Triads in Services Outsourcing: Bridge, Bridge Decay and Bridge Transfer


  • *Like all invited papers and invited notes, the original version of this manuscript underwent a double-blind review process.


Typically, a triad of actors is involved in any outsourcing situation: the buyer, the supplier and the buyer's customer. In manufacturing, the buyer acts as a bridge between its supplier and its customer and maintains this bridge position before, during and after the outsourcing. However, in services, the relationship structures among the three actors change before, during and after the outsourcing. Before outsourcing (i.e., during the contract negotiation stage), the buyer is the “bridge” between its supplier and its customer. During implementation, this bridge position begins to “decay” as its supplier comes in direct contact with the buyer's customer. Post-implementation, the bridge position is intended to be “transferred” to the supplier. However, if left unmanaged, this state of transferred bridge position has serious performance implications for the buyer. The supplier is now the bridge and thus enjoys the leverage inherent in being a bridge. This point has been missed in many services outsourcing ventures by major multinational corporations. To mitigate this effect, we propose that the buyer should continue to actively interact with its customer and closely monitor the supplier in order to prevent the supplier from solidifying its bridge position.


Services outsourcing is more ubiquitous than ever (Grossman and Helpman 2002; Roth and Menor 2003; Amiti and Wei 2005). A recent CAPS Research study on services outsourcing (Tate and Ellram 2006) showed that more organizations outsource services than those that outsource direct or indirect materials. Services outsourcing is expected to expand in the future — in the same CAPS study, companies expected the outsourcing of services to become even more pervasive.

Yet, popularity does not guarantee good results. The outcome of services outsourcing is mixed at best (Baitheiemy 2003; Aubert and Croteau 2006). According to a DiamondCluster report in June 2005, the number of buyers prematurely terminating an outsourcing relationship has doubled to 51 percent while the number of buyers satisfied with their providers has “plummeted from 79 percent to 62 percent” (Weakland 2005, p. 2). One such example is Dell Inc., which outsourced its technical support for business customers to a supplier in Bangalore, India, in 2001, and not long after, in 2003, moved it back to the United States (InfoTech 2003; Aubert and Croteau 2005). Other companies such as AT&T, Capital One, and J. P. Morgan Chase have also taken their outsourced services functions back in house (Overby 2003, 2005; McLaughlin and Peppard 2006; Pfeffer 2006). A 2005 Gartner study predicted that 60 percent of organizations that have outsourced services functions would see significant numbers of frustrated customers switching to competitors (Chon 2005). As columnist Jeffrey Pfeffer (2006) puts it, “a penny saved, a customer spurned” (p. 1).

Why have there been so many failed attempts in services outsourcing? In general, the existing literature in services outsourcing has focused on two key questions —“should we outsource” and “what to outsource.” The former has generally been addressed by economists interested in the impact of services outsourcing on the national economy (Fixler and Siegel 1999; Amiti and Wei 2004, 2005; Gorg and Hanley 2004). The latter was studied by scholars looking for the key drivers of outsourcing decisions building on transaction cost economics (TCE) (Williamson 1975; Jensen and Meckling 1976; Milgrom and Roberts 1992; Lacity and Willcocks 1995; Ang and Straub 1998; Williamson 2008), core competence theory (CCT) (Venkatesan 1992; Quinn and Hilmer 1995; Jennings 2002) or the resource-based view (RBV) (Wernerfelt 1984; Harrigan 1986; Barney 1991, 2001; Jennings 2002).

Although these scholars have addressed their questions well, a careful examination reveals a serious omission. The question of how to manage the process of services outsourcing is missing. An implicit assumption is if we followed the theoretical guidelines of “what should we outsource,” we would be fine after the outsourcing. However, in the business world, this simplistic view has not been matched by the reality. In this study, we intend to examine the process of the relational structures involved in services outsourcing —“What was the relationship structure prior to, during, and post the services outsourcing?” and “What are their performance implications?” We will endeavor to understand better, using a social network lens, the shifting of relationship structures among the services supplier, the buyer, and the buyer's customer, and suggest better strategies for managing the relationships from the buyer's perspective.

This study has profound theoretical and managerial implications. First, it breaks away from the “once and for all” static view of the services outsourcing approach and delves deeper into the dynamics of the underlying relationship structures at different stages of the services outsourcing process. Second, it extends social network theory into the services outsourcing context. As such, it offers insights not available in other outsourcing theories such as TCE, CCT and the RBV. The explanatory power of the social network theory and the insights it offers also lends itself to the very prevalent and pragmatic managerial problems we are facing today.

In the following sections, we will first explore the nature of services and then discuss relational structures in terms of how services outsourcing is different from manufacturing outsourcing. Next, we will present social network theory as a new lens to examine the process and consequences of structural shifting. We will then synthesize the literature on services operations and the structural hole concept and present our conceptual model linking each stage of the services outsourcing process with a set of unique relationship structures among the services buyer, its supplier and its customers. We will close with a discussion of managerial implications, limitations of the present study, and directions for future research.


Services have one enduring characteristic — the customer interacts with the service provider during the process of delivery (Gaither and Frazier 1999; Sampson 2001; Sampson and Froehle 2006). For instance, Sampson and Froehle (2006) reviewed the services literature and highlighted this interaction and the role of customer inputs in this interaction. Chase and Aquilano (1977, p. 17) state, “the main feature that sets a service system apart from a manufacturing system is the extent to which the customer must be in direct contact.” In this regard, the services we are referring to in this study are in fact “pure services” based on the terminology of CCT (Chase and Aquilano 1977; Kellogg and Chase 1995). It refers to the operations where “major production is carried out in the presence of the customers” (Chase 1981, p. 701). Compared with other types of services such as “mixed services” and “quasi-manufacturing services,” pure services emphasize the importance of the real-time interaction that occurs between the service providers and the customers. Mixed services “commonly involve a mix of face-to-face contact and loosely coupled back office work” (Chase 1981, p. 701), while quasi-manufacturing “entails virtually no face-to-face contact” (Chase 1981, p. 702). The outsourcing of these other types of services shares characteristics common to manufacturing outsourcing and is not the focus of this study.

Services outsourcing is typically defined as the conscious choice of replacing internal service functions with the use of external agents to perform one or more service activities. It seems then services outsourcing should be similar to manufacturing outsourcing: in services outsourcing we just outsource intangible goods, such as customer service; while in manufacturing outsourcing we outsource tangible goods such as parts and raw materials. Contrary to our initial intuition, there are fundamental differences underlying the two types of outsourcing arrangements and it is important to understand these differences.

Hewlett-Packard (HP), one of the world's top manufacturers of notebook PCs, was outsourcing its manufacturing in the 1990s. When a retail shop ordered Pavilion zd8000 laptops, this shop interacted with HP but had no contact with the suppliers. It was HP that interfaced with the suppliers — an assembly supplier in Taiwan, a graphic chip supplier in Markham, Ontario, liquid crystal display screens and memory chips suppliers in South Korea and Taiwan, and a hard-disk drives supplier in Japan (Dean and Tam 2005). In this type of manufacturing outsourcing arrangement, the customer is not in direct contact with any of HP's suppliers. HP, acting as a go-between or bridge, controls the direct information and product flows between its suppliers and its customers prior to, during, and after the outsourcing.

Such relational dynamics change when it comes to a services setting. Besides outsourcing its manufacturing operation, HP also outsourced some service work — a portion of its software support activity. In this case, the customer service representatives from the supplier company were in direct contact with the customers of HP when offering their services. HP had no ready measure to control this interaction while its suppliers delivered their services to HP's customers.

Figure 1 depicts these two different structural arrangements. The linear diagram on the top represents a manufacturing setting wherein manufacturing buyers can effectively control the interaction between its suppliers and its customers. The triangular diagram on the bottom of Figure 1 depicts the services setting where the services supplier and the buyer's customer contact each other directly. The triadic relationship structure among the services buyer, services supplier and buyer's customer in Figure 1 serves as the foundation of our study. Specifically, we will frame services outsourcing from the lens of social network theory and examine the shifting of the triadic relationship structures in a services supplier network.

Figure 1.

Comparison of Supply Chain Triadic Relationship Structures in Manufacturing vs. Services


A network is made of elements and the links that connect these elements (Borgatti and Li 2009). When these elements represent “agents” that are able to make volitional choices, such as individuals or organizations, the network is called a social network (Wellman and Berkowitz 1988; Burt 1992; Watkins 2003; Borgatti and Li 2009). Our model builds on two key perspectives from the social network literature: social capital (Coleman 1988a, b, 1990) and the structural hole concept (Burt 1992, 1997).

Social Capital and Social Networks

Social capital derives from an agent's relationship with other agents in the same network through which the agent gains access to resources (Coleman 1990). One particular type of relational arrangement is called a structural hole, wherein an agent works as a bridge between two other agents or networks with no links between them (Burt 1992, 1997; Gargiulo and Benassi 2000).

The most prominent growth area in social network research has been the study of social capital (Borgatti and Foster 2003). Bourdieu (1985) first noted the positive effects for an individual who is tied to a group of other individuals. Other scholars (Coleman 1988a, b; Baker 1990; Schiff 1992; Portes 1998) continued this research stream by considering power and influence associated with being in social networks. Later, Burt (1992, 1997) elevated the ideas of social capital in a novel direction. Instead of emphasizing the benefits of being closely associated to a social group, he studied the leverage gained when individuals span across structural holes among social networks. His concepts stressed the advantage of bridging two disconnected networks.

Structural Hole Concept

A structural hole is defined as the lack of connections between agents or groups that are not directly linked together (Burt 1992, 2000a). The structural hole concept is closely related to the concept of a bridge. A bridge is the agent that is positioned on a structural hole. In the absence of a connection between two isolated agents, a bridge acts as a go-between and the gatekeeper of information.

The underlying premise of the structural hole theory is that the structure of the network is what offers an agent the social capital. In other words, the structural position within a network is what determines the dynamics among actors because social actors in some positions in the networks are better off than those in other positions (Burt 1992). Specifically, social actors that occupy the bridge position enjoy brokerage opportunities. Figure 2 offers a pictorial rendition of the structural hole and bridge concepts.

Figure 2.

Illustration of the Structural Hole Concept

Figure 2 shows three agents. Agent B and Agent C are linked to Agent A but they are not directly linked to each other. This lack of connection forms a structural hole. A structural hole “creates a competitive advantage for an individual whose relationships span the holes” (Burt 2000a, p. 6). Agent A spans the structural hole between Agent B and Agent C and is in the bridge position and reaps benefits that come with this position.

Benefits of Bridge Position

Burt (2000a, b, 2002) explains the advantages that come with being a bridge. First, there is the information benefit (Burt 1992, 2000a, 2002). Because a structural hole can be viewed as the gap between two nonredundant agents or networks, agents who occupy the bridge position can benefit from additional information from nonredundant sources. Second, there is the control benefit. Simmel (1955) and Merton (1968) introduced the ideal type of people who derive control benefits from the structural hole — the tertius gaudens or a third who benefits from brokering the connection between others (Burt 1992, 2000a). The bridge can negotiate and exploit information to its advantage (Burt 2000a; Zaheer and Bell 2005).

Zaheer and Bell (2005) explicated further the advantages of the bridge. By accessing information from unique parts of the network, the bridge can hear about the impending threats or opportunities before other actors who are not in the bridge position. The bridge can also enjoy privileged information about the quality of possible exchange partners and potential allies (Powell and Smith-Doerr 1994; Uzzi 1996; Zaheer and Bell 2005).

Bridge Decay and Bridge Transfer

Being a bridge may be a good thing, but it does not last forever. A firm's bridge position is not a permanent state and is subject to change (Burt 2000b, 2002; Soda, Usai and Zaheer 2004). First, the two agents that are otherwise isolated by the bridge can begin making connections and reduce the leverage of a bridge, creating the state we call “bridge decay.” This happens once the agents on each side of the structural hole are able to link directly and the connection to the agent that used to be the bridge becomes redundant and loses its value (Johnson 2004). In fact, such bridge decay may be more prevalent than one may realize. Burt (2001) observed that compared with other kinds of relationships, the bridge position shows a faster rate of decay over time. Second, a bridge can willingly or unwillingly relinquish its position to one of the other two agents and create a state we call “bridge transfer.” Structurally, bridge transfer happens if the existing structural hole is filled and one of the other two links is disconnected.

Bridge decay has implications for the stability of social capital. With the loss of bridge position, a social agent also loses the advantage of social capital it once enjoyed. Therefore, it is seldom to the benefit of agents who hold the bridge position to encourage others to join them in linking groups, because then they become redundant and lose their value (Johnson 2004). This puts the burden on the social agent to make extra effort to maintain the bridge position (Burt 2001). Even after the decaying of the bridge, the shifting of the relationship structure is by no means over. This structural shifting is bridge transfer. Compared with bridge decay, bridge transfer is rarely studied.

Figure 3 depicts the concepts of bridge decay and bridge transfer. During the initial stage for both bridge decay and bridge transfer, Agent A is in the bridge position across the structural hole created between two nonredundant agents, B and C. During the transformed stage, differences begin to emerge. In the bridge decay scenario, Agents B and C establish a direct link with each other, effectively eliminating the structural hole and nullifying Agent A's bridge position. In the bridge transfer scenario, Agent A loses its connection with C while Agent B establishes its connection with C. Agent B is the new bridge now and assumes the benefits of social capital brought by its position and power and influence in the network of agents.

Figure 3.

Bridge Decay versus Bridge Transfer

Triads in Services Supply Networks

Services outsourcing offers a unique context for studying how service supply networks evolve through different stages of outsourcing. As mentioned before, in manufacturing, the buyer typically retains its bridge position before and after the outsourcing. It often consciously creates barriers between the suppliers and its customers, thereby continuing to act as an intermediary. When its bridge position is lost, it is called “supply chain disintermediation” (Rossetti and Choi 2005). In a services context, however, the buyer has no choice but to permit a supplier to directly interface with its customer. Building on Figure 3, Figure 4 depicts these changes of network positions among supply chain partners. The network states of bridge, bridge decay and bridge transfer describe the triadic structures before, during and post services outsourcing.

Figure 4.

Shifting Relationship Structures

Triad before Outsourcing: Buyer as Bridge

During the negotiation stage before outsourcing, two dyadic relationships exist, one between the buyer and its potential supplier and another between the buyer and its customer. The buyer is positioned as the bridge between the supplier and its customer, and thus the supplier and its customer lack a direct link to each other.

Consider the case where a major computer company outsourced to a services supplier in India a call center for its business customers. The service work entailed supporting a series of software tools used in supply chain operations. Before outsourcing, the internal IT team of this company was in direct contact with its customers and provided the call center support. The same team also was involved in the selection of the services supplier and thus was in contact with the prospective suppliers. At this stage, the customers were not in direct contact with the potential services suppliers in India. Positioned as the bridge, this buying company had full control of what information to share and how to share it with its potential suppliers and customers. For example, this company could strategically deemphasize the complexity involved in the application support tasks to the potential suppliers in order to negotiate a lower price or higher service. It could also strategically emphasize the complexity involved in the application support tasks to its customer in order to negotiate a better contract.

As shown in the top line of Figure 4, one might say the buyer is “central” in this triad (Bavelas 1950; Burt 1980; Borgatti and Li 2009). The buyer has access to information about the supplier's capability and capacity and also maintains information on the customer's service performance requirements. As long as there is a structural hole between the supplier and buyer's customer, the buyer can strategically control the flow of this type of critical information. This leads us to the first proposition:

Proposition 1: Prior to services outsourcing, the services buyer acts as a bridge between the services supplier and the buyer's customer and therefore enjoys information and control benefits associated with the bridge position.

Triad during Outsourcing: Bridge Decay

The structure of our triad changes after the negotiation is done and when the supplier and the buyer's customer are brought together. As we stated earlier, the key characteristics that separate services operations from manufacturing operations is the direct contact between the services provider and the customer (Chase and Aquilano 1977; Kellogg and Chase 1995). In a services outsourcing context, the buyer has no effective measures to block this interaction and at the same time expect the supplier to deliver services to its customers. While outsourcing agreements are implemented, the services supplier comes in direct contact with the buyer's customer. As these two entities begin to interact with each other, the bridge position of the buyer is gradually eroded. As the bridge position becomes diminished, the leverages that the buyer used to have are weakened (Burt 2002). The buyer would not be able to control how information flows between the supplier and its customers. Eventually, the buyer's bridge position is no longer there and the triad would decay, as shown in Figure 4.

Proposition 2: Services outsourcing causes the buyer to lose its bridge position, leading to the erosion of the information and control benefits it used to enjoy.

Once the services supplier and the buyer's customer begin to interact directly, a new relationship between them would start to form. A critical question that emerges is how strong this new tie might be. We posit that the strength of the new tie is influenced by the past relationship or experience between the buyer and its services supplier. For instance, the buyer may have had prior experience with the supplier in another functional area. Or, the supplier could have been subject to certain treatment from the buyer during the supplier selection and negotiation process.

There are two major types of buyer–supplier relationships — adversarial and collaborative (Auster 1994; Gulati 1998; Humphreys, Shiu and Chan 2001). An adversarial relationship is characterized by a competitive price-driven arrangement (Lamming 1993), which is a common approach practiced in the commodity market. On the opposite spectrum is the collaborative type of relationship, which has received considerable academic attention (Monczka, Petersen, Handfield and Ragatz 1998). A collaborative relationship is characterized by the close cooperation between a buyer and a selected group of suppliers typically based on a long-term agreement. Trust and open communications are two characteristics associated with the collaborative buyer–supplier relationship (Chen, Paulraj and Lado 2004). In a collaborative relationship, the buyer and supplier would engage in frequent communication as part of their daily routines.

When the buyer engages in services outsourcing and the supplier is entrusted to offer a consistent level of services to its customers, it is reasonable to expect that the supplier will replicate the existing communication patterns with its new customers. Also, trust in a collaborative relationship tempers the buyer's opportunism or the possibility of intentionally abusing its power of being the bridge. The result is a more transparent information flow between the supplier and the buyer, and subsequently this type of practice would spill over into the new relationship between the supplier and the buyer's customer (Balakrishnan, Mohan and Seshadri 2008). Fewer “surprises” would occur in the resulting contractual relationship among the supplier, the buyer, and the buyer's customer. A cooperative relationship between the buyer and supplier before the outsourcing is expected to continue fostering trust and open communication between the supplier and the buyer's customer. Similarly, an adversarial relationship between the services supplier and the services buyer before the outsourcing arrangements would tend to extend an adversarial relationship between the services supplier and the buyer's customers.

Proposition 3: How the new relationship is formed between the services supplier and the buyer's customer depends largely on the type of relationship the buyer has established with the supplier prior to bridge decay. A collaborative relationship between the services buyer and services supplier tends to foster a collaborative relationship between the services supplier and the buyer's customer. An adversarial relationship between the services buyer and services supplier tends to foster an adversarial relationship between the services supplier and the buyer's customer.

The following example illustrates such a path dependent nature of a newly formed relationship. Aviva, a British insurance firm, outsourced its software development and claims administrative services to three supplying companies in India (Bhalla, Sodhi and Son 2008). To secure services, Aviva formed a strategic alliance with these three services supplying firms, and in 2003, it deepened its commitment to them as it opted for the build–operate–transfer model. As a result, the services outsourcing effort became a total success and Aviva claimed how outsourcing enabled it to improve the service levels to its customers and attain higher customer satisfaction (Aviva 2004). Aviva's outsourcing success has been attributed to the collaborative relationship it had with its suppliers, which in turn worked as a foundation for the resulting relationship between the services supplier and its customers.

Triad after Outsourcing: Bridge Transfer

Once the services supplier is in direct contact with its customers, the buyer will gradually cease to be involved in the creation of services, per its outsourcing intention. When outsourcing a service, the goal of the buyer is to relinquish the outsourced services operation to the supplier and for this supplier to deliver services to its customers without its direct involvement. Otherwise, it will continue to incur administrative costs. Therefore, it is only natural for the buyer to withdraw its involvement and let the supplier take over. In fact, