Understanding the Implementation of Social Customer Relationship Management in the North African Context: An Integrated Theory Perspective

ABSTRACT

Social customer relationship management (CRM) deals with integrating social media data with traditional CRM in order to engage customers. This paper extends the Technology-Organization-Environment model with the diffusion of innovation theory to obtain insights into the factors and differences driving the use of social CRM in firms in Morocco, Algeria and Tunisia (n = 211). We find that the key factors driving the use of social CRM are organizational, through top management support and employee IT skills. Environmental factors also play a role, particularly through competitive pressure Compatibility, as a technology factor, is apparently not an antecedent to use. Our research also finds differences in social CRM use according to gender and industry, with male managers and those in ICT, media, training, and consulting more likely to use social CRM. This study differs from past studies by focusing on use rather than on adoption while examining a new context to provide additional insight for theory and practice.

KEYWORDS:

• Social CRM
• use
• North Africa
• survey
• TOE model
• DOI theory

1. Introduction

Social media has changed how customers search for information, communicate, and interact between them or with brands (Wang, Shi, Ma, Shi, & Yan, 2012), and it is not surprising that firms adopt social media to embrace this change in customer behavior and be present where customers are. Almost all marketing activities have become digitalized as marketing increasingly evolves to become digital marketing, which has blossomed into a fruitful research domain (Kannan & Li, 2017), including a comprehensive body of literature about different aspects of social media marketing (Appel, Grewal, Hadi, & Stephen, 2020). Notwithstanding, many important facets of social media marketing still remain underexplored. For example, the added value of social media to customer relationship management (CRM), known generally as Social CRM, needs additional research (Faase, Helms, & Spruit, 2011; Kannan & Li, 2017). Further, it appears that customer relationships in general require further investigation (Lemon & Verhoef, 2016).

Social CRM as a research theme typically focuses on studying the integration of social media with CRM systems. The theoretical debate and conceptualization of social CRM have already begun, but there is a need for further research into social CRM, particularly focusing on empirical research (Lehmkuhl & Jung, 2013; Yawised & Marshall, 2015). In this paper, our objective is to understand the antecedents (determinants) of social CRM technology use. Adoption cannot be used as a substitute for use (Hasani, Bojei, & Dehghantanha, 2017) and does not represent a guarantee of use (Shih, Venkatesh, Chen, & Kruse, 2013). Therefore, this research is differentiated and makes a contribution by virtue of its original focus on social CRM use.

This study is carried out within three North African countries: Algeria, Morocco and Tunisia. The scarcity of research in emerging markets persists; marketing literature generally devotes little attention to the African context (Shamah, Mason, Moretti, & Raggiotto, 2018). However, social media is managed and measured differently in emerging and developed markets (Medjani, Rutter, & Nadeau, 2019) and studies from different regions focus on different CRM aspects (Liu, Wang, & Zhao, 2020). Thus far, no cross-country study has examined social CRM in North African countries at an organizational level. This is an important contribution; it further adds an international dimension enabling the comparison of countries (Xu, Zhu, & Gibbs, 2004). As Shamah et al. (2018) point out, “there is a need to fill the gap regarding management and marketing research for African contexts: this need urges both scholars, and market practitioners.” Understanding and addressing this gap will contribute to advancing marketing theory development and managerial practice. The study makes a further contribution by examining differences according to gender, industry and country: aspects that are not well understood in the previous social CRM literature.

The remainder of this paper is organized as follows. In the next section, we briefly review the salient literature pertinent to our investigation. The subsequent section is devoted to conceptual and hypothesis development. Subsequently, we detail the research methods used, and this is followed by the empirical analysis and presentation of results. Finally, we juxtapose our findings with related literature in the discussion section, before concluding with theoretical and managerial implications in the final section.

Literature Review

In this section, we examine the literature foundation for the study. We review IT adoption in the North African context, CRM and social media, and social CRM.

IT Adoption and Use in North African Countries

North African countries are lagging in IT adoption (Bagchi & Udo, 2010); even though they are progressing in adopting it, they are still not efficient in using it (Kayisire & Wei, 2016). Due to unavailability of data on technology adoption and use in African countries, especially at the enterprise level, studies that debate this topic are still scarce (Mouelhi, 2009). One of the rare studies that covered four North African countries (Algeria, Egypt, Morocco and Tunisia), encourages partnerships with multinational firms to support adoption and use in these countries, fostering value creation and foreign investment (Bruno, Esposito, Iandoli, & Raffa, 2004). Youssef, Hadhri, and Maherzi (2015) emphasize the role of managers and workers’ IT knowledge in this process within Tunisian firms. This study further finds that market forces stimulate Tunisian firms to adopt IT and new organizational practices. For instance, Bahri-Ammari and Nusair (2015) advocate for the importance of management systems in determining CRM use in Tunisia.

CRM and Social Media

The concept of “CRM” emerged during the 1990’s among IT solutions vendors (Payne & Frow, 2005). There is a disagreement in the literature about the similarities of CRM and Relationship Marketing (Sheth, 2017). Consequently, CRM can be defined from different perspectives (Zablah, Bellenger, & Johnston, 2004).

Technology transformed traditional CRM into Electronic (e-) CRM, Mobile CRM and more recently Social CRM with the advent of social media. Social media refers to “a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0, and that allow the creation and exchange of User Generated Content [UGC]” (Kaplan & Haenlein, 2010). By allowing interaction between users and between users and firms, social media has proven to be a revolutionary step. Customers are able to generate their own content – UGC (Presi, Saridakis, & Hartmans, 2014) – and have more power through digital communication (Labrecque, vor dem Esche, Mathwick, Novak, & Hofacker, 2013).

