The impacts of supply visibility and demand visibility on product innovation: the mediating role of supply chain integration

Abstract

Purpose

Drawing on absorptive capacity theory, this study explores the impacts of supply visibility and demand visibility on product innovation (i.e. exploratory and exploitative innovation), and it examines how supplier integration, customer integration and internal integration mediate these impacts.

Design/methodology/approach

The authors employ empirical survey data from 200 Chinese manufacturers and use structural equation modeling to test the proposed relationships.

Findings

The results show that supply visibility is positively related to supplier integration and internal integration and that demand visibility is positively related to customer integration. Furthermore, only customer integration and internal integration positively relate to exploratory and exploitative innovation.

Originality/value

First, this study emphasizes that supply visibility and demand visibility are important sources of a firm's innovation performance and that supply chain integration increases focal firms' capability of exploiting information and facilitates product innovation. Second, the study shows that supply visibility and demand visibility have distinct effects on three dimensions of supply chain integration and exploratory and exploitative innovation. The study also provides significant managerial guidelines for effectively leveraging supply chain visibility and integration in the promotion of product innovation.

Citation

Hu, M., Jiang, S. and Huo, B. (2024), "The impacts of supply visibility and demand visibility on product innovation: the mediating role of supply chain integration", The International Journal of Logistics Management, Vol. 35 No. 2, pp. 456-482. https://doi.org/10.1108/IJLM-01-2021-0033

1. Introduction

Supply visibility and demand visibility have emerged as two powerful instruments for improving focal firms' innovation performance by increasing the effectiveness and efficiency of information sharing in supply chain networks and developing a long-term relationship between a focal firm and its supply chain partners (Vanpoucke et al., 2014; Gao et al., 2015; Somapa et al., 2018; Swift et al., 2019). However, recent studies in the supply chain management literature have found that the dark sides of supply visibility and demand visibility (e.g. information overload and supplier opportunism) may hinder product innovation (Karhade and Dong, 2021; Skowronski et al., 2020). To reconcile this paradox, scholars should explore the underlying mechanisms that allow supply visibility and demand visibility to influence product innovation to better leverage supply chain visibility for innovation performance promotion.

Drawing on absorptive capacity theory, we argue that unless focal firms have the absorptive capacity of information exploitation, supply visibility and demand visibility do not necessarily result in better innovative outputs (Zahra and George, 2002). Supply visibility and demand visibility represent a focal firm's ability to capture supply-side and demand-side information (Somapa et al., 2018; Williams et al., 2013; Srinivasan and Swink, 2018). However, possessing these two abilities does not mean that the focal firm can exploit external information and transform it into innovative outcomes. Strategic supply visibility (or demand visibility) promotes frequent and intense communication between focal firms and their suppliers (or customers); hence facilitating supplier integration (or customer integration). Moreover, focal firms with the two types of visibility contribute to internal integration by coordinating different departments to respond to the changes of business environments (Prajogo and Olhager, 2012). The three types of supply chain integration build inter-organizational processes, enhance collaborative initiatives among supply chain partners and improve a focal firm's absorptive capacity to exploit strategic supply chain information (Schoenherr and Swink, 2012). This absorptive capacity helps transform supply-side and demand-side information into product innovation (Stock, 2014). Therefore, based on absorptive capacity theory, we propose that supply visibility and demand visibility can improve focal firms' innovation performance by distinctly supporting supplier integration, customer integration and internal integration.

The practices of Cisco are a case in point of this phenomenon. As the world's largest networking equipment company, Cisco has established a global online network to capture information on its component suppliers and customers. Thanks to this online network, Cisco can better integrate with its supply chain partners and centralize its internal business activities (Kraemer and Dedrick, 2002). Specifically, the firm's supply-side information (e.g. production schedule, inventory, quality, performance and capacity information) can promote supplier integration and improve Cisco's ability to respond rapidly to supply changes in the supply chain (Zhou and Benton, 2007; Dong et al., 2009). Importantly, suppliers can join Cisco's product design process through a virtual manufacturing model in the global online network (Zhou and Benton, 2007). Furthermore, the company's demand-side information (e.g. delivery status information, product offerings and pricing, technical assistance and customer service) can promote customer integration and a deeper understanding of customer needs, which helps to develop new products (Kraemer and Dedrick, 2002). The benefits of inter-organizational and intra-organizational integration allow Cisco to better exploit the strategic information, focus on its core competencies, and reduce the time to introduce new products (Kraemer and Dedrick, 2002). However, the mechanisms underlying these advantages remain unclear. The current study, therefore, seeks to bring clarity to this issue.

Our study contributes to the literature on supply chain visibility and product innovation in two ways. First, drawing on the absorptive capacity perspective, our study suggests that supply visibility and demand visibility provide focal firms with opportunities to acquire novel external information. However, to transform this information into innovative outcomes, focal firms must have the capability to exploit the information (Delen and Zolbanin, 2018). The two types of visibility facilitate supply chain integration, which develops a focal firm's absorptive capacity that can improve product innovation. This study, therefore, deepens our understanding of the benefits of supply and demand visibility in innovation performance. Second, our study uncovers the mediating roles of three types of supply chain integration in the relationship between supply visibility (or demand visibility) and the firm's innovation activities. To explore the different mediating effects, we divide supply chain integration into three dimensions: supplier integration, customer integration and internal integration. This study, therefore, provides a more holistic picture of how focal firms use supply-side and demand-side information to improve their innovative outcomes by integrating their supply chain partners.

The remainder of this paper is organized as follows. First, we describe the theoretical background, propose the hypotheses and present the conceptual model. Then, we describe the sampling and research method. We discuss the study's theoretical contributions and managerial implications based on the findings. Finally, we explain the study's limitations and offer some suggestions for future research.

2. Literature review and hypothesis development

2.1 Key constructs

2.1.1 Supply visibility and demand visibility

Supply visibility and demand visibility are two important dimensions of supply chain visibility (Srinivasan and Swink, 2018). Supply visibility indicates a focal firm's capability of acquiring supply-side information (e.g. production planning and capacity, order completion status, back order status, delivery schedule and lead time) (Somapa et al., 2018). Accordingly, demand visibility emphasizes the ability to acquire demand-side information (e.g. promotional forecasting, point-of-sale [POS] or actual sales data and customer inventory levels) (Somapa et al., 2018). Firms can achieve the two types of visibility through information systems and technologies. With information system supports, focal firms with high levels of supply visibility and demand visibility can develop more extended contracts among supply chain partners; they also increase flexibility and responsiveness, manage conflict and interdependence and reduce opportunism in the supply chain network (Barratt and Oke, 2007). Prior studies have explored the effects of supply visibility and demand visibility on supply chain performance separately (e.g. Brandon-Jones et al., 2015; Kraft et al., 2018; Yang et al., 2021). Furthermore, considering that the two types of visibility represent a firm's capabilities of acquiring upstream and downstream information flows, a few scholars analyze their effects within single models (e.g. Williams et al., 2013; Srinivasan and Swink, 2018; Sharma et al., 2022). Motivated by these studies, we explore the underlying mechanisms of the influences of supply visibility and demand visibility on innovation in a single model to provide more insights into supply chain visibility. We summarize the literature on the two types of visibility in Appendix 1.

