1 Introduction

Supply chain integration (SCI) has received considerable attention as an essential means of generating material and knowledge flows and leveraging the core competencies embedded in a supply chain (Narasimhan et al. 2010; Swink et al. 2007). Researchers and practitioners have put much emphasis on the potential of SCI. It is increasingly recognized that a competitive advantage can be gained through a firm’s efficient internal operations and solid supply chain networks (Olhager and Prajogo 2012). Leading companies tend to work frequently with their supply chain partners to achieve exceptional synergies and benefits through integration activities, such as accurately identifying customer demand and promoting a mutual exchange of information with suppliers (Koufteros et al. 2010; Allred et al. 2011). Similarly, researchers have provided empirical evidence about the variety of impacts that SCI has on performance. The empirical studies generally support the positive relationship between SCI and performance over the past several years (e.g., Villena et al. 2009; Cao and Zhang 2011). Little attention, however, has been directed to review themes or omissions in the empirical studies and reconstruct findings in an integrated way.

There are a few studies that have reviewed the literature on SCI and performance. Fabbe-Costes and Jahre (2008), for example, conducted a systematic review of quantitative papers, including mathematical simulation and conceptual papers, and they argued that unclear definitions and measures of SCI result in conflicting research findings. Similarly, Van der Vaart and Van Donk (2008) reviewed survey-based research published in ten journals and found that much more work is needed to examine the effect of buyer-supplier relationships on performance. In a review comparing integration strategies in US and East Asian companies, Zailani and Rajagopal (2005) concluded that the potential benefits of SCI can be realized through an appropriate alignment among partners and the recognition of this interrelationship. While the researchers made significant progress in theoretical arguments on the benefits of SCI, there remain shortcomings that should be addressed. First, the previous studies offer little insight into an in-depth analysis on the role of internal, customer, and supplier integration in performance improvement. Van der Vaart and Van Donk (2008) restrict their discussion to external factors and do not consider internal organizational aspects, such as internal collaboration and integration. The study by Fabbe-Costes and Jahre (2008) focuses on presenting definitions and measurement items of SCI. Second, findings of previous studies are based on relatively old references published between 2000 and 2006 (e.g., Van der Vaart and Van Donk 2008; Fabbe-Costes and Jahre 2008). Little attempt has been made to update the literature review on studies published since 2007. Further, although the link between SCI and performance has been considered as one of the most promising topics in survey-based research, some studies select mathematical simulation, case study, and pure literature review papers for their analysis (e.g., Fabbe-Costes and Jahre 2008). Third, researchers apply their own criteria for selecting journals, which might be subjective and omit important studies. Some researchers only consider journal ranking (e.g., Fabbe-Costes and Jahre 2008) although journal rankings have significantly changed over recent years. Others do not involve even SCI-oriented journals, such as the Journal of Supply Management (e.g., Van der Vaart and Van Donk 2008). There is no study that used the selection criteria proposed by other researchers. Researcher justification as to why they chose some journals and did not consider other journals remains unclear.

This study reviews survey-based research on the benefits of SCI and discusses methodological issues in the SCI literature. I focus on analyzing the potential of internal, customer, and supplier integration, rather than exploring the aggregate impact of SCI to which previous studies have already paid attention. This is because many researchers support the importance of the role of the three dimensions of integration and give a high priority to research (Huo 2012; Danese and Romano 2011; Koufteros et al. 2005, 2010; Flynn et al. 2010; Swink et al. 2007; Kim 2006; Gimenez and Ventura 2005; Vickery et al. 2003; Rosenzweig et al. 2003). Further, this study updates the previous literature review by dealing with survey-based research published from 1990 to 2012. To develop selection criteria of journals, I consider journal rankings and a journal list proposed by Rungtusanatham et al. (2003). This study addresses three important research questions: What are the definitions and characteristics of internal, customer, and supplier integration discussed in the previous studies? What are the methodologies that have been broadly applied? What are the weaknesses and the contradictions in the SCI literature? I believe that a comprehensive review can offer a holistic picture of findings from individual studies and stimulate future researchers to explore problematic aspects of existing studies. The remainder of this study is organized as follows. The next section outlines the methodology utilized in this study. The following section discusses findings of existing studies, investigates methodological issues, and proposes future research agendas. Finally, this study concludes with contributions and limitations.

