“E-marketing strategy: to improve customer preference for local brand over foreign brand in the era of a developing country”

This research will determine the factors that influence customer preference for local brands over foreign brands in the era of a developing country. This is a quantitative research design. Questionnaires have been used to collect the data on product standards, social media engagement, local positive testimonials, brand application and brand trust. The sample size was 80 – randomly selected. Organizations should improve quality, design, compatibility, consistency, attitude and online appearance on YouTube, Facebook, Instagram, and LinkedIn. Celebrity testimonials have more influence on consumers than a non-celebrity testimonial. The research is original in that it was conducted from scratch. Moreover, the researchers sought help from previous research in the field to define technical terms, of which have been cited accordingly.


INTRODUCTION
The marketing environment is rapidly shifting from traditional forms of marketing to electronic marketing (e-marketing). E-marketing is the process of designing and executing the pricing of services and products, promotion, distribution, and conception in an automated networking environment such as the world wide web, the internet, email, and wireless media to satisfy the consumer demands and facilitate the exchanges (Walrave & Heirman, 2012). Types of e-marketing include content marketing, email marketing, blogging, affiliate marketing, video marketing, and article marketing. Customer preferences are the likes, dislikes, inclinations, motivations, and expectations that influence the decision of a customer of whether or not they should purchase a product or service (Safitri, Do, & Irawanto, 2017). Consumer preferences can vary from product to product, and the ingredients, components, or location of manufacture of each product can affect the preference. There are consumers who like locally manufactured products, and there are those who like foreign products. This paper will use quantitative analysis to determine the factors that influence the customer preference for local brands over foreign brands in the era of a developing country.

SIGNIFICANCE OF THE STUDY
Customer preference is a strong determinant of whether an individual will buy a product or not. The findings will be significant to society and the corporate sector. The findings will be significant to the society in that members of the society will be conscious of product designs, quality, consistency, safety, and compatibility. This research will be significant to the corporate sector, because they are the ones struggling to create awareness about local products over foreign products. The research will give them ideas on the best ways to persuade the customers to purchase local products.

Research model and hypotheses
This study has four independent variables, one mediating variable and one dependent variable. Independent variables -often represented by xare the variables whose changes do not depend on the variation of another variable. Mediating variables are variables that seek to explain or identify the relationship between the independent variable and the dependent variable. Dependent variables -often represented as y -are the variables whose variation depends on another variable. The independent variables are improved product standards, local social media, positive testimoni-als, and brand application. The mediating variable is e-brand trust, and the dependent variable is a preference for a local brand.
Globalization has been considered as the inevitable phenomenon, because it is connecting the entire world to become one market, which can be also said as a global village. While the world is becoming a single market, various companies from different regions are expanding their businesses in different countries and become successful in attracting the local customers towards their products and services. In addition, because of getting closer and expanding the business operations beyond the borders, different brands are getting popular in different contexts. The brand has been consistently reviewed and redefined in the marketing literature owing to which there are different definitions for the brand. In 1960, American Marketing Association (AMA) defined the brand as a name, symbol, sign, term or design or combination of all of them that intends to determine products and services of a particular seller and differentiates one from the others.
According to Amaldoss and Jain (2005), customers with high requirements for uniqueness tend towards adopting new products and brands more quickly as compared to that with low requirement for uniqueness. This shows that the local products   are getting ignored by most of the customers because of either not showing some uniqueness or being unable to market those products in the local markets. Moreover, there are several reasons behind customer's preferences towards foreign brands as compared to local brands. Some of these can be the price of the brand, brand image, brand name, availability, consumers' ethnocentrism, advertising campaigns and much more.
According to Karoui and Khemakhem (2019), customers of developing countries are getting highly preoccupied with foreign brands as compared to the local brands. This situation in the developing markets has raised a question over the behavior of the customers; Are local brands struggling to create awareness about local products over foreign products?
Previous researches have shown that as compared to the developing countries, developed countries are more willing to purchase local products rather than foreign brands. On the other hand, it has been evident in a few studies that customers from developing or less developed nations have been preferring imported products over local ones. Moreover, it has been anticipated that customers of less developed or developing countries only purchase the local products when they have high emotional value towards them. On the other hand, customers do not seem to perceive local brands as offering a positive brand image. This shows that the customers who have attitudes towards foreign brands do not perceive local brands as taking emotional value. Overall, it shows that the customers that have higher emotional value towards local brands will purchase the products and services from local brands, whereas customers having a high affinity or emotional value towards foreign brands will purchase only those products. This shows that emotional value also plays a significant role in improving the customer preference for the local brand over foreign brands in less developed and developing countries.
The study of Karoui and Khemakhem (2019) shows that in the developing countries, the government is required to engage in advertising campaigns where it can effectively promote domestic or local products. For example, it had happened in Tunisia at the end of the 1990s. In Tunisia, the government started a movement towards building a patriotic feeling among citizens and convince them to purchase nationally manufactured goods. Moreover, several other major factors play a significant role in the e-marketing of local products rather than foreign products. Some of these factors are product standards, social media, local testimonials, brand application, and e-brand trust.

