Aligning digital investment with CSR in dual-channel operations: a game-theoretic analysis of supply chain performance Open Access

The advancement of information technology and the development of digital technologies have driven the rapid growth of e-commerce. Traditional single sales channels are gradually unable to meet consumers' diverse demands. To expand market share, enterprises have gradually opened online channels alongside traditional retail, forming a dual-channel supply chain model that has become a standard model in current supply chain management (Zhang et al., 2024a). The advantage of this model lies in its ability to meet the needs of a wide range of consumers, providing a flexible, convenient purchasing experience and expanding enterprises' market coverage. At the same time, both upstream and downstream enterprises in the supply chain can benefit from the new channels and achieve a win-win situation under specific coordination (Cai, 2010). The coexistence of online and offline channels has intensified competition and conflict among channel members, rendering the coordination of dual-channel systems a significant challenge in management practice. In response, enterprises are actively exploring optimization pathways, beginning with internal digital investments to enhance operational capabilities. In the current digital era, digitalization has also created more opportunities to generate value in supply chains, especially in e-commerce. An increasing number of enterprises are focusing on digital development and attempting to establish their own competitive advantages. Meanwhile, integrating digitalization into supply chains can help enhance perception capabilities and better coordinate the supply chain (Jena and Singhal, 2023; Agrawal et al., 2019). Xiao et al. (2020) contend that enterprises must invest substantially in digital technologies to secure and expand their market share. In online shopping, consumers rely entirely on product information presented on websites to obtain product insights, which also enables enterprises to provide consumers with better consumption scenarios and experiences through digital technology (Zhao et al., 2024). Digital investment refers to the amount of investment in digital technology applications made by manufacturers or retailers (Zhang et al., 2024a). Driven by such investment, brand owners enhance user experience through the application of digital technologies. For example, Nike's official website launches Nike Fit and Nike By You, using AR size measurement to customize shoes for users; Tesla customers can view the production stage, shipping location, and expected delivery time of the vehicle in real time after placing an order on the official website. However, while digital investment enhances efficiency, it simultaneously reshapes the existing channel ecosystem. While digital empowerment bolsters online platforms, it potentially marginalizes offline channels, intensifying channel conflict. Consequently, firms must leverage brand equity to maintain consumer loyalty amidst these technology-driven pressures.

Following a period of intense focus on economic performance, environmental performance, and social welfare, these areas have gradually gained prominence. The theory of creating shared value (CSV) proposed by Michael E. Porter and Mark R. Kramer holds that enterprises should set their goals to create shared value, which means generating both economic and social value (Porter and Kramer, 2011). Corporate social responsibility (CSR), as a vital practice in modern enterprise management, precisely becomes the key “soft power” for reshaping competitive advantages in the digital era. It not only shapes a company's image but also drives product sales through consumer brand recognition and loyalty. With growing public concern for environmental protection and social equity, research increasingly indicates that consumers are willing to pay a premium for products associated with CSR attributes (Panda, 2014; Modak et al., 2014, 2019). Research indicates that corporate CSR implementation positively impacts business performance (Wang et al., 2015), making such practices a strategic means to enhance competitiveness. At the same time, digital supply chain announcements that disclose corporate social responsibility information have been found to elicit positive and significant market responses (Liu et al., 2023). Donation is one of the core corporate social responsibility activities that enterprises are increasingly participating in. Enterprises can build a strong corporate image and enhance their reputation through donations (Modak et al., 2019). Increasingly, enterprises are making charitable donations. TOMS has significantly improved the living conditions of children worldwide through its iconic giving programs, such as the “One for One” model and its subsequent evolution into a profit-sharing donation framework. ANTA established its own charitable foundation to provide regular donations to impoverished regions. Similarly, Patagonia maintains its “1% for the Planet” environmental initiative, pledging 1% of total sales to global environmental organizations.

Although digital investment or digital transformation brings many advantages and enhances enterprise competitiveness, it also introduces new challenges. When manufacturers choose to enhance supply chain efficiency and meet new demands through digital investment, they inevitably intensify conflicts of interest and competition for resources with offline channels. For instance, during Nike's digital transformation towards a DTC model, it focused on developing its own app, thereby significantly reducing the resources allocated to traditional wholesale channels. Under such circumstances, Corporate social responsibility has become a key strategic tool for reshaping consumers' willingness to pay. However, the decision-making process between digital investment and CSR is no longer isolated. The interaction between the two has begun to emerge in real-world enterprises. For example, Alibaba has established a digital traceability system to support its charity initiatives; while Patagonia is engaged in environmental protection efforts, its endeavors in mobile application innovation and enhanced e-commerce platforms have also provided consumers with a seamless shopping experience and personalized marketing. Therefore, integrating digital investment and CSR practices and exploring the synergy between the two across different fulfillment entities has become an important issue to address in supply chain management in the digital era. Based on the above background, this study explores the following problems:

This study aims to discuss the impact of CSR behavior and digital investment on dual-channel supply chain management, considering two behavioral images of manufacturers. Digital investment directly affects the online channel, while CSR practices, such as social donations, improve brand image and impact both online and offline channels. Therefore, this study constructs a “manufacturer-operated online + retailer” dual-channel pricing, service, and digital investment framework that integrates two factors typically discussed independently into a single model. Using the research method of Stackelberg game theory, this study explores the relationships between CSR practices and the decisions and profits of supply chain members, as well as between the level of digital investment and their decisions and profits. Then, a wholesale price contract is adopted to coordinate the supply chain, thereby further achieving a win-win situation. Furthermore, this study extends a model of retailers' CSR fulfillment. In addition, it explores whether the impact of digital investment on decision-making changes when the CSR fulfillment subject is replaced, as well as which subject is better at fulfilling CSR.

This study fills a research gap in the dual-channel supply chain by examining the combination of CSR practices and digital investment. The core innovation lies not only in exploring the multi-dimensional impacts of the two behavioral strategies on the supply chain but also delving deeper into the influence of different CSR fulfillment entities on digital investment, member decisions, and supply chain performance. Thus, it provides theoretical guidance and decision-making references for enterprises to plan for commercial efficiency and social value in the complex, changing channel game environment. Specifically, the key findings are concluded as follows. Manufacturer's CSR donations influence profits and decisions depending on consumer CSR sensitivity; manufacturer's digital investment in online channels raises online prices, weakens offline channels, and reduces retailer profits; manufacturer's profits correlate positively with digital investment under certain CSR sensitivity conditions. Decentralized decision-making leads manufacturers to develop online channels at the expense of offline demand; compared to centralized decisions, manufacturer decisions improve, whereas retailer decisions decline. When consumers have low sensitivity to CSR, retailer-led CSR can increase the overall supply chain profits and encourage manufacturers to invest in digitalization; otherwise, manufacturer-led CSR is optimal. The relationship between digital investment and CSR, however, is the opposite of how manufacturers implement CSR.