2 Methodology

The literature review method was used to determine the scope of this study and analyze empirical studies. Three steps were employed. I first determined a list of journals. While survey-based studies appeared in a variety of journals, I selected 12 journals for this study. I included five journals by following the selection criteria proposed by Rungtusanatham et al. (2003), who argued that these journals are often ranked in the upper echelon in the operations management field. The journals were Journal of Operations Management, Decision Sciences, Management Science, International Journal of Production Research, and Production and Operations Management. The selection criteria also involved impact factors, journal rankings, and specialized areas such as supply chain management. The following journals were added to the list: Journal of Supply Chain Management, Supply Chain Management: An International Journal, International Journal of Operations and Production Management, International Journal of Production Economics, Journal of Purchasing and Supply Management, International Journal of Logistics Management, and OMEGA. The assessment period chosen was from 1990 to 2012 because researchers have been significantly involved in theorizing supply chain management since 1990 (Cousins and Menguc 2006). The Business Source Complete, Google Scholar, and Science Direct databases were used to select the studies using keywords, such as supply chain integration, supply chain interaction, and supply chain collaboration. The studies selected were characterized by the use of survey-based methods and the examination of organizational benefits of internal, customer, or supplier integration. I reviewed the titles and abstracts of the studies selected. Some studies were discarded because they modeled SCI as a dependent variable or provided insufficient information about theory or methodology. In the last step, a matrix was used to categorize key findings or methodological issues. Thirty six papers were selected for review.

3 Review results and discussion

Appendix Table 1 summarizes the studies selected for the analysis. It should be noted that a majority of studies operationalize SCI as multiple constructs (e.g., Moyano-Fuentes et al. 2012; Droge et al. 2012) while there are a few studies that conceptualize SCI as a single construct (e.g., Liu et al. 2012; Danese and Romano 2012). Performance is measured at the operational level, financial level, or at multiple levels (Flynn et al. 2010; Handfield et al. 2009; Chen et al. 2007). The structural equation modeling approach is mainly employed to analyze data (e.g., Prajogo et al. 2012; Cao and Zhang 2011; Koufteros et al. 2010) while regression analysis or correlation analysis is also used in a few studies (e. g., Olhager and Prajogo 2012; Droge et al. 2012; Das et al. 2006). Research findings are based on a variety of sample sizes ranging from 38 to 980. Some studies rely on data collected from multiple respondents, including CEOs, vice presidents, directors, or managers (e.g., Flynn et al. 2010; Sanders 2008; Devaraj et al. 2007; Koufteros et al. 2005). In other studies, managers are only considered as key respondents (e.g., Danese and Romano 2012; He and Lai 2012; Handfield et al. 2009). Appendix Table 1 also offers conflicting findings on the direct effect of SCI on performance. In the rest of this section, I discuss research findings of the selected studies and opportunities for future research. Methodological issues are discussed at the end of this section.

3.1 Definition of SCI

Researchers have considered SCI to be one of the most important activities when leveraging their internal and external networks (He and Lai 2012; Cousins and Menguc 2006). While researchers define SCI in different ways, I found that there are three important characteristics that should be considered when defining SCI. SCI can be defined as a formalized process that connects not only one department’s processes with those of other departments but also one firm’s processes with the processes of other firms. This perspective focuses on answering how supply chain partners integrate unique and inimitable capabilities to leverage their expertise and core competency (He and Lai 2012; Allred et al. 2011; Zhao et al. 2011). Swink et al. (2007) argued that SCI is not an outcome, but instead a process to acquire and consolidate strategic knowledge. Cao and Zhang (2011) defined SCI as a partnership process where firms execute and plan supply chain operations toward mutual benefits and goals. Pagell (2004) viewed SCI as a process of collaboration and interaction where firms work together for mutual outcomes. Similarly, researchers often underline the transaction cost theory to explain why a firm collaborates with other firms and how integration activities reduce transaction costs and result in superior performance (Saeed et al. 2005; Sanders 2008; Cao and Zhang 2011). It can be argued that SCI represents the exchange mechanism of resources and knowledge in a supply chain.