Since customers create relationships, interact with other customers and with firms on social media (Wang et al., 2012), firms must learn to effectively manage customer relationships on this new media. If this is successfully achieved, the benefits can prove very significant. From a relationship marketing perspective, social media use increases brand attachment (Gensler, Völckner, Liu-Thompkins, & Wiertz, 2013), intention to buy and sales (Kozinets, Patterson, & Ashman, 2016).

Social CRM

Mohan, Choi, and Min (2008) offer one of the first papers examining social CRM and proposing a conceptual “social CRM system.” Their conceptual system combines Web 2.0 functionalities, social media, and CRM practices. From this perspective, Orenga-Roglá and Chalmeta (2016) validated a new methodology to implement social CRM in firms while integrating Web 2.0 and big data technologies to enhance engagement.

To define social CRM most researchers adopt Greenberg’s (2010) definition that refers to social CRM as “a philosophy and a business strategy, supported by a system and a technology, designed to engage the customer in a collaborative interaction that provides mutually beneficial value in a trusted and transparent business environment.” Academics and practitioners agree that social CRM is an extension and not a substitution to traditional CRM (Yawised & Marshall, 2015). Social CRM is about connecting social data with existing CRM to get new insights (Woodcock, Broomfi, Downer, & Starkey, 2011b). It is not just an additional channel, it extends CRM capabilities (Mosadegh & Behboudi, 2011). It can be used along different steps of the customer activity cycle, including information gathering and decision-making, sales and transactions, customer service and after sales service (Alt & Reinhold, 2012).

Examining social CRM definitions, it appears to refer to a new paradigm (Askool & Nakata, 2011) and a transition to CRM 2.0 that is not limited to a combination of social media and CRM (Mohan et al., 2008). Social CRM is differentiated in that it requires a philosophy, an organizational culture and a strategy (Choudhury & Harrigan, 2014). It is important to mention that all definitions of social CRM focus on “engagement” as a new concept, as the general objective of social CRM is to construct a strategy of consumer engagement (Lu & Miller, 2019).

Many studies combine social media and CRM to examine social CRM, which emerged as a popular topic in the CRM literature since 2013. There are various studies in East Asia, Europe, and North America (Liu et al., 2020), but we have only few insights into the situation in the African context.

Conceptual Framework and Hypotheses

Our study is at the crossroads of research into technology acceptance, social media and CRM. From this perspective, our research model and hypotheses emanate from each of these research streams.

In this section we justify the hypotheses and research model for the explanation of the use of Social CRM.

Technology Use

Several theories exist to explain the adoption and use of technology. To study “use” of technology, many terms are employed, including intention to use, perceived use, continuous use, actual use, in addition to adoption, diffusion and acceptance (Williams, Dwivedi, Lal, & Schwarz, 2009). Technology use can be studied according to several levels, but we distinguish two particular levels: individual-level (use by a person) and organizational (enterprise) level (Gangwar, Date, & Raoot, 2014; Oliveira & Martins, 2011; Williams et al., 2009).

The Technology Acceptance Model (TAM) (Davis, 1989), and its variants, is the most popular model in the literature (Williams et al., 2009). However, this model is widely criticized for its simplified view of technology (Legris, Ingham, & Collerette, 2003) and for reaching saturation (Bagozzi, 2007). Alternative theories such as the diffusion of innovations (DOI) or Technology-Organization-Environment (TOE) model (Tornatzky & Fleischer, 1990) are more suitable to study the organizational level (Baker, 2012; Gangwar et al., 2014; Oliveira & Martins, 2011). TOE may be more “complete” as it includes the “environment” dimension that is not included in the DOI. Moreover, it has a solid theoretical basis, consistent empirical support, and more potential for application (Oliveira & Martins, 2011).

TOE Framework

In simple terms, TOE assumes that three pillars of a firm’s context (Technological, Organizational, and Environmental) impact the technology adoption and use decision. The technological context is associated with existing and future technology characteristics that are susceptible to influence technology use. Likewise, the organizational context is related to organizational aspects such as structure, size and culture that determine adoption. The environmental context comprises competitors, government and other external actors.

This selection of theory is apt for the north African context. The three key aspects of the TOE may present differences in Africa: (1) the technology infrastructure is underdeveloped, resulting in barriers to using ICT developments (Ponelis & Holmner, 2015); (2) regarding organizational context, management practices, including social media, should be reevaluated in the context of developing nations (Medjani et al., 2019; Ocloo, Xuhua, Akaba, Shi, & Worwui-Brown, 2020); (3) the environment and market forces act differently in emerging markets – while competition leverages superior technology and brand appeal in developed markets, it is based on cost and distribution systems (cf. Honda example in Kumar & Srivastava, 2019). The environment can also be affected by sociopolitical governance problems and unbranded competition (Sheth, 2011).

The TOE model is emerging as a theoretical perspective increasingly used in innovation adoption studies (Gangwar et al., 2014) including social CRM (Jalal, Bahari, Tarofder, & Musa, 2019). It is applicable in many technological, industrial and cultural contexts (Baker, 2012; Oliveira & Martins, 2011). Given the above-mentioned considerations, we assume adopt TOE as the most appropriate theory for our study of the use of social CRM technologies. We further integrate DOI variables so as to make an association between TOE and DOI (Zhu, Dong, Xu, & Kraemer, 2006a).

Let us now consider each of the hypotheses within our research model.

Technological Context

Business must take in consideration the organizational changes created by technology use. This includes all technologies related to the business, whether already used by the business or available in the market (Baker, 2012).

Compatibility

Roger (2003) defines compatibility in DOI theory as “the degree to which an innovation is perceived as consistent with the existing values, past experiences, and needs of potential adopters.” Venkatesh and Bala (2012) suggested naming the variable “process compatibility” and define it as “the degree to which an innovation is perceived as being consistent with precursor methods for executing interorganizational processes.” We will adopt this latter definition for our research.