The organizational structure of the subsequent part of this article is as follows: The next section will conduct a literature review. The third section introduces model symbols, assumptions, and model construction. The fourth section will examine the impact of CSR and digital investment under decentralized decision-making and compare it with centralized decision-making. The fifth chapter explores the coordination mechanism based on wholesale price contracts. The sixth chapter constructs an expanded model for retailers to fulfill CSR and conducts a comparative analysis. The seventh chapter summarizes the entire content and looks forward to future research.

This research is closely related to digital investment in supply chains, CSR, dual-channel supply chains, and supply chain coordination. Consequently, this section reviews relevant studies across these four streams.

With technological innovation and development, digital investments and big data applications have become key drivers of improvements in supply chain efficiency and cost reduction. In recent years, more scholars have conducted research on this topic. Xin et al. (2024) noted that digital investment is key to driving enterprises' green transformation, enabling them to reduce costs, increase sales, and improve overall supply chain performance through upstream and downstream cooperation. Digital empowerment also improved service quality and collaborative innovation, thereby positively impacting the profits of the platform service supply chain (Peng et al., 2025). Regarding the performance impact mechanism, Wang and Prajogo (2024) found that supply chain digitization is positively correlated with enterprise performance, and this relationship is influenced by the supply chain's flexibility and innovation capabilities. Kang et al. (2024) explored the low-carbon environmental impact of digital supply chains, where digital investment primarily manifests as investments in blockchain technology and green production technologies. The research results showed that applying blockchain technology increased information transparency and digital operations, and, under government incentives, could significantly increase supply chain demand and profits. In the context of the cold chain supply chain, Zhang et al. (2023) also focused on digital investment in blockchain applications. The latest research turns the perspective of digital investment to digital twin technology. Yang et al. (2026) studied the introduction of digital twin technology into the secondary return supply chain and found that a high product mismatch rate and return loss would prompt retailers to use digital twin technology. And only when the cost of implementing digital twin technology is low, or the efficiency is high, can the win-win result of the supply chain be achieved. From the perspective of game decision-making, the selection of digital investment subjects is also key to influencing pricing and profits. Fan et al. (2024) found that when digital investment is incorporated into pricing and operational decisions, the investment and use of big data are conducive to alleviating information asymmetry and releasing more demand potential. In the dual-channel context, Zhang et al. (2024a) used a differential game to study three different scenarios of digital investment entities. The results showed that when the manufacturer leads the investment, it usually brings higher profits. However, it compresses the retail end's revenue and introduces cost- and information-sharing contracts to improve coordination and achieve Pareto optimality. Li et al. (2025) focused on digital investment integrating blockchain and big data and found that blockchain investment enhances demand and creates value through a trust mechanism, while the dominant party can leverage its power advantage to capture most of the incremental benefits. The analysis of the electric vehicle supply chain also points out that digitalization and investment in emission reduction can create a demand premium but require cost control, and their benefits are highly dependent on the supply chain structure and external policies (Chakraborty et al., 2025). In addition, many scholars take the long-term impact of digital investment into account when conducting research on sustainability. Jena and Singhal (2023) explored how digitalization, sustainability, and CSR affect supply chain pricing decisions, where digital investment decisions represent a series of investments and applications of digital technologies, such as artificial intelligence, big data, and blockchain; their research found that under digital transformation, the overall profits of the supply chain increased significantly, and the profits of retailers and manufacturers also improved under different power models. Wei et al. (2024) found that manufacturers' investments may not yield immediate results. However, in the long term, through digital technology investment, it can significantly increase product market share and overall profits. Ouyang et al. (2025) used digital investment as a mediating variable in their empirical study and found that digital investment could significantly enhance the positive promotion effect of corporate competition risks on CSR, but it could not alleviate the negative impact of financial risks on CSR. Research on the role of government shows that government subsidy policy can effectively promote manufacturers to apply blockchain technology in a dual-channel supply chain, which has a positive role in promoting the strategies and profits of all parties. Benefiting from the improvement of trust brought by information transparency, social welfare has also been significantly improved under the government innovation subsidy. This conclusion has been verified in the studies of Zhong et al. (2023) and Kang et al. (2024). Currently, most research on digital investment treats it as a single decision factor in the model and conducts studies. Meanwhile, in the research on the combination of digital investment and CSR, scholars also treat digital investment as a decision. At the same time, CSR is primarily presented as consumer surplus. There is a lack of studies that incorporate CSR behaviors, especially social donations, into the model.

CSR activities, as part of corporate social responsibility, not only impact the enterprise's social image and brand value but also indirectly influence various supply chain links by influencing consumers' purchasing decisions. In the CSR literature, most scholars categorize enterprises' CSR fulfillment into two types: one involves decision-making as an investment behavior, and the other focuses on consumer surplus. Modak et al. (2019), in their research on closed-loop supply chains, specifically defined CSR activities as social donations by enterprises, incorporated them into the profit formula for modeling and analysis, and concluded that social welfare can affect, and even significantly improve, channel performance. Liu et al. (2021) further compared the dual-retailer competition model across different CSR investment entities. They concluded that, compared with no CSR investment or only one party's investment, when both parties invest in CSR, the overall system performance is better. Zheng et al. (2023) found that CSR investment plays a key role in retailer decisions, especially in the manufacturer-recycling closed-loop supply chain. At the same time, retailers are more inclined to invest in CSR under centralized decision-making, thereby creating a win-win situation. Xin et al. (2022) in the dual-channel research regarded CSR as a decision variable and found that CSR investment would affect the pricing and service levels of supply chain participants, and the CSR level and the profits of participants showed a positive correlation; in the consumer group with high CSR sensitivity, the impact of CSR would be more significant. Wang et al. (2024) incorporated a joint strategy of CSR and carbon emissions into a dynamic supply chain game model. They found that CSR investment helps reduce supply-chain emissions and further improves overall social welfare. Jena and Singhal (2023) integrated CSR in the form of consumer surplus into the profit function, and the study showed that CSR increases the manufacturer's profits by enhancing consumer surplus; at the same time, CSR investment in centralized decision-making also significantly improved the overall performance of the supply chain. Cong et al. (2023) also examined CSR in terms of consumer surplus and reached similar conclusions. Cheng et al. (2025) studied the CSR focus of e-commerce platforms. They concluded that the CSR focus of e-commerce platforms significantly influences the introduction of store brands, and that CSR investment can mitigate potential adverse effects, thereby achieving a win-win situation for the platform and the manufacturer. Ni et al. (2025) focused on the closed-loop supply chain with asymmetric demand information and found that the manufacturer is responsible for both CSR and recycling to achieve the highest level of CSR engagement, but the retailer is responsible for CSR engagement, and the manufacturer is responsible for recycling, which is the best strategy. Most previous studies have examined CSR activities in decision-making through CSR investment or consumer surplus, but have not examined their combination with digital investment behaviors, especially in the context of concretizing CSR activities as social donations.

