The Acceptance Technology Model for Adoption of Social Media Marketing in Jabodetabek

The research aimed to find out the influence and impacts of social media marketing (SMM) in promoting micro, small, and medium enterprises (MSMEs) in Indonesia. The approach of research was a quantitative research with data collection method using electronic questionnaires of Google forms obtaining 163 respondents who were owners of MSMEs in Jakarta and Tangerang. The measurement model of outer and inner model was analyzed by SmartPLS software. Structural Equation Model was being used to test the relations of each construct. The findings show that SMM positively mediates: 1) the relationship between perceived usefulness and impact on business; 2) the relationship between perceived ease of use and impact on business; 3) the relationship between compatibility and impact on business; and 4) the relationship between cost and impact on business, which conclude that SMM has positive impacts on business. On the other hand, SMM negatively mediates the relationship between facilitating conditions and impact on business. The research implies that the adoption of social media marketing by MSMEs is encouraged by such factors as usefulness, ease of use, compatibility, and cost. Hence, it is advisable that MSMEs facilitate the employees with technical infrastructures and training to increase productivity and performance.


I. INTRODUCTION
The Indonesian government considers micro, small, and medium enterprises (MSMEs) as one of the most important structures for the Indonesian economic system and a mainstay of the government (Purnomo, 2011;Tambunan, 2019). This type of business model is independently owned by an individual which meets the criteria set in the Indonesian law (Tambunan, 2019). Data from Ministry of Cooperatives and SMEs of Republic Indonesia, and The Economic Census from The Central Bureau of Statistics in 2016 shows the larger contribution of MSMEs in Indonesia towards Indonesian economy includes the MSMEs contributions in absorbs 89,2% of the total workforce, provides 99% employment, contributes 60,43% in Indonesian Gross Domestic Product (GDP), contributes 14,17% in exports, and contributes 58,18% in investment.
The criterias of MSMEs in form of capital are: 1) the micro business with net asset equal or less than Rp 50 million and annual sales equal or less than Rp 300 million; 2) the small business with net asset more than Rp 50 million to Rp 500 million and annual sales from Rp 300 million to Rp 2,5 billion; and 3) the medium business with net asset more than Rp 500 million to Rp 10 billion and annual sales more than Rp 2,5 billion to Rp 50 billion (Suci, 2017;Tambunan, 2019). All criteria of net asset does not include land and buildings for business premises.
To increase the number of MSMEs for using Information and Communication Technology (ICT), the government has initiated a program called 8 million MSMEs Go Online. The purpose of MSMEs going online is to provide opportunities for a wider market share for MSMEs in Indonesia, both local and global, to be able to increase sales and revenue both locally and globally. Expectation from the increase sales of MSMEs using both offline and online transactions is to attract more consumers due to the increased promotions, increased product sales, market share control, and increased profits. Sales in online platforms provides convenience in reaching consumers without physically meeting them, and reduces e-MSMEs expenses such as rental fee (Nilasari et al., 2019).
In 2020, a pandemic COVID-19 forces all business processes to take place online, including MSME in Indonesia. The Minister of Cooperatives Small and Medium Enterprises states that right now is the perfect time for MSMEs to switch their operation to online platforms. Bank of Indonesia reported an increase in the sales of e-commerce to 18% in May 2020. Unfortunately, only 13% of MSMEs are connected to this online marketplace, or around 8 million MSMEs (Catriana, 2021).
Electronic commerce has become a new accessible channel where MSMEs can globally improve the business scale (Kanchanatanee, Suwanno, & Jarernvongrayab, 2014). The vast development of e-commerce in Indonesia is influenced by the millennial lifestyle, in which online shopping is more preferred than conventional shopping. E-commerce also provides access for MSMEs to expand and reach other markets and retain relationship with customers and facilitate business transactions by utilizing internet and website technology (Mumtaha & Khoiri, 2019). Therefore, this situation drives MSMEs to utilize e-marketing. Furthermore, MSMEs in the rural area have now the chance to upgrade their business and compete with other competitors with low administration cost (Kanchanatanee et al., 2014).
Due to changes in the digital era, there is an increase in the number of individual or companies who choose to advertise in social media. GetCraft (2020) reports that Indonesian consume social media more than TV. The figure reaches to 3 hours 16 minutes compared to 2 hours 23 minutes. As a consequence, social media marketing (SMM) is considered to be truly influential with its commonly acceptable technology to be used in daily life. Usually, marketers use this marketing method to increase the level of engagement with their customers by exposing their brand. Marketers utilize SMM to attract consumers to purchase their products. Marketers consider social media not only as a mediator to enhance the brand's image but also as consumers' problem solver with abundance of information and insights.