SCI can be described by a routinized practice that is generated to share resources and information across internal departments or external organizations. Researchers emphasize that the sharing effort is a critical prerequisite for building a seamless supply chain and creating a synergy. Swink et al. (2007) described SCI as a combination of various practices to integrate inputs into internal outcomes and share external practices among participants. Lockström et al. (2010) depicted SCI as an important practice that connects external works into a seamless congruency with internal processes. The practice-driven definition assumes that a firm cannot expect the benefits of SCI unless they share their own resources and capabilities with their partners. The resource-based view has been applied to the studies of SCI, emphasizing the activities of sharing resources and information in supply chains. Researchers often discuss the resource-based view highlighting that competitive potential can be obtained through a collection of resources, capabilities, or assets accumulated within an organization or across a supply chain (Das et al. 2006; Cousins and Menguc 2006; Devaraj et al. 2007; Allred et al. 2011).

SCI can be depicted as a performance-oriented effort to create tangible or intangible values, such as efficient flows of products and operational performance. Flynn et al. (2010) and Zhao et al. (2011) define SCI as the degree to which a firm strategically collaborates with other firms to achieve an efficient flow of information, cash, decisions, and products. This view argues that a level of integration is the most important indicator to show the success of a partnership and predict the potential of SCI. On the basis of the above discussion, I propose the following operational definition of SCI: organizational practices for developing value-added seamless processes across a supply chain, for sharing resources and knowledge among participants, and for transforming firms’ capabilities into synergistic values.

Researchers mainly classify SCI according to a stream-based perspective that involves internal integration within a firm, upstream integration with the supplier, and downstream integration with the customer (Danese and Romano 2011; Flynn et al. 2010; Swink et al. 2007; Kim 2006; Koufteros et al. 2005). The underlying premise of this perspective is that the integration activities can be managed forward from a supplier to a buying firm or backward from a customer to a buying firm (Cousins and Menguc 2006). Since the direction of the integration is associated with the flow of material and information, participants in the flow play a large role in dealing with the flow and generating value. While some studies classify SCI into internal and external integration (e.g., Droge et al. 2004; Gimenez and Ventura 2005; Sanders 2007), a common view is that external integration should be split into customer and supplier integration for the managerial purpose (Koufteros et al. 2010; Flynn et al. 2010; Zhao et al. 2011; Devaraj et al. 2007). Thus, this study focuses on discussing the impact of internal, supplier, and customer integration on performance.

3.2 Internal integration and performance

Internal integration refers to organizational practices of combining and improving internal resources and information in order to generate knowledge sharing beyond the boundaries of individual functions or departments, to assist external integration initiatives, and to achieve organizational goals (Zhao et al. 2011; Koufteros et al. 2010; Sanders 2007; Germain and Iyer 2006). Internal integration stresses that a competitive participant should know ways to manage conflicts across departments and convert individual abilities into organizational capabilities. It is posited that the isolated use of capabilities could not aid in the construction of a seamless value chain where all teams work together and accumulate a bundle of resources (Droge et al. 2004). This is because the individual capabilities are significantly interconnected and should be coordinated to ensure high efficiency and effectiveness (Grewal and Slotegraaf 2007). It is critical not only to emphasize joint problem-solving initiatives and systematic coordination among functional areas, but also to eliminate traditional functional boundaries (Germain and Iyer 2006).

Internal integration has the potential to affect customer and supplier integration. Internal integration serves as a foundation where a firm can more rapidly absorb, interpret, and apply external information (Marquez et al. 2004; Zhao et al. 2011). Some external information may be overlapping or useless. From a strategic view, a firm should internally evaluate the information and transform the external data into economically useful information by using data management systems and learning mechanisms. Efforts for building internal integration rely on several techniques, such as cross-functional teams, concurrent engineering, design for manufacturability, standardization, and enterprise resource planning (Vickery et al. 2003; Droge et al. 2004; Koufteros et al. 2005). Despite the different objectives in the approaches, practitioners often apply these approaches to drive the strategic adjustment of individual goals, the stimulation of knowledge sharing, and the establishment of cooperative culture. Researchers reported that knowledge sharing and value creation through internal integration positively affect a level of external collaboration and competitive performance (Droge et al. 2004; Gimenez and Ventura 2005; Koufteros et al. 2005; Flynn et al. 2010; Allred et al. 2011).