Consumers are increasingly conducting multidimensional research and discussions on dual-channel supply chains, covering aspects such as pricing strategies, service levels, channel preferences, and showroom effects. Li et al. (2024) examined the selection of online sales formats for green products in dual-channel settings, focusing on whether choosing an agency sales model versus a resale model affects green product pricing and sales performance. Their findings indicate that, at low agency fees, agency sales enhance product greenness and benefit manufacturers, whereas the resale model boosts product greenness across both channels. Wang et al. (2025) found that the cost of warranty services and the proportion of direct channel buyers significantly affect pricing strategies. Guo et al. (2020) analyzed the impact of pre-sale services and online delivery cycles on the design of dual-channel supply chains. They considered three supply chain designs: opening only an online channel, opening only an offline channel, and opening both channels simultaneously. The research found that in a decentralized setting, the dual-channel strategy is not always optimal, and factors such as retailer costs and customers' acceptance of the online channel must be considered. Manufacturers strategically utilize online channels to adjust pricing and service strategies to increase overall supply chain profits. Barman and Sana (2025) introduced carbon emissions, risk aversion, and government subsidies into the dual-channel sustainable supply chain for research. Additionally, Bian and Xiao (2024) proposed different types of return freight strategies based on showroom-effect research and demonstrated that these strategies can effectively mitigate the showroom effect and encourage consumers to purchase products through traditional channels. In Li et al. 's (2019) research, the showroom effect was also considered, and the conclusion was that Service efforts can regulate the showroom effect. Zhao et al. (2023) compared the dual-channel retail strategy and the omnichannel BOPS model and analyzed the impact of freshness perceptions on consumer behavior and pricing; the study showed that the BOPS model can effectively alleviate the negative impact of freshness perceptions. Zhang et al. (2024a) explored the impact of digital technology on the dual-channel supply chain. They concluded that sharing digital technology investment can enhance the overall performance of the supply chain. Zhong et al. (2023) introduced blockchain technology into the dual-channel supply chain and found that both blockchain and government innovation subsidies can enhance the supply chain's overall performance. Most current studies treat digital investment and CSR as a single consideration in the dual-channel supply chain, lacking research on their combined impact, especially in contexts where manufacturers adopt CSR practices, such as social donations.

Supply chain members often make decisions to maximize their own interests, which may negatively affect the supply chain's overall performance. Therefore, to achieve optimal supply chain coordination and a win-win situation, contracts are often designed to deliver the best results for the supply chain system. Supply chains usually adopt mechanisms such as cost and revenue sharing to achieve Pareto improvements in profits, enabling members to achieve optimal outcomes under decentralized decision-making. Yu et al. (2024) proposed a cost-sharing mechanism to coordinate decision-making in a dual-channel closed-loop supply chain, helping both parties achieve optimal recovery and pricing decisions and ensuring win-win outcomes and collaboration among supply chain members. Zhang et al. (2024a) used their research to better balance profit distribution between manufacturers and retailers, adopting not only cost-sharing coordination but also information sharing, thereby achieving Pareto optimality for the supply chain as a whole. Modak et al. (2019) used a two-part coordination contract to explore in depth the channel coordination issue between manufacturers and retailers. While Yang et al. (2025) proposed a hybrid contract combining cost sharing and two-part charging to address the negative impact of farmers' overconfidence on the supply chain. This contract better achieves benefit coordination and avoids decision-making deviations caused by overconfidence. Wu et al. (2025) introduced an improved two-part charging contract that features a traceability system and fairness considerations. The traceability mechanism increases transparency and participation in recovery behavior, thereby enhancing the sustainability and environmental benefits of the supply chain. At the same time, this contract focuses on the fairness of profit distribution for retailers, avoiding the negative impact that excessive focus on fairness can cause. In contrast, traditional two-part charging contracts usually focus only on the allocation of fixed and variable costs, without addressing fairness issues in profit distribution. Panda (2014) explored the channel coordination mechanism for manufacturers and retailers that incorporate corporate social responsibility, adopting a revenue-sharing contract to achieve it. Liu et al. (2024) also proposed a revenue-sharing contract, aiming to coordinate the interests of manufacturers and cloud service providers in the cloud service supply chain. This contract encourages both parties to cooperate to ensure data security, improve overall efficiency, and foster coordination. Zhao and Ma (2022) proposed a combined revenue-and-cost-sharing mode to coordinate the battery supply chain; this approach can reduce the dual marginal effect arising from decentralized decision-making, increase recovery volume and demand, and better meet environmental requirements. Ren and Hu (2024) adopted a Nash bargaining-based cooperation model to coordinate the interests of supply chain members, especially to address wholesale price constraints, enabling the supply chain to achieve optimal profit distribution with government subsidies. Dong et al. (2023) proposed a cooperative-competitive newsboy model and introduced a wholesale price contract to coordinate the competition and cooperation among supply chain members. This study will use the wholesale price contract to coordinate the supply chain and identify a feasible range of wholesale prices to achieve a win-win outcome.

Most of the existing literature regards digital investment and CSR practices in the supply chain as two independent research lines, or regards CSR as a focus on consumer surplus, and only considers its impact on the profit function. Furthermore, most demand functions directly incorporate either digital investment or CSR as factors within linear demand models. Most existing research lacks exploration of the combined effects of CSR behavior and digital investment in dual-channel supply chains, particularly when manufacturers engage in CSR activities, such as social donations. This research differs from prior studies by first integrating digital investment and CSR into a unified model. CSR implementation is defined concretely as social donation behavior, while manufacturers' digital investment focuses specifically on online channels, including, but not limited to, blockchain technology applications. Second, considering dual-channel supply chain design, we incorporate consumers' perceived product value. By accounting for differences in consumer utility across channels, we derive demand functions for online and offline channels. Notably, manufacturers' digital investments impact only the online channel, while CSR implementation influences all channels. At the same time, this study does not stop at a single scenario analysis. Through the comparative analysis of the benchmark model of manufacturers' performance of CSR and the expansion model of retailers' performance of CSR, it portrays the impact of changes in CSR subjects on the overall supply chain and digital investment decision-making, and analyzes the advantages and disadvantages of different subjects' performance of CSR. Thus, within the CSR and digital investment context, we construct a manufacturer-led dual-channel pricing-service-digital investment framework: “Manufacturer-operated online + offline retailers.” To address differences between decentralized and centralized models, we coordinate the entire supply chain through wholesale price contracts.

