The study is founded on the principles of the SET and the RMT in relation to the constructs explored and the proposed relationships between the constructs. SET is founded on the principle of voluntary exchange of value between individuals (business-to-consumer) and organisations (B2B) in the relational process. As a result, it (SET) embraces the norm of mutual exchange, where value creation is to the benefit of all parties involved (Tanskanen 2015:579). SET further proposes that the building of relationships is founded on the basis of a subjective cost–benefit analysis as well as the evaluation of options (Liu et al. 2016:54). This implies that the parties to a relationship will evaluate the future of a relationship on the value that is still to be accrued from a partner (e.g. financial rewards, the trustworthiness of a partner and the level of future satisfaction from the relationship) (Sierra & McQuitty 2005:393). Chen and Choi (2005:2) and Ward and Berno (2011:1557) concur and state that social exchange is not always founded on financial rewards, although the continuation of the relationship will be based on views regarding the comparative costs and benefits of the relationship, past experience of satisfaction and its implication(s) for future relationship satisfaction. Considering this, SET has been found to produce outcomes such as trust, informal commitment and satisfaction (Lawler 2001:326; Lioukas & Reuer 2015:1826, 1829).

In terms of RMT, Stavros and Westberg (2009:308) and Lui, Wong and Liu (2009:1216) argue that relationship marketing embraces a customer-centric approach with the primary outcomes being increased customer retention, greater loyalty, lower marketing costs and increased profit levels. Ballantyne (2003:1255) argues that a relationship improvement strategy is initiated by the following question: ‘What is of value and to whom?’ Nicholson, Lindgreen and Kitchen (2009:195) state that the manner in which a relationship is managed by the parties involved will have a direct influence on a partner’s perception of trust in the relationship, which ultimately will influence the level of commitment towards the continuation of the relationship. This is an argument that has been posited in the seminal work of Morgan and Hunt (1994) in their trust-commitment model, perceived as one of the most applied models in inter-organisational relationship-building in relationship marketing. Therefore, the theory of relationship marketing is founded on the principles of trust between parties, the cultivation of mutual understanding between such parties, the ability to deliver on expectations based on previous cooperative history, to eventually secure customer cooperation as an outcome (Chang et al. 2015:869–870).

Considering the discussion above, the authors draw on the foundations of these theories to hypothesise the relationships between the constructs of the study, in the South African B2B environment. No previous study in the context of the South African B2B environment has applied the SET or the RMT to propose that satisfaction is an outcome of trust and commitment on the one hand, and an antecedent of coordination, cooperation and continuity on the other.

The current, competitive business environment necessitates an understanding of the changing nature of B2B relationship-building dynamics. The supplier no longer has the principal influence in the relationship because the buyer or customer is increasingly playing a key role in the establishment and management of relationship commitment and trust to secure satisfaction (Brodie 2017:22). Knox and Gruar (2007:115) concur and argue that RMT proposes the establishment and building of mutually beneficial value-add relationships between all relevant stakeholder groups, based on the principles of trust and commitment, to secure long-term satisfaction. RMT, therefore, argues that a continuous focus on customer relationship development secures the development of customer value strategies. Such strategies create a platform for the creation of a sustainable competitive advantage and increased profitability in the long-term (Theron & Terblanche 2010:384).

Relationship marketing is perceived as an influential strategy for companies aiming to distinguish their product offerings in the market environment. It has also become a growing area of international research considering the globalisation of markets (Samaha, Beck & Palmatier 2014:93). The supporters of the relationship marketing philosophy argue that business survival is no longer dependent on traditional marketing approaches of developing, selling and delivering products to customers without securing their retention for future purchases. Relationship marketing is encompassing a more inclusive focus that is founded on the establishment, growth and management of relationships that are both beneficial and satisfying to all parties involved (Leahy 2011:1). According to Malhotra, Uslay and Ndubisi (2008:213), relationship marketing is the foundation on which B2B relationships are built and is a growing area in the domain of business relationships. Theron and Terblanche (2010:386) also argue that the relevance of a relationship-orientated approach to marketing was highlighted in 2004 by the American Marketing Association (AMA) with the introduction of a new definition of marketing management:

Marketing is an organizational function and a set of processes for creating, communicating, and delivering value to customers and for managing customer relationships in ways that benefit the organization and its stakeholders. (Harker & Egan 2006:217)

Sheth and Parvatiyar (1995:399) further posit that in a B2B relationship, it is key for all parties to secure elevated levels of satisfaction with every single business interaction.

