this study.
Another finding in this study is that the people variable also has a significant effect on marketing performance, as indicated by the loading factor value of 0.435, the C.R. value of 2,011 (> 2.00), and a P value of 0.044 (< 0.05), so the research hypothesis (H1e) is accepted. This shows that the people strategy in the form of responses to customer questions, concern for customer complaints, and personal selling has a major impact on the marketing performance of food MSMEs in the e-marketplace. These results are consistent with a study conducted by Ferdinand and Wahyuningsih (2018) that the people strategy through sales force innovation will increase positive sales atmosphere initiatives, which in turn will improve marketing performance.
CONCLUSION AND RECOMMENDATIONS
This research is a development of empirical model in the context of the MSMEs snacks industry in the E-Marketplace, so that it cannot be generalized to the e-commerce market and other industries, or the population in general. A more representative sample population
needs to be searched and tested to generalize the findings. The main limitation of this study is only using quantitative methods to examine relatively complex phenomena. Methods such as confirmatory factor analysis and structural equation modeling are useful for testing measurement and model structure, but they cannot answer the "why" question when results deviate from what was expected. For example on product and process variables, which in this study have a negative influence on marketing performance. Future research could focus on confirming the empirical models developed and replicating this research in other contexts.
Despite these limitations, the research has important implications. The results of research testing the effect of the e-marketing mix on the marketing performance of food MSMEs in the e-marketplace show that the place and people variables have a significant effect on marketing performance. Meanwhile, other e-marketing mix variables, namely product, price, promotion, and process have no significant effect on the marketing performance of food MSMEs. Based on these findings, food MSMEs can prioritize the implementation of place strategies in e-marketplaces, namely maintaining product availability/stock, selecting keywords/ product search words, complete delivery support including providing on-site payment options (Cash on Delivery ) for consumers, as well as people strategies such as responding quickly to customer inquiries, high concern for customer complaints, and personal selling to be able to produce high marketing performance, such as sales growth, profit, customers, and market share.
REFERENCES
1. Agarwal, R. and Venkatesh, V. (2002) 'Assessing a firm's Web presence: A heuristic evaluation procedure for the measurement of usability', Information Systems Research, 13(2), pp. 168-186. doi: 10.1287/isre.13.2.168.84.
2. Aladwani, A. M. and Palvia, P. C. (2002) 'Developing and validating an instrument for measuring user-perceived web quality', Information and Management, 39(6), pp. 467476. doi: 10.1016/S0378-7206(01)00113-6.
3. Allen, E. and Fjermestad, J. (2001) 'E-commerce marketing strategies: an integrated framework and case analysis', Logistics Information Management, 14(1/2), pp. 14-23. doi: 10.1108/09576050110360106.
4. Anand, B. N. and Khanna, T. (2000) 'Do Firms Learn To Create Value? The Case Of Alliances', Strategic Management Journal, 21, pp. 295-315.
5. Arnott, D. C. and Bridgewater, S. (2002) 'Internet, interaction and implications for marketing', Marketing Intelligence & Planning, 20(2), pp. 86-95. doi: 10.1108/02634500210418509.
6. Badan Pusat Statistik (2019) Statistik E-Commerce 2019, Badan Pusat Statistik. Available at: https://www.bps.go.id/publication/2019/12/18/fd1e96b05342e479a83917c6/statistik-e-commerce-2019.html.
7. Bakos, Y. and Brynjolfsson, E. (1999) 'Bundling information goods: pricing, profits, and efficiency', Management Science, 45(12), pp. 1613-1630. doi: 10.1287/mnsc.45.12.1613.
8. Baourakis, G., Kourgiantakis, M. and Migdalas, A. (2002) 'The impact of e-commerce on agro-food marketing: The case of agricultural cooperatives, firms and consumers in Crete', British Food Journal, 104(8), pp. 580-590. doi: 10.1108/00070700210425976.
9. Barratt, M. and Rosdahl, K. (2002) 'Exploring business-to-business marketsites', European Journal of Purchasing and Supply Management, 8(2), pp. 111-122. doi: 10.1016/S0969-7012(01)00010-7.
10. Brush, G. J. and McIntosh, D. (2010) 'Factors influencing e-marketplace adoption in agricultural micro-enterprises', International Journal of Electronic Business, 8(4/5), p. 405. doi: 10.1504/ijeb.2010.035295.
11. Cheung, C. M. K., Lee, M. K. O. and Thadani, D. R. (2009) 'The impact of positive electronic word-of-mouth on consumer online purchasing decision', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 5736 LNAI, pp. 501-510. doi: 10.1007/978-3-642-04754-1_51.
12. Daniel, E., Wilson, H. and Myers, A. (2002) 'Adoption of E-Commerce by SMEs in the UK: Towards a Stage Model', International Small Business Journal, 20(3), pp. 253-270. doi: 10.1177/0266242602203002.
13. Dermawan, A. A., Nasution, H. and Sitepu, M. H. (2020) 'The impact of branding on purchasing decision-making in mall shopping and online shopping', IOP Conference Series: Materials Science and Engineering, 801(1). doi: 10.1088/1757-899X/801/1/012146.
14. Dominici, G. (2009) 'From Marketing Mix to e-Marketing Mix: a literature overview and classification', International Journal of Business and Management, 4(9), pp. 17-24. doi: 10.5539/ijbm.v4n9p17.
15. Egan, J. (2011) Relationship Marketing: Exploring Relationship Strategies in Marketing. 4th Ed. Pearson Education Limited.
16. Eid, R. and Trueman, M. (2002) 'The Internet: New international marketing issues', Management Research News, 25(12), pp. 54-67. doi: 10.1108/01409170210783278.
17. Fatonah, S. (2009) 'Pengaruh Bauran Pemasaran dan Orientasi Pasar terhadap Keunggulan Bersaing dalam Meningkatkan Kinerja Pemasaran pada Perusahaan Batik Di Surakarta', Jurnal Ilmu Ekonomi dan Manajemen, 6(No. 1).
18. Febriantoro, W. (2018) 'Kajian Dan Strategi Pendukung Perkembangan E-Commerce Bagi Umkm Di Indonesia', Jurnal MANAJERIAL, 17(2), p. 184. doi: 10.17509/manajerial.v17i2.10441.
19. Ferdinand, A. T. and Wahyuningsih, W. (2018) 'Salespeople's innovativeness: A driver of sales performance', Management and Marketing, 13(2), pp. 966-984. doi: 10.2478/mmcks-2018-0016.
20. Firmansyah, I. (2021) Survei: Masyarakat Indonesia Minati E-Commerce yang Cepat dan Nyaman, lifestyle.sindonews.com. Available at: https://lifestyle.sindonews.com/read/418252/166/survei-masyarakat-indonesia-minati-e-commerce-yang-cepat-dan-nyaman-1620140735 (Accessed: 5 May 2021).
21. Ghozali, I. (2014) Model Persamaan Struktural: Konsep dan Aplikasi dengan Program AMOS 22.0 Update Bayesian SEM. Semarang: Badan Penerbit Universitas Diponegoro.
22. Guisi, Z. Z. (2018) 'The Effect of Marketing Mixed Strategy on Marketing Performance and Competitive Excellence', JAGADHITA: Jurnal Ekonomi & Bisnis, 5(1), pp. 33-40. doi: 10.22225/jj.5.1.438.33-40.
23. Hair Jr, Joseph, F. et al. (2009) Multivariate Data Analysis 7th Edition. 7th editio. Pearson Prentice Hall.
24. Holzmüller, H. H. and Schlüchter, J. (2002) 'Delphi study about the future of B2B marketplaces in Germany', Electronic Commerce Research and Applications, 1(1), pp. 219. doi: 10.1016/S1567-4223(02)00003-0.
25. Jiang, P. and Balasubramanian, S. K. (2014) 'An empirical comparison of market efficiency: Electronic marketplaces vs. traditional retail formats', Electronic Commerce Research and Applications, 13(2), pp. 98-109. doi: 10.1016/j.elerap.2013.11.003.
26. Khanna, R. and Awal, G. (2019) 'Consumer Online Purchasing Decision and its Influencing Factors in Uttrakhand: An Exploratory Study of Selected Districts of Garhwal Division', SSRN Electronic Journal. doi: 10.2139/ssrn.3383352.
27. Kotler, P. and Keller, K. L. (2012) Marketing Management. Fourteenth Global Edition. 14th Ed, Pearson Education Limited. 14th Ed. Pearson Education Limited. doi: 10.1017/cbo9781139174749.022.
28. Kurniawan, A. (2021) Survei Membuktikan: Marketplace Bantu UMKM Bertahan Saat Pandemi dan Menembus Ekspor, ekbis.sindonews.com. Available at: https://ekbis.sindonews.com/read/414520/34/survei-membuktikan-marketplace-bantu-umkm-bertahan-saat-pandemi-dan-menembus-ekspor-1619805901 (Accessed: 1 May 2021).
29. Loiacono, E. T., Watson, R. T. and Goodhue, D. L. (2007) 'WebQual: An instrument for consumer evaluation of web sites', International Journal of Electronic Commerce, 11(3), pp. 51-87. doi: 10.2753/JEC1086-4415110302.
30. Murphy, B. et al. (2005) 'Stakeholder perceptions presage holistic stakeholder
relationship marketing performance', European Journal of Marketing, 39(9-10), pp. 1049-1059. doi: 10.1108/03090560510610716.
31. Ng, E. (2013) 'Making strategic decisions on B2B E-commerce models: An empirical study on Australian agribusinesses', International Journal of Electronic Commerce Studies, 4(1), pp. 1-20. doi: 10.7903/ijecs.995.
32. Nunnally, J. C. and Bernstein, I. H. (1994) Psychometric Theory 3rd edition. 3rd editio. Mc, Graw-Hill.
33. Permana, S. H. (2017) 'Strategi Peningkatan Usaha Mikro, Kecil, Dan Menengah (Umkm) Di Indonesia', Aspirasi, 8(1), pp. 93-103.
34. Petre, M., Minocha, S. and Roberts, D. (2006) 'Usability beyond the website: An empirically-grounded e-commerce evaluation instrument for the total customer experience', Behaviour and Information Technology, 25(2), pp. 189-203. doi: 10.1080/01449290500331198.
35. Pogorelova, E. V. et al. (2016) 'Marketing mix for e-commerce', International Journal of Environmental and Science Education, 11(14), pp. 6744-6759.
36. Rahayu, R. and Day, J. (2015) 'Determinant Factors of E-commerce Adoption by SMEs in Developing Country: Evidence from Indonesia', Procedia - Social and Behavioral Sciences, 195, pp. 142-150. doi: 10.1016/j.sbspro.2015.06.423.
37. Renna, P. (2010) 'Negotiation policies and coalition tools in e-marketplace environment', Computers and Industrial Engineering, 59(4), pp. 619-629. doi: 10.1016/j.cie.2010.07.007.
38. Robins, F. (2004) 'The E-Marketing Mix', The Marketing Review, 1(2), pp. 249-274. doi: 10.1362/1469347002529134.
39. Rudansky-Kloppers, S. (2017) 'Key factors influencing the online buying decision when purchasing products in specific product categories', African Journal of Science, Technology, Innovation and Development, 9(1), pp. 43-54. doi: 10.1080/20421338.2016.1258026.
40. Sin, K. Y. et al. (2016) 'Relative Advantage and Competitive Pressure towards Implementation of E-commerce: Overview of Small and Medium Enterprises (SMEs)', Procedia Economics and Finance, 35(October 2015), pp. 434-443. doi: 10.1016/s2212-5671(16)00054-x.
41. Suyanto, U. Y. and Purwanti, I. (2020) 'Pengembangan Model Peningkatan Daya Saing UMKM Berbasis E-Commerce ( Studi Pada UMKM Kabupaten Lamongan)', Jesya (Jurnal Ekonomi & Ekonomi Syariah), 4(1), pp. 189-198. doi: 10.36778/jesya.v4i1.299.
42. Tälpäu, A. (2014) 'The marketing mix in the online environment', Bulletin of the Transilvania University, 7(2), pp. 53-58.
43. Turban, E. et al. (2018) Marketing and Advertising in E-Commerce. Springer International Publishing AG. doi: 10.1007/978-3-319-58715-8_10.
44. Yadav, R. and Mahara, T. (2019) 'Factors affecting e-commerce adoption by handicraft SMEs of India', Journal of Electronic Commerce in Organizations, 17(4), pp. 44-57. doi: 10.4018/JEC0.2019100104.
45. Zakiah, A., Ekawijana, A. and Laksana, E. A. (2019) 'Implementasi Metode Action Research untuk Peningkatan Daya Saing Umkm Melalui E-Commerce', Jurnal Penelitian Komunikasi Dan Opini Publik, 23(1). doi: 10.33299/jpkop.23.1.1727.
46. Zhu, L., Zhang, W. and Zhu, Y. (2012) 'Research on the influence of online reviews on internet consumer purchasing decision', Proceedings - 2012 International Conference on Management of e-Commerce and e-Government, ICMeCG 2012, pp. 38-41. doi: 10.1109/ICMeCG.2012.97.