2.1.2 Three dimensions of supply chain integration

Supply chain integration has three dimensions, i.e. supplier integration, customer integration and internal integration. Specifically, in a buyer–supplier context, supplier integration is the degree to which a focal firm with its suppliers' structures inter-organizational strategies, practices and processes into collaborative and synchronized processes (Vanpoucke et al., 2014). Accordingly, customer integration refers to the degree to which a focal firm cooperates with its customers to structure its inter-organizational strategies, practices, procedures and behaviors into collaborative, synchronized and manageable processes (Zhang et al., 2022). Internal integration is the degree to which a focal firm structures its functional strategies, practices and processes into collaborative and synchronized processes (Huo et al., 2016). Recent studies suggest that internal integration may influence supplier and customer integration (Huo et al., 2016; Jajja et al., 2018; Munir et al., 2020). Scholars find that the three dimensions of supply chain integration have different effects on operational and business performance (Flynn et al., 2010; Wong et al., 2013; Zhao et al., 2020). In this study, we test how supplier, customer and internal integration differently mediate the relationship between supply chain visibility and product innovation to provide more evidence regarding their distinct effects.

2.1.3 Exploratory innovation and exploitative innovation

Focal firms typically engage in two types of product innovation: exploratory innovation, which means creating new technologies, products or services, and exploitative innovation, which involves recombining existing knowledge and technologies (March, 1991). Exploratory innovation represents a higher-order outcome of absorptive capacity than exploitative innovation (Jansen et al., 2006). Specifically, exploratory innovation involves searching, risk-taking, experimenting, discovering and making products more flexible (March, 1991). Exploitative innovation entails exploiting current knowledge, experience and technologies to enhance efficiency, reduce costs and improve product design during new product development (March, 1991). Both exploratory and exploitative innovation are essential for firms to gain competitive advantages (Luger et al., 2018).

2.2 Absorptive capacity theory

Cohen and Levinthal (1990, p. 1) define absorptive capacity as a firm's “ability … to recognize the value of external information, assimilate it, and apply it to commercial ends.” According to them, absorptive capacity consists of four processes: acquisition, assimilation, transformation and exploitation. In a follow-up study, Zahra and George (2002) suggest that when absorptive capacity involves knowledge creation and exploitation processes, it represents an information-driven capability. To effectively leverage knowledge, firms need to develop the ability to absorb information from external entities (Cohen and Levinthal, 1990) and combine it with internal knowledge to develop new knowledge (Henderson and Cockburn, 1994). Generally, absorptive capacity is a dynamic process of knowledge creation and utilization that enhances a firm's ability to gain and maintain a competitive advantage. Following the dynamic perspective of absorptive capacity, we suggest that supply visibility and demand visibility represent a firm's capacity to acquire and assimilate external information. The two types of visibility help firms sense the changes in the market and promote supply chain integration to respond to environmental changes. Supplier integration, customer integration and internal integration are important strategic initiatives that influence the firm's absorptive capacity by promoting transformation and exploitation of supply chain information (Schoenherr and Swink, 2012). Firms cannot exploit supply chain information without first acquiring it. Relatedly, firms can acquire and assimilate information but might not be able to transform and exploit it for profit generation. Therefore, we propose that the three dimensions of supply chain integration improve a firm's capabilities of transforming and exploiting the assimilated supply visibility and demand visibility by incorporating them into the firm's operations, thereby improving its innovation performance. Figure 1 shows the proposed conceptual model.

2.3 The relationship between supply chain visibility and supply chain integration

Supply visibility is critical for process coordination and joint practices between a focal firm and its suppliers. Supply visibility helps focal firms sense the changes in supply markets, costs and materials inventories (Srinivasan and Swink, 2018). These changes facilitate focal firms to form a closer collaboration with their suppliers through different operational practices, e.g. integrating production planning, coordinating activities and implementing joint decision-making with their suppliers, thereby promoting supplier integration (Danese, 2013). For example, to reduce the uncertainties and risks caused by the changes in supply markets, the focal firms integrate with their suppliers to improve the decision quality of collaborative planning, forecasting and replenishment (CPFR) (Wu et al., 2014). We thus offer the following hypothesis:

H1a.

Supply visibility is positively related to supplier integration.

Demand visibility enables focal firms to sense the changes in customer and competitor markets, including demands, pricing and promotional actions, and product inventories. To respond to the above changes, the focal firms can facilitate the synchronization of the focal firms' and customers' activities, which include the transfer of ordering responsibility and retailer inventory replenishment from customers to manufacturers (i.e. vendor-managed inventory), the use of jointly developed closed-loop or reverse logistics mechanisms (i.e. the coordinated handling of surplus, used or defective products), and the coordination of the design, development, and introduction of new products and services (Kulp et al., 2004). Therefore, demand visibility creates a closer relationship between focal firms and their customers, thereby promoting customer integration. We thus offer the following hypothesis:

H1b.

Demand visibility is positively related to customer integration.

We posit that focal firms have more opportunities and capabilities to achieve internal integration if they have access to acquire timely, complete and useful information from the supply and demand sides. Supply visibility and demand visibility help employees detect uncertainties in the market and quickly adjust and coordinate product design, manufacturing and other internal activities to react to supply chain risks. Specifically, when focal firms sense supply changes through supply visibility, they will reduce costs and create greater value by promoting collaboration among different functions (Flynn et al., 2010). For example, an integrative inventory management system can help the relevant functions acquire knowledge, such as order picking and inventory and improve inventory management decisions (Huo et al., 2021). Similarly, demand visibility can help focal firms sense changes in the customer markets and facilitate their understanding of customers' needs. The focal firms will better structure their functional strategies, practices, and processes into collaborative and synchronized activities (developing production plans and producing goods on time) to improve delivery performance (Flynn et al., 2010). Therefore, we propose the following hypotheses:

H2a.

Supply visibility is positively related to internal integration.

H2b.

Demand visibility is positively related to internal integration.

2.4 The relationship between supply chain integration and product innovation

A focal firm's supply chain integration comprises inter-organizational (supplier integration and customer integration) and intra-organizational integration (internal integration). The three types of integration promote social interactions in the supply chain network and trigger inter-organizational reciprocity. Supply chain partners will share their privileged knowledge and experiences and engage in the focal firms' new product development practices (Zahra and George, 2002; Zhang et al., 2018). Moreover, the three types of integration help focal firms improve their absorptive capacities by transforming external information into knowledge and exploiting the knowledge for complex and innovative processes (Srinivasan and Swink, 2018). Specifically, focal firms can refine their existing skills and knowledge (exploitative innovation) and use external knowledge to create new abilities and knowledge (exploratory innovation) for better product innovation (Luger et al., 2018). Based on prior studies, e.g. Flynn et al., (2010), Wong et al., (2013) and Jajja et al., (2018), we assume that supplier integration, customer integration and internal integration positively influence innovation performance.