To facilitate a more thorough description of the model and subsequent research on related issues, we have defined all relevant symbols in Table 1.

We consider a supply chain consisting of manufacturers and retailers, where manufacturers sell products through both online and offline channels. In the offline channel, manufacturers sell products to retailers at a wholesale price $w$⁠, and retailers sell them to consumers at a price $pr$ through physical stores. Simultaneously, manufacturers operate an online channel to sell products directly to consumers at a price $pm$⁠, delivering them via logistics at a unit transportation cost $τ$⁠. We assume consumers are aware of both online and offline channels and assign value to products purchased offline, denoted by $v$⁠. Following Chiang et al. (2003) and Guo et al. (2020), $v$ is uniformly distributed over [0, 1] with density 1. To capture channel preference, the total utility derived from the online channel is $δv$⁠, where $δ$ ranges from 0 to 1 and represents the consumer's preference for the online channel. For simplicity, $δ=12$ is adopted in this study (Gao et al., 2025). This study examines the CSR behavior of manufacturers, focusing on the practical manifestation of social donations made by enterprises. Huang et al. (2026) demonstrated that corporate social responsibility initiatives would stimulate consumers' socially responsible consumption, especially when enterprises make social donations. Moreover, in line with the research model of Modak et al. (2019), the impact of social donations is incorporated into the consumer utility function: for each product sold, a donation of $λ$ proportion of the amount (the level of CSR investment). This study assumes that the manufacturer acts as the leader in the game. Panda et al. (2017) noted that when the manufacturer is the channel leader, its corporate social responsibility goals align with channel performance. And the CSR behavior expressed through social donations is based on the brand-level investment. Therefore, to simplify the model and highlight the core variable research, we set the sensitivity of consumers to CSR in both channels as the same parameter and use $ρ$ to represent it. Simultaneously, manufacturers make digital investment decisions $d$ targeting online channels, such as VR virtual experiences, personalized push notifications, and AI-powered customer service assistants. This investment is directly perceptible only to online channel consumers, and manufacturers bear the associated digital investment $cd=12k1d2$ (Jena and Singhal, 2023), where $k1$ represents the digital investment cost coefficient. Retailers invest in offline service levels $s$⁠. Consumers in the retail channel can experience products and services immediately, and retailers bear the service level cost $cs=12k2s2$⁠, where $k2$ represents the service cost coefficient. This assumption is widely used in literature, such as in Gu et al. (2023), Xin et al. (2022), and Chen and Wu (2024).

Based on the above discussion, we model the total utility of consumers' offline purchases as

Meanwhile, the total utility of consumers' online channel purchases is

Where consumers' sensitivity to service level and digital investment is represented by $θ$ and $γ$⁠, respectively, and sensitivity to CSR is denoted by $ρ$⁠. We assume online and offline consumers have identical sensitivity to CSR. When $Ur>Um$⁠, consumers choose offline purchasing; they opt for online purchasing. We define v₁ and v₂ as the indifference values of v when $Ur=Um$ and $Um=0$⁠, respectively. Calculations yield $v1=γd−θs+pr−pm1−δ$ and $v2=pm−γd−ρλδ$⁠. To ensure the product is sold through both channels, $0<v2<v1<1$ must hold, satisfying the following inequality:

Finally, the purchase demand expressions for the offline and online channels are expressed as follows:

The game sequence for this model is as follows: the manufacturer decides on the wholesale price, the digital investment level, and the online price. After learning the manufacturer's decision results, the retailer decides on the offline price and service level. To facilitate discussion, we assume that the manufacturer's unit production cost is zero (Hsieh and Lathifah, 2024). Under decentralized decision-making, the profit functions of the manufacturer and the retailer are, respectively:

When the manufacturer engages in CSR behavior (⁠$λ$ >0), the optimal decision solution and optimal profit of the model are as follows:

Proof of Proposition 1. The complete proof is relegated to Appendix A1.

By observing the optimal decision solution and the optimal profit, it can be seen that the sales price, the level of digital investment, the service level, and the profit are all related to the coefficient of the cost item, consumer sensitivity, and CSR, among other factors. However, the optimal wholesale price is only related to CSR and the consumer's CSR sensitivity, shows a positive correlation, and is not affected by digital investment or service level.

Based on Proposition 1, we conducted in-depth research and analysis of the relationships between CSR and decisions/profits, and between digital investment and decisions/profits.

1. When $0<ρ<2k1(k2−θ2)+γ2(−5k2+4θ2)k2γ2+2k1(−k2+θ2)$⁠, $pmD*$ decreases with $λ$⁠; when $ρ>2k1(k2−θ2)+γ2(−5k2+4θ2)k2γ2+2k1(−k2+θ2)$⁠, $pmD*$ increases with $λ$⁠.

2. When $0<ρ<1$⁠, $prD*$ and $sD*$ increase with $λ$⁠, $dD*$ decreases with $λ$ ; when $ρ>1$⁠, $sD*$ decreases with $λ$⁠, $dD*$ increases with $λ$⁠.

Proof of Corollary 1. The complete proof is relegated to Appendix A2.

Online and offline prices, retailer service levels, and manufacturer digital investments all exhibit nonlinear relationships with corporate social responsibility (CSR) fulfillment, with specific impacts contingent upon consumer CSR preferences. When consumer CSR preferences are low, $pm$ and $d$ correlate negatively with $λ$⁠, while pr and s correlate positively with $λ$⁠. Conversely, when consumer CSR preferences are high, pm and d correlate positively with $λ$⁠, while s correlates negatively with $λ$⁠. Wholesale prices consistently exhibit a linear negative correlation with $λ$⁠. Conclusions align with findings from Qi et al. (2025), indicating that high CSR preferences tend to drive price increases and sustained investment.