A close examination of existing studies indicates two important findings. First, research shows that shared knowledge and values through internal integration help firms strengthen cooperation with suppliers or customers. Collaboration across departments seems to be an important factor for generating a partnership with external partners. Second, studies provide evidence that internal integration has an impact, either direct or indirect, on operational or financial performance. The moderating effect of internal integration is also found in a few studies. In a survey of 64 manufacturing firms, Gimenez and Ventura (2005) found that internal integration positively generates external integration, while external integration mediates the relationship between internal integration and performance. Later research by Koufteros et al. (2005) expanded the research boundary by examining inter-relationships between internal and external integration in the new product development context. Koufteros et al. (2005), in examining 244 firms, found that a high level of internal integration activities, such as concurrent workflow and early involvement, are critical enablers that facilitate a timely exchange of key information about competitive capabilities between supplier and customers. Studies by Allred et al. (2011) and Droge et al. (2004) provided similar results showing that collaboration across different functions in a firm enable that firm to improve productivity, customer satisfaction, product development time, and financial performance.

It is valuable to note that there is a need for more research in this area. Although Flynn et al. (2010) explored the interaction effects among different types of integration, there has still been little work exploring how internal integration interacts with other types of integration. There are also mixed findings on which type of integration should be implemented first. For example, Devaraj et al. (2007), in a study of 120 firms, concluded that supplier integration should be developed first before investing in customer integration practices. In contrast, Droge et al. (2004) reported that adopting various types of integration concurrently is better than implementing a single type of integration. Further study is needed to carefully examine the interaction effect of internal integration and external integration. This effort will enhance our understanding of whether various integration practices should be implemented sequentially or simultaneously.

3.3 Supplier integration and performance

Supplier integration is defined as the organizational practice of a buying firm and its suppliers sharing and applying operational, financial, and strategic knowledge in order to generate mutual benefits (Narasimhan et al. 2010; Swink et al. 2007; Das et al. 2006; Germain and Iyer 2006; Petersen et al. 2005). Internal integration is necessary but not sufficient to continually improve competitive advantage. In a turbulent environment, firms need to gain more accurate information to leverage supplier resources and networks and enhance customer satisfaction (Petersen et al. 2005). It is essential for a buying firm to communicate with its suppliers and to continually upgrade information accumulated through internal integration practices. This is because the buyer’s information may not perfectly reflect new or ongoing issues in a real market (Narasimhan et al. 2010; Das et al. 2006). A supplier collaborates with a buying firm in two ways: as a seller providing components and parts, and as a strategic collaborator sharing expertise and knowledge. From the seller perspective, a supplier is simply involved in a buyer’s purchasing process and has the sole responsibility to produce a product (Koufteros et al. 2010). It is required that a buying firm pay more attention to the tasks of selecting a proper supplier, inspecting products delivered, and controlling relevant processes. In this respect, this type of integration is called supplier product integration, or a black box approach, because a supplier is mainly considered as a provider of a product and is trusted in respect to the quality, cost, and performance of a product (Koufteros et al. 2007; Kim 2009). It is hard to expect that the suppliers jointly work and share ideas or suggestions with a buyer. A supplier’s competence is not subject to sharing with a buyer and is often protected by an intellectual property law.

By contrast, suppliers play a vital role as a strategic collaborator allowing buyers to access their technological and operational resources (Droge et al. 2004; Narasimhan et al. 2010). Since suppliers work closely with a buyer in various processes, this type of integration is called supplier process integration, or a gray-box approach (Koufteros et al. 2005). This integration aims at generating communication, leveraging a supplier’s expertise, and fulfilling mutual goals. Using suppliers’ critical technological capability and expertise, a buying firm can minimize any change in design, prevent delays, and have chances for parallel processing (Droge et al. 2004). Suppliers can also assist a buying firm in several product development stages, such as idea generation, preliminary technological assessment, concept development, and testing (Petersen et al. 2005). A qualified supplier tends to possess innovative capacity, technical skill, and a dynamic business network. From a long-term perspective, a competitive supplier develops their capability through supplier development programs, namely a certification program, a site visit by a buying firm, and a feedback loop related to performance evaluation (Droge et al. 2004).