Product standards
The first independent variable is improved product standards. Product standards are the standards relating to the manufacturing and design of consumer products to ensure that they in no way represent any hazards or harm to the consumers. In other words, product standards have something to do with the features or characteristics of services and goods with respect to fitness for purpose, safety, and quality. An example of a product standard in the United States is the restrictions put in place by the United States Food and Drug Administration (FDA) (Mcgrath, 2018). In the United Kingdom, products and standards are regulated by the Office for Product Safety and Standards (Mccormick, 2019). In Australia, products standards are controlled by the Australian Competition and Consumer Commission (ACCC) (Raghuvanshi, 2016). The World Trade Organization (WTO) is the global authority that dictates product standards.
Product standards do not just involve matters pertaining to the health and safety of consumers. They also involve usability of the product/service, reliability, quality, compatibility, transparency of product labelling and information, protection from misleading and false claims, fair competition, processing and handling claims, consistency in service delivery and whether the product is suitable for vulnerable populations such as persons with disability, the elderly and children (Janiszewska & Insch, 2012). Product standards also address contemporary issues such as customer satisfaction management, sustainable consumption, design accessibility, privacy and personal data issues, social security and responsibility, and ethical practices. Product standards are the basic building blocks for product and process development leading to universal acceptance (Laher, 2010).
Customers -especially in a developing countryare usually in a dilemma of choosing between a local brand and a foreign brand because of the difference in product standards. Consumers would prefer a product that would guarantee the quality, safety, ease of use, and compatibility. In the developing world, consumers are beginning to become conscious of the standards of the goods and services they buy (Charry, Coussement, Demoulin, & Heuvinck, 2016). This trend has forced companies to improve the design, quality, safety, consistency and compatibility of their products lest they fall out of business.
H 1 : Improving product standards increases customer preference for local brands in the era of developing countries.

Social media
Social media refers to applications and websites that enable users to engage in social networking or to share and create the content. Examples of social media sites include Facebook, YouTube, Instagram, Twitter, Google Plus, WhatsApp, and so on. Social media has overtaken mainstream media and become the primary tool for marketing and promotions. All companies are rushing to grab a fan base on social media before it is too late ( Twitter is one of the social networks with the fewest users -only 330 million monthly -but with the highest influence. Influence can flow from Twitter into other social networks. 66% of American companies with over 100 employees use Twitter for promotion and marketing (Watfa, Najafi, & Bakkar, 2013). The site is mostly used by individuals between the age of 18 and 29. Even though Instagram is popularly known as a photo-sharing network, it is still influential in the marketing and promotion world, because the use of pictures is an excellent way of marketing. One billion people use Instagram monthly with 88% coming from the United States, and 71% of them are under the age of 35. YouTube is an excellent platform for video marketing. The site has over 1.95 billion users monthly. Every day, the platform reaches more 18 to 49-year-olds than any other cable television program or broadcast. LinkedIn is the perfect platform for interacting with corporate professionals. However, with 106 million users monthly, it is not only a place to interact with corporate professionals, but also a good platform for marketing (Bowie, Paraskevas, & Mariussen, 2014).
Local social media promotional engagement can improve the preference for local brands by increasing brand reputation, brand awareness, brand loyalty, and customer interaction. Companies can improve their reputation by instantly responding to developments in the industry. Local social media can improve brand awareness by increasing the visibility of the brand and grabbing the attention of potential customers. Through social media, companies can interact with customers by providing feedback and responding to negative feedback. Through building relationships on social media, companies can increase advocacy and brand loyalty (Bowie, Paraskevas, & Mariussen, 2014).
H 2 : Social media promotional engagement can improve the preference for local brands in the era of a developing country.

Local testimonials
In marketing, advertising and promotion, a testimonial is a written or spoken statement or recommendation from a satisfied customer or a celebrity affirming the quality, value, and performance of a product or service. Product and service testimoni-als are one of the most powerful marketing tools. Product testimonials can be divided into two: ordinary testimonials and celebrity endorsements. Ordinary testimonials are testimonials from consumers who are not famous, and celebrity endorsements are when celebrities (well-known and influential individuals) issue a testimony about a product or service. Using celebrities generates awareness, relevance, and appeal to a brand image. Local celebrities can influence tens of thousands or even millions of their followers to purchase a product. For example, Ryan Seacrest -an influential American TV broadcaster -is a brand ambassador for Ford products.
Testimonials are usually effective and efficient, because it only takes someone's belief that the testimonial is true for consumers to purchase the given product or service. Testimonials reached an all-time high with the advent of the internet and social media. It can also be in the form of product reviews; the effect is the same. However, the effectiveness and efficiency of testimonials depend on the level of influence of the product with respect to product features and quality. Individuals -especially young people -with high susceptibility to norm influence (SNI) put much emphasis on testimonials. This means that the younger generations are the target audience of testimonials (Drechsler, Natter, & Leeflang, 2012).
Local online positive feedback about the brand improves consumer preference for local brands in the era of a developing country.

Brand application
Brand application is the idea of using the company's brand on all customer touch points and marketing materials. It is required that a company's brand be the foundation of everything a company does as part of its marketing efforts and plans. Facets of brand application include the company's logo, website, business forms, letterhead, blogs, and so on. Also known as brand implementation, it is basically the consistent application and physical representation of brand identity across verbal and visual media. This can include uniforms, signage, interior designs, liveries, and branded merchandise in visual terms. Brand application is a very important part of the branding cycle, and companies need to activate it during the brand development and design stage. Brand application is a continuous process that requires manipulating the brand presence and image despite the company structure and changes in the market. Brand application is likely to influence a customer's preference towards a product (Drechsler, Natter, & Leeflang, 2012).
H 4 : An application by the name of the brand influences the consumers' preferences for local brands in the era of a developing country.

E-brand trust
E-branding is a marketing term used to refer to the sum total of an organization's mission, vision, attitude, values, appearance and personality, shown to the online audience. E-branding is the next trend in online development. An e-brand is a company's online digital representation and what the company stands for. E-branding offers a channel of direct communication between brand owners and customers, increases brand loyalty and sales, strengthens brand identity, values and voice, and delivers service experience (Field, 2013). A vibrant e-brand can enable a company to retain and attract top talent, enhance the overall organizational brand, and reduce employee turnover.
A good e-brand also enables a company to survive its competitors, build loyalty and familiarity with customers, gain a reputation among them, while at the same time acting as the digital asset of the company. An e-brand does not only popularize or advertise a company's service or product; it also expands a company's customer base and builds a strong relationship by communicating with customers.

METHODOLOGY
This section will explain the data collection process, the population, instrumentation, the variables and measurements. The data were collected through the use of questionnaires. The researchers chose a population size of 100 from the general population. The population was randomly selected, and a sample was obtained. The sample size was found to be 80 at 95% confidence level and 5% margin of error. Therefore, the researchers proceeded with 80 participants. The researchers had prepared 100 questionnaires, even though the sample size emerged to be 80. The extra 20 questionnaires catered for errors made by the respondents when answering the questions.
Four variables were studied under the independent variables section: product standards, social media, testimonials, and brand application. E-brand trust was the only mediation variable, i.e., acting as the bridge between the independent variables and the dependent variables. Preference for the local brand was the only dependent variable. In each variable, the researchers asked the participants more than four questions. The questions were answered on a 5-point Likert scale with 1 -strongly agree, 2agree, 3 -don't know (neither agree nor disagree), 4 -disagree, and 5 -strongly disagree.
Before asking the main questions pertaining to the study, the researchers asked for the demographic information of the participants, including gender, age, occupation, level of education, and level of income. Table 1 shows the respondents' profile.

