Keywords

1 Introduction

By leveraging the Internet, many firms are able to introduce and sell their products to customers all over the world through their websites. Global e-commerce is increasingly important for not only firms but also consumers. The sales of global B2C e-commerce is estimated around $1.5 trillion and accounts for nearly 6.7 % of worldwide retail in 2015 (Rueter 2014). Correspondingly, more and more consumers are engaging in global e-commerce. According to the reports (Miglani 2012; Nielsen 2008), the world online users will reach 3.5 billion by 2017 and more than 85 % of the world’s online population have purchased online in 2008.

There are two general ideas for designing websites for different countries: standardization and localization. First, website standardization refers to using same website content with only language translation for every country (Singh and Pereira 2005). In contrast, website localization refers to customizing a website for a specific country. Website localization makes a website seem natural or “local” to their users (Singh et al. 2009). Emerging studies have investigated the effectiveness and importance of website localization (Chen et al. 2009; Singh et al. 2004; Tixier 2005; Vyncke and Brengman 2010) and we think that website localization is significant trend for global e-commerce and more research of website localization is necessary for successful global e-commerce.

Website localization has received attention from researchers. Although previous international marketing literature also has discussed adaptation or localization strategies, they focus on marketing mix(Solberg 2000), resource allocation (Szymanski et al. 1993), environmental and market determinants (Yip 1997), advertising and communication (Solberg 2002) and performance measurement (Lages et al. 2008). Some of studies on website localization literature have focused on establishing website localization frameworks and validating them through comparing and analyzing different countries’ websites localization (Baack and Singh 2007; Gibb and Matthaiakis 2007; Singh and Matsuo 2004; Singh et al. 2009). Moreover, some of studies have explored the related factors which influence website localization. The factors studied include immigrants or minorities’ acculturation level (Singh et al. 2008), culture, geography and infrastructure (Shneor 2012), managerial attitudes (Singh et al. 2010). There are also a few studies that identify and discuss several localization elements for website localization practice, like symbols and graphics, color preferences, links, maps, search functions, page layout, language, and content (Cyr et al. 2005), presentation of human image and environment types (Singer et al. 2007), access to product, logical presentation of product information, and information professional design (Cyr et al. 2005).

Although some related factors have been explored in previous literature, other exogenous factors, such as history length, online transaction availability and product types, have not been discussed. These factors provide a way to predict the usage of website localization strategies and decide the degree of website localization. For example, the longer history the foreign companies operate business in local countries, the more knowledge and motivation the foreign companies may have to localize their websites. Therefore, studying these factors into more detail could further our understanding of website localization. Besides, compared with other exogenous factors, information about history length, online transaction availability and product types is easier to access through public reports and foreign companies’ websites. Thus, the objectives of this study are:

What exogenous factors influence website localization? How?

To achieve the objectives of this study, we structure the rest of this paper as follows. In the second section, we discuss the theoretical background and develop the hypotheses about the relationships among website localization strategies, website localization degree and their exogenous factors. In the third section, we empirically test the hypotheses through a content analysis. In the fourth section, we report the analysis results. In the fifth section, we discuss the implication, limitation and future research of this study.

2 Theoretical Background and Hypothesis Development

2.1 Website Localization Strategies

Previous literature discusses the specific localization strategies mainly based on a culture framework (Baack and Singh 2007; Singh et al. 2008; Singh and Matsuo 2004). The website localization strategies are proposed according to the definitions of different cultural dimensions like Masculinity for quizzes and games or Uncertainty avoidance for customer service (Singh and Matsuo 2004). The studied cultural dimensions included: individualism–collectivism, uncertainty avoidance, power distance, masculinity- femininity (Hofstede 1980), high/low context (Hall 1976), intellectual autonomy, hierarchy and affective autonomy (Schwartz 1994). Although several strategies have been put forward based on cultural frameworks, very few studies have actually tested the effectiveness of strategies.

Wu et al. (2015) also identified several website localization strategies to localize foreign companies’ websites based on social identity perspective. In line with the social identity perspective, website localization strategies can localize foreign companies’ websites through creating similarities between foreign companies’ websites and local consumers or through providing information about foreign companies’ beneficial actions towards local consumers or their communities. We do further investigation based on Wu et al.’s (2015) website localization strategies. Given website localization strategies (e.g., Local Symbol and Corporate Social Responsibility) involve local group’s features or attributes to make the websites be similar to the local group, or convey foreign companies’ kindness and benefit to local group through the websites, they could increase websites’ localization degree. Therefore, we can hypothesize as:

H1: Website localization strategies associate with website localization degree positively.