Compatibility is one of the most used antecedents of technology adoption, especially at the organizational level (e.g., Awa, Ojiabo, & Orokor, 2017; Venkatesh & Bala, 2012; Wang, Li, Li, & Zhang, 2016; Zhu et al., 2006a). Companies are more likely to use a technology if it is compatible with existing processes (the full system). However, compatibility is associated with the cost of switching technology. If changing technology induces costs (procedural, financial or relational), users will tend to keep using the existing option (Blut, Evanschitzky, Backhaus, Rudd, & Marck, 2016). Interestingly, Pick and Eisend (2014) meta-analysis shows a weak relationship between switching costs and switching intention. Additionally, employee IT skills (that will be discussed hereafter) will help to overcome the switching cost barrier. IT-expert employees will be less reluctant to use the new technology (Hyun & Pae, 2005).

Compatibility is relevant in explaining social media adoption (Tajudeen, Jaafar, & Ainin, 2018). It is the most important variable explaining e-CRM adoption (Sophonthummapharn, 2009). Further, the results of Ahani, Rahim, and Nilashi (2017) and those of Hasani et al. (2017) validate its significance in Social CRM adoption. Based on the above support, we posit that:

H1: Social CRM technologies compatibility with existing CRM processes has a significant impact on social CRM use by companies.

Organizational Context

Organizational factors play a greater role in technology use at firm-level in developing countries. Business characteristics and resources (e.g., structure, process, and size) may affect technology use in many ways; some business characteristics may, for example, encourage innovation, while others may not (Baker, 2012). In this study we focus on two key organizational aspects, top management support and employee IT skills.

Top Management Support

Top management support is one of the most critical factors of IS implementation in both academic and practitioner literature (Hameed, Counsell, & Swift, 2012). It refers to the level of senior management engagement with technologies. Top management can create a favorable environment for technology use in many ways; e.g. by allocating budgets and resources, overcoming resistance to change, and being active in implementation efforts (Elbanna, 2013). Consistent with these descriptions, we define top management support as the degree to which top management supports social CRM technology implementation (Hung, Huang, Lin, Chen, & Tarn, 2016). On the other hand, a lack of top management support can cause IT project failure (Liu, Zhang, Keil, & Chen, 2010) and hinders the adoption of social software in developing economies (Mukkamala & Razmerita, 2014).

Top management support can be complex to apprehend; it may vary in importance from one project to another in a multi-project context; supporting one particular project can lead to less support for another (Elbanna, 2013). However, many studies regularly show that top management support is an important determinant for technology adoption at firm level (Awa et al., 2017). For example, it was a key determinant factor for successful ERP implementation at Cisco (Datta, 2009) and in other cases such as in B2B e-commerce (Ocloo et al., 2020), knowledge management systems (Hung et al., 2016) and social CRM (Ahani et al., 2017). Thus, we propose the following hypothesis:

H2: Top management support has a significant impact on social CRM use by companies.

Employee IT Skills

This variable is referred to in the literature with various labels, such as IT/IS expertise, technology competence, savvy employees, IS knowledge of employees, inter alia. Hameed et al. (2012) define employee skills as the prior experience of IT employees in terms of knowledge and skills. In our research, we define it as the professional knowledge and technical capacities, in terms of technologies/information systems, of employees (Hung, Hung, Tsai, & Jiang, 2010). This includes skills such as social media management, communication and analytics.

In the social CRM context, companies need employees with skills in social media and data analysis (Malthouse, Haenlein, Skiera, Wege, & Zhang, 2013). Typically, companies adopt the technology only once the IT skills necessary for its use have been acquired by its employees. Previous studies agree that employee IT skills have a positive impact on technology adoption and use (Hung et al., 2010; Lin & Lee, 2005; Premkumar & Roberts, 1999; Sophonthummapharn, 2009; Wang et al., 2016; Zhu et al., 2006a; Zhu & Kraemer, 2005). Conversely, a lack of IT skills has a negative impact on technology adoption (Rana, Barnard, Baabdullah, Rees, & Roderick, 2019). Employee IT skills is an important factor for adoption in many CRM contexts (Ahani et al., 2017; Hung, Chang, Lin, & Hsiao, 2014; Sigala, 2011). We therefore posit that:

H3: Employee IT skills has a significant impact on social CRM use by companies.

Environmental Context

Environmental refers to external factors that may encourage or discourage technology adoption. The environment contains industry structure, presence or absence of technology suppliers, and regulatory situation (Baker, 2012). For our context, we focus on pressure from competition and consumers. Sometimes these variables are grouped under the same variable, external pressure (e.g., Hasani et al., 2017), but we examine them separately.

Competitive Pressure

Competitive pressure can be expressed in terms of intensity, as “the degree that the company is affected by competitors in the market” (Zhu, Kraemer, & Dedrick, 2004). It can be defined as “the degree of pressure that the company feels from competitors within the industry” (Zhu & Kraemer, 2005). In a CRM context, it refers to “the level of competition that a company feels when competing with other similar companies” (Hasani et al., 2017). Competition affects business with different levels of intensity; technology changes industry structure and allows the creation of competitive advantage by providing new ways to surpass competitors or by creating new businesses that may become competitors (Porter & Millar, 1985).

