The concept of ‘complementarity’ was initially introduced by Edgeworth (1881), who defined activities as complementary, ‘if doing (more of) any one of them increases the returns to doing (more of) the others’. The concept also identifies resources as co-existing and as collectively helping to achieve performance enhancement (Man Tang et al. 2022). According to Zhao et al. (2023), the synergistic effect derived from applying practices jointly can give rise to overall performance that is more significant than the sum of outputs of each of its parts. The literature further highlights the fact that complementarity is supplementary on the one hand and inciting on the other (Liao & Li 2018). The concept of supplementarity denotes that one variable accounts for the deficiencies of the other (Deeg 2007), while it is also grounded on the logic of synergy, which advocates that the presence of one variable spurs that of another. This mutually influential relationship allows an entity to benefit from the mutually reinforcing effects (Deeg 2007).

The notion of complementarity between practices has been widely accepted in management literature (Liao, Chen & Dong 2022; Man Tang et al. 2022). In addition, the concept of supplier involvement is also sustained in the literature. Wang, Modi and Schoenherr (2021) define supplier involvement as the extent to which practices are administered together with the supplier. This collaboration echoes the complementarity theory, which advocates that activities and/or practices are complementary and tend to be applied collectively. However, research highlights the fact that supplier involvement is contingent upon information sharing in order to fulfil its goals (Hwang et al. 2019), while e-procurement is a contemporary practice that has been introduced to make the supplier’s involvement more technically efficient (Mavidis & Folinas 2022). Together, these practices complement supplier involvement, which in turn influences a firm’s performance. This hypothesis is in accordance with the complementarity theory, which contends that activities and/or practices are symbiotic and increase performance in unison (Man Tang et al. 2022).

The level of success achieved in a relationship is largely moderated by sharing ‘knowledge’ (Kang, Wang & Cao 2024). The term ‘knowledge’ is often used interchangeably with the term ‘information’, and it is understood to relate to contexts pertaining to supply chain management. It is significant for measuring intrinsic value (Wan, Huang & Holtskog 2020). Information sharing refers to the essence of a buyer–supplier relationship and is a key requirement for developing its foundation. The concept incorporates two major aspects, namely communicating information to others and accepting information that has been transferred by the information giver (Moshood, Rotimi & Rotimi 2022). Having access to pertinent information helps one to understand what is actually happening, and how this might develop so that decisions and actions taken are right for the situation. Research, however, cautions that information sharing should be understood and examined appropriately, as the construct is characterised by diverse concepts such as information exchange, information transfer, information diffusion and information dissemination – therefore suggesting that information sharing cannot be understood as a generic concept of giving and receiving information (Savolainen 2017). For example, Hersberger, Rioux and Cruitt (2007) argue that information exchange entails a mutual, multidirectional information exchange, while information sharing represents a single, unidirectional activity. In light of this, and with reference to the buyer–supplier relationship that forms the context of this study, this study defines information sharing according to Kembro and Naslund (2014) as, ‘the inter-organisational sharing of data, information and/or knowledge in supply chains’.

Research into supplier involvement was originally initiated by the observation that Japanese automotives perform better than their Western counterparts in time-to-market and development cost in terms of supplier participation in new product development (Iansiti & Clark 1994). Subsequent research has yielded additional literature and insights on supplier involvement (e.g. Oh & In 2023). Moreover, several terms have been used in defining the concept, including integration, coordination, cooperation, collaboration and interaction (Wlazlak et al. 2018). However, involvement is the term that has emerged as dominant in the literature (e.g. Ayala et al. 2021; Suurmond et al. 2020; Wang et al. 2021). This term has been used in a number of studies to address the interface between buyers and suppliers (Flankegård et al. 2021). A further review of the literature has also brought to light the fact that there are different types of supplier involvement which authors describe differently (e.g. Cheng & Krumwiede 2017; Wieteska 2020). This classification helps to describe the degree of involvement, which is the level of responsibility assigned to the supplier and the associated risks. This study defines supplier involvement, in accordance with Wang et al. (2021), as the degree to which practices are performed together with the supplier.