Specifically, integrating with suppliers can help focal firms transform supply-side information into knowledge and exploit the knowledge to improve innovation performance. Suppliers provide focal firms with access to advanced technologies and components and engage in the focal firms' new product development practices (Gao et al., 2015). Thus, focal firms can fully understand the supply market and current technologies to make better R&D investment decisions and mitigate R&D risk. Moreover, by closely working with innovative suppliers, focal firms can acquire or learn new skills or new technology and upgrade their knowledge to produce exploratory innovation and exploitative innovation, e.g. improving productivity, optimizing new product development processes, reducing new product development costs and time, and improving product design (Cheng and Krumwiede, 2018; Wang et al., 2017). These arguments lead to the following hypotheses:

H3a.

Supplier integration is positively related to exploratory product innovation.

H3b.

Supplier integration is positively related to exploitative product innovation.

Moreover, customer integration can help focal firms transform demand-side information into knowledge and exploit the knowledge for making better decisions on product innovation. Customers specify their requirements and needs and provide strategic insights into market expectations and future opportunities (Jha et al., 2017). Focal firms can exploit the knowledge to identify appropriate innovation directions, improve innovation novelty, shorten innovation cycles and reduce the innovation failure rate (Wong et al., 2013). Thus, focal firms' exploratory and exploitative innovation performances have been improved significantly. Moreover, customer integration can promote customers' engagement in focal firms' various activities of new product development, from new idea generation to prototyping. Indeed, extensive literature has shown that customers play a critical role in improving new product development and product innovation (e.g. Wong et al., 2013; Gao et al., 2015; Golini and Gualandris, 2018). Therefore, we propose the following hypotheses:

H4a.

Customer integration is positively related to exploratory product innovation.

H4b.

Customer integration is positively related to exploitative product innovation.

Internal integration can help focal firms identify valuable knowledge from their supply chain partners and exploit this knowledge to create innovation (Zahra and George, 2002; Zhang et al., 2018). Exploiting knowledge to create innovation requires intensive interaction and collaboration among different departments within firms (Stock, 2014; Zhang et al., 2018). Internal integration provides integrated internal routines and processes, such as interdepartmental communication and cross-functional teams, that encourages employees to engage in knowledge transformation and exploitation. Moreover, internal integration can increase the interdepartmental consistency of innovation goals and enhance interdepartmental knowledge regarding the productivity of new product development processes, novel methods and technology when producing exploratory and exploitative innovation (Zhang et al., 2018). Therefore, internal integration is important to facilitate the transformation and exploitation of knowledge into innovation outputs (Lane et al., 2006). We thus propose the following hypotheses:

H5a.

Internal integration is positively related to exploratory product innovation.

H5b.

Internal integration is positively related to exploitative product innovation.

3. Methodology

3.1 Data

To test our model, we collected data from 200 manufacturing firms in China. First, we designed the English version of the questionnaire, which two Chinese researchers then translated into Chinese. The questionnaire was then translated back into English by another Chinese researcher. We checked the original English version to verify the accuracy of the translation. To assess the clarity of the wording, we sent the survey instrument to 40 supply chain managers. Based on the managers' feedback in this pretest, we modified the questionnaire to ensure that all the questions were unambiguous from a practical standpoint. These steps ensure the content validity of our survey instrument.

To ensure our data quality, we collaborated with All China Marketing Research (ACMR), a national research company with considerable expertise and a strong track record in data collection in China (e.g. Bai et al., 2016). ACMR owns a proprietary database called ACMR EIS (Enterprise Information System), which integrates several other databases developed by the Chinese central government to provide basic information on the country's firms (e.g. size, industry and age) and the contact details of their executives. We designed a cover letter that briefly introduced the project and the expected research report and explained how confidentiality would be guaranteed. The cover letter also included the study’s objectives and some guidelines on how to answer the questions. Two respondents were selected from each manufacturer for the period 2018–2019: one top manager (e.g. CEO, VP or general manager) and one middle manager (e.g. operations manager or supply chain manager). We only targeted senior managers who were familiar with the firm's daily processes and were more likely to have a deep understanding of its operations and innovation practices. The top manager provided information on innovation. The middle manager answered questions regarding supply chain visibility, supply chain integration and control variables. In this way, we avoided potential common method bias.

At first, the ACMR recruiters randomly contacted 2,483 firms from the firm's EIS database to inquire about their willingness and qualification to participate. Eventually, 346 qualified firms agreed to participate in the survey and put us in touch with their managers. Eventually, we received 200 usable samples, with a response rate of 8.1%. The profiles of the responding firms are shown in Table 1. We assessed non-response bias following the recommendations of Wagner and Kemmerling (2010), who compared respondents to non-respondents based on their characteristics (i.e. company size, turnover and industry). No significant differences were found in the t-test results (p > 0.05), which indicates that non-response bias was not a threat.

3.2 Measurements

In this study, supply visibility and demand visibility are constructed as second-order multi-item reflective scales. Reflective scales were originally constructed by Williams et al. (2013) and Srinivasan and Swink (2018). We use the same scales to ask the respondents to assess the general degree to which visibility is created with their major suppliers and customers; thereby, we can reduce the biases resulting from the different visibility levels of each major supplier and customer. Regarding supply visibility, there are five first-order factors: supplier inventory information, overall market-level supply information, order information, advance shipment notices, and finished goods location status. Each first-order factor has four items: timeliness, accuracy, completeness and usefulness. The scale of demand visibility consists of five first-order factors: sales information, forecast information, market-level demand information, customer inventory information and promotional information. Each first-order factor contains four items, which ask the respondents to assess the timeliness, accuracy, completeness and usefulness of the demand-side information. The measurement of all the scales in this study relies on a seven-point Likert scale (from strongly disagree to strongly agree). Table 2 summarizes the results of the confirmatory factor analysis (CFA) for supply visibility and demand visibility and their respective first-order constructs. We find that timely, accurate, complete and usefulness are generally equal for each dimension of supply visibility and demand visibility. In the following analysis, we use the four items' average scores for measuring the first-order factors of supply visibility and demand visibility.

We use the measures of supplier integration, customer integration and internal integration from Zhao et al. (2011). We revise these measures to emphasize the interaction and collaboration between focal firms and their major suppliers and customers. Specifically, to represent supplier integration, we capture the level of strategic partnership, the level of participation in procurement and production processes by major suppliers, the sharing of the focal firm’s demand forecast and inventory level, and the help provided to major suppliers to improve their processes to better accommodate the focal firm's needs. To represent customer integration, we capture the level of strategic partnership, the frequency of communication and periodical contact, system connections and the follow-up feedback with major customers. We measure internal integration by asking respondents to assess the level of periodic interdepartmental meetings among internal functions and cross-functional teams, as well as the real-time integration and connection among all internal functions.

We adopt the measures for exploratory product innovation and exploitative product innovation from Zhou and Wu (2010). Specifically, we assess exploratory innovation with items that reflect whether the manufacturing technologies, skills and processes for new product development are entirely new to the respondent's firm. We assess exploitative innovation with items that reflect whether the innovation is related to upgrading current knowledge and practices.