Measurements
This section will display all the variables, type, and scale of measurement. Table 2 shows the information.

Ethical considerations
Ethical considerations are very important in descriptive and correlation design. Ethical considerations are the global standards that determine what is right and what is wrong in a research study. Ethical considerations that the researchers followed include free from harm, informed consent, confidentiality and anonymity, the dignity of the participants, exaggerations, and acknowledgement of previous research studies (Kumar, 2017;Faisal, 2016). The researchers assured and ensured that the participants were no in way harmed whether psychologically or physically. The researchers asked for the consent of the participants before engaging them in the questionnaire. This ensured that the respondents participated intelligently, willingly, knowingly, and voluntarily and in and clear way. The researchers communicated to them the purpose of the experiment and the source of funding (Kotler & Armstrong, 2013).
The researchers ensured the confidentiality and anonymity by not including a space for the name of the participant and their contact information in the questionnaire. By doing this, the information in the questionnaire can never be traced to the source. By assuring the respondents' confidentiality, the participants felt obliged not to withhold the information (Chaudhuri & Holbrook, 2012). The dignity of vulnerable groups like the physically disabled was observed to the latter. The researchers also avoided deceptions, exaggerations, misleading information, and misinterpretation of facts regarding the objectives of the experiment and also assured transparency and honesty. Lastly, any information obtained from previous research studies has been cited accordingly.

DATA ANALYSIS AND RESULTS
The information in the questionnaires was coded into IBM SPSS version 23, and the data analysis of this study contained five parts: exploratory study, confirmatory study (measurement model), testing the structural model, and mediation.

Exploratory study
This subsection will display and explain mean, standard deviation, skewness, kurtosis, and Cronbach alpha (reliability). Table 3 shows the descriptive statistics of the collected data, which is majorly used to show the measure of central tendency and measure of dispersion. From Table 3 we can see that the measure of central tendency of collected data is shown with the help of mean and measure of dispersion is shown with the help of standard deviation. The output of in the descriptive statistics table shows that the mean of quality, design, safety, consistency, compatibility, Facebook, YouTube, Instagram, Twitter, celebrity testimonials, mission, attitude, and appearance is less than two and greater than one. Moreover, the mean of LinkedIn, Consumer Testimonials, Logo, Website, Blog, Business Forms, and Online personality is in between two and three. However, the mean of Letter Head is only greater than three.
On the other hand, the standard deviation of quality, design, safety, consistency, compatibility, Facebook, YouTube, Instagram, business forms, mission, vision, attitude, and appearance is less than one, which shows that the responses between the respondents are not highly dispersed in terms of these variables. Moreover, the standard deviation of Twitter, LinkedIn, consumer testimonials, celebrity testimonials, logo, website, blog, letterhead, and online personality is more than one, which shows that the responses of participants are varied highly. Table 3 shows the deviations from the means range from 0.5 to 1. Skewness indicates where the bulk of the data lies -a negative skewness in an indication that the bulk of the data is on the left. A zero skewness indicates a perfect symmetry (normal distribution). Positive skewness indicates that the bulk of the data is on the right. Table 3 shows that quality, consistency, YouTube, Twitter, LinkedIn, celebrity testimonials, website, business forms, letterhead, mission, vision, attitude, appearance, and online personality are positively skewed. Design, safety, compatibility, Facebook, Instagram, consumer testimonials, logo, blog, and letterhead are negatively skewed.
Kurtosis can be explained in terms of the peak of a curve. Larger values indicate a sharp, high peak and smaller values indicate a lesser distinct peak. In other words, higher kurtosis is an indicator that there are extreme standard deviations from the meanwhile, a lesser distinct kurtosis indicates modestly-sized deviations (Sharma & Baoku, 2012). A normal distribution has a kurtosis of 3 and is referred to -in statistical terms -as mesokurtic. A distribution with kurtosis less than 3 is referred to as platykurtic, and its tails are thinner and shorter, and its peak broader and lower. A distribution with a kurtosis larger than 3 -referred to as leptokurtic -has a fatter and longer tail and its central peak is sharper and higher. Table 3 shows that all the variables are platykurtic (Hundekar, 2010).
Before a Cronbach's alpha test of reliability was conducted, the individual sub-variables were computed in the five initial variables, that is, product standards, social media, testimonials, brand application and e-brand trust. The table below shows the Cronbach's alpha results. According to the table above, Cronbach's alpha is 0.901, which indicates a high level of internal consistency for our scale with this specific sample. Based on standardized items, the Cronbach's Alpha is 0.936, which indicates an even higher level of internal consistency.
The table above shows the inter-item correlation matrix. In other words, it shows the correlation between the variables. A negative value indicates that there is a negative correlation (an increase in one variable leads to a decrease in another variable) (Bowie, Paraskevas, & Mariussen, 2014). A positive value means that an increase in one variable leads to a decrease in another variable. All the correlation values in our correlation matrix are positive, which means that an increase in one variable leads to an increase in the corresponding variable. However, the strength of the correlation differs. The closer the correlation coefficient is to 1, the stronger the positive correlation. The closer the correlation coefficient is to -1, the stronger the negative correlation (Bowie, Paraskevas, & Mariussen, 2014). This indicates that a majority of the correlations are strong correlation apart from a few correlations with medium strength (testimonials-product standards, testimonials-social media, and product standards-testimonials).
From Table 5, we can see that the correlation coefficient of social media is the highest among all the variables. Its correlation coefficient is near to one, which shows that it is strongly correlated to the brand application as compared to other variables. Moreover, the correlation coefficient of brand application is also higher than 0.75, which shows that brand application is also strongly correlated with the product standards. In addition, the correlation coefficient of testimonial and e-branding is above 0.05, which shows that these variables are also positively correlated with the brand application, but not as strong as the other two variables.

Confirmatory study (measurement model)
This subsection will show the exploratory factor analysis, fit indices, and discriminant analysis.

Exploratory factor analysis
The exploratory factor analysis (EFA) is a statistical technique used to reduce a data set to a set of smaller summary variables. EFA identifies the na-ture of the correlation between the participant and the variable (Chaffey & Smith, 2008). The following is the factor analysis for this study.
The first four factors have an eigenvalue of greater than 1. This means that four factors can be used to explain a majority of the variability in the dataset. The percentage of variability explained by factor 1 is 57.2%. The percentage of variability explained by factor 4 is 4.2%. The scree plot in Figure 1 shows that the first four points account for most of the variability, because the difference in variability between component 1 and component 4 is 53%.
The scree plot in Figure 2 helped the researcher to determine the number of factors to retain. It can be noticed that the curve starts to flatten at point 4. This means that only four factors can be retained. In other words, the components that have an eigenvalue equal to or greater than 1 are considered strong factors and those with eigenvalues less than 1 are considered weak factors.

Fit indices
Chi-square goodness of fit test is used to determine the fit indices. This is a non-parametric test used to determine how the observed value of a given scenario significantly differs from the expected values (Singh, 2017;Amelia, 2014). This test divides data into intervals, and the numbers of points that fall within the intervals are compared with values of the individual intervals. Table 7 provides the results of the Chi-square goodness of fit test. It can be seen from the table that the test performed is statistically significant, because all the asymptotic significance levels are equal to or less than 0.05. When the p-values are less than the level of significance, the null hypothesis is rejected, and a conclusion is made that there are significant differences between observed and expected values

Discriminant analysis
A discriminant analysis test is used to find out the variables that discriminate between two or more groups that occur naturally. Table 8 shows the test of equality of groups means (Beyer, 2018). The output shows that Wilks' lambda is not statistically significant for each of the four predictor variables. This is because all the levels of significance are greater than 0.05. Therefore, the null hypothesis is rejected.  Note: a -0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 13.3, b -0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 10.0, c -0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 8.9, d -0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 8.0.