2.2 Exogenous Factors of Website Localization

Global marketing literature revealed that companies which produced the different types of products localized their businesses differently. The literature suggested that industrial and high-tech products were more appropriate for standardization rather than consumer products because industrial and high-tech products tended to meet universal needs and consumers were more rational in purchasing such products. Compared with industrial and high-tech products, consumer products are more suitable for localization because they appeal to consumers’ hard-to-change tastes, habits, cultures and customs and foreign companies should adapt their marketing mix to be consistent with local consumers’ characteristics (Cavusgil et al. 1993; Jain 1989). Similar to the relationship between consumer and industrial products, non-durable products require higher adaptation rather than durable products in consumer products (Malhotra 1993). Given that companies which produce different product types localize their marketing activities differently and website localization strategies are likely to help localize foreign companies’ website, we can hypothesize that:

  • H2a: Product types associate with website localization degree positively.

  • H2b: Product types associate with usage of website localization strategies positively.

In e-commerce era, many companies do the transaction through their websites and trust is a widely accepted essential ingredient for developing successful business relationship between companies and consumers (Gefen et al. 2003). Previous studies have revealed consumers’ willingness to accept vulnerability if the consumers trust the website and its vendor. The acceptance of vulnerability facilitates consumers’ engagement in the online transaction (Mayer et al. 1995). According to Kramer et al. (1996), trust can be built on shared features or common values between consumers and the website vendors in e-commerce context. Meanwhile, localized website makes consumers feel that the foreign companies’ website is naturalized to them and helps to identify foreign companies and their websites as local ones. In other words, website localization can develop and establish commonalities between local consumers and website vendors (Singh et al. 2009). Therefore, website localization strategies and high website localization degree can promote trust and improve the usage of online transaction. Thus, we can hypothesize that:

  • H3a: Online transaction availability associates with website localization degree positively.

  • H3b: Online transaction availability associates with website localization strategies positively.

Strategic choice perspective suggests organizations respond to the external environment changes by adapting their strategies (Chaffee 1985). Such adaptation seeks for a balance between organizations’ behavior systems and their external environment (Alderson 1965). Boeker (1989) studied the effect of organization history on organization’s strategies’ change and uncovered its significant role. To be specific, the external conditions or events in an organization history limit or encourage organization’s strategy choice. As an important channel of companies’ global business, the choice of website design method is an important strategic decision. Given history impacts organizations’ strategy choice, more adaptation in their websites will be done and more website localization strategies will be used in line with longer history length. Therefore, we can hypothesize that:

  • H4a: History length associates with website localization degree positively.

  • H4b: History length associates with website localization strategies positively.

3 Methodology

In this study, we test the relationships among website localization strategies, website localization degree and their three exogenous factors. Website localization strategies are from Wu et al. (2015) and the designed website features to localize foreign companies’ websites, while website localization degree is the extent to which local people perceive the localization of foreign companies’ websites. Towards the three exogenous factors, product types are about main products provided by foreign companies to consumers and contain two types: consumer product and industrial product, online transaction availability is about whether foreign companies’ websites provide the channel to do online business, and history length is the length foreign companies do business in local countries. The sample of this study and the ways to measure the variables are described as follows.

3.1 Sample

For this study, we choose foreign companies’ (Fortune 1000) websites in China as the object of our research. The local people are Chinese people. To do the analysis, the first step is to sort out companies that have Chinese websites/pages and companies that do not have Chinese websites/pages. Among the Fortune 1000 American Companies, 729 companies do not have specific websites/pages for the Chinese market and only 271 companies have Chinese websites/pages. Thus, the final sample for this study is the 271 companies that have Chinese websites/pages.