We also include several control variables that may influence innovation. We control for firm size (logarithm of total assets in a million yuan) because larger firms are more capable of investing in supply chain visibility technologies and innovation activities than smaller ones. We control for firm age (logarithm of years since the firm was founded) because older firms tend to have more complicated structures and inertia than smaller ones, which may dampen innovation. We control for organizational structure by using two dummies (centralization and centrally led structure) with the baseline group-decentralized firm. Previous studies have shown that centralization helps to minimize coordination costs when firms conduct innovation activities, such as R&D, and improve product innovation (Leiponen and Helfat, 2011). Detailed information about the measurement items for the related constructs is shown in Appendix 2.

3.3 Measurement model

Covariance-based structural equation modeling (SEM) is used for the empirical examination. Overall, the values of the fit indices are acceptable (χ² (515) = 890.438, RMSEA = 0.06, CFI = 0.92, TLI = 0.91, SRMR = 0.06). All the constructs are subjected to CFA. The results for reliability and convergent validity of the reflective constructs are shown in Table 3. The unidimensionality of the constructs is measured with Cronbach's alpha. As shown in Table 3, the values of Cronbach's alpha vary between 0.82 and 0.92, which are greater than the threshold value of 0.70. Composite reliability (CR) estimates the extent to which construct items share in their construct measurement. As shown in Table 3, our CR measurements are within the recommended threshold (0.7), and each of the constructs in the study is sufficiently reliable. Furthermore, average variance extracted (AVE) estimates are also calculated to confirm convergent validity. As shown in Table 3, the AVE measurements satisfy the recommended threshold (0.5), which means that each construct in the study is sufficiently reliable and has adequate convergent validity. We test discriminant validity by examining the confidence intervals around the latent correlation estimates (Rönkkö and Cho, 2022). We obtain the confidence intervals by estimating a CFA model that includes all the scales. According to Rönkkö and Cho (2022), discriminant validity is not an issue if the confidence intervals around the latent correlation estimates are below 0.8 for positive correlations (above 0.8 for negative correlations). As shown in Table 4, this criterion is fulfilled for the relationships of each pair of constructs. Thus, we can assume that discriminant validity is not a problem for this study. The descriptive statistics and inter-correlations among the variables are shown in Table 5.

To mitigate common method bias, we first collect information on the dependent and independent variables from different informants (Podsakoff et al., 2003). Moreover, we follow Podsakoff and Organ (1986) and separate conceptually related variables in the questionnaire to control the consistency of respondents' answers. Second, we conduct Harman's one-factor test with exploratory factor analysis (EFA) (Podsakoff and Organ, 1986; Podsakoff et al., 2003). The results reveal seven factors with eigenvalues above 1.0, which explain 70% of the total variance; the first factor did not explain most of the total variance (37.1% < 50%) (Hair et al., 2006). Third, we use the respondent’s degree of satisfaction with the personal characters as a marker variable and employ the partial correlation procedure recommended by Lindell and Whitney (2001). This marker has no theoretical relationships with our constructs. We use the minor positive correlation (r = 0.1) between the marker variable and other latent variables as the value to adjust correlations among the variables. We compare this adjusted matrix with our unadjusted correlation matrix. After adjustment, all inter-item correlations remain statistically significant, which indicates that our results are not threatened by the common method bias (Lindell and Whitney, 2001).

4. Results

To test our hypotheses, we use SEM. The estimates are generated with Stata 17. The goodness-of-fit indices of our model are χ² (634) = 1049, RMSEA = 0.06, CFI = 0.91, TLI = 0.90, SRMR = 0.06, which are better than the commonly accepted thresholds (Hu and Bentler, 1999). The results of the hypothesis testing are presented in Figure 2. We find that supply visibility is positively and significantly related to both supplier integration and internal integration (supplier integration: b = 0.492, p < 0.01; internal integration: b = 0.634, p < 0.01), while demand visibility is positively and significantly related only to customer integration (b = 0.729, p < 0.01). Furthermore, customer integration is positively and significantly related to exploitative innovation (b = 0.450, p < 0.01) and exploratory innovation (b = 0.196, p < 0.05). Internal integration is positively and significantly related to exploitative innovation (b = 0.250, p < 0.05) and exploratory innovation (b = 0.196, p < 0.10). However, supplier integration does not significantly relate to exploitative and exploratory innovation. Therefore, hypotheses H1a-b, H2a, H4a-b and H5a-b are supported.

We run the Monte Carlo simulations in AMOS to test the significance of all indirect paths in the proposed model. Preacher and Hayes (2008) emphasize that testing various mediators in one model can effectively reduce bias in parameter estimates by controlling for the other effects included in the model. We test indirect effects with Monte Carlo simulations that generate 95% confidence intervals (CIs) using 20,000 resamples. Monte Carlo simulation with biased corrected confidence intervals provides the most accurate confidence limits and the greatest statistical power for the validity of the mediation effects (MacKinnon et al., 2004; Preacher and Selig, 2012). Statistical significance is assessed by checking whether zero is included in the 95% CI. The results of the mediation test are shown in Table 6. We find that the indirect effect of supply visibility on exploitative innovation via internal integration is 0.086, with a 95% CI [0.011, 0.176], which does not contain zero. Therefore, internal integration mediates the relationship between supply visibility and exploitative innovation. The total indirect effect of demand visibility on exploitative innovation via customer integration and internal integration is 0.195, with a 95% CI [0.087, 0.315]. We find that the indirect effect of demand visibility on exploitative innovation via internal integration is 0.024, with a 95% CI [0.001, 0.074], which does not contain zero. Moreover, we find that the indirect effect of demand visibility on exploitative innovation via customer integration is 0.156, with a 95% CI [0.054, 0.269], which does not contain zero. Therefore, the indirect effect from demand visibility to exploitative innovation through customer integration is significantly greater than that through internal integration (differences = 0.132, 95% CI [0.017, 0.249]).

We address possible endogeneity concerns in the following ways. Since firms with a high level of external integration may have a higher ability to develop supply chain visibility, one concern is potential reverse causality. We thus conduct a two-stage least square (2SLS) regression and use business analytical capabilities as our instrument variable for demand and supply visibility separately. A valid instrumental variable should satisfy relevance and exclusion restriction assumptions. Specifically, the instrumental variable should be uncorrelated with the error (customer integration, supplier integration and internal integration) and correlated with the endogenous regressor (supply visibility and demand visibility). In other words, the instrument should explain the outcome variable only through the endogenous regressor. A firm's business analytical capability is defined as the methodologies, processes, architectures and technologies that transform data into meaningful and useful information (Teo et al., 2016). This capability helps to identify environmental changes and capitalize on emerging opportunities, the need for new products, or an opportunity to penetrate new markets and gather new, external information, thus realizing its value and assimilating it into its functioning (Teo et al., 2016). Furthermore, business analytical capabilities can provide the necessary mechanisms and support to analyze and transfer information across the supply chain and make it available to decision-makers (de Oliveira and Handfield, 2019). Therefore, a firm's business analytical capability may increase supply chain visibility. Regarding supply chain integration, prior studies have confirmed that the intensity of information technology connection and information sharing between firms and their suppliers positively affects logistics integration (Prajogo and Olhager, 2012). In addition to information sharing, supply chain integration requires synchronized planning, operational coordination and strategic partnership (Liu et al., 2016). Therefore, only firms with high business analytical capabilities do not directly affect supply chain integration if their partners in the supply chain have low business analytical capabilities or coordination. A firm's business analytical capability can thus be used as an instrumental variable to address the possible endogeneity concern regarding the relationship between demand visibility and supply visibility.