Structural model
Structural modelling is a statistical technique used to define and analyze the structural relationships. It also analyzes the relationship between latent construct variables and measured variables. Figure 3 shows a structural equation model.
A one sample t-test was carried out to determine whether there was statistically significant between the four independent variables (product standards, social media, testimonials, brand application, and e-branding) and customer preference at 95% confidence interval. There was extreme significance in all the cases, meaning that all the hypotheses are upheld. Improving product standards increases customer preference for local brands in the era of developing countries, t(79) = 29.799, p = .000. Social media promotional engagement can improve the preference for local brands in the era of a developing country, t(79) = -26.280, p = .000.
Local online positive feedback about the brand improves consumer preference for local brands in the era of a developing country, t(79) = 21.061, p = .000. Application by the name of the brand influences the consumers' preferences for local brands in the era of a developing country, t(79) = 43.913, p = .000. A trusted e-brand improves consumer's preference for local brands in the era of a developing country, t(79) = -26.931, p = .000.

Mediation analysis
A mediation model is a model that seeks to identify and explain the process or mechanism that underlies an observed correlation between a dependent and an independent variable by including a third variable called a mediating variable. In this case, the mediating variable is e-brand trust.
All the correlations are significant at 0.01 level. This means that there is a mediation effect of e-branding trust on customer preference for locally manufactured goods over foreign imported goods. Table 11 is a linear regression for further proof.    Note: a -dependent variable, e-branding.
The fact that two variables suggest that there is no statistically significant difference means that there is a mediation going on.

Practical
The findings of this study showed 55% of participants strongly agreed and 45% agree that quality influences preference. Overall, it shows that the maximum population believes that the preferences regarding purchasing the products highly depend on the product standards, which includes the quality of products. Moreover, 40% of the participants strongly agreed and 60% agreed that design influences preference regarding purchasing the products. It means that most of the people purchase the products by observing the product design rather than the region of the brand that whether it is the local or foreign product.
In addition, the result from the study also shows that 44 percent of people strongly agreed and 56 percent agreed that safety needs to be given as a preference while purchasing the products. It means that most of the population prefers the safety of the products before purchasing. Similarly, 56% strongly agreed and 44% agreed that consistency influences preference regarding purchasing any product or services.
Moreover, the study also revealed that the participants agreed in one way or another with compatibility, Facebook, YouTube, and Instagram. Only 16% disagreed that Twitter influences customer preference. 59% agreed in one way or another that LinkedIn affects consumer preferences. Facebook, YouTube, and Instagram have a high influence on the population in terms of e-marketing. On the other hand, Twitter is considered as having a low influence on people in terms of building customer preferences. Overall, it shows that e-marketing through Facebook, Instagram, and YouTube is highly influential as compared to Twitter. The reason behind it can be a lower amount of content on Twitter for in- Note: ** Correlation is significant at the 0.01 level (2-tailed).
fluencing the customers' preferences towards local brands. The content quantity on Facebook, YouTube and Instagram can be higher as compared to Twitter due to which customer's preferences can be easily influenced.
Similarly, LinkedIn also does not restrict the quantity of data to be shared for marketing purposes due to which it has also influenced a higher number of people. Moreover, most participants disagreed that consumer testimonials influence the customer preference. However, most of them agreed that celebrity testimonials have a significant effect on consumer preference.

Implication
Most respondents either disagreed or are undecided whether the logo is important. 73% agree that a company's website plays a major role; however, most of them disagreed that a company's blog is important. 67.5% said that business forms are vital, but the letterhead is not. The company's mission, vision, attitude, appearance and online personality play a huge role in ensuring that consumers choose local goods over foreign products. There is a strong relationship between the individual variables and also between the mediation variable and the individual variables.

CONCLUSION AND FUTURE STUDIES
A company should improve the quality of their goods because when they do so, the consumers will not hesitate to purchase local products. Another thing a company should concentrate on include design, consistency and compatibility. It can also be concluded that the best social media sites to promote and advertise products are Facebook, YouTube, Instagram, and LinkedIn. Twitter has fewer users. Companies should use celebrity testimonials rather than consumer testimonials, because celebrities have high convincing power. An organization should also put much emphasis on websites and business forms and little emphasis on blogs, letterheads, and logos. A strong mission, vision, attitude, appearance and online personality is required if a company wants to convince consumers to buy local products. It is recommended that future studies concentrate on the effect of online marketing on rural areas that have limited internet coverage.