E-procurement has become appreciated by businesses, industries and governments. It is an instrumental tool that is used to increase effectiveness and efficiency and also to improve the service quality of its adopters (Emery, Mélon & Spruk 2023). This method of procurement became recognised only following the inception of the Internet, and it is now identified as a tool that drives procurement reform by means of this e-commerce technology, including the Internet, Extensible Markup Language (XML) and, in some instances, key civic infrastructure (Vaidya & Campbell 2016). Its adoption is on the rise, despite resistance from some organisations in the past (Singh & Chan 2022). While the advantages of adopting e-procurement (e.g. increased efficiency in procurement processes, reduction in inventory costs and decreased order fulfilment time – see Bag et al. 2020; Nawi et al. 2016) have been expressed and acknowledged, scholars like Vaidya, Sajeev and Callender (2006) hold that its benefits had been exaggerated, perhaps because challenges in technology, infrastructure and legislation, resource constraints and organisational and management characteristics remain ever-present. However, a standard report by the Aberdeen Group reinforced the understanding that e-procurement was ‘here to stay’ and can do ‘more with less’ (Aberdeen 2008). Such conceptions have been criticised by the likes of Nawi et al. (2016), who argue that it is common for larger organisations to support such systems given that they are well capacitated and would benefit optimally from such systems because of their large volume of trade and numerous transactions. Thus, the significance of e-procurement is still subject to scrutiny.

The concept of e-procurement is a function of the e-commerce supply chain (Taghipour, Murat & Huang 2021). It has been acknowledged as strategically significant in inter-organisational relationships and as an incentive for some e-commerce supply chain practices (Singh & Chan 2022). At the minimum, there are three different ways in which the concept can be understood by academics and practitioners. The first is to reflect on the description of e-procurement, which explains its key characteristics (Davila, Gupta & Palmer 2003). A second approach is to take note of the various technologies used to explain the concept (De Boer, Harink & Heijboer 2002). The third approach is to be cognisant of the concept ‘public e-procurement’, which also has different levels that exist in its development (Mohungoo, Brown & Kabanda 2020). This study defines the concept of e-procurement as the ‘electronic integration’ and administration of all procurement-related activities between the buyer and the supplier (Chen et al. 2021).

A firm’s performance is highly likely to be influenced by how effectively and efficiently actions are implemented to reinforce one another. Despite financial reforms, emerging markets, developments in communication and information technology and changing customer behaviour, organisational practices affect firms’ performance, and also determine performance differences within the industry. Organisations therefore need to be decisive on how these aspects of the business should be applied, because this determines their efficiency. In the literature of strategic management, firm performance is one of the most relevant constructs. However, research into the construct is not without criticism. In fact, as early as the 1990s, firm performance has been described as ‘generally problematic’ with ‘no consistent approach, that clear definitions are lacking and that identical indicators represent alternative concepts simultaneously’ (Shenhav, Shrum & Alon 1994:771). More than two decades later, the construct was once again criticised, with Hult et al. (2008) and Bouland-van Dam et al. (2021) pointing to its conceptual and measurement issues, respectively. Miller, Washburn and Glick (2013), in their turn, documented three approaches towards conceptualising performance. Firstly, the ‘latent multidimensional approach’ acknowledges performance in both theoretical and empirical studies. The main argument is that the different dimensions of performance are interrelated but inadequately demonstrate overall success. Secondly, the ‘separate constructs approach’ necessitates using the exact properties of performance when developing theory and analysing empirical results. For example, the argument may be focused on improvement, while the increase in purchases is measured as the empirical construct. Thirdly, the ‘aggregate construct approach’ involves using a definite composite from different performance constructs consistently. Collectively, these constructs measure overall firm performance in general but should not be regarded as correlated. This study conceptualises firm performance as a unidimensional construct and defines it as the aptitude and ability of an organisation to efficiently use the available resources to realise accomplishments coherent with the set objectives of the organisation (Nguyen & Dao 2023).