The measures for the instrumental variable are adopted from de Oliveira and Handfield (2019). Respondents are asked to rate the extent to which professionals in their company (1) use available data to derive business insights, (2) understand statistics, (3) employ statistical analysis to describe business scenarios and problems, (4) make use of statistical analysis to forecast business scenarios and anticipate problems, (5) use statistical analysis to drive actions, (6) exploit the company's stored data and (7) ensure that the quality of the stored data is protected.

To conduct our analysis, we use the “ivreg2” command in Stata. The results are shown in Table 7. In the first stage, we regress demand visibility on business analytical capability and all the control variables. The results show that business analytical capability positively relates to demand visibility (b = 0.284, p < 0.01). The following estimation shows that business analytical capability is an effective instrumental variable for demand visibility (the Kleibergen–Paap rk LM statistic for the underidentification test is 11.505 (p < 0.01), the Cragg–Donald Wald F statistic for the weak identification test is 17.647 and the Stock–Yogo weak ID test critical values: 10% maximal IV size is 16.38). Model 2 shows the results of the second stage, which regresses customer integration on the predicted value of the independent variables obtained from the first stage and the control variables. The results show that demand visibility is positively related to customer integration (b = 0.874, p < 0.01). As shown in Model 3, demand visibility is also positively related to internal integration (b = 1.033, p < 0.01). We conduct a similar analysis for supply visibility, supplier integration and internal integration. The majority of the results are strongly consistent. Table 8 summarizes all the results for the proposed relationships.

5. Discussion and conclusion

In this study, we empirically examine the influences of supply visibility and demand visibility on product innovation (exploratory and exploitative innovation) and explore how supplier integration, customer integration and internal integration mediate these influences. Based on absorptive capacity theory, we propose that supply visibility and demand visibility represent a focal firm's ability to acquire different supply chain information flows, which may not necessarily contribute to innovation performance if the firm cannot exploit this information. Supplier integration, customer integration and internal integration increase a firm's absorptive capability to exploit information and transform supply chain information into innovation. Based on data from 200 Chinese manufacturers, our findings shed light on the relationship between supply chain visibility and innovation performance. Specifically, we find that supply visibility is positively related to supplier integration and internal integration and that demand visibility is positively related to customer integration. Supply-side information reflects the changes in the supply market and facilitates the collaboration between the focal firm and its suppliers and the collaboration between different department. Compared with external integration, internal integration is more flexible and requires fewer communication and transaction processes. Therefore, supply visibility is more effectively for internal integration. Our finding that demand visibility does not significantly influence internal integration is somewhat surprising. One possible explanation is that demand-side information is often of great volume, variety, variability and complexity, which makes it difficult for the focal firms to transform the information into knowledge and exploit the knowledge to update internal information (Yang et al., 2022). Internal integration requires necessary internal information sharing to obtain consistency across different departments. Thus, demand visibility does not have significant influences on internal integration. Furthermore, customer and internal integration positively affect exploratory and exploitative innovation. However, supplier integration is not significantly related to exploratory and exploitative innovation. The literature has shown similar findings of the insignificant effects of supplier integration on innovation, e.g. Primo and Amundson (2002) and Perols et al., (2013). Suppliers' involvement in the new product development practices requires a number of activities, such as negotiating, planning, coordinating, monitoring and enforcing buyer–supplier relationships, which cause high transaction costs, distract the focal firms from internal tasks and halt the innovation process (Perols et al., 2013). Moreover, suppliers' opportunism behavior also reduces the efficiency of a focal firm’s innovation practices (Skowronski et al., 2020). Our work extends the findings of Williams et al. (2013) and Srinivasan and Swink (2018), who emphasize the benefits of supply visibility and demand visibility in terms of efficiency and responsiveness, and it provides new evidence that the two types of visibility serve as critical drivers for product innovation.

5.1 Theoretical contributions

Our findings significantly contribute to the supply chain management and product innovation literature. First, drawing on absorptive capacity theory, our study deepens the understanding of supply chain visibility and confirms that supply visibility and demand visibility influence focal firms' innovation performance through supply chain integration. Although previous studies have long recognized and examined the benefits of supply visibility and demand visibility in operations management (see Appendix 1), few of them have paid attention to their roles in driving product innovation (Somapa et al., 2018). Our findings further suggest that capturing the value of information in the supply chain is not easy and that collaborative practices with supply chain partners are critical for achieving product innovation (Kembro and Selviaridis, 2015). We provide evidence that a focal firm's capability to search, capture and exploit supply-side and demand-side information is an important source of product innovation.

Second, our study has theoretical implications for the supply chain integration literature as it confirms the distinct mediating effects of supplier integration, customer integration and internal integration. The existing literature highlights that the three dimensions of supply chain integration represent collaborative processes within and between firms and their supply chain partners (Williams et al., 2013). Scholars have found that supplier integration, customer integration and internal integration play different roles in supply chain management (e.g. Flynn et al., 2010; He et al., 2014). However, these roles have not been fully examined in the relationship between supply chain visibility and innovation performance. Our findings confirm that customer integration and internal integration enable firms to interpret and synthesize different external information flows across organizational boundaries and promote innovative performance. These findings are strongly consistent with those of Wong et al. (2013) and Ayoub et al. (2017). The comparison of the mediating roles of customer integration and internal integration in the relationship between demand visibility and exploitative innovation further deepens our understanding of the different dimensions of supply chain integration. Supply-side information drives supplier and internal integration mechanisms, while only customer integration and internal integration play important roles in transforming external information into innovation performance. This result confirms prior studies, such as Koufteros et al. (2007), Primo and Amundson (2002) and Perols et al., (2013), which find that supplier integration does not always have a positive effect on innovation performance. We also enrich the existing literature by showing that supply visibility and demand visibility represent two significant antecedents for internal integration.

5.2 Managerial implications

Our study has significant implications for firms leveraging supply chain visibility and integration to improve innovation performance. Our findings confirm that a focal firm’s supply visibility and demand visibility can help integrate supply chain partners. Managers should efficiently invest in technology to capture supplier inventory information and improving market research capabilities. With high supply visibility and demand visibility levels, firms can sense the changes in the supplier and customer markets. To respond to these changes, firms improve their capabilities of using external information to enhance their innovation ability by cooperating more closely with their suppliers and customers and improving internal integration. Managers choose different integration to exploit different types of external information better. For example, managers can find out the most popular products and the new market trends based on purchase order information. To increase the ability of exploiting the demand-side information, firms integrate with their major customers. After discussing their major customers, firms can deeply understand the new trend and add the features of the most popular products to their existing products. In addition, inventory levels provide information on available raw materials and new raw materials. To increase the ability of exploiting the supply-side information, firms integrate with their major suppliers and improve internal integration. Firms can discuss product improvement online with their major suppliers with the new raw materials information. The available raw materials information can also facilitate collaboration among different departments to improve the efficiency of production and operational efficiency. To increase levels of supply visibility and demand visibility, managers should invest in information technology infrastructures, e.g. blockchain, radio frequency identification (RFID), Bluetooth, long-range (LoRa) technology and ultra-narrowband technology (Kalaiarasan et al., 2022). Information technology should be equipped with the functions of promoting data connectivity, achieving real-time data generation and collection, and facilitating data management (Swift et al., 2019).