In a competitive environment, technology benefits are important in encouraging adoption. Businesses pay attention to competitor actions; they are more likely to adopt technology when it is used by competitors. It may be seen as a social need (to be conform to the group) expressed as a technical need (Thong, 1999). Firms that do not use a technology may perceive that their organizational capabilities are less than those of competition using that technology. In fact, technology enhances competitiveness (Oliveira & Martins, 2011) of companies that successfully transform technology into a competitive advantage (Thong, 1999). From this perspective, social CRM use can potentially enable a competitive advantage (Ahani et al., 2017). Competitive pressure is confirmed as an important predictor in many contexts, including in information systems (Lin, 2014; Premkumar & Roberts, 1999; Soliman & Janz, 2004; Thong, 1999), e-business (Jia, Guo, & Barnes, 2017; Zhu et al., 2006a; Zhu & Kraemer, 2005), B2B e-commerce (Ocloo et al., 2020), social media (Braojos-Gomez, Benitez-Amado, & Llorens-Montes, 2015), and CRM (Ahani et al., 2017; Avlonitis & Panagopoulos, 2005; Hasani et al., 2017; Sophonthummapharn, 2009). Thus, in our study, we hypothesize that:

H4: Competitive pressure has a significant impact on social CRM use by companies.

Customer Pressure

Customer pressure can be defined as “customer demands and behaviors that make companies adopt new technologies” (Hasani et al., 2017). Sophonthummapharn (2009) defines customer pressure as “the behaviour and demand of customers that force a firm to adopt a techno-relationship innovation in order to keep and satisfy customers..” This latter definition is more elaborate and closer to our context: social CRM can be considered as a techno-relationship innovation. Moreover, the definition is proposed in an e-CRM study.

Customer pressure can manifest as the result of customer expectation that firms will be present and active on social media (Braojos-Gomez et al., 2015). This encourages more pre-purchase and customer service requests. Firms must be sensitive to their customers’ needs and demands. These customers may expect that the firm will be using a specific technology. If customers do not ask the firm to use a technology, it could potentially be perceived as useless investment that engenders loss of resources. The lack of external pressure (including customer pressure) is a barrier to technology adoption (Rana et al., 2019). As customers become more and more social, firms are encouraged to rethink their CRM and move toward using social CRM (Ahani et al., 2017).

Although Askool and Nakata (2012) do not support the impact of customer pressure in social CRM adoption in Saudi Arabia, Sophonthummapharn (2009) classifies it as the third most important factor explaining e-CRM adoption. Furthermore, three studies (Ahani et al., 2017; Hasani et al., 2017; Zheng, 2011) confirm its significance in the different CRM contexts. Therefore, we posit:

H5: Customer pressure has a significant impact on social CRM use by companies.

Country, Industry and Gender Differences

The hypothesized relationships may be different across countries, regions, industries or products (Kumar, 2014). To capture some of these important variances, we will study country, industry and gender differences in social CRM adoption.

Country Differences

Cultural differences are a critical factor to understand technology diffusion levels (Choden, Bagchi, Udo, Kirs, & Frankwick, 2019; Hwang, 2011; Kumar, 2014). Xu et al. (2004) and Zhu and Kraemer (2005) studied e-business adoption and use under the TOE framework. The two studies exhibit the same results: there are differences between developed and developing countries, even though there might be similarities in some contexts (Oliveira & Martins, 2010). This is consistent with other studies comparing China and Australia (Wang & Zander, 2018), Egypt and the US in B2B e-commerce (Elbeltagi, Hamad, Moizer, & Abou-Shouk, 2016), or comparing the US, Sweden and India regarding cloud computing (Shih et al., 2013). Finally, Kumar, Sunder, and Ramaseshan (2011) reveal differences, in CRM systems adoption between North America, the Asia-Pacific and Europe.

Sabi, Uzoka, Langmia, Njeh, and Tsuma (2018) report that countries within sub-Saharan Africa have different technology adoption and use levels (Sabi et al., 2018). Hence, this can be applicable to our context as each North African country has a different perspective on technology development; the three countries have different rankings regarding Network Readiness Index (NRI, 2020). Morocco has more favorable conditions than others in terms of IT adoption (Bruno et al., 2004). It is the second most efficient African country in IT adoption and use, while Algeria and Tunisia are not found to be efficient (Kayisire & Wei, 2016). Thus, in the context of our study, we posit that:

H6: Social CRM technology use by managers varies across North African countries.

Industry Differences

Industry effects are widespread in the information systems literature (Martins, Oliveira, & Thomas, 2016; Oliveira & Martins, 2010; Zhu et al., 2006a). Firms from different industries will have different desires to use a technology (Sun, Cegielski, Jia, & Hall, 2018). Those activating in some industries are more likely to use e-business for example (Hsu, Kraemer, & Dunkle, 2006). These results are obtained in many contexts, such as differences between manufacturing and service industries (Oliveira, Thomas, & Espadanal, 2014;). We further argue that industry differences are important in social CRM use. In particular, we hypothesize:

H7: Social CRM technology use by managers varies across industries.

Gender Differences

Gender differences regarding technology behavior are widely discussed in the academic literature. Marketing and IS research demonstrate that there are disparities between male and female individuals regarding technology acceptance (e.g., Ameen, Willis, & Shah, 2018; Venkatesh, Thong, & Xu, 2012), and different social media behaviors (e.g., Park, Kim, Cho, & Han, 2019).

Gender differences are relevant in studies from an organizational perspective (Anwar et al., 2017). Ahuja (2002) find that gender is important is determining technology use within companies. Di Tommaso et al. (2020) confirm that male and female employees use enterprise social media platforms differently. In addition, Dong and Zhang (2011) examined employees of a firm introducing a CRM system and identified significant gender differences. Morris, Venkatesh, and Ackerman (2005) found similar results for technology adoption and use in the workplace based on the theory of planned behavior.

Given the scarcity of research examining gender differences in social CRM context and based on the above-mentioned information systems studies, we hypothesize that:

H8: Social CRM technology use by managers varies across males and females.

Overall, while Figure 1 presents the research model, Table 1 summarizes the key theoretical constructs employed in the study.

Table 1. Summary of theoretical constructs

	Variable	Definition	Previous studies
Technology	Compatibility	The degree to which an innovation is perceived as being consistent with precursor methods for executing inter-organizational processes.	Ahani et al. (2017), Hasani et al. (2017), Awa et al. (2017), Martins et al. (2016), Premkumar and Roberts (1999), Rana et al. (2019), Sophonthummapharn (2009), Tajudeen et al. (2018), Thong (1999), Venkatesh et al. (2012), Wang et al. (2016) and Zhu et al. (2006a)
Organization	Top management support	The degree to which business top management supports social CRM technology use.	Ahani et al. (2017), Awa et al. (2017), Lin 2014), Martins et al. (2016), Premkumar and Roberts (1999), Soliman and Janz (2004), Thong (1999) and Wang and Lai (2014).
	Employee IT skills	The professional knowledge and technical capacities, in terms of technologies/ information systems, of the employees.	Ahani et al. (2017), Hung et al. (2010), Lin and Lee (2005), Martins et al. (2016), Premkumar and Roberts (1999), Rana et al. (2019), Sigala (2011), Sophonthummapharn (2009), Wang et al. (2016), Zhu and Kraemer (2005) and Zhu et al. (2006a).
Environment	Competitive pressure	The level of competition that a company feels when competing with other similar companies.	Ahani et al. (2017), Braojos-Gomez et al. (2015), Hasani et al. (2017), Jia et al. (2017), Lin (2014), Premkumar and Roberts (1999), Soliman and Janz (2004), Sophonthummapharn (2009), Thong (1999), Zhu and Kraemer (2005) and Zhu et al. (2006a)
	Customer pressure	The behavior and demand of customers that force a firm to adopt a techno-relationship innovation in order to keep and satisfy customers.	Ahani et al. (2017), Hasani et al. (2017) and Sophonthummapharn (2009).

Figure 1. The research model

Research Method

To test our hypotheses, a quantitative research method was used based on data collection via questionnaire and statistical testing via a generalized linear model (GLM) (Nelder & Wedderburn). In this section we examine aspects of the research method in more detail.

Sample and Data Collection Procedure

The target population of respondents are marketing and CRM managers at middle and senior management level in three North African countries: Algeria, Morocco, Tunisia. The study was conducted, during the period from October 2019 to December 2019, using an anonymous online survey and followed a convenience sampling procedure and the human ethics requirements of HEC Algiers. A total of 1600 managers were contacted, presented with the study context and invited to answer the questionnaire. No payment was provided, and respondents were free to leave the survey at any time. In total we received 255 surveys; 211 of them were complete and usable for the analysis. The sample included managers with various characteristics. Industry was captured using an open-text question. The coding of categorical variables used in testing differences between gender, industries and countries is shown in Table 2.

Table 2. Sample characteristics for categorical variables

Variable	Category	Frequency No. (%)
Country	Algeria	98 (46.4%)
	Morocco	66 (31.3%)
	Tunisia	47 (22.3%)
Gender	Male	105 (49.8%)
	Female	106 (50.2%)
Industry	ICT, media, training and consulting	96 (45.5%)
	Manufacture, construction, transport and energy	58 (27.5%)
	Retail, tourism, banking and financial services	57 (27.0%)

Two further control variables were included in our analysis, firm size and years of industry experience. The mean of firm size was 5945 employees (SD = 33,270). The mean of experience was 7.69 years (SD = 7.42).

Operationalization of the Variables

We use Jarvis, MacKenzie, and Podsakoff (2003) decision rules for determining whether a construct is formative or reflective. We choose to rely on existing scales to enhance construct validity. As our study employs GLM, all measures are converted to summative variables prior to analysis.

All our independent variables are reflective and the items were adjusted from prior studies as illustrated in the Appendix A. We employ Premkumar and Roberts (1999) scales for top management support and competitive pressure. For compatibility and customer pressure we use Sophonthummapharn’s (2009) scale. Finally, employee IT skills are measured via Hung et al.’s (2010) items.

Concerning our dependent variable, social CRM technology use, Küpper, Jung, Lehmkuhl, and Wieneke (2014) noted a lack of a robust and valid construct to measure it. From this perspective, Küpper, Lehmkuhl, Wittkuhn, Wieneke, and Jung (2015) developed a formative measurement model that we use in our study. Social CRM technology use is conceived a first-order construct composed by six dimensions representing second-order constructs (monitoring & capture, analysis, exploitation, IS integration, communication and management). All the items were measured using a 5-point Likert scale from 1 = strongly disagree to 5 = strongly agree.

Measurement Validity and Reliability

Reliability for reflective measures is systematically evaluated vis internal consistency using Cronbach’s alpha. As recommended by many authors (e.g., Fornell & Larcker, 1981; Hair, Ringle, & Sarstedt, 2011; Hair, Sarstedt, Ringle, & Mena, 2012) we also examine composite reliability. As shown in Table 3, all constructs exhibit composite reliabilities above the recommended level of 0.7, while Cronbach’s alpha is above 0.7 for all but Competitive Pressure (0.670).

Table 3. Reliability and convergent validity

Variables	Cronbach’s Alpha	Composite Reliability	AVE
Compatibility (Comp)	0.721	0.841	0.640
Top Management Support (TMS)	0.848	0.898	0.687
Employee IT Skills (ES)	0.734	0.850	0.656
Competitive Pressure (CPR)	0.670	0.853	0.744
Customers Pressure (KPR)	0.829	0.921	0.853

Convergent validity is assessed based on Average Variance Extracted (AVE), while discriminant validity is examined via the Fornell-Larcker criterion (Hair, Risher, Sarstedt, & Ringle, 2019; Hair et al., 2012; MacKenzie, Podsakoff, & Podsakoff, 2011). As we can see from Table 3, all AVEs are above 0.5. Further, as Table 4 demonstrates, each variable demonstrates discriminant validity (square-root of AVE for each variable is greater than its inter-correlations with other variables).

Table 4. Inter-correlations (square-root of AVE on diagonal)

	CPR	Comp	ES	KPR	TMS
Competitive Pressure (CPR)	0.863
Compatibility (Comp)	0.497	0.800
Employee IT Skills (ES)	0.317	0.430	0.810
Customers Pressure (KPR)	0.518	0.496	0.317	0.924
Top Management Support (TMS)	0.293	0.439	0.544	0.315	0.829

Reliability based on internal consistency is not applicable for formative measures, where formative measurements are not expected to be correlated. Here, validity can be evaluated using collinearity and via loadings or weights (Hair, Hult, Ringle, & Sarstedt, 2017; Hair et al., 2011, 2019; MacKenzie et al., 2011). Collinearity is assessed using the Variance Inflation Factor (VIF) approach. In our study, the VIF for each item was well below the recommended level of 5, ranging from 1.356 to 3.335. Furthermore, all of the items of the formative construct were found to load at the p < .001 of significance.

Analysis

The Generalized Linear Model (GLM) method was adopted for analysis (Nelder & Wedderburn,) to test relationships between model variables and examine differences in SPSS 25.0. GLM is a more flexible extension of ordinary least squares that allows many types of response variables. Moreover, the model allows for the application of the maximum likelihood approach. After coding the categorical variables, as explained above, we created composite variables for each construct of our model. We also use standardized measures to reduce statistical bias. We choose a normal distribution and identity link function to run the analysis.

Results

The results of testing the full research model (with control variables) are reported in Tables 5 and 6. The Likelihood Ratio Chi-Square of 109.916 (df = 12, p < .001) indicates that overall, the model is a good fit and outperforms the null (intercept) model. The Pearson Chi-Square is 116.107(df = 198; value/df = 0.586), while the Akaike’s Information Criterion (AIC) is 500.752 and Bayesian Information Criterion (BIC) is 547.678. The results were further confirmed by entering only the significant variables into an additional GLM. The significance of the different variables remained unchanged. Again, the Likelihood Ratio Chi-Square is significant (χ² = 111.428; df = 15, p < .001), while the AIC falls to 498.575 and the BIC to 525.390, confirming a stronger fit for those variables. Here we report the results of testing the full model.

Table 5. Tests of model effects: full model

Source	Wald Chi-Square	df	p.value
(Intercept)	.003	1	.958
Gender	4.470	1	.034
Industry	6.521	2	.038
Country	4.252	2	.119
Size	.182	1	.670
Experience	2.156	1	.142
Compatibility	1.691	1	.193
Top Management Support	19.041	1	.000
Employee IT Skills	12.897	1	.000
Competitive pressure	4.495	1	.034
Customer pressure	.153	1	.695

Type III model effects.

Table 6. Parameter estimates

Parameter	B	Std. Error	Hypothesis Test
			Wald Chi- Square	df	Sig.
(Intercept)	−.109	.1528	.510	1	.475
Male	.224	.1061	4.470	1	.034
Female	** Reference category **
ICT, media, training and consulting	.301	.1260	5.697	1	.017
Manufacture, construction, transport and energy	.076	.1410	.290	1	.590
Retail, tourism, banking and financial services	** Reference category **
Algeria	−.272	.1386	3.855	1	.050
Morocco	−.105	.1436	.531	1	.466
Tunisia	** Reference category **
Size	.024	.0553	.182	1	.670
Experience	−.079	.0538	2.156	1	.142
Compatibility	.086	.0662	1.691	1	.193
Top management support	.290	.0665	19.041	1	.000
Employee IT skills	.231	.0642	12.897	1	.000
Competitive pressure	.138	.0652	4.495	1	.034
Customer pressure	.026	.0655	.153	1	.695
(Scale)	.550^b	.0536

The statistical tests show that the technological context variable in our model is not significant. compatibility (β = 0.086, p = .193). Thus, H1, is not supported. It seems that while these variables may influence social CRM adoption, they do not determine the post-adoption (use) of these technologies in the North African context.

In contrast to technological context, the organizational context is a very significant determinant of social CRM usage. Top management support is the most significant (β = 0.290, p < .001) determinant of social CRM use, supporting H2. Furthermore, from the other perspective, employee IT skills is the second most significant variable (β = 0.231, p = .001), providing strong support for H3. This appears to show that the decision of using social CRM is based mainly on the support of top management and for employees to have the IT skills and knowledge to use these technologies.

The environmental context results were mixed. While competitive pressure is a significant determinant (β = 0.138, p = .034), providing support for H4, customer pressure showed a non-significant effect (β = 0.026, p = .695), offering no support for H5. Thus, it appears that firms use social CRM technologies in response to their use by competitors rather than to satisfy customers’ behavior encouraging them to use social CRM; they are more preoccupied by competitors than customers.

The results also show significant effects in the model for gender (χ² = 4.470, p = .035) and industry (χ² = 6.521, p = .038), but not for country (χ² = 4.252, p = .119). However, there were significant differences in social CRM use between genders, industries and countries, supporting H6, H7 and H8. In support for H6, comparing countries results indicates a significant difference between Algeria compared to the reference category Tunisia (β = −0.272, p = .05). This suggests that Algeria had a significantly lower use of social CRM than Tunisia. Similarly, social CRM use in ICT, media, training and consulting is significantly greater than the reference category of Retail, tourism, banking and financial services (β = 0.301, p = .017). This result supports H7. Finally, the data suggests that male managers have a significantly greater use of social CRM in the workplace compared to female managers (β = 0.224, p = .034). This provides support for H8.

Our control variables were not significant in the analysis: firm size (β = 0.024, p = .670) and industry experience (β = −0.079, p = .142). Table 7 summarizes all the hypothesis tests.

Table 7. Hypotheses test

Hypothesis	Decision
H1. Compatibility ➔ Social CRM use	Rejected
H2. Top Management Support ➔ Social CRM use	Supported
H3. Employee IT Skills ➔ Social CRM use	Supported
H4. Competitive pressure ➔ Social CRM use	Supported
H5. Customer pressure ➔ Social CRM use	Rejected
H6. Differences between Countries	Supported
H7. Differences between Industries	Supported
H8. Differences between males and females	Supported

Discussion

In this research, we study the determinants of social CRM technology use. The academic literature tends to develop constructs and test them in developed countries without verifying their transposition to emerging markets such as those in Africa. To help ameliorate this problem, we studied three north African countries, responding to many calls for more contributions emanating from these countries (e.g., Shamah et al., 2018).

First, we found that compatibility, as a technological factor, is non-significant. Whereas previous studies confirmed that compatibility is the strongest variable for e-business use (Zhu et al., 2006b), e-CRM (Sophonthummapharn, 2009) and social CRM adoption (Ahani et al., 2017; Hasani et al., 2017), its impact was not proven for social CRM technology use. One possible reason is that North African companies do not complete systems and manage each IS individually. Based on this assumption, compatibility of social CRM with existing processes will not be an issue for its implementation as it will be managed separately from other systems. From this perspective, this result can be applicable in countries where the digital ecosystem is fragmented, where each IS is used independently from others.

Consistent with previous social CRM studies (Ahani et al., 2017; Malthouse et al., 2013; Sigala, 2011)., the organizational factors showed the most significant effects. Top management support and employee IT skills are found to be the most significant variables impacting social CRM use. The support of top management is crucial for the success of IS projects. Top managers decisions influence implementation in many ways; for instance, they can guarantee the required funding and resources, support employees to engage in implementation efforts, and lead actively the implementation project. Employees from their side, will contribute and facilitate social CRM use if they have IT skills allowing them to take advantage from social media in order to draw customer insights. These findings imply that regardless of the context (developed vs. developing) organizational factors are at the heart of ICT use (Arslan et al., 2019). Even if top managers may not perceive social CRM advantages in developing countries, their support remains essential for social CRM use. Likewise, employee IT skills can be recognized as a universal factor facilitating any technology use, whatever the context.

The environmental context findings are mixed. While competitive pressure is significant, customer pressure has apparently no effect on social CRM use. Askool and Nakata (2012) studied social CRM adoption in Saudi banks and found the same results. Similar findings were drawn for competitive pressure in Ahani et al. (2017) and Hasani et al. (2017) for social CRM adoption by SMEs. Companies using social CRM may gain considerable competitive advantage and surpass their rivals. Thus, firms monitor competitors’ actions and try to be as technologically equipped as other firms competing with them to prevent competitive decline (Zhu & Kraemer, 2005). Notwithstanding, the present study contradicts previous results regarding customer pressure (Ahani et al., 2017; Sophonthummapharn, 2009). The challenge for firms today is to join customers’ conversations on social media and social CRM use is an excellent way to respond to this challenge. While managers in developed countries rethink customer management using social CRM tools, firms in north Africa appear to neglect these aspects.

Overall, our findings appear consistent with those of Xu et al. (2004), who state that the key determinant of usage shifts from technology to organizational capabilities. This confirms that digital transformation in general, is not just about technology, it is about strategy (Kane, Palmer, Phillips, Kiron, & Buckley, 2015). Social CRM implementation is the result of a philosophy, an organization and a strategy (Choudhury & Harrigan, 2014).

The results make an interesting contribution; they partially support previous studies while revealing some differences calling for a deeper analysis of causes. We can speculate theoretically that differences are due to the shift from adoption to use from one hand, and to the different geographical context from the other hand. Moreover, previous social CRM adoption studies were essentially done with SMEs while our sample is composed of firms of all sizes.

In addition, this study is the first, to the best of our knowledge, that examines differences between genders, industries and countries in the context of social CRM. Regarding countries, we confirm that managers in Algeria use social CRM less than those of Tunisia. In other words, it is a less efficient use of technology (Kayisire & Wei, 2016). This means that the Algerian context tends to hinder social CRM use. This confirms country differences reported in previous studies (e.g., Kumar et al., 2011). However, our findings did not indicate that country as a variable provided a significant effect on use, partly contradicting the findings of Bruno et al. (2004) and Kayisire and Wei (2016) for the context of social CRM.

Hsu et al. (2006) found that different industries have different desires to use technologies. Our research supports these findings; industry has a significant effect on social CRM use. Moreover, we find that IT, media, training and consulting industries are also more active in using social CRM studies, this might be due to the customer orientation and/or technology orientation of these industries compared to others.

Finally, our results indicate that male managers are more likely to use social CRM. This finding is consistent with previous studies reporting women managers have lower levels of technology use (Anwar et al., 2017).

Conclusions

The current paper contributes to the marketing and IT literature by examining the antecedents of social CRM use across three developing countries. The social CRM literature has typically examined its adoption from an organizational perspective in the SME context (e.g., Ahani et al., 2017; Harrigan & Miles, 2014; Hasani et al., 2017). In addition, these studies are largely focused in America, Europe or Asia (Liu et al., 2020). Thus, it is important to replicate them in other countries to validate, refine and expand the findings. Building on previous studies, we contribute to the literature by going beyond mere adoption to assess determinants of social CRM use in a new context (i.e., three North African countries) including both SMEs and large companies. By doing so, we answer Ahani et al.’s (2017) call to extend research into social CRM in other country and firm contexts.

Our first main theoretical contribution is to show that technological factors are not significant when we shift the focus from adoption to use. Notwithstanding, organizational factors represented by top management support and employee IT skills remain the most significant determinants of social CRM technology use, and this is supplemented by competitive pressure. This provides a first view to understand social CRM and related IT use at enterprise level in the specific context of North Africa.

Our second major contribution is to add an international dimension to social CRM adoption and use studies. Our results are obtained by analyzing three countries in North Africa. This is different from other social CRM studies focusing on a unique country. From this perspective, the current work is an initiative to join researchers’ efforts to investigate IT adoption and use so as to collectively build a cumulative body of knowledge (Zhu et al., 2006a).

Our third important theoretical contribution lies in highlighting gender, industry and country differences. The three variables are relevant to explain different levels and differences in social CRM use from an organizational perspective.

From a practical perspective, managers must understand that social CRM is not just about technology; it is predominantly about organization. Professionals need to realize the importance of top management support and employee IT skills in maintaining social CRM use in their companies. Firms need to ensure that senior management commit and become engaged with social CRM, providing adequate resources for training and recruiting staff with the appropriate skills to use it effectively.

Similarly, the findings with respect to environmental factors reveal that firms in developing economies may be missing the opportunity to learn about their customers by responding only to competitive pressure. Engaging with customers is expected to provide data to make relevant decisions that can be crucial for their CRM and marketing.

The findings offer another important practical implication for social CRM solutions vendors. In order to encourage social CRM use in these countries, they must give priority to organizational factors rather than to technological factors as the essence of concept of social CRM is strategy (Ahani et al., 2017). Accordingly, Social CRM solutions providers have to rethink their strategies to penetrate emerging markets. Moreover, as employee IT skills are a key factor, firms providing social CRM technologies must think to complement their solutions with detailed information according to targeted employees’ skills.

Despite this paper’s theoretical and managerial implications, the results should be interpreted considering the study’s limitations. Firstly, the TOE framework has been criticized; Gangwar et al. (2014) consider it as a taxonomy and not a real theory as it does not include specific variables. Consequently, TOE variables change across contexts. Second, our study variables are not exhaustive, the reality is complex and many variables from technology acceptance and use theories could be significant in explaining social CRM use. Thus, other theories and variables can serve in future studies to have a deeper understanding of factors affecting the use of social CRM. Third, the generalizability of our results is limited; the data was collected from a self-reported survey and the sample size by country is relatively small and further studies should consider collecting real social CRM use data from larger samples. This will help to increase the limited degrees of freedom in the model. Further, our study is limited in that it examines North African countries; we encourage others to retest our model in other regions (such as South America and other parts of Africa) to further extend generalizability.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Fares Medjani

Fares Medjani ([email protected]) received his PhD from HEC Alger, Algeria. His research concentrates on social media and digital marketing with an emphasis on social CRM. He has presented and published conference papers at the International Conference on Digital Economy and published in International Journal of Business and Emerging Markets.

Stuart J. Barnes

Stuart J. Barnes ([email protected]) is Chair in Marketing at King’s College London, United Kingdom and Director of the CODA Research Centre. He has published and presented widely in leading outlets in a variety of disciplines, including information systems, computing, environmental science, tourism, health, and marketing.

Appendix A.

List of measurement items by construct

Table

Construct	Dimensions	Items
Compatibility (Sophonthummapharn, 2009)	-	Using social CRM is compatible with most aspects of the firm’s work. Using social CRM fits with the firm’s work style. I think that using social CRM fits well with the way I like to work.
Top Management Support (Premkumar & Roberts, 1999)		The owner or manager enthusiastically supports the adoption of social CRM. The owner or manager has allocated adequate resources to the adoption of social CRM. Top management is aware of the benefits of social CRM. Top management actively encourages employees to use social CRM in their daily tasks.
Employee IT Skills (Hung et al., 2010)	-	Our employees were all familiar with social CRM. There was at least one employee who was a social CRM expert. I would rate my employees’ understanding of social CRM as very good as compared with competitors.
Competitive Pressure (Premkumar & Roberts, 1999)	-	We believe we will lose our customers to our competitors if we do not adopt social CRM. We feel it is a strategic necessity to use social CRM to compete in the marketplace.
Customer Pressure (Sophonthummapharn, 2009)	-	Customers’ requirements indicate that a company needs to use social CRM. Customer behavior indicates that a company needs to use social CRM.
SCRM Use (Küpper et al., 2015)	Monitoring & Capture	The company utilizes a tool to … collect and store unstructured social media information, e.g., about the company and product, on their social media platform(s). collect and store unstructured information about a single artifact (e.g., consumer or a single event) on their social media platform(s).
	Analysis	The company utilizes a tool to … analyze and assess different types of content in real time. analyze unstructured social media data across various criteria (e.g., consumer segmentation) in order to identify general trends, profitable consumers, and so on. analyze unstructured data for a single consumer (e.g., a high potential influencer) across one (or more) social media platforms in order to understand their social behavior or motivations.
	Exploitation	The company utilizes a tool to … forecast consumer behavior and trends, and enhance their predictive modeling. create a network map of consumers and the relationships between them. support product purchasing, increase sales, cross- and upsell (e.g., social advertising campaigns). prepare summary statements, evaluate user activity and their loyalty, and/or prepare management reports,
	IS Integration	The company utilizes a tool to … integrate social media data with an existing CRM system. integrate other information systems, sales processes and existing technologies, and other tools along the project lifecycle (excluding a CRM system).
	Communication	The company utilizes a tool to … interact personally, via one-to-one communication, with a single consumer. communicate with an entire community and/or multiple consumers. communicate with other employees throughout the organization.
	Management	The company utilizes a tool to … manage their social media accounts, communities and forums, such as moderation and internal process management. allocate employee access rights. apply different engagement features (e.g., gamification).