When consumers show no significant preference or sensitivity toward corporate CSR behavior, manufacturers' social donations increase costs while moderately enhancing brand image. Offline retailers may leverage this to improve service levels and consumer experience, thereby better and more directly showcasing these socially responsible brand images to consumers, who become willing to pay additional fees. Manufacturers' online channels, however, face intense price competition pressure. They attract consumers through price reductions while controlling other investment expenditures to manage costs. Conversely, when consumers exhibit clear preferences or heightened sensitivity to CSR actions, manufacturers are motivated to increase their social contributions to enhance brand image and market share. Consumers become willing to accept and pay higher prices, while manufacturers also gain the incentive and resources to invest in digitalization. Due to heightened consumer focus on social contributions, the influence of retailers' additional services on consumer purchasing decisions weakens, resulting in a negative correlation.

When $0<ρ≤1$⁠, $ΠrD*$ increases with $λ$⁠, $ΠmD*$ decreases with $λ$ ; when $ρ>1$⁠, $ΠrD*$ decreases with $λ$⁠, $ΠmD*$ increases with $λ$⁠.

Proof of Corollary 2. The proof of Corollary 2 is similar to that of Corollary 1.

When consumers have a low level of CSR preference, social donations will reduce manufacturers' profits and increase retailers' profits; when consumers have a high level of CSR preference, social donations will increase manufacturers' profits and decrease retailers' profits. CSR has a positive impact on profits, but the specific impact depends on market demand and CSR perception preferences. The conclusion of Xin et al. (2022) also states that when consumers are highly sensitive to CSR, investing in CSR increases profits.

When consumers are not sensitive to CSR behavior, manufacturers cannot offset the cost of donations through higher prices, and social donations will negatively impact manufacturers' profits; however, retailers may attract consumers through higher service levels or product value-added, thereby increasing sales. In this case, social donations help enhance brand image and increase consumers' recognition of the brand. When consumers have a clear preference or are relatively sensitive to CSR behavior, consumers are willing to pay higher prices to support brands with social responsibility, and manufacturers can thus obtain higher gross margins; in this case, manufacturers may raise prices to compensate for the cost of social donations, while retailers face higher purchase costs and pricing pressure. High prices will inhibit the consumption of price-sensitive groups, and the influence of additional services and product add-ons on consumers' purchase decisions will also weaken, resulting in compressed profits for retailers.

Incorporating the manufacturer's digital investment level d as a known variable into the model yields the following expression for the optimal solution of $d$⁠:

Proof of Proposition 2. The proof of Proposition 2 is similar to that of Proposition 1.

$pr(d)$ and $s(d)$ decrease with $d$⁠, $pm(d)$ increases with $d$⁠.

Proof of Corollary 3. The proof of Corollary 3 is similar to that of Corollary 1.

Offline prices and retailer service levels exhibit a negative correlation with the manufacturer's digital investment level, while online prices show a positive correlation. Findings from Jena and Singhal (2023) align with this conclusion.

In manufacturer-dominated dual-channel supply chains, enhanced digital capabilities simultaneously reshape channel value and competitive dynamics. On one hand, it intuitively enhances the perceived value of the online channel, causing the optimal online price $pm$ to rise with $d$⁠. On the other hand, it strengthens the substitution effect of online for offline channels, as some consumers shift from offline to online channels for purchases. This also reduces the conversion efficiency of offline services, forcing retailers to lower offline prices to regain customer traffic and, to some extent, reduce service levels to control costs.

$Πr(d)$ always decreases with $d$⁠; When any of the following conditions is met, $Πm(d)$ increases with $d$⁠.

• $ρ>1and0<k1<γ2(1+k22k2−2θ2)and0<τ≤12+k2−3k2+2θ2andk2>2θ2$

• $0<ρ<1and0<λ<1−6τ4−4ρand0<τ<16andk2>2θ2(1+2λ(−1+ρ)−2τ)1+4λ(−1+ρ)−6τand0<k1≤3γ22$

Proof of Corollary 4. The proof of Corollary 4 is similar to that of Corollary 1.

Retailer profits exhibit a negative correlation with manufacturers' digital investment levels; under certain conditions, the profit of manufacturers is positively correlated with the level of digital investment, and these conditions are determined based on the sensitivity of consumers to CSR.

The digital investment by manufacturers to enhance the matching efficiency and experience on the direct sales end may draw some consumers with high payment willingness away from offline channels. As a result, the demand for offline channels decreases, and retailers are forced to lower prices to attract consumers. This leads to a compression of the retailers' profit margins. When manufacturers operate in markets where consumers recognize CSR behaviors, investing in CSR can effectively enhance brand image and demand. In this case, digital investment can form synergy with the CSR strategy. At this time, it may only be necessary to ensure the efficiency of digital investment and control the transportation costs of online channels, while the service costs of retailers reach the minimum requirements. The manufacturer's profit will increase due to the improvement in the level of digital investment. When consumers are not sensitive to CSR behaviors, manufacturers need to control the CSR investment and transportation costs, and at the same time, retailers' service level costs also have more stringent requirements.

In the previous section, we conducted the model solution and analysis under the decentralized decision-making scenario. At this time, the optimal decisions and profits of the supply chain members are all obtained as local optimal solutions based on maximizing their respective interests. However, these local optima do not necessarily lead to the optimal overall system. To achieve optimal overall supply chain profit, we need to conduct model-based analysis and solutions under centralized decision-making to ensure the rational allocation of resources and the maximization of overall benefits. Under centralized decision-making, we consider the manufacturer and the retailer, analyze the problem from a global perspective, and determine the online price, offline price, digital investment level, and service level. The overall profit function of the supply chain is:

1. $pmD*>pmC*;dD*>dC*;sD*<sC*$

2. When $γ21+k2k2−θ2<k1<2γ2k2−θ2k2−2θ2$⁠, we have $prD*<prC*$⁠; when $k1>2γ2(k2−θ2)k2−2θ2$⁠, we have $prD*>prC*$⁠.

3. $ΠC>ΠmD*+ΠrD*$

Proof of Corollary 5. The complete proof is relegated to Appendix A4.

Under centralized decision-making, the online prices and the level of digital investment by manufacturers are both lower than those under decentralized decision-making. Retailer service levels and the supply chain's total profits are higher under centralized decision-making. The comparison of offline prices under centralized and decentralized decision-making depends on the range of $k1$⁠.

Under decentralized decision-making, both manufacturers and retailers aim to maximize their own profits, which often leads to the double markup effect. Manufacturers tend to raise prices and investment levels to increase profits, while retailers, to maintain their own earnings, are also intentionally controlling expenditures on service inputs. When the digital investment cost coefficient falls within a specific range, the online channel is highly attractive. To avoid intense competition eroding offline retailers, centralized decision-making will raise offline prices to balance the channels. As digitalization costs increase significantly, the advantage of the online channel weakens. When manufacturers are under centralized decision-making, they will reduce digital investment to cut costs. To maintain market share, centralized decision-making will shift to relying on offline channels and lowering prices to maintain demand and total profits. Therefore, there is a “turning point” in the difference in offline prices between centralized and decentralized decisions, which is determined by the digital investment cost (⁠$k1$⁠).