We suggest that managers should proactively acquire information from supply chain partners and set collaborative practices to better exploit the value of shared information. Specifically, managers should focus on the customer and internal integration to promote the firm's product innovation. Our findings show that compared to supplier integration, the efficacy of customer and internal integration on product innovation is stronger. Both the supply-side and the customer-side information are important sources of innovation performance. However, the possibility of transforming external information into product innovation mainly depends on the customer and internal integration. We suggest that managers keep a long-term and reciprocal relationship with their major customers (e.g. automatic replenishment programs, vendor-managed inventory and customer services) and build an integrated internal structure (e.g. use of enterprise resources planning, lean production systems, searching real-time inventory and operating data). For example, firms can provide incentives for customers to offer suggestions regarding product design and component simplification to promote exploitative innovation in new product development. Managers can improve the usage effectiveness of computer-mediated communication channels, which promote information sharing and collaboration across functional departments and improve the effectiveness of utilizing existing knowledge and resources for new product development. These practices can help integrate and exploit external information, develop innovative products to satisfy customers' needs and achieve the desired benefits of supply chain visibility.

Regarding the mediating effects, to produce exploitative innovation, managers should closely collaborate with customers, integrate their departments to increase the exploitation of demand-side information or integrate their departments using internal integration to exploit supply-side information. Moreover, the comparison of the indirect effects indicates that managers should strategically choose integration mechanisms when exploiting different types of external information for exploitative innovation. Specifically, managers should preferentially choose a customer integration mechanism for exploiting demand-side information and internal integration mechanism for exploiting supply-side information to produce exploitative innovation.

5.3 Limitations and future research

This study has some limitations. First, it uses cross-sectional survey data, which limits the understanding of causality and hides the dynamic nature of supply chain visibility, supply chain integration and product innovation. The degree of supply and demand visibility can change over time. Therefore, longitudinal data with objective measures would provide more solid results concerning the role of the two types of visibility on product innovation. Although this study takes many steps to address the potential reverse causality of the proposed relationships, a longitudinal design and more sources of data would help to further evaluate causality. Second, our exploratory and exploitative innovation measures are based on managers' perceptions. Although these measures have been widely used in previous studies, they may involve social desirability bias. Future research should obtain objective measures (e.g. patents) to corroborate our findings. Third, this study uses data from Chinese manufacturers, which may limit the generalizability of its findings. In China, the role of supply and demand visibility may differ from that in developed economies. Future research should thus collect data from other countries to validate our results. Finally, we focus on the supply visibility and demand visibility induced by a focal firm, which are different from the visibilities created by supplier integration and customer integration. Therefore, future research can examine what type of supply chain visibility is more effective in improving innovation performance.

Figures

Figure 1

Conceptual model

Figure 2

SEM results

Table 1

Profiles of responding firms

Industry	Percentage	Structure	Percentage
Metal, Mechanical and Engineering	29.5	Decentralized	27.5
Electronics and Electrical	17.5	Centralized	39.5
Publishing and Printing	2.0	Center-led	33.0
Chemicals and Petrochemicals	6.0
Building Materials	6.0	Ownership
Rubber and Plastics	3.0	State-owned	25.5
Food, Beverage and Cigarettes	10.5	Privately-owned	46.5
Pharmaceutical and Medicals	6.0	Foreign-owned	16.0
Others	19.5	Joint venture	12.0
Informant profiles
Tenure of the middle manager in the firm (years)		Tenure of top manager in the firm (years)
1–5	38.5	1–5	19.0
6–10	39.5	6–10	29.5
>10	22.0	>10	51.5

Source(s): Author's own work

Table 2

CFA results for demand visibility and supply visibility

Indicator	Direction	Construct	Loadings	S.E.	p
Panel A: Second-order conceptualization of demand visibility and supply visibility
SV1_timely	←	SV1	0.897	0.014	p < 0.01
SV1_accurate	←	SV1	0.868	0.017	p < 0.01
SV1_complete	←	SV1	0.873	0.017	p < 0.01
SV1_useful	←	SV1	0.785	0.027	p < 0.01
SV2_timely	←	SV2	0.847	0.020	p < 0.01
SV2_accurate	←	SV2	0.873	0.017	p < 0.01
SV2_complete	←	SV2	0.869	0.017	p < 0.01
SV2_useful	←	SV2	0.836	0.020	p < 0.01
SV3_timely	←	SV3	0.868	0.017	p < 0.01
SV3_accurate	←	SV3	0.853	0.019	p < 0.01
SV3_complete	←	SV3	0.879	0.016	p < 0.01
SV3_useful	←	SV3	0.841	0.021	p < 0.01
SV4_timely	←	SV4	0.838	0.210	p < 0.01
SV4_accurate	←	SV4	0.867	0.018	p < 0.01
SV4_complete	←	SV4	0.842	0.021	p < 0.01
SV4_useful	←	SV4	0.846	0.020	p < 0.01
SV5_timely	←	SV5	0.873	0.017	p < 0.01
SV5_accurate	←	SV5	0.881	0.016	p < 0.01
SV5_complete	←	SV5	0.870	0.017	p < 0.01
SV5_useful	←	SV5	0.871	0.017	p < 0.01
DV1_timely	←	DV1	0.878	0.016	p < 0.01
DV1_accurate	←	DV1	0.896	0.014	p < 0.01
DV1_complete	←	DV1	0.870	0.017	p < 0.01
DV1_useful	←	DV1	0.821	0.023	p < 0.01
DV2_timely	←	DV2	0.874	0.017	p < 0.01
DV2_accurate	←	DV2	0.888	0.015	p < 0.01
DV2_complete	←	DV2	0.840	0.021	p < 0.01
DV2_useful	←	DV2	0.868	0.017	p < 0.01
DV3_timely	←	DV3	0.895	0.014	p < 0.01
DV3_accurate	←	DV3	0.918	0.011	p < 0.01
DV3_complete	←	DV3	0.875	0.017	p < 0.01
DV3_useful	←	DV3	0.845	0.020	p < 0.01
DV4_timely	←	DV4	0.896	0.014	p < 0.01
DV4_accurate	←	DV4	0.913	0.018	p < 0.01
DV4_complete	←	DV4	0.884	0.016	p < 0.01
DV4_useful	←	DV4	0.882	0.016	p < 0.01
DV5_timely	←	DV5	0.921	0.011	p < 0.01
DV5_accurate	←	DV5	0.934	0.009	p < 0.01
DV5_complete	←	DV5	0.909	0.012	p < 0.01
DV5_useful	←	DV5	0.894	0.014	p < 0.01
SV1	←	SV	0.734	0.037	p < 0.01
SV2	←	SV	0.810	0.030	p < 0.01
SV3	←	SV	0.848	0.025	p < 0.01
SV4	←	SV	0.769	0.033	p < 0.01
SV5	←	SV	0.852	0.025	p < 0.01
DV1	←	DV	0.848	0.025	p < 0.01
DV2	←	DV	0.852	0.025	p < 0.01
DV3	←	DV	0.865	0.023	p < 0.01
DV4	←	DV	0.717	0.039	p < 0.01
DV5	←	DV	0.651	0.046	p < 0.01
Panel B: Using average scores for measuring the five dimensions of demand visibility and supply visibility
SV1	←	SV	0.733	0.037	p < 0.01
SV2	←	SV	0.810	0.030	p < 0.01
SV3	←	SV	0.848	0.025	p < 0.01
SV4	←	SV	0.769	0.033	p < 0.01
SV5	←	SV	0.852	0.025	p < 0.01
DV1	←	DV	0.848	0.025	p < 0.01
DV2	←	DV	0.851	0.025	p < 0.01
DV3	←	DV	0.864	0.023	p < 0.01
DV4	←	DV	0.717	0.039	p < 0.01
DV5	←	DV	0.651	0.046	p < 0.01