In addition, marketers use SMM due to its ability to listen, fix, and avoid another mistake. Social listening helps professional marketers by providing some enlightening information. SMM also can help company achieve their ultimate goals, build brand loyalty and increase sales. However, if customers see social media outreach programs only as a disguise to sell, they are likely to be alienated. Instead, marketers should use social media to engage with consumers and consider it as a long-term strategy to increase sales. Despite the significant benefits of SMM, there is no absolute guarantee that MSMEs are fully aware of the various factors that need to be utilized to provide maximum business benefits.
There are factors that can support MSMEs in adopting the SMM such as social media usefulness, perceived ease of use, compatibility, facilitating condition, cost reduction, and the positive influence of SMM which impact the business of MSMEs. The research tests these factors to determine the significance of their impact on business.
According to Davis (1989), technology acceptance model (TAM) consists of three variables: 1) perceive usefulness (PU), 2) perceived ease of use (PEOU), and 3) attitude toward using (ATU). Previous researchers have found that TAM helps researchers to understand consumer behavior towards the use of technology (Rahayu & Day, 2015).
Another model of technology acceptance is Unified Theory of Acceptance and Use of Technology (UTAUT), with four main constructs such as, performance expectancy, social influence, effort expectancy, and facilitating conditions. These constructs influence the usage behavior and intention to use a technology (Venkatesh et al., 2003).
The research is to determine whether the perceived usefulness (PU), perceived ease of use (PEOU), compatibility (COM), facilitating conditions (FCO), and cost (COS) have positive impacts on MSMEs to adopt SMM; and whether SMM has a positive influence on the impact on business (IOB) of MSMEs. All these variables are extracted from both TAM and UTAUT.
The research examines and explores two research questions which are later elaborated using quantitative approach: 1) Does each perceived usefulness (PU), perceived ease of use (PEOU), compatibility (COM), facilitating condition (FCO) and cost (COS) have a positive impact on MSMEs to adopt SMM?; 2) Does SMM has a positive influence on the impact on business (IOB) of MSMEs?
The research results will be beneficial for all stakeholder of MSME, especially owner and government. Owners can develop SMM strategy and government can facilitate the willingness of MSMEs owner by developing soft infrastructure (regulation and policies) and hard infrastructure (broadband infrastructure). Perceived usefulness is the degree where a particular system is believed to improve individual's job performance. Alduaij (2019) and Davis (1989) argue that work environment with integrated information system will be more effective since the worker finds it to be helpful by optimally using its features and functions. Furthermore, social media is considered inexpensive, easy to learn and master compare to other expensive and complex technologies (Atanassova & Clark, 2015). Karahanna and Straub (1999) conclude that perceived usefulness (PU) is an important factor in improving work performance. Based on the hypothesis, the research paradigm is shown at Figure 1. Discovered by Davis (1989), perceived ease of use refers to the degree where a particular system is free from difficulties or great effort and it appears in technology acceptance model (TAM) as PU. This belief states that the adoption of a technology is not complex yet useful (Venkatesh et al., 2012). Rusmana, Bawono, and Indriyani (2018) have found that the system usage is influenced by perceived usefulness (PU) and perceived ease of use (PEOU). According to Alduaij (2019), most respondents agree that social media is a communication tool that is easy to use since users do not require a lot of efforts to create and use it. Besides, they believe that social media is understandable. It indicates that the use of ICT is not complex since users do not require any special IT skills and knowledge to operate it. Thus, the perceived ease of use has a special relation SMM. Compatibility refers to what extent that innovation is compatible with the user's prior experiences (Karahanna & Straub, 1999). The research concerns with the compatibility of the innovative technology strategy with the business practices of MSMEs. If MSMEs adopt a compatible SMM system with the MSMEs work system, they might consider the possibility and adopt the method (Hung & Lai, 2015). Adopting SMM in MSMEs systems would be considered perfect to professionally reach the prospective consumers and improve the business condition (Derham, Cragg, & Morrish, 2011). Facilitating condition is an individual perception of technological availability or organizational resources to create a system that can remove the barrier (Isiyaku et al., 2018). The technical infrastructure exists to support new technology use a new system (Venkatesh et al., 2003;Yang & Forney, 2013 (Accenture, 2014;Kim & Shin, 2015). Low barrier of low cost, low technology knowledge, and skills motivate MSMEs to adopt SMM in their business strategy (Derham et al., 2011). Thus, social media is a cost-effective technology, which also help MSME communicate with its consumers (Kaplan & Haenlein, 2010;Zhang et al., 2019).  Social media impacts the business operational, financial, and performance-social capital, social marketing, social corporate networking. With the increased engagement between company and consumers, consumers are willing to pay and participate in increasing company's revenue and brand awareness (Paniagua & Sapena, 2014).
H 6 : Social media marketing (SMM) has a positive influence on the impact on business (IOB).