To enrich the results, we conducted case studies on the model. Based on the model's parameter assumptions and prior research (Guo et al., 2020; Hsieh and Lathifah, 2024; Zhang et al., 2024b), we selected the following baseline parameter settings: $k1=2.4,k2=8,γ=θ=1,τ=0.1,λ=0.6$⁠. The setting $τ=0.1$ aligns with Guo et al. (2020). $θ=1$ indicates consumers exhibit balanced responses to service levels, without strong preferences or aversions. This application follows the research settings of Zhang et al. (2024b) and facilitates model simplification. Other parameter settings were adjusted within the fundamental range to establish the basis for this model.

Figure 1 shows that the total profit of the supply chain decreases monotonically as the digital investment cost coefficient $k1$ increases under decentralized decision-making. Moreover, under the same cost coefficient, the higher the consumers' sensitivity to digital investment, the higher the total profit of the supply chain. This indicates that the recognition of digital investment by consumers can effectively compensate for the high expenditure on digital construction. Figure 2 shows that the supply chain's total profit increases as consumers' sensitivity to CSR increases, and, within a given sensitivity, the higher consumers' sensitivity to digital investment, the higher the total profit. This means that the more sensitive consumers are to digital investment, the higher the profit conversion rate brought by CSR fulfillment. Digital investment makes it easier for enterprises' CSR behaviors to be perceived by consumers and converted into actual purchasing power.

Figure 3 illustrates the impact of consumers' CSR preferences on the supply chain's overall profit under decentralized and centralized decision-making. Within the feasible region of $ρ$⁠, we observe that $ρ$ is positively correlated with the profits under both decision-making scenarios, and the overall profit under centralized decision-making is higher than that under decentralized decision-making. This also verifies the profit comparison results between centralized and decentralized decision-making as stated in Corollary 5.

An analysis of centralized decision-making solutions and a comparison with decentralized decision-making reveals that optimal decisions under decentralized decision-making fail to maximize overall supply chain profits. Therefore, supply chain coordination is necessary to ensure the effective implementation of centralized decision-making solutions. To this end, this study designs a coordination contract for wholesale price agreements. Its objective is to enhance the profit distribution among supply chain members by adjusting wholesale prices, thereby increasing profits compared to decentralized decision-making. The coordinated profits for manufacturers and retailers are denoted as $ΠmH(w)$ and $ΠrH(w)$⁠, respectively. We derive the following profit functions:

If there exists an interval $W=[wmin,wmax]$ such that for any $w∈W$⁠, $ΠrH(w)>ΠrD$ and $ΠmHw>ΠmD$ are hold, then $W$ is termed a feasible wholesale price interval, and this scenario is referred to as “win-win coordination.”

The feasible range of wholesale prices $W=[wmin,wmax]$ defines the boundary conditions and game space for the supply chain to achieve win-win coordination. The lower limit of the interval $wmin$ is the bottom line of the price acceptable to the manufacturer. When the wholesale price is at this level, the manufacturer's coordinated profit equals the profit under decentralized decision-making. The excess profit obtained after the coordination of the supply chain is obtained by retailers, which is the most favourable situation for retailers. The upper limit of the interval $wmax$ indicates the highest wholesale price acceptable to the retailer. At this price, the retailer's profit is equal to the profit under decentralised decision-making. The excess profit is obtained by the manufacturer, which is the most favourable situation for the manufacturer. The width of the interval reflects the negotiation space between the two sides. Only when $w$ is within the open interval $(wmin,wmax)$ does the profit of both parties achieve Pareto improvement; that is, their profits are higher than under decentralized decision-making, thereby improving overall supply chain efficiency. In practical business applications, the final wholesale price within the range depends on the bargaining power between manufacturers and retailers. $w$ close to $wmax$ indicates that manufacturers dominate the channel and can retain more coordinated profits; $w$ close to $wmin$ reflects that retailers have a stronger market say and bargaining power, such as large retail terminals.

Figures 4 and 5, respectively, illustrate the impact of different levels of w on the manufacturer's and retailer's profits after coordination and are based on their respective profit levels under decentralized decision-making. From the figures, when the wholesale price exceeds a certain threshold, the profit of the coordinated manufacturer will be higher than that under decentralized decision-making; at the same time, when the wholesale price exceeds a certain threshold, the profit of the retailer after coordination begins to be lower than the result under decentralized decision-making.

Figure 6 combines Figures 4 and 5 into a unified view, which clearly shows that there exists a feasible range of $[wmin,wmax]$ such that the profits of the coordinated manufacturer and retailer are both higher than the results of decentralized decision-making. This also further verifies the mathematical derivation results of the coordination part. The specific pricing of w depends on the bargaining power of both parties.

To fully investigate the combined effect of CSR and digital investment, this study expands on the situation where retailers make social donations, that is, where retailers fulfill their CSR obligations, and explores how digital investment affects member decisions and profits in the context of CSR fulfillment by retailers. At the same time, it compares the overall profit situation of the supply chain in the two CSR fulfillment scenarios. When retailers make social donations, the CSR decision only affects the offline channels. Therefore, the utility functions that consumers obtain on the online and offline channels are: $Ur=v−pr+θs+ρλ$ and $Um=δv−pm+γd$⁠. Therefore, the demand functions are: $Dr=1−pm−pr−dγ+sθ+λρ−1+δ$ and $Dm=pm−pr−dγ+sθ+λρ−1+δ−pm−dγδ$⁠. The profits of the manufacturer $ΠmR$ and the retailer $ΠrR$ are as follows:

The game sequence of the extended model is consistent with the model where the manufacturer conducts CSR. The manufacturer still takes the lead and makes the decision first. Using the backward-solving method, the following optimal decisions and profit solutions are obtained:

Proof of Proposition 4. The proof of Proposition 4 is similar to that of Proposition 1.

To further explore the impact of digital investment levels on the decision-making and profits of manufacturers and retailers, we internalize the digital investment levels and obtain the following optimal solution:

Proof of Proposition 5. The proof of Proposition 5 is similar to that of Proposition 1.