The key focus of a relationship marketing strategy is, therefore, to establish relationships at all points of interface with the customer. The outcome of this should be to establish benefits for both the organisation and the customer (Theron & Terblanche 2009:384). Managing B2B relationships is intricate because these relationships are often close, complex and long-term (Sarmento, Simões & Farhangmehr 2015:131). Therefore, an understanding of the management of B2B relationships can be attained once clarity has been obtained on the different elements contributing to them (Theron, Terblanche & Boshoff 2010:198). In the marketing literature, there is wide recognition of the importance of trust, commitment and satisfaction in the relationship-building process (Brito, Brito & Hashiba 2014; Jarratt & Ceric 2015; Ndubisi 2011; Morgan & Hunt 1994).

Table 1 serves as a support for this statement reflecting current studies on trust, commitment and satisfaction and why trust is an antecedent of commitment.

Altinay et al. (2014:724) and Laeequddin et al. (2012:551–552) state that in a buyer–seller relationship, trust is perceived as a key element in the relationship-building process. If the level of trust between partners is perceived as high, the willingness to commit to such a relationship is also stronger, eventually leading to higher levels of satisfaction. Despite this, there is limited knowledge on how to manage B2B relationships from a comprehensive perspective to secure positive relationship-building with the outcome of coordination, cooperation and continuity. The reason for including coordination, cooperation and continuity as outcome variables of interest in this study is that they are key characteristics of strong relationships in B2B marketing (Payan et al. 2016:64; Razzaque & Boon 2003:30–31; Shamdasani & Sheth 1995:8–9; Theron & Terblanche 2010:389–390). RMT argues that strong exchange relations are key to business success (Ng 2012:161; Wilson & Nielson 2001:3). According to Chang and Chuang (2016:519), cooperation represents an engaging process whereby the buyer actively interacts with the product innovation processes of the supplier to drive participation. Satisfied engagement experiences, therefore, become a driver for further cooperation between a buyer and a supplier. Coordination reflects engagement in joint activities relating to structure or process, between organisations (Keung et al. 2015:1). Coordination is key to ensuring high levels of efficiency and achievement of objectives in B2B relations. It helps to reduce conflict and disagreements, thereby ensuring smooth working relations. Continuity, on the other hand, is concerned with the duration of the business relationship. Success in relationship marketing is therefore characterised by loyalty, which includes commitment to continued business relations (Gaurav 2016:5). The various constructs depicted in Figure 1 form the focal point of this article, and different hypotheses are formulated that relate to these constructs.

Rauyruen and Miller (2007:24) state that trust is a key component in the relationship-building process when companies make promises to customers and then keep to their promises as well. Wang et al. (2014:1) emphasise that the increased acknowledgement of relationship marketing in the buyer–supplier relationship-building process has enhanced interest in the area of buyer trust. From a B2B perspective, trust is perceived as the degree to which a company views a business partner as being honest. Trust is also considered as the conviction of one partner that the other partner will conduct business that will secure positive benefits to the company and not conduct activities that will create a negative outcome to the business relationship (Hung, Cheng & Chen 2012:667).

According to Fullerton (2011:95), the willingness of one party in a relationship to be associated with another will be guided by the principle of shared values such as trust. Laeequddin and Sardana (2010:355–356) concur and argue that a partner’s level of satisfaction with a relationship will be influenced by previous engagements based on trust expectations. Satisfaction with previous experiences will influence the trust expectation of one partner to another in the B2B relationship, and the ultimate decision to continue or discontinue with the relationship (Sarmento et al. 2015:133). Voldnes et al. (2012:1082) confirm that trust is an antecedent of satisfaction because higher levels of trust between parties enhance overall satisfaction expectations and strengthen long-term commitment. Jiang, Henneberg and Naudé (2011:6) concur and state that trust is a primary precursor of satisfaction. They argue that when high levels of trust exist between the buyer and the seller, the former is usually satisfied with the relationship as there is trust that the activities of the supplier will secure positive results for all parties involved. This argument is further supported by Izogo (2016:377) who states that customers are more willing to be supportive of an organisation that establishes trust. Therefore, trust has a positive relationship with satisfaction.