3.2 Measurement

To measure each websites’ localization degree, several measurement frameworks are proposed in previous literature. Among the measurement frameworks established in previous studies, Singh (2012) comprehensively covered and tested the dimensions of website localization. To apply this framework in our context, we adapted it according to our sample’s traits in this research. The adapted framework contains four dimensions: Translation quality (How will the foreign language translate into Chinese), Content localization (How the content in website can be understood by local people), Cultural customization (How the website reflect local culture) and Local Gateway (Whether there’re links to Chinese pages from English home pages) would be primarily considered to compose the measurement framework in this research. Each dimension has different items and totally ten items are included in the framework. Every item is ranked on a scale of 1 to 5, based on the level of localization degree. We add up all the scores of certain website obtained in each item to reflect its website localization extent.

To decide whether the websites make use of website localization strategies, we analyze Fortune 1000 companies’ Chinese websites and code the website as “1” if it leverages some specific website localization strategy and “0” if it does not utilize it. We choose four website localization strategies proposed in Wu et al. (2015) for the purpose of this study. They are Special Plan (Information about a special plan of product/services provided for local people by foreign companies), Corporate Social Responsibility (Information about how the foreign companies undertake their social responsibilities for the local people), Local History (Information which depict the development history of foreign companies in the local country) and Local Authentication (Symbols denoting that the website is authenticated by a local organization as a trusted, safe and legal website).

Regarding the product types being marketed by the companies, we analyze the public financial information of companies and decide the product type based on their main revenue sources. If the companies’ revenue mainly comes from producing and selling consumer product, we code the companies’ website as “1”. If the main revenue sources are industrial product, we code their websites as “0”. Because some of the companies are service companies, we do not consider them for this variable and 181 websites are included in this variable. Similarly, for availability of online transaction channel, we code the websites as “1” if we can do online transaction with the companies through websites and “0” if not. With regard to how long Fortune 1000 companies operate in China, we also search the public information of the companies and compute the history length by using the number of current year minus the number of year they entered into China.

3.3 Data Analysis

This paper uses content analysis to judge the localization extent of websites in our sample by using a measurement scale, analyze whether the websites contain some specific website localization strategies and figure out whether there are transaction channels in their websites. The reason why we use content analysis is because the data of this study is limited to the content of foreign companies’ website and language of website is crucial (Kassarjian 1977). Meanwhile, one type of content analysis is to quantify or index the content according to some principles and this study belongs to this type (Neuendorf 2002). To test the quality of content analysis, two bilingual raters who have website localization and design related knowledge and background analyzed the 271 American companies’ Chinese websites. The inter-rater reliability test among raters is used to check to judge whether the analysis is reliable. The percentage of agreement of identifying website localization strategies and three exogenous factors is 79.5 %.

Given website localization strategies are binary variables, binary logistic regression technique is adopted to analyze their relationships with product types, online transaction availability and history length. As the data of website localization degree is continual, multiple linear regression technique is utilized to analyze its relationship with website localization strategies, product types, online transaction availability, and history length. Both regressions are implemented by PASW 18.0.

The descriptive statistics of variables in this study are listed in Table 1.

Table 1 Descriptive Statistics of Variables
Full size table

4 Results

4.1 Regression Analysis of Website Localization Degree

To test the relationships between the exogenous factors, website localization strategies and website localization degree, we analyze the data by using Multiple Linear Regression technique. The estimation results are in Table 2, which show the regression coefficients and standard errors. In model 1, we find that website localization strategies: Local History, Local Authentication, Corporate Social Responsibility and Special Plan are all significantly and positively associated with website localization degree. Thus, the regression results support H1. In model 2, we find that only product type is significantly associated with website localization degree, while history length and availability of online transaction channel are not significantly associated with website localization degree. Thus, H2a is supported, while H3a and H4a are not supported.

Table 2. Regression results of website localization degree
Full size table

4.2 Regression Analysis of Website Localization Strategies

Given the important role of website localization strategies, to know the relationships between exogenous factors and website localization strategies can provide valuable guidance to enhance websites’ localization degree. Considering the nature of data of website localization strategies, we utilize Binary Logistic Regression to analyze the relationships. The results are in Table 3 which exhibit the regression coefficients and P-values. The results indicate that availability of online transaction channel only positively associates with Local Authentication significantly, while product types positively associate with Local History, Corporate Social Responsibility and Special Plan significantly. History length also positively associates with Local History and Corporate Social Responsibility. Thus, H2b, H3b and H4b are all partially supported.