The influence of $d$ on $pmR(d)、prR(d)、sR(d)$ and $ΠrR(d)$ is consistent with the results in Corollary 3 and Corollary 4. The influence of $d$ on $ΠmR(d)$ is still related to parameters such as $k1、k2$⁠, and the specific range of $d$⁠. When $ρ>1,0<λ<5−12τ-4+4ρ,k2>2θ2−3+4τ−5+4λ−1+ρ+12τ,0<k1<γ2k25−4λ−1+ρ−12τ+2θ2−3+4τ2k2−θ2$⁠, and $14γ<d≤1+3λ(−1+ρ)4γ$⁠, $∂ΠmR(d)∂d>0$⁠.

Proof of Corollary 6. The complete proof is relegated to Appendix A5.

The impact of manufacturers' digital investment on their own profits depends on the CSR and service cost coefficients of retailers, as well as the digital investment cost coefficient and investment level. Firstly, the CSR investment of retailers should be within a limit. Moderate CSR investment will expand the offline market and provide a potential demand market for the manufacturers' online channels. Secondly, if retailers' offline service costs are too low, when facing competition from manufacturers' online channels, retailers can easily consolidate their position by improving services, leading to intense channel competition. Finally, there is an optimal range for the level of digital investment. Only when the investment reaches the minimum investment threshold can it have market impact and scale effects; at the same time, manufacturers should not over-invest, as excessive investment will cause a squeeze-out effect on the offline market, negating the positive regulatory effect brought by retailers' CSR investment.

When $0<ρ<1$⁠, the optimal digital investment level of the manufacturer when the retailer performs CSR will be higher than the optimal digital investment level when the manufacturer performs CSR. When $ρ>1$⁠, the opposite is true.

Proof of Corollary 7. The complete proof is relegated to Appendix A6.

When consumers are insensitive to CSR, if the manufacturer implements CSR, not only is the demand brought by CSR limited, but also the manufacturer shoulders the additional costs of donations, which diminishes its marginal investment capacity. If retailers fulfill their CSR obligations, manufacturers do not need to bear the cost of CSR, but they can still benefit from the weak demand that CSR brings. Thus, enhance the competitiveness of online channels. When consumers are sensitive to CSR, manufacturers, as the direct implementers of CSR, gain the corresponding brand reputation and premium ability, and CSR and digital investment have a synergistic effect, thus further driving the demand. If the retailer performs CSR, although the demand will also increase significantly, the manufacturer cannot directly control the CSR input, which may lead to free-rider behavior.

Proof of Corollary 8. The complete proof is relegated to Appendix A7.

When $0<ρ<1$⁠, the retailer's CSR level is positively correlated with digital investment; when $ρ>1$⁠, the situation is the opposite. When consumers respond to offline CSR activities with indifference or mild enthusiasm, the retailer's CSR fulfillment plays a role in enhancing the overall brand image but has little impact on the online channel. To prevent excessive traffic from shifting to the offline channel and to capitalize on the positive brand effect driven by the retailer's CSR, the manufacturer may choose to increase digital investment to solidify its online advantage. However, when consumers are highly sensitive to CSR, the retailer's CSR activities make the offline channel more attractive, eroding the demand for the online channel. The manufacturer may find that the efficiency of digital investment declines at this point and may choose to reduce investment to protect its own profits.

This conclusion is contrary to that in Corollary 1. When the retailer is responsible for fulfilling CSR, CSR and digital investment are separate entities. The retailer's CSR primarily affects offline channels and serves as a means to compete with the manufacturer's online channels for attracting consumers. This may lead to an investment suppression. The sensitivity of consumers to CSR also serves as a tool for demand conversion across the two channels. The more sensitive consumers are to CSR, the more customer traffic there is in the offline channels, and the more the manufacturer is inclined to control investment. When the manufacturer fulfills CSR, both CSR and digital investment are under the manufacturer's control. At this time, the social reputation and brand influence brought by CSR are global. Consumer sensitivity to CSR becomes the main driver of premium pricing. The more sensitive consumers are to CSR, the stronger the manufacturer's brand image through CSR becomes. To maximize capture of this traffic, the manufacturer is more motivated to invest in digital channels, thereby improving the efficiency and experience of those channels.

When $0<ρ<1$⁠, the manufacturer's profit is higher in the case where the retailer fulfills CSR, while the retailer's profit is the opposite; when $ρ>1$⁠, the manufacturer's profit is higher in the case where the manufacturer fulfills CSR itself, while the retailer's profit is the opposite.

Proof of Corollary 9. The complete proof is relegated to Appendix A8.

When consumers have a lower sensitivity to CSR, the manufacturer can avoid bearing the cost of CSR investment and can also take advantage of the market benefits brought by the retailer's CSR investment, thereby obtaining higher profits than when the manufacturer fulfills CSR itself. However, the retailer must bear the cost of CSR investment and, due to the low sensitivity of consumers to CSR, the retailer's CSR returns are not high. Therefore, the retailer's profit will be lower than the profit of the manufacturer when the manufacturer fulfills CSR. When consumers have a higher sensitivity to CSR, the return benefits of CSR are higher than the cost of investment. Both the manufacturer and the retailer tend to make CSR investments themselves, grasp the CSR initiative, and thus gain market initiative.

To conduct a more intuitive study on which CSR implementation model is beneficial for the entire supply chain, this research employs numerical analysis to present the overall profit situation of the supply chain under the two models of the manufacturer fulfilling CSR and the retailer fulfilling CSR. Among them, the total profit under the manufacturer model $ΠcM=ΠrD+ΠmD$⁠, total profit under the retailer model $ΠcR=ΠrR+ΠmR$⁠. Based on the model's parameter assumptions and prior research (Guo et al., 2020; Hsieh and Lathifah, 2024; Zhang et al., 2024b), we select $k1=1.684,k2=8,γ=1,θ=1,τ=0.1$⁠; The setting $τ=0.1$ aligns with Guo et al. (2020). $γ=θ=1$ indicates consumers exhibit balanced responses to digital investments and service levels, without strong preferences or aversions. This application follows the research settings of Zhang et al. (2024b) and facilitates model simplification. Other parameter settings were adjusted within the fundamental range to establish the basis for this model.

Figures 7 and 8 show that in markets where consumers are not sensitive to CSR, fulfilling CSR by retailers can lead to higher total supply chain profits; conversely, in markets where consumers are sensitive to CSR, fulfilling CSR by manufacturers can achieve a better result for the total supply chain profits.