Empirical findings on supplier integration can be summarized into two categories. First, researchers generally support that collaborating with suppliers can lead to various types of operational performance. In a study of 322 firms in 23 countries, Frohlich and Westbrook (2001) found that a high level of integration with suppliers is positively related to performance improvement. A study by Saeed et al. (2005) examined the impact on process efficiency of the information exchange between a buying firm and its suppliers. They reported that a level of information sharing with suppliers is a critical determinant of operational performance, especially process efficiency. The second finding is that a contingency perspective is needed when interpreting the potential of supplier integration. Contextual factors may play a crucial role in the implementation of supplier integration. Collaboration with suppliers may not always be beneficial; they depend on the contextual factors, such as the firm size, a level of uncertainty, and the scope of cooperation. In a study of 157 firms, Koufteros et al. (2007) found that the positive effects of supplier integration on product innovation are different for large firms and small firms. Their analysis showed that large firms realize a high level of innovation success while small firms do not achieve any significant improvement in innovation as a result of strengthening the supplier integration. Koufteros et al. (2007) interpreted these results to indicate that the small firms’ insufficient resources and expertise may lead to different results. Similarly, using empirical data collected from 205 incremental projects and 110 radical product development projects, Song and Thieme (2009) investigated the different impacts of supplier involvement on success in incremental and radical product innovation. Song and Thieme (2009) found that, depending on incremental and radical innovations, supplier involvement for gathering market knowledge may influence a buyer’s market share and performance in different ways.

While researchers offer valuable insights on the role of supplier integration, important questions remain unanswered. First, there is little empirical evidence to firmly support the notion that the potential of supplier integration is context-dependent. For example, by considering strategic goals or operational issues, a buying firm may decide in which stage to involve a supplier. A supplier who holds an extremely unique technology may need to be involved in the very early stage of product development. Another supplier offering a fairly well-known standard technology could play a key role in reducing cost and improving quality in a variety of development stages (Ragatz et al. 2002). Consistent with the views of Song and Thieme (2009) and Koufteros et al. (2007), I believe that future research should address whether other contextual factors may produce different effects of supplier integration on performance. Second, further research should assess whether a buying firm’s second- or third-tier suppliers have a positive effect on their performance improvement. Researchers failed to explore how and why a supplier’s structural embeddedness can lead to a buyer’s superior performance. Much of the research placed heavy reliance on a simple structural framework dealing with only first-tier suppliers. However, the attention to multiple tiers of suppliers has increased remarkably over recent years. A practical rationale of the trend is that most buying firms are relying on a complex business network where second- or third-tier suppliers deliver the buying firms various components, parts, and knowledge (Gnyawali and Madhavan 2001; Oh and Rhee 2008). It is likely that suppliers at different tiers could play significant roles in ensuring product quality, tracing products, and planning business strategies (Bi and Lin 2009). Therefore, a supplier’s structural embeddedness could be considered as a critical determinant of performance and a vital indicator in evaluating a supplier’s real competence (Choi and Kim 2008; Galaskiewicz 2011). To fully explore the benefits of supplier integration, future research should carefully examine the effect of the supplier’s structural network.

3.4 Customer integration and performance

Customer integration refers to organizational practices of identifying, interpreting, and utilizing customer needs in order to produce a customer-defined product and increase customer satisfaction (Zhao et al. 2011; Koufteros et al. 2010; Lau et al. 2010; Devaraj et al. 2007; Swink et al. 2007; Droge et al. 2004). From a marketing view, a customer is a decision maker who has a potential purchasing power and evaluates the characteristics of products. Integrating with customers mainly involves sharing a set of information between a customer and a firm. Customers offer a firm their perception and judgment on a product through a survey or in person, whereas a firm provides customers with operational information, such as production schedule, inventory status, and sales forecast (Lau et al. 2010). A customer-driven firm tends to be in regular contact with customers, to encourage customers to be involved in most product development stages and to develop a feedback mechanism (Swink et al. 2007; Koufteros et al. 2010). Communication with customers mainly relies on the technological infrastructure, such as a point of sale system, an inventory management system, and a customer ordering system (Flynn et al. 2010). In terms of benefits, a firm can increase the accuracy of demand forecasting and quickly detect demand changes (Vickery et al. 2003; Flynn et al. 2010). The demand-oriented efforts also provide an opportunity to reduce market uncertainty and avoid costly mistakes (Koufteros et al. 2005).