II. METHODS
The population involves MSMEs owners in Jakarta and Tangerang. The research has a criteria that respondents are the owners of micro, small, or medium enterprise in Jakarta and Tangerang who use social media as their marketing strategy. Double sampling procedure is used to eliminate some of the unnecessary questions or invalid questions to provide more accurate answer for the hypothesis.
The researcher has provided 32 questions in the preliminary test which have been distributed to 30 respondents. The next step is to analyse the responses and eliminate the invalid indicators.
In the actual test, several adjustments in the questions or indicators have been made to assure that only the valid indicators are presented. There are 32 valid indicators based on the calculation on preliminary test and those presented in the test.
Partial Least Square (PLS) used as an analysis tool is a method to implement SEM (Susanti & Kuntadi, 2016) which works effectively with small sample size. Meanwhile larger sample size is distribution free which increases the precision of the estimation and accommodates measurement models that are both reflective and formative (Hair et al., 2017). The total number of 163 respondents exceeds the required minimum range of respondents, namely 30-100 respondents for the PLS SEM method (Chin, 2000in Zuhdi et al., 2016. The causal research explains the position of each variable that drives a causal relationship, where the independent variables (PU, PEOU, COM, FCO, COS) influence the dependent variable (IOB) by including the mediating variable (SMM). Moreover, causal research allows the researcher to test the proposed hypotheses partially and simultaneously.
The seven variables consist of five independent variables (PU, PEOU, COM, FCO, & COS), one dependent variable (IOB), and one mediating or intervening variable (SMM). Operationalization variable is performed to ensure the accuracy and coherency of question or statement in questionary, as seen in Table 1. Data are collected by distributing electronic questionnaires (Google Form) to MSME owners in Jakarta and Tangerang. The questionnaires contain statements regarding the research problem and using the Likert Scale: 1 = strongly disagree, 2 = disagree, 3 = slightly agree, 4 = agree, and 5 = strongly agree.
The research uses SmartPLS software to evaluate the measurement model results and test its indicators of latent constructs' validity and reliability. In addition, the structural and measurement model are evaluated to examine the influence of each construct or variable to another.  (Alduaij, 2019).

Compatibility (COM)
The extent of where innovation is compatible with user's prior experiences (Karahanna & Straub, 1999 Hair et al. (2017) describe structural model as the relationship between the exogenous variables and other latent variables. There are several stages in evaluating the inner model using R-squared (R²), collinearity statistics (VIF), and T-statistics for each path to test the significance between constructs. Hair et al. (2017) describe measurement model as the connection between latent variables with the indicator (manifest variable). The two types of outer model are formative indicator model and reflective indicator model. The outer model will measure the validity of the construct (particular concept) and make sure to measure the right concept. Meanwhile, the reliability is an instrument that measures the concept to make sure a stable and consistence measurement (Sekaran, 2016).
The research conducts a preliminary test to ensure the validity and reliability of the data. For Preliminary Convergent Validity Test, the result shows all variables including PU, PEOU, COM, FCO, COS, SMM, and IOB have value above 0,5 which means all variables are valid The next step is checking the value in outer loadings. Despite all valid variables, there might be several invalid indicators in the outer loadings. Indicator COM3, FCO5, FCO6, IOB2, PU5 are below 0,7. However, these indicators will not be eliminated due to the information they can provide for the research.
The preliminary test comes up with the Average Variance Extracted (AVE) value results. It is found that all indicators are valid (>0,5), which can be provided: 1) PU has a value of 0,673; 2) PEOU variable has a value of 0,608; 3) COM is 0,588; 4) FCO has a value of 0,639; 5) COS is 0,757; 6) SMM value is 0,772; and 7) IOB is 0,650. They are considered as valid since each of the variables has a value above 0,5 so that it meets the constraints of convergent validity. Therefore, each variable that has been tested for validity will be reused in the actual test.
For preliminary discriminant validity test, the discriminant validity of all variables can be identified using the Fornell-Larcker. Table 2 shows that each indicator does not have a high correlation with other indicators. Each indicator is higher than other indicator, and each value is the result of square root of AVE.
The reliabilities of the preliminary test are based on the composite reliability in SmartPLS software. The result shows that all indicators are reliable since their value are above 0,7, which are provided: 1) PU value is 0,909; 2) PEOU is 0,885; 3) CO value is 0,838; 4) FCO has a value of 0,913; 5) COS value is 0,925; 6) SMM is 0,910; and 7) IOB is 0,900. The summary of a preliminary test is shown in Table 3.