Information sharing has been conceptualised as an independent variable in previous studies (Nabila et al. 2022). This implies that as a variable, it has the potential to influence another variable in a meaningful way. This insight is especially important for processes like e-procurement that are imposed and expedited in businesses to make them more efficient (Emery et al. 2023). E-procurement is a function that ensures that the necessary processes are performed electronically without manual intervention – the latter has been found to be costly and delaying the procurement process (Osei-Tutu et al. 2019). The correlation between information sharing and e-procurement is evident in the sense that during the procurement process, information is communicated between the buyer and the supplier (Osei-Tutu et al. 2019). However, capabilities such as the Internet and Information and Communication Technology are required to run these and to facilitate the sharing of information during the e-procurement process. In light of this, the present study therefore proposes that there is a relationship between information sharing and e-procurement. In particular, it is hypothesised that:

Information sharing has also been found to play a key role in supplier involvement (Zhang, Zou & Wang 2023). This insight is also propounded by Petersen, Handfield and Ragats (2003), who, drawing from 17 case studies, found that information sharing improves supplier involvement by assisting the firm in meeting its objectives and realising better outcomes. Kankam et al. (2023) further emphasise inter-organisational information processing in achieving project or goal fulfilment. However, Zhang, Wang and Gao (2017) argue that the information shared should be clear and sufficient. This is understandable, given that shared information that is meaningless and inadequate does not have value. Against this backdrop, this study therefore proposes that there is a relationship between information sharing and supplier involvement. In particular, it hypothesises that:

The relationship between e-procurement and supplier involvement can be acknowledged in this study’s adopted definition. Mavidis and Folinas (2022) add that e-procurement supports the integration and coordination with the supplier. Moreover, commitment has been identified as significant in the influence of e-procurement by Nawi et al. (2016). These authors highlight that this is especially true among SMEs. Therefore, if there is a lack of commitment, it becomes difficult for e-procurement to contribute to supplier involvement in a meaningful way. In light of this and with reference to relevant literature, this study proposes that there is a relationship between e-procurement and supplier involvement accordingly. In particular, it is hypothesised that:

Firm performance is a construct that is often analysed as the final dependent variable (Raj et al. 2023; Tajvidi & Karami 2021). This approach has been adopted in this study, and the construct is hypothesised to be influenced by supplier involvement. The correlation between the constructs has been acknowledged in the literature. For example, Delgado-Verde and Diez-Vial (2024) highlight that a dedicated effort by suppliers guarantees quality improvements, while Bahemia, Squire and Cousins (2017) report supplier involvement as contributing to improvements in innovation capabilities, increased resource flexibility and a reduction in risks, respectively. However, supplier involvement has been largely acknowledged as an interconnected process (Wang et al. 2021) and an ongoing integration of resources (Lusch 2011). This implies that supplier involvement requires equal commitment from the buyer in order to influence firm performance in a significant way. Therefore, in light of this and the literature, this study proposes that there is a relationship between supplier involvement and firm performance. In particular, it hypothesises that:

Supplier involvement is an interdisciplinary phenomenon that has been studied in the field of supply chain management (Suurmond et al. 2020; Wang et al. 2021). It is a construct that can explain the significance of information sharing, illustrating that supplier involvement can only meet its mandate of supply (e.g. manufacturing or transforming of raw materials and their delivery as finished goods) effectively through information sharing with the buyer (Mathu 2019). This implies that the value of supplier involvement is influenced by information sharing, a notion that is further supported by Narkhede et al. (2013) as generating performance improvements for the firm. In light of this, the present study proposes that supplier involvement mediates the relationship between information sharing and firm performance. In particular, it hypothesised that:

Firms have been acknowledging the significance of information technology in terms of realising improved performance (Sánchez-Rodríguez, Martínez-Lorente & Hemsworth 2020). It is likely that this acknowledgement has also contributed to the increased study of e-procurement (Bag et al. 2020; Emery et al. 2023). In the process of working towards supplier involvement, e-procurement facilitates material cost savings and productivity – which, in turn, stimulate improvements in key performance indicators such as quality, responsiveness and risk management (Högel et al. 2018). This insight suggests that the value of e-procurement on firm performance can also be realised and appreciated in terms of the impact it has on supplier involvement. For this reason, this study therefore proposes that supplier involvement mediates the relationship between e-procurement and firm performance. In particular, it is hypothesised that:

Cronbach alpha tests were run to ascertain the reliability of measurement scales. The results presented alpha coefficients that varied between 0.828 and 0.918, which signified high reliability in light of the ≥ 0.7 threshold recommended by Green and Salkind (2016). Tests for composite reliability (CR) were further performed to verify reliability (Haji-Othman & Yusuff 2022). The results revealed CR indices that ranged between 0.717 and 0.942. According to Chinomona (2011), scores greater than 0.6 are indicative of internal consistency.

Factor loadings (or standardised regression weights) were observed to determine convergent validity (Cheung et al. 2024; Sarstedt et al. 2014). Notwithstanding FP1, measurement items appeared to load strongly on their common constructs as they were greater than 0.5. This implies that items measured at least 50% of their respective constructs (see the seminal work of Anderson & Gerbing 1988). Tests for the average variance extracted (AVE) indicated that all constructs except firm performance were represented well by their measurement items, because they exhibited a variance extracted estimate that is greater than 0.5, respectively (Sarstedt et al. 2014). To further substantiate validity, discriminant validity was examined. The results are presented in Table 3.

The square root of each AVE value was determined and a higher index when compared against correlation coefficients signifies discriminant validity (Yusoff et al. 2020). The results presented in Table 3 indicate and substantiate the presence of discriminant validity.

As a standard procedure, explanatory power assessment, predictive accuracy assessment and path analysis were undertaken when testing the structural model (Hair et al. 2019). The coefficient of determination (R²) was measured to ascertain explanatory power. The results indicate that e-procurement (0.583) and supplier involvement (0.555) carry moderate explanatory power (+0.50) while firm performance (0.104) exhibits a weak one (Johannesson, Ohlson & Zhai 2024). This implies that the conceptualisation and measure of firm performance in this context need to be revised, given that it fails to represent the research model well. Figure 2 shows the structural model on which R² coefficients are presented.

The blindfolding-based cross-validated redundancy measure (Q²) was performed to determine predictive accuracy. The results are presented in Table 4.

Firm performance exhibited a model parameter that is without predictive accuracy, while e-procurement and supplier involvement displayed medium to large predictive relevance, because they exhibited model parameters that are between 0.025 and 0.50 (Hair et al. 2019). The results indicate again that the operationalisation of firm performance in this study’s context needs to be revised. Although this is yet another indication of the construct’s conceptual and measurement issues (see Bouland-van Dam et al. 2021; Hult et al. 2008), e-procurement and supplier involvement are significant constructs to measure within this context and can be applied in expanding and validating the complementarity theory. Following the test for predictive accuracy, path analysis was conducted. This involved examining the direct and indirect relationships between latent constructs (Ruan et al. 2024). The results from this procedure are presented in Table 5.

T-statistics were compared against a threshold of ≥ 1.96, while p-value tests were run at a ≤ 0.05 significance level (i.e. 1% – 95%) to avoid false positives (Kock 2016; Winship & Zhuo 2020). Two of the direct relationships, namely H1 and H2, exhibited positive and significant correlations, while H3 and H4 were rejected. It is important to note that H3 demonstrated a statistically insignificant relationship, while H4a was positive and yet insignificant. Mediation was also tested using SmartPLS bootstrapping with 5000 samples to assess the indirect effects of information sharing and e-procurement on firm performance via supplier involvement. As shown in Table 5, the indirect effect of information sharing and e-procurement on firm performance via supplier involvement was not statistically significant (H4b: β = −0.250, p = 0.284; H4c β = −0.013, p = 0.737), indicating no mediation.