Farrelly and Quester (2005:211) state that a customer with a strong trust in the supplier will expect the company to have open communication channels to enhance satisfaction experiences. Such open communication is critical in creating and sustaining positive buyer–supplier exchanges. Kaur, Sharma, and Mahajan (2012:282) concur and argue that buyers are more willing to continue with a relationship if the level of trust between the buyer and the seller is increased. This argument is further supported by Hansen, Morrow and Batista (2002:45) who state that satisfaction is a direct outcome of trust and increased levels of trust in the supplier can lead to higher levels of buyer satisfaction.

Considering the discussion above, it is argued that trust can be perceived as an antecedent to satisfaction. Therefore, the ensuing hypothesis is formulated:

H₁: Trust positively influences satisfaction.

Commitment is seen as a key element in securing long-term B2B relationships. It is described as a partner’s willingness to establish a long-term relationship with another partner and to continue with such an association, inclusive of an emotion of psychological attachment (Sung & Choi 2010:1051). However, the level of commitment of a customer will depend on the customer’s perception of the amount of effort that the seller puts into the relationship (Human & Naudé 2014:921). Dai, Haried and Salam (2011:4) state that the growth of continuous exchange relationships with customers offers a supplier a customer base that secures reliable income. In the field of relationship marketing, the significance of establishing, developing and maintaining the commitment of the customer to the supplier is highlighted (Wu, Zhou & Wu 2012:1760). Kaur et al. (2012:285) and Theron and Terblanche (2010:388) state that the stronger the level of commitment becomes in a seller–buyer relationship, the more stable the relationship becomes. Espejel, Fandos and Flavián (2011:209) agree and propose that customer satisfaction is a direct outcome of buyer’s commitment to the relationship with the seller, based on evaluating the difference between expectations and results. In addition, the commitment of employees towards the relationship-building initiatives of the supplier will strengthen customer relationship management initiatives, which could increase customer satisfaction levels (Ata & Toker 2012:504). Kim (2014:2) also established that commitment has a substantial influence on customer satisfaction. This argument is supported by Richard and Zhang (2012:573) asserting that a satisfied buyer has a stronger commitment towards a supplier than an unsatisfied customer, which enhances loyalty in the long-term. Akman and Yörür (2012:220–221) concur and argue that a buyer will remain committed to the supplier if the former is satisfied with the value offered in the relationship. Therefore, a high level of commitment gives rise to increased satisfaction with the relationship. There is general agreement by some researchers in the field of relationship marketing that commitment can be perceived as an antecedent to satisfaction (Anderson & Narus 1990; Johnson, Sividas & Garbarino 2008; Mohr & Spekman 1994; Wong & Zhou 2006). Fullerton (2011:95) concur and state that satisfaction can only be secured if both parties are fully committed to the relationship. Therefore, committed can be perceived as an antecedent of satisfaction in a buyer–supplier relationship.

Morgan and Hunt (1994:22), in their seminal work on the trust-commitment theory, argue that commitment is central to the establishment and management of supplier–customer relationships. The reason is that it motivates customers not to consider short-term options as being more beneficial than the long-term opportunities that a relationship with an existing supplier might provide. Therefore, for the purpose of this study, commitment is defined as an ‘implicit or explicit pledge of relational continuity between exchange partners’, involving an ‘enduring desire to maintain a valued relationship’ (Sarmento et al. 2015:133). Rutherford (2012:961) also notes that a higher level of commitment towards a supplier can result in higher levels of customer satisfaction, a lower propensity by the buyer to spread negative word-of-mouth opinion, and lower levels of customer defection. Richard and Zang (2012:573) concur and state that the level of customer satisfaction experienced is determined by the level of commitment between a buyer and a seller. It is, therefore, theorised that commitment is an antecedent of customer satisfaction in B2B relationships:

H₂: Commitment positively influences satisfaction.