Table 3. Regression Results of Website Localization Strategies
Full size table

5 Discussion and Conclusion

5.1 Discussion

In this study, we explore the relationships between website localization and their exogenous factors. The exogenous factors contain product types, online transaction availability and history length, while we distinguish website localization design strategies from localization degree of each website. Based on the nature of data of different variables, we analyze the relationships by employing multiple linear regression and binary logistic regression. The results imply that the more we leverage website localization design strategies, the higher website localization degree is, while websites which belong to consumer product companies require higher website localization degree than the other websites.

To provide more insights to design a highly localized website, we also discuss the relationships between exogenous factors and specific website localization strategies. Online transaction availability only associates with Local Authentication which is used to endorse websites as safe and secure websites. The result implies that the utmost variable for online transaction is security. Product types link to more website localization strategies than other exogenous factors. Given consumer product companies need to interact with local consumers online and website localization strategies can raise positive attitude of local consumers, consumer product companies have stronger motivation to localize their websites as much as possible rather than industrial product companies. Although history length does not relate to website localization degree directly, it still associates with two website localization strategies: Local History and Corporate Social Responsibility. Taking full advantage of these two strategies needs a deep understanding of local culture and accumulation of interactions with local consumers. History length is a precondition of the understanding and accumulation.

The unsupported hypotheses include the relationships between history length, online transaction availability and website localization degree. Although organizations should be adaptive to the environment as time passes based on strategic choice perspective, the current result shows their websites design do not follow this practice. We explain this result in that the role of websites in different companies’ strategies is different. For some companies, websites are an important channel to communicate with local consumers and building brand image, while websites are just a way to deliver basic information for other companies. For online transaction availability, the utmost and essential concerns for people are privacy and security (Belanger et al. 2002), may not be website localization. Although website localization can convey some quality information about the website, it may still not be enough to ensure the security.

5.2 Limitations and Future Research

Although we explore the relationships among exogenous factors, website localization strategies and website localization degree, we still have some limitations which provide many opportunities for the future research. First of all, we demonstrate the positive relationship between website localization strategies and website localization degree, but empirical experiments which test consumers’ response towards website localization strategies directly will help us understand the effectiveness of website localization strategies deeper. Therefore, future research can conduct experiments that emphasize rigor to test the effectiveness of website localization strategies.

Secondly, we do not explore the underlying working mechanisms of different website localization strategies. As we can see from the discussion of this study, different website localization strategies work through different underlying mechanisms and these mechanisms may determine their effectiveness. To explore such mechanisms could be the key to understanding why some website localization strategies are more effective than the others. It may also lead to substantial theoretical contribution through addressing this research question because it can open the “black box” and establishes an opportunity for theory building. Therefore, future research also can explore the underlying processes of different website localization strategies through implementing controlled experiments.

Thirdly, since the study target is Chinese consumers who are collectivistic, consumers who belong to other cultures like individualistic culture may value website localization differently from Chinese consumers. We think this difference will make individualistic consumers focus on different website localization strategies from collectivistic consumers. For example, previous studies revealed that individualistic culture treats in-groups and out-group more equally than collectivistic culture (Lee and Ward 1998), and strategies which involve in-group features like Local Symbol for individualistic culture may not be as effective as it is for collectivistic culture. Therefore, future research can analyze the effectiveness of website localization strategies in individualistic context and compare the results in this study, which is in the collectivistic context.

5.3 Implications

This study contributes to the literature in several ways. Firstly, we explore and confirm the relationship between exogenous factors and website localization strategies/degree. The exogenous factors make us understand and predict the usage of website localization strategies and website localization design. Secondly, we understand the effect of different website localization strategies on website localization degree. This could trigger further studying of the website localization strategies. At last, we adapt and validate the measurement framework of Singh (2012) in our study context, and develop measurement framework of website localization furtherly. To establish an appropriate website localization measurement instrument is always meaningful for improving effectiveness of website localization.

Meanwhile, our study also contributes to website localization practice. At first, we investigate the current extent of website localization of Fortune 1000 companies. This result can help the companies to check whether their websites’ localization degree help or impede achieving their strategic goal in China. Second, exogenous factors of website localization we explored can be applied directly to predict the usage of website localization strategies and decide website localization degree. These three factors are quite operable and practical. Last but not the least, we validate the effectiveness of website localization strategies and foreign companies can better understand how to design their websites when they decide to promote the localization of their websites.