When $ρ$ is low, having retailers fulfill CSR enables manufacturers to focus on enhancing overall efficiency through digitalization, thereby optimizing resource allocation. Although the profits of retailers may be affected, the increase in manufacturer profits can compensate for this, thus raising the overall supply chain profits. However, as can be seen from the figure, the total profits increase slowly with the investment in CSR, indicating that the marginal benefit of CSR is low at this time. When $ρ$ is high, CSR becomes a strong demand-driven factor. Fulfilling CSR by manufacturers can generate a synergy effect with the digitalization investment in online channels. For example, digitalization can enhance transparency, making CSR efforts more transparent and public, and having a certain promotional effect. Manufacturers at the upstream of the supply chain fulfilling CSR can enable the entire channel to be empowered through brand image, achieving the highest total supply chain profits.

Based on the above in-depth discussion and model analysis, we have reached the following conclusions.

Firstly, the social donations made by manufacturers in response to the call to fulfill their corporate social responsibility affect member decisions and profits, as they are related to consumers' sensitivity to CSR. When consumers are more sensitive to CSR, manufacturers' profits increase as they fulfill CSR, while retailers' profits are squeezed and service decisions at the retail end are reduced; manufacturers' digital investment will increase accordingly. Secondly, when manufacturers invest in digitalization of the online channel, it will raise online prices and squeeze the offline channel. The investment and service levels of retailers in the offline sector will be inversely affected; moreover, retailers' profits will decrease due to manufacturers' digital investments. Under certain conditions, manufacturers' profits are positively correlated with the level of digital investment, and these conditions depend on consumers' sensitivity to CSR. Finally, the results of decentralized decision-making show that the investment and development of manufacturers' online channels will erode their offline channels, directly diverting some consumer demand from offline to online. By comparing with the results of centralized decision-making, it can be found that the decisions of manufacturers under decentralized decision-making will be improved. Retailers, to protect their own interests, must reduce investment and expenditure, which will lead to lower decision outcomes under decentralized decision-making than under centralized decision-making. However, the comparison of offline prices in decentralized and centralized situations mainly depends on the size of the cost coefficient for online channel digitalization investment, which also involves coordination issues in the centralized decision-making situation. By constructing an expansion model for retailers' fulfillment of CSR, we found through comparative analysis that in markets where consumers are not sensitive to CSR, fulfilling CSR by retailers can lead to higher total supply chain profits, and the optimal digital investment level of manufacturers will be higher in this scenario; conversely, fulfilling CSR by manufacturers can maximize the total profits. Furthermore, when retailers fulfill their CSR obligations, the relationship between manufacturers' digital investments and CSR is still related to consumers' CSR sensitivity, but the outcome is the opposite of what it is when manufacturers fulfill their CSR duties. To alleviate the imbalance in development between the two channels, the article also conducts supply chain coordination through wholesale price contracts, thereby obtaining a feasible range of wholesale prices. Within this range, wholesale prices can enable retailers and manufacturers to achieve effective profit improvement in a decentralized decision-making environment, resulting in a win-win outcome.

According to these findings, the following management implications emerge. Firstly, CSR needs to be visualized, and the performance metric should be flexibly switched based on market sensitivity. Investment should match the intensity of consumers' preferences for CSR. For consumers with a weaker preference for CSR, manufacturers should adopt the “authorizing retailers for donations” model. Retailers can make small community donations or fund local schools and display the results on the product page. According to the research conclusion, this “delegation of power” can bring higher total supply chain profits in low-sensitive markets. For consumers with a stronger preference for CSR, manufacturers should “fulfill themselves” with large-scale social donations. This can ensure the effectiveness of digital investments through global reputation. Enterprises can establish a donation flow-tracking system that allows consumers to see where their contributions go on their mobile devices. For example, Alibaba's “Goods for Good” links each transaction to a small donation and sends a public welfare bill to consumers, converting the willingness to pay a premium into actual growth in the online channel. Secondly, for digital investment, manufacturers need to continuously monitor and evaluate consumers' awareness and willingness to pay for corporate social responsibility and adjust the investment scale promptly based on the CSR fulfillment entity to address the channel encroachment effect. For highly sensitive consumer markets, manufacturers' digital investments should align with their own CSR. In this situation, digital investment should focus on maximizing the overall premium of brand reputation, such as implementing full-chain traceability through embedding blockchain-based product identity codes in apps; or using VR technology to create a “cloud visit” of the environmental protection factory in the online flagship store, thereby ensuring that digital investment does not shrink due to the enhanced attractiveness of offline channels; for the less sensitive consumer market, manufacturers' digital investment efforts should shift towards channel collaboration, focusing on attracting customers to offline channels or assisting in the operation of offline channels. For example, Uniqlo's palm flagship store attracts traffic through precise online push notifications and then directs it to the store for purchase; for new members, Sam will not only provide free online app shipping vouchers but also offer large vouchers for offline stores. Enterprises should utilize digital tools to enhance the efficiency of the entire chain rather than merely expanding online. Furthermore, attention should be paid to the protection of retailers' profits and market competitiveness. Manufacturers should enhance retailers' competitiveness by optimizing wholesale prices, providing reasonable price support, and necessary resources, especially when the level of digital investment fluctuates significantly in response to the CSR environment. Manufacturers can use promotional activities, price discounts, and other measures to alleviate retailers' operational pressure. At the same time, manufacturers can also help retailers improve service levels by sharing market information and providing marketing support. When promoting social donation results online, manufacturers should list retailers as “joint donors” and leverage improvements in brand reputation to help retailers boost offline conversion rates, thereby increasing consumer satisfaction and brand loyalty. The overall optimization and coordination of the supply chain should be achieved as much as possible. Finally, manufacturers and retailers attempt to establish cooperation, adopting different strategic partnerships in response to changes in consumers' sensitivity to CSR, and completing the transformation from “who will do it” to “when to do it”. Business leaders should be aware that digital investment is not an isolated decision; it is deeply influenced by the CSR environment. For the low-sensitive market, manufacturers actively promote and assist retailers in making CSR investments, while digital tools are used to improve the full link turnover efficiency. In highly sensitive markets, manufacturers strive to fulfill their CSR commitments in person and integrate digital displays online. Ultimately, through the rational design of contracts, the continuous growth of profits for both parties was achieved in the dynamic CSR environment. Ultimately, through the rational design of contracts, the continuous growth of profits for both parties was achieved in the dynamic CSR environment.

Finally, this paper has several limitations. First, digital investment considerations are confined to online channels, neglecting scenarios in which retailers invest in digital channels. Secondly, the supply chain structure is considered to involve only a single manufacturer and a single retailer. Therefore, future research directions can be expanded to include closed-loop supply chains with recycling and remanufacturing production, and to consider the specific impact of digital investment in such closed-loop supply chains, as well as to expand the structure of supply chain members.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

The supplementary material for this article can be found online.