A review of empirical research shows that researchers obtained contradictory findings. Some researchers found that customer integration drives product innovation and performance, such as productivity and marketplace (e.g., Frohlich and Westbrook 2001; Droge et al. 2004; Koufteros et al. 2005). Others, however, reported that focusing on customer needs negatively influences performance (e.g., Swink et al. 2007; Devaraj et al. 2007; Flynn et al. 2010; Koufteros et al. 2010). Swink et al. (2007) found that strategic customer integration has a positive effect on customer satisfaction but that there is a negative association between customer integration and market performance. Swink et al. (2007) noted that too much focus on customer needs may lead to declining overall market share and poor profit. Devaraj et al. (2007) found that customer integration is not effective in enhancing operational performance. Devaraj et al. (2007) offered interpretations that the potential of customer integration is contingent on a level of supplier integration and that supplier integration assists a firm’s capability to interact with customers and increase operational performance. Similarly, Flynn et al. (2010) reported that a high level of customer integration is not positively associated with business performance while customer integration contributes directly to operational performance. Flynn et al. (2010) argued that this non-significant finding is attributed to the fact that the effect of customer integration on business performance is mediated by operational performance. Koufteros et al. (2010) found that collaboration with customers does not have a positive influence on a buying firm’s ability to reduce glitches and to quickly change engineering procedures. Based on this result, Koufteros et al. (2010) noted that customer input in the product development process may be negligible. Lau et al. (2010) argued that customers tend to request that a firm continue producing the current familiar products and, thus, inhibit a firm’s innovation. Thus, it would be a high priority for future researchers to explore why and how customer integration could be or not be effective in enhancing firm performance using a longitudinal data or a large sample size.

3.5 Methodological issues

This section discusses the methodological weaknesses of existing studies and provides practical guidance about how to increase the generalizability and the validity of studies. Following the five-step approach proposed by Vokurka and O’Leary-Kelly (2000), I assessed the studies in terms of five aspects: operationalization of SCI and performance, generalizability, informant bias related to data collection, the unit of analysis, and statistical power issues. The methodological issues are summarized in Appendix Table 2.

Researchers mainly operationalize SCI and performance constructs using perceptual and multiple-item scales. Very few studies employ objective and single-item scales (e.g., Saeed et al. 2005). Most studies use a key informant method. It is also common that studies test unidimensionality, convergent validity, and discriminant validity (e.g., Allred et al. 2011; Cao and Zhang 2011; Das et al. 2006). Examining the validity makes it possible to check and reduce the substantial risk of measurement error that the perceptual measures may face (Vokurka and O’Leary-Kelly 2000). The use of multiple-item scales helps to describe a construct’s domain more accurately and enhance the reliability and validity of constructs (Villena et al. 2009). It can be argued that researchers tried to capture the complex nature of SCI by using perceptual and multiple scales as well as testing the validity.

The generalizability of research findings has been considered to be one of the major issues in empirical studies (Cousins and Menguc 2006). While various approaches have been proposed, collecting data across multiple industries is the best way to strengthen the generalizability (Schilling and Phelps 2007). The literature review indicates that a majority of studies are conducted in multiple industrial sectors – such as chemicals, metal products, and transportation industries. Only a few studies are based on a single industry, such as electronic computer industry or automotive industry (e.g., Vickery et al. 2003; Droge et al. 2004; Sanders 2008). The automotive industry is the most represented data source in the studies (e.g., Cousins and Menguc 2006; Villena et al. 2009). This may be because the automotive industry has relatively well-developed supply chain structures. Researchers have increased the generalizability of research findings by collecting data in multiple industrial sectors.

It is critical to test the informant bias in empirical studies that have mainly relied on single informants and perceptual measures. The single informant data may lead to a serious informant bias that threatens the validity of research findings (Vokurka and O’Leary-Kelly 2000). Multiple informants are better than a single informant when reducing respondent bias and increasing the validity (Lau et al. 2010). Reviewing existing studies shows that most studies are based on a single informant’s self-reported data. It was also found that many researchers appropriately examined the single informant bias through Harman’s one-factor test that has been widely used to check for common method variance (e.g., Devaraj et al. 2007; Flynn et al. 2010; Cao and Zhang 2011).

Regarding the unit of analysis, researchers mainly conduct studies at the firm level (e.g., Gimenez and Ventura 2005; Koufteros et al. 2007; Sanders 2008). There are relatively few studies examining the role of SCI at the plant or project level. For example, in a study of 224 plants, Swink et al. (2007) investigated the link between strategic integration and performance at the plant level. They found that the product-process and strategy integration initiatives result in increasing manufacturing competitive capabilities and that effective knowledge integration is a key factor for competitive gain. Similarly, a study by Koufteros et al. (2010) of 191 product development projects examined whether integration activities are effective in improving glitches and timeliness. They demonstrated that customer integration and supplier process integration are positively related to market success or product development outcomes. While it is also important to consider research contexts, such as global sourcing strategies and the plant’s role in a supply network, future studies may need to examine the benefits of SCI at an operational or plant level.

Statistical power involves a process to estimate the probability of rejecting a false null hypothesis (Vokurka and O’Leary-Kelly 2000). The purpose of a power analysis is to evaluate the adequacy of sample size and make a reliable decision in a hypothesis testing procedure. Reviewing the literature indicates that studies rarely discuss the statistical power issue. There are few studies that apply the power analysis. In a study of 120 firms, Devaraj et al. (2007) examined statistical power to decide whether the sample size is statistically adequate. Similarly, Rosenzweig et al. (2003) considered statistical power as a measure to assess the adequacy of sample size. They suggested that a sample size is adequate when the power is greater than .80. Future researchers may need to make efforts to apply various methods to their studies and prove that the sample size is adequate (MacCallum et al. 1996).

4 Conclusion

This study makes value-added contributions to the literature in three ways. The first contribution is that this study provides a critique of existing studies and presents research agendas that can stimulate future researchers to explore the link of SCI and performance. The analysis indicates that much work is still needed to address the potential of SCI and provide useful in-depth insights into the management of integration activities. Regarding internal integration, there is a need for further examination of the interaction or moderating effect of internal integration. Moreover, researchers failed to investigate the effect of second- and third-tier suppliers on performance. To fully understand the role of suppliers, researchers need to study a supplier’s structural embeddedness. Additionally, researchers provide conflicting findings on benefits through customer integration. Future researchers may need to utilize longitudinal data or a large sample size to carefully assess the customer role. Regarding a research framework, there has been a limited effort to examine contextual factors in supplier integration. More research is needed to explore the contingency perspective. Further, most studies consider only operational or business performance as a benefit of SCI. Future research needs to assess whether SCI activities are effective in enhancing other types of performance such as innovation.

Another contribution is that this study expands current understanding of the effectiveness of SCI on performance by reorganizing diverse perspectives and findings. The analysis offers researchers useful insights into complex aspects of SCI and takes a further step to create a better theory. I found that internal integration is a foundation for collaborating with external participants. As a strategic collaborator, a supplier can play a role in achieving mutual goals by proactively working with partners and suggesting ideas and technologies. A customer is a vital information source in the product planning and designing stages. The third contribution is that this study employs the five-step approach proposed by Vokurka and O’Leary-Kelly (2000) to systematically evaluate methodological issues of existing studies. I believe that the use of this approach enabled this study to use the rich quality of the analysis results. The analysis reveals that multiple-item and perceptual scales were mainly used to operationalize variables, while data were collected in multiple industrial sectors to enhance the generalizability. The questionnaires were mainly completed by single informants. Most studies were conducted at the firm level and rarely tested statistical power analysis.

Despite the important contributions, there are a few notable limitations of this study. The study is based on survey-based research published in a limited number of journals. It may be difficult to conclude that the analysis result reflects the findings of all existing studies. Future researchers need to utilize studies published in other journals, conference proceedings, or books. Moreover, the research scope of this study does not include determinants of SCI. It may be interesting to also review empirical studies on the determinants of SCI. In conclusion, I hope that this study will serve as a theoretical reference for researchers and practitioners exploring the relationship between integration practices and performance.