Hong and Cho (2011:473) and Hess et al. (2011:15) state that trust is a central element in buyer–supplier relationship-building strategies. It is a key antecedent to commitment and is perceived as the most influential instrument available to a supplier in the relationship-building process. Hartmann, Klink and Simons (2015:110) support this argument by stating that a direct relationship exists between trust and commitment, because a higher level of trust between parties in a relationship implies greater commitment of one party to the other. Read (2009:27) also argues that a buyer will only commit to a trustworthy supplier as commitment involves vulnerability and could expose the buyer to opportunism. Nguyen and Mutum (2012:401–402) concur by arguing that trust is directly related to commitment because a buyer will be more willing to commit, if the supplier delivers on its promises and can, therefore, be trusted. The strengthening of customer commitment in a relationship can result in the buyer developing more trust in the supplier. Hung et al. (2012:667) further argue that when trust is formed between the buyer and the seller, there is a greater willingness to commit, securing a long-term orientation towards the relationship. This implies that being committed to a supplier is positively related to customer trust (Chang et al. 2012:942; Watkins & Hill 2009:994).

Hasche, Linton and Öberg (2017:33) highlight the relationship between trust and commitment by stating that trust is a key construct influencing buyer commitment. These authors further state that the existence of trust between parties in a buyer–seller relationship implies a greater willingness between the parties to commit to the relationship. This will enable the parties to develop a relationship that goes beyond economic exchange and that could secure long-term partnering even when the economic benefits are no longer existing (Parra et al. 2011:608). Stein, Smith and Lancioni (2013:856) also note that the interrelatedness of the parties in a buyer–seller relationship will define their level of commitment and the eventual duration of the relationship. Morgan and Hunt (1994:23) state clearly that ‘the major differentiation of these exchange relationship types … is the mutual social trust and the resultant commitment on the part of the individuals to establish and maintain exchange relationships’. This argument by Morgan and Hunt (1994) supports the view proposed in this article that the construct trust is an antecedent to commitment. The following hypothesis is therefore formulated:

H3: Trust positively influences commitment.

Customer satisfaction has been a topic of research interest for academics over the past two decades (Akman & Yörür 2012; Bejou, Ennew & Palmer 1998; Bowden-Everson, Dagger & Elliot 2013; Cheng et al. 2017; Garbarino & Johnson 1999; Ndubisi 2012). A reason for this is the benefits that customer satisfaction holds for an organisation, such as relationship continuity and an increased willingness to remain with the organisation (Naumann, Williams & Sajid Khan 2009). Since the early 1990s research has emphasised a shift in focus from a transactional to a relational approach towards relationship-building (Grönroos 1994; Morgan & Hunt 1994; Sheth & Parvatiyar 1995). This has resulted in customer satisfaction being identified as a key element in the development of relationship marketing strategies (Sun & Kim 2013:68). In the context of this study, satisfaction is defined as ‘a positive affective state resulting from the appraisal of all aspects of a firm’s working relationship with another firm’ (Sarmento et al. 2015:133). Satisfaction is largely acknowledged in the academic literature as an antecedent of behavioural outcomes such as customer retention (Trasorras, Weinstein & Abratt 2009), customer loyalty (Bowen & McCain 2015); Seto´-Pamies 2012) and word-of-mouth Ulaga and Eggert (2006). Satisfaction has also been identified as a key variable in the decision of a business to remain in a B2B relationship, where parties secure high levels of satisfaction during each business transaction (Sheth & Parvatiyar 1995; Theron & Terblanche 2010; Ulaga & Eggert 2006). In addition, this study also hypothesises that cooperation, coordination and continuity are outcomes of satisfaction.

Biggemann and Buttle (2009:551) state that coordination in a relationship will be present when both parties understand the act of exchange, empowering them to believe that the establishment of interaction is a possibility. Skinner, Gassenheimer and Kelley (1992:180) concur by arguing that there is a positive relationship between satisfaction and cooperation and that cooperation can be an influential factor on the level of satisfaction experienced through a buyer–supplier interaction. Mohr and Spekman (1994:139) also agree and state that the satisfaction of mutual expectations in the relationship strengthens the need for enhanced cooperation between parties. In terms of coordination, Guiltinan, Rejab and Rogers (1980:41) already argued in the beginning of the 1980’s that coordination can only be secured if optimal system performance (inclusive of satisfaction) is secured. Mohr, Fisher and Nevin (1996:105) and Payan (2007:218) established that satisfaction is a precursor to coordination and the probability of continuity. This finding supports the illustration in Figure 1 showing that both coordination and continuity are proposed outcomes of satisfaction. The study further posits coordination as a behavioural outcome because of its conceptual similarity to both cooperation and continuity. Woodside and Baxter (2015:104) propose that in both new and established relationships, coordination through social bonding assists in strengthening the relationship, thereby securing its continuity expectancy. This argument is in line with the study of Palmatier et al. (2006:138–139) stating that satisfaction can be viewed as an antecedent to both coordination and continuity in the relationship-building process. Therefore, to strengthen relationship continuity through coordination and a positive satisfaction experience, both parties must illustrate a desire to continue with the relationship, the relationship must be important to both parties and it must provide a growing basis for personal and business satisfaction to the parties involved. Shamdasani and Sheth (1995:8) concur and state that the willingness of parties in a relationship to continue with the alliance depends on the level of satisfaction experienced with the partnership. Faryabi, Sadeghzadeh and Zakeri (2015:39) also agree and state that satisfaction is perceived as a direct influencer of the decision to continue with the relationship.

Therefore, for the purpose of this study, the concept ‘continuity’ relates to the duration of the relationship, and the communal undertakings by both parties are mirrored by coordination. In addition, the willingness of one party to work with the other will be the basis of cooperation (Evensen & Hansen 2016; Faryabi et al. 2015; Samaha, Palmatier & Dant 2011; Voldnes et al. 2012).

The academic argument proposed by this study is that the three outcomes of satisfaction, namely cooperation, coordination and continuity, are a reflection of the intent and actions that relate to the interactions between two organisations. Considering this, three hypotheses are formulated:

H4: Satisfaction positively influences coordination.

H5: Satisfaction positively influences continuity.

H6: Satisfaction positively influences cooperation.

When conducting research, one can take the cross-sectional time horizon or the longitudinal time horizon. The cross-sectional time horizon data on variables of interest are collected only at a single point in time. The longitudinal time horizons involve collection to data with the purpose of observing the same variables over time to determine whether trends can be identified (Chisnall 2005). While in some longitudinal studies the same individuals serve as respondents over time, in business research, it is not always feasible to ensure that the same individual respondents take part in a study over time. This is because of varied reasons including movement of individuals to different positions or changes in company circumstances that make them fall out of the target population.

The target population of interest in this study was top companies in South Africa by revenue. Both the original and replication study followed a quantitative approach. The replication study was conducted in 2014/2015, while the original study was conducted in 2011/2012. In both cases, a total of 500 largest companies in South Africa were included in the sample frame. This frame was derived from a Topco list of the top 500 companies operating in the private sector of South Africa, based on their turnover. The Topco list was chosen because it was the only list that provides details of all the top 500 companies in South Africa. Either the purchasing manager or the procurement manager was contacted by phone to establish the suitability of the respondent to answer the questionnaire. The reason for selecting the purchasing or procurement managers as respondents is that the study focused on business relationships with suppliers. Purchasing or procurement managers are the individuals who deal with suppliers in a company, which makes them more knowledgeable about relationship issues with their company suppliers. However, in a case where the respondent was not qualified to answer the questionnaire, another individual who was better qualified to answer was identified and used as a respondent. Computer-assisted telephone interviews were conducted with the respondents. Respondents were asked to keep one supplier that they were familiar with, in responding to the questions. In the current study, a total of 250 usable questionnaires were returned, which produced a response rate of 50.0%. This compares well with the original study in which 232 usable questionnaires were returned, representing a response rate of 46.4%.

Two items, namely, (1) how much the respondent knew about his or her company’s perspective on the study topics and (2) how much the respondent knew about specific experiences with the supplier, were included for the purpose of checking the competency of the respondents used in both the original study and the replication study. This is in line with Campbell’s (1955) recommendations that respondents used in a study need to be competent enough to answer questions relating to the subject matter under investigation. The findings of the original study and the replication study, respectively, showed that 96.6% and 94.8% of the respondents had knowledge of their company’s perspective regarding the study topics and that 98.7% and 98.4% had knowledge regarding experiences with the supplier. The mean value for the first item in the original study was 4.24 and for the replication study it was 4.26. Regarding the second item, the mean value of the item in the original study was 4.31 and for the replication study it was 4.33. The mean values for both the original and replication studies illustrate that the respondents had appropriate knowledge about their company’s perspective on the topics under study and their company’s experiences with a supplier. The question on specific experiences with the supplier was aimed at capturing levels of knowledge through specific and not just general experiences with the supplier that the respondents considered when answering the questionnaire.

Measures that support the proposed conceptual model are depicted in Figure 1. Both the original and replication studies were based on exactly the same items used by Svensson, Mysen and Payan (2010) and others as discussed next. The choice to make use of pre-existing scales was because such scales help to enhance validity (Churchill 1979; Peter 1979). The items used to measure satisfaction were borrowed from the original work by Andaleeb (1996). Items used to measure commitment were borrowed from the original work by Morgan and Hunt (1994) and Anderson and Weitz (1992), while items used to measure trust were borrowed from the work of Zaheer, McEvily and Perrone (1998).

The items each to measure each of the proposed outcomes of satisfaction, namely, continuity, cooperation and coordination, were developed as follows: (1) continuity – the items for this construct were borrowed from the work of Lusch and Brown (1996); (2) cooperation – the items for thisconstruct were borrowed from Skinner et al. (1992); and (3) coordination – the items for this construct were borrowed from the works of Guiltinan et al. (1980) and Heide and John (1988). A five-point Likert-type scale was used for all items in the original study and the replication study, using ‘strongly agree’ (5) and ‘strongly disagree’ (1) as the end points (refer to Table 2).

The researchers used confirmatory factor analysis (CFA) and structural equation modelling (SEM) (Jöreskog & Sörbom 1976) to test the measurement model and assess the structural relationships. Initially, the researchers performed CFAs of the measurement model (i.e. 18 indicator variables as in Figure 2) based upon six constructs, applying the SPSS/AMOS 22.0 software. Our testing of the model in the original study and the replication study generated consistent and satisfactory findings through time. The goodness-of-fit measures in the original study and the replication study were all acceptable (Hair et al. 2006:745–749), as shown in Table 3.

Results on testing of construct reliability and validity are presented in Table 4 for the original study and in Table 5 for the replication study. According to the results, the variance extracted from items of all constructs exceeds 50% in both the original study and replication study, which indicates convergent validity through time. The average explained variance is slightly higher in the original study (75.2%) compared with the replication study (68/2%). The composite trait reliability levels of all included constructs are also above 0.7 (Hair et al. 2006) in both the original study and the replication study.

Furthermore, the researchers compared the variance extracted to the squared inter-construct correlations to examine whether the model measures different constructs (Hair et al. 2006). In the original study, the variance extracted for all constructs was equal to or greater than the corresponding squared inter-construct correlations (see Table 4), while it was not greater for all constructs in the replication study (see Table 5). The reason is that the constructs were more strongly correlated to each other in the replication study than in the original one, which is a limitation. The researchers therefore argue that the tested structural model indicates satisfactory discriminant validity in the original study (see Figure 2), while it is less satisfactory in the replication study. The hypothesised relationships of the model were all significant in both the original study and the replication study as expected by theory, which indicates consistent and satisfactory nomological validity through time.

Our testing of the model in the original study and replication study based upon South African business relationships accomplishes satisfactorily the requirements for convergent and nomological validity, as well as for construct reliability. Discriminant validity between constructs is, however, not fully accomplished in the replication study because of stronger relationships between each of them. The researchers therefore conclude that the measurement and structural metrics of the model do not indicate full consistency of validity and reliability between the original study and the replication study.

The researchers decided to test the structural model as illustrated in Figure 2 because of the satisfactory results from running the CFA of the measurement model in the original study and the replication study based on the South African business relationships.

The results of the goodness-of-fit measures of the structural model in the original study and the replication study are illustrated in Figure 2 (see Table 6). The results are consistent and satisfactory through time.

Furthermore, the structural model’s hypothesised relationships (see Figure 2) were all significant in the original study and the replication study, as presented in Table 7. Accordingly, the results support consistently all six hypotheses of the model tested on South African business relationships in the original study and the replication study.