Convergent Validity Test
To determine the convergent validity of an indicator Overall, all the indicators AVE value are above 0,5. All indicators are valid.

Discriminant Validity Test
Using the Fornell-Larcker, it identifies the discriminant validity of all variables. It measures the correlation of one indicator to others.
The correlation of latent variables and each of its indicators are higher than the correlation with other latent variables. All latent variables are valid.

Reliability Test
It ensure the consistent measurements over time and with each item in the instrument.
The value of all variables composite reliability are above 0,7. Therefore, the instrument is reliable Source: The Results of Data Processing Using SmartPLS (2020)

III. RESULTS AND DISCUSSIONS
For the sample, the research takes 163 respondents of MSME owners in Jakarta and Tangerang who use social media as their marketing strategy. As provided in Table 4, the information about the respondents is divided into seven categories, namely: 1) gender, 2) age, 3) latest education, 4) domicile, 5) social media being used, 6) annual sale, and 7) type of business.
The analysis starts with outer model which calculates and declares the convergent validity, discriminant validity, and composite reliability of each indicator in the actual test, whether the indicators and data are relevant to the research or not. To determine the convergent validity of an indicator, the value of the Average Variance Extract (AVE) have to be above 0,5 and the value of factor loading should be above 0,7. The value of factor loading from each indicator in the research variables shows that there are several invalid indicators with values below 0,7 such as PU1, PU5, and COM3. However, these indicators are not omitted since they can be used as research information.
The result shows that all variables are valid because each AVE values are above 0,5. Therefore, it can be concluded that the every variables have good convergent value.
The reliability test measures the reliability of the data. The method used to test the reliability of an indicator is composite reliability. SmartPLS software can calculate the composite reliability. The criteria to determine the reliability of the data is to have value above 0,7. Result shows that all indicators values such as PU (0,851), PEOU (0,869), COM (0,852), FCO (0,912), COS (0,923), SMM (0,873), and IOB (0,892) are above 0,7 meaning all indicators are reliable.
For inner model test, R-squared test result shows the R-squared value of SMM is 0,458 and IOB R-squared value is 0,389, which are considered as moderate. Therefore, this concludes that SMM variable is moderately influenced by perceived usefulness, perceived ease of use, compatibility, facilitating conditions, and cost variable by 45,8% and impact on business variable is moderately influenced by SMM variable by 38,9%.
Multicollinearity test is to find out whether a correlation between interdependent variables in the regression model is found by looking at the Variance Inflation Factor (VIF) value which must be lower than 10. A good regression model should not have multicollinearity, so if the VIF value is higher than 10, it indicates the existent of multicollinearity in the indicator. Therefore, the VIF value should remain below 10. Table 5 shows the result of multicollinearity test. As shown in Table 5, all the VIF value from the independent variables are below 5. Therefore, the independent variables in the research are not correlated, which means that there is no occurrence of multicollinearity. Outer and inner model test summary results are provided in Table 6.
There are two types of hypothesis testing, namely direct effect hypothesis and indirect effect hypothesis. Table 8 shows that there are five indirect effect hypothesis testing, where SMM mediates the relationship between: 1) PU and IOB, 2) PEOU and IOB, 3) COM and IOB, 4) FCO and IOB, and 5) COS and IOB. Furthermore, there is one direct effect hypothesis testing towards IOB (Table 9). To test the hypothesis, the t-statistics uses t-table value (one-tail) which is 1,65 with significant level of 0,05. The t-table value will be used to determine whether the hypothesis is significant or not. Table 7 shows the original sample (O), sample mean (M), standard deviation (STDEV), t-statistics, and p-values of the processed data using SmartPLS. These numbers can determine whether the hypotheses are accepted or rejected. The result for 3 types of MSME in Indonesia is later presented in Table 10. The correlation of latent variables and each of its indicators are higher than the correlation with other latent variables.

Reliability Test
It ensures the consistent measurements over time and with each item in the instrument.
The value of all variables composite reliability are above 0,7.

R-Squared
It measures how much the dependent variable is influenced by the independent variable.
The value of R-squared are in the moderate category.
All independent variables are moderately affect the dependent variable.

Multi-collinearity Test
To find out whether or not a correlation between independent variables in the regression model is found by looking at the Variance Inflation Factor (VIF) value.
The VIF values of all independent variables are below 10.
The VIF values of all independent variables indicates that there is no multicollinearity among independent variables. The result of hypothesis testing H 1 shows that Perceived Usefulness (PU) has a positive impact on the MSMEs to adopt SMM, which is considered as a partial mediator in the relationship between PU and Impact on Business (IOB). The result is supported by Siamagka et al. (2015) who find that adoption of social media is significantly affected by organizational innovativeness and perceived usefulness. The use of social media as part of MSMEs business strategy gives them more access to conduct a market research and branding (Kim et al., 2013). Chatterjee and Kar (2020), and Davis (1989) support this hypothesis. Chatterjee and Kar (2020) find that PU impacts SMM since PU impacts performance, effectiveness, risks, and trust. Siamagka et al. (2015) have found that the adoption of social media enhances organizations competitiveness, improve cost-effectiveness, builds customer relationship, gives business exposure, and receives feedback from customers.
However, Table 10 shows that micro and medium enterprises have insignificant relationship. Isiyaku et al. (2018) find that the ability of the ICT user is affecting the perceived usefulness of technology where they specifically mention results Daramola, Yusuf, and Oyelekan (2015) stating that the teachers' ability for using ICT is low. Meanwhile the teacher self-efficacy can affect the perceived usefulness of ICT. Therefore, the result of perceived usefulness in Daramola et al. (2015) is insignificant. Even though the result of hypothesis 1 shows that SMM is a partial mediation, it is also important for businesses to include and use SMM in their strategy. Based on the answers of the questionnaire about PU, the average respondents strongly agree that social media is useful for business. It is believed that social media becomes a valuable tool for marketing which enhances the productivity of the business, helps better query management, and helps businesses to satisfy more customers.
Based on hypothesis testing H 2 , Perceived Ease of Use (PEOU) has a positive impact on the MSMEs   (2020) to adopt SMM. Table 10 shows that all categories have insignificant relationship between PEOU and impact on business. Setiawan, Setyohadi, and Pranowo (2018) have found that PEOU is not significant with their research findings. It is possibly because there are other technologies that are easier to use. However, the ease of use would have less or no impact on usage. Gefen and Straub (1997) find that there is an insignificant relationship between PEOU and technology about e-mail acceptance. Their research result turns to be the opposite direction of their hypothesis, that women is not significant with perceived ease of use in using e-mail. In addition, it is mentioned that men are proven to have the tendency and knowledge on how to use computers. Chatterjee and Kar (2020) point out the opposite findings that PEOU is significant due to its simplicity and self-efficacy. There is a linkage between PU and PEOU. Moreover, it shows that the average respondent agrees that it is easy to learn SMM, identify new customers using social media, find customer information with SMM, and advertise products on social media platforms. Most respondents strongly agrees that social media provides an easier way to identify customers' demand.
The result of hypothesis testing H 3 shows that compatibility (COM) has a positive impact on the MSMEs to adopt SMM since all categories are significant, as seen in Table 10. Ainin et al. (2015) state that compatibility is a significant factor in the adoption of technology. For instance, the use of Facebook is compatible with the internet connection and technology which makes it simple and easy to use.
Compatibility is one of the important factor in adoption of innovation. The company is likely to consider adopting the new technology when technology is proved to be compatible with the business systems. Some companies also indicate their intention of adopting technology if it is compatible with their values and beliefs. Nevertheless, some research find the insignificant impact due to the incompatibility between the company's system and the technology (Ainin et al., 2015). Therefore, SMM is considered as partial mediation in this hypothesis since they would have to put more effort to expand their business without social media. The average respondents strongly agree that they use social media for different purposes-regularly for business and marketing purposes, and that the organization have provided support for training on social media.
Based on hypothesis testing H 4 , SMM negatively mediates the relationship between facilitating conditions and impact on business. Table 10 shows that all enterprises have insignificant relationship between facilitating conditions and impact on business. Sichone et al. (2017) find that there is an insignificant relationship between facilitating conditions and e-filling of tax return as the participants had already started to file their tax return electronically. Therefore, the issuance of technology device such computer and other support devices seem unnecessary. Chatterjee & Kar (2020) find a positive and significant relationship between facilitating conditions and SMM. However, it is mentioned that their hypothesis testing results contradict with previous studies since not many businesses presumably support the use of SMM in MSMEs business activities, so company may not encourage employees to use SMM. The external problem may be due to the low bandwidth internet connectivity. Therefore, the business cannot effectively use social media as their marketing tool. In addition, Hung & Lai (2015) examine the relationship between aggregate rating, customer comment (Facebook likes) and purchase intentions. It is found that participants with high purchase intention choose to see the high numbers of Facebook likes than the aggregate rating. In this research, aggregate rating can be considered as the facilitating conditions variable, which might be significant if it is consistent with customers' comments. The average respondents agree that: 1) they have adequate infrastructure for using social media, 2) they use social media to promote their business, 3) they have enough trained manpower dealing with SMM, and 4) all of their employees are provided with training to use SMM. The respondents highly agree that they adequately invest for SMM and have inhouse training facility to learn different aspects of social media.
Based on H 5 result, cost (COS) has a positive impact on the MSMEs to use SMM. Based on Table  10, only medium enterprises have insignificant result while others have significant relationship between cost and impact on business. Chatterjee & Kar (2020) explain that SMEs in India have limited resources and people are cautious if they have to pay some extra cost in using SMM. Therefore, they point out the SMM have positive impact in the relationship between cost and impact on business. The average respondents agree that: 1) SMM has been reducing their cost in dealing with customer enquiries, 2) respondents are able to reduce the cost of identify new customers using social media, 3) customer awareness and training cost have been diminished due to SMM, and 4) advertising and promotion cost have decreased since using SMM.
The result of H 6 test shows that SMM has a positive impact and significant influence on impact on business. Table 10 shows that all enterprises have significant relationship between SMM and impact on business. Saravanakumar and Lakshmi (2012) mention that that the social media can be expected to impact on businesses now. Many big brands use social media to promote their products and services, so they can show their strong existence in the society. Businesses can promote their products by giving discounts or other participatory promotions to increase customers' excitement to purchase online and inform their relatives. SMM has many advantages, one of which is free or low-cost that in turn will increase the profit margin of the business. Descriptive results show that most respondents strongly agree with the SMM concept and believe that: 1) SMM is helpful to advertise their products and services, 2) the increase number of competitors drives respondents to use the SMM, and 3) the usage of SMM technique is good for business.

IV. CONCLUSIONS
Social media marketing positively mediates the relationship of perceived usefulness, perceived ease of use, compatibility, and cost towards impact on business. In contrast, social media marketing negatively mediates the relationship between facilitating conditions and impact on business. Finally, social media marketing has a positive and direct effects towards impact on business.
The research implies that the adoption of social media marketing by MSMEs is encouraged by such factors as usefulnes, ease of use, compatibility and cost. Adoption of social media marketing enables the company to expose its business more widely, receive feedback from customers, and reduce expenses especially in promotion cost. MSMEs should facilitate the employees with technical infrastructures and Social Media Marketing training, which is essential to increase productivity and performance.
The research has contributed to enrich existing research related to MSME in Indonesia, especially in convincing MSME owners to digitize their business processes, especially on marketing and service processes.
Variables in the research are dominated by internal views related to attitude or perception of MSME owners toward SMM. Therefore, it is suggested that future research include more variables that capture external views such as MSME's competitiveness, marketing effectiveness and branding. The additional variables will complement this research from external views-competitor and customers.