Source(s): Author's own work

Table 3

CFA results

Constructs	Items	Loadings	Cronbach's alpha	CR	AVE
SV	SV1	0.8101	0.9002	0.9266	0.7163
	SV2	0.8552
	SV3	0.8767
	SV4	0.8196
	SV5	0.8682
DV	DV1	0.8593	0.8873	0.9209	0.7000
	DV2	0.8558
	DV3	0.8715
	DV4	0.8245
	DV5	0.7679
SI	SI1	0.6545	0.8268	0.8778	0.5915
	SI2	0.7974
	SI3	0.8582
	SI4	0.7777
	SI5	0.7430
CI	CI1	0.7889	0.8871	0.9175	0.6900
	CI2	0.8652
	CI3	0.8288
	CI4	0.8256
	CI5	0.8430
II	II1	0.8003	0.8283	0.8795	0.5941
	II2	0.7734
	II3	0.7033
	II4	0.7619
	II5	0.8103
Exploratory innovation	Exploratory1	0.8777	0.9186	0.9395	0.7566
	Exploratory2	0.8751
	Exploratory3	0.8635
	Exploratory4	0.8593
	Exploratory5	0.8733
Exploitative innovation	Exploitative1	0.7958	0.8565	0.8971	0.6359
	Exploitative2	0.8343
	Exploitative3	0.8067
	Exploitative4	0.7384
	Exploitative5	0.8088

Note(s): Supply visibility (SV); Demand visibility (DV); SI (Supplier integration); Customer integration (CI); II (Internal integration); Average variance extracted (AVE); Composite reliability (CR)

Source(s): Author's own work

Table 4

Discriminant validity

	SV	DV	CI	II	SI	Exploratory	Exploitative
SV	–
DV	0.616 [0.515, 0.717]	–
CI	0.674 [0.581, 0.766]	0.761 [0.685, 0.838]	–
II	0.687 [0.590, 0.783]	0.488 [0.361, 0.615]	0.585 [0.470, 0.700]	–
SI	0.606 [0.493, 0.719]	0.583 [0.468, 0.698]	0.501 [0.372, 0.629]	0.517 [0.384, 0.650]	–
Exploratory	0.415 [0.287, 0.543]	0.460 [0.338, 0.583]	0.364 [0.229, 0.499]	0.387 [0.250, 0.525]	0.323 [0.178, 0.468]	–
Exploitative	0.514 [0.393, 0.635]	0.508 [0.386, 0.630]	0.594 [0.484, 0.705]	0.484 [0.351, 0.616]	0.318 [0.169, 0.468]	0.672 [0.577, 0.767]	–

Note(s): SV=Supply visibility, DV = Demand visibility, CI=Customer integration, II=Internal integration, SI=Supplier integration

Source(s): Author's own work

Table 5

The descriptive statistics and inter-correlations among the variables

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
(1) SV	1
(2) DV	0.562***	1
(3) CI	0.611***	0.688***	1
(4) II	0.596***	0.434***	0.510***	1
(5) SI	0.551***	0.545***	0.459***	0.464***	1
(6) Exploratory innovation	0.363***	0.414***	0.331***	0.340***	0.303***	1
(7) Exploitative innovation	0.449***	0.454***	0.518***	0.411***	0.280***	0.601***	1
(8) Marker	0.284***	0.243***	0.282***	0.326***	0.109	0.242***	0.230***	1
Mean	0	0	0	0	0	0	0	6.005
SD	1	1	1	1	1	1	1	0.916
Max	1.684	1.576	1.266	1.656	1.592	1.356	1.550	7
Min	−4.371	−3.809	−3.331	−3.372	−3.528	−3.930	−2.719	2

Note(s): ^*p < 0.10, ^**p < 0.05, ^***p < 0.01; SV=Supply visibility, DV = Demand visibility, CI=Customer integration, II=Internal integration, SI=Supplier integration

Source(s): Author's own work

Table 6

Monte–Carlo simulation for mediation effect

	Total effect	Direct effect	Indirect effect	95% CI
SV-SI/II-Exploratory	0.205**	0.124	0.081	[−0.031, 0.204]
SV-SI-Exploratory			0.016	[−0.057, 0.090]
SV-II-Exploratory			0.059	[−0.022, 0.156]
DV-CI/II-Exploratory	0.297***	0.284***	0.013	[−0.109, 0.133]
DV-CI-Exploratory			−0.005	[−0.116, 0.103]
DV-II-Exploratory			0.017	[−0.003, 0.064]
SV-SI/II-Exploitative	0.223***	0.147	0.075	[−0.032, 0.200]
SV-SI-Exploitative			−0.040	[−0.108, 0.026]
SV-II-Exploitative			0.086	[0.011, 0.176]
DV-CI/II-Exploitative	0.376***	0.181*	0.195	[0.087, 0.315]
DV-CI-Exploitative			0.156	[0.054, 0.269]
DV-II-Exploitative			0.024	[0.001, 0.074]

Note(s): *p < 0.10, **p < 0.05, ***p < 0.01; SV=Supply visibility, DV = Demand visibility, SI=Supplier integration, CI=Customer integration, II=Internal integration

Source(s): Author's own work

Table 7

The results of 2SLS by using business analytical capabilities as an instrumental variable

	First stage	Second stage		First stage	Second stage
	(1) DV	(2) CI	(3) II	(4) SV	(5) SI	(6) II
DV		0.874***	1.033***
		[0.092]	[0.282]
SV					0.742***	0.844***
					[0.173]	[0.207]
BAC	0.284***			0.350***
	[0.081]			[0.072]
Firm size	0.182	−0.153	0.315	0.223	−0.050	0.459***
	[0.234]	[0.129]	[0.222]	[0.230]	[0.137]	[0.169]
Firm age	0.031	−0.062	−0.101	0.043	0.025	−0.057
	[0.089]	[0.459]	[0.119]	[0.097]	[0.111]	[0.099]
Firm structure dummies	Included	Included	Included	Included	Included	Included
N	200	200	200	200	200	200
adj. R2	0.120	0.448	−0.157	0.160	0.270	0.315

Note(s): (1) SV=Supply visibility, DV = Demand visibility, SI = Supplier integration, CI = Customer integration, II = Internal integration; (2) ^*p < 0.10, ^**p < 0.05, ^***p < 0.01; (3) Standard errors in brackets

Source(s): Author's own work

Table 8

Summary of hypothesis tests

Hypothesis	Results
The relationship between supply chain visibility and supply chain integration
H1a. Supply visibility is positively related to SI	Supported
H1b. Demand visibility is positively related to CI	Supported
H2a. Supply visibility is positively related to II	Supported
H2b. Demand visibility is positively related to II	Rejected
The relationship between supply chain integration and innovation
H3a. SI is positively related to exploratory product innovation	Rejected
H3b. SI is positively related to exploitative product innovation	Rejected
H4a. CI is positively related to exploratory product innovation	Supported
H4b. CI is positively related to exploitative product innovation	Supported
H5a. II is positively related to exploratory product innovation	Supported
H5b. II is positively related to exploitative product innovation	Supported

Note(s): SV=Supply visibility, DV = Demand visibility, SI=Supplier integration, CI=Customer integration, II=Internal integration

Source(s): Author's own work

Table A1

Literature summary of demand visibility and supply visibility

Studies	Definition	Visibility type	Antecedent	Outcome	Theoretical lens	Method
Kraft et al. (2018)	The extent of disclosing supply chain information to consumers	Visibility to consumers		Social responsibility		Behavioral experiment
Srinivasan and Swink (2018)	the availability of rich, timely data	Demand visibility and supply visibility		Cost performance and delivery performance	Organizational information processing theory	Survey
Swift et al. (2019)	“The ability to trace the points of origin of materials used in a product”	Supply visibility		Profitability		Secondary data
Williams et al. (2013)	The extent of access to high-quality information describes various demand and supply factors	Supply visibility, demand visibility, and market visibility		Supply chain responsiveness	Organizational information processing theory	Survey
Kim et al. (2011)	The extent to which partner firms' information/knowledge related to supply chain cooperation is visible to the focal firm through inter-organizational information systems	Buyer's IOIS visibility, supplier's IOIS visibility	Supplier's internal IS integration, buyer's internal IS integration, inter-organizational IT infrastructure compatibility	Buyer's expectation of relationship continuity, supplier's expectation of relationship continuity, joint profit performance	–	Survey
Sharma et al. (2022)	The ability to share or access or gather useful information, for supply chain operations	Information sharing with customers, Information sharing with suppliers		Supply chain performance	Resource-based theory	Survey
Småros et al. (2003)		Demand visibility		Production and inventory control efficiency		Simulation model
Lehtonen et al. (2005)		Demand visibility		Product introductions		Simulation model
Yang et al. (2021)	Manufacturer's access to accurate, timely, complete, and useful information about the major supplier's inventory levels, lead times/delivery dates, advanced shipment notices and inventory levels in the distribution network	Supply visibility	Supply-side digitalization	Supplier opportunism	Social exchange theory, transaction cost economics	Survey
Brandon-Jones et al., (2015)	Identifying and understanding inventory and demand levels across the upstream supply chain, improves an organization's capability to process this information	Supply visibility		Plant performance	Organizational information processing theory	Survey

Source(s): Author's own work

Table A2

Measurement items

Construct and items
Please indicate the extent to which decision-makers in your supply chain organization have access to important operational information of the major suppliers in the following categories
Supply visibility
SV1: The supplier inventory information is …
… timely (current)
… accurate
… complete (all the information we need)
… in a useful format
SV2: The overall market level supply information is …
… timely (current)
… accurate
… complete (all the information we need)
… in a useful format
SV3: The order information we receive from major suppliers is …
… timely (current)
… accurate
… complete (all the information we need)
… in a useful format
SV4: The advanced shipment notice information we receive from major suppliers is …
… timely (current)
… accurate
… complete (all the information we need)
… in a useful format
SV5: We have information regarding finished goods' location status in the distribution network (e.g. distribution centers, transportation) is …
… timely (current)
… accurate
… complete (all the information we need)
… in a useful format
Please indicate the extent to which decision-makers in your supply chain organization have access to important operational information of the major customers in the following categories
Demand visibility
DV1: The sales information we receive from major customers is …
… timely (current)
… accurate
… complete (all the information we need)
… in a useful format
DV2: The forecast information we receive from major customers is …
… timely (current)
… accurate
… complete (all the information we need)
… in a useful format
DV3: The market-level demand information we gather is …
… timely (current)
… accurate
… complete (all the information we need)
… in a useful format
DV4: The customer inventory information is …
… timely (current)
… accurate
… complete (all the information we need)
… in a useful format
DV5: The promotional information we receive from major customers is …
… timely (current)
… accurate
… complete (all the information we need)
… in a useful format
Please indicate the extent of integration between your organization and your major customers in the following areas
Customer integration
CI1: The level of a strategic partnership with major customers
CI2: The level of communication with our major customers
CI3: The establishment of a quick ordering system with our major customers
CI4: Follow-up with our major customers for feedback
CI5: The frequency of periodical contacts with major customers
Please indicate the extent of integration between your organization and your major suppliers in the following areas
Supplier integration
SI1: The level of the strategic partnership with our major suppliers
SI2: The participation level of our major suppliers in the process of procurement and production
SI3: We share our demand forecast with our major suppliers
SI4: We share our inventory level with our major suppliers
SI5: We help our major suppliers to improve their process to meet our needs better
Please indicate the extent of integration of your organization in the following areas
Internal integration
II1: Integrative inventory management
II2: Real-time searching of logistics-related operating data
II3: The utilization of periodic interdepartmental meetings among internal functions
II4: The use of cross-functional teams in new product development
II5: Real-time integration and connection among all internal functions from raw material management through production, shipping, and sales
Please indicate the degree to which you agree or disagree with your firm's innovation performance statements
Exploitative innovation
Exploitative1: In the new product development processes, our firm has upgraded current knowledge for familiar products
Exploitative2: In the new product development processes, our firm has invested in exploiting mature technologies that improve the productivity of current innovation operations
Exploitative3: In the new product development processes, our firm has enhanced abilities in searching for solutions to customer problems that are near to existing solutions
Exploitative4: Upgraded skills in product development processes in which the firm already possesses rich experience
Exploitative5: Strengthened the knowledge and skills to improve the efficiency of existing innovation activities
Exploratory innovation
Exploratory1: Acquired manufacturing technologies and skills entirely new to the firm
Exploratory2: Learned product development skills and processes entirely new to the industry
Exploratory3: Acquired entirely new managerial and organizational skills that are important for innovation
Exploratory4: Learned new skills in funding new technology and training R&D personnel
Exploratory5: In the new product development processes, our firm has strengthened innovation skills in areas where it has no prior experience
Firm size
Log of total assets in a million yuan
Firm age
Log of years since the firm was found
Organization structure
How would you describe your organization structure (Choose one of the three)
Decentralized
Centralized
center-led

Source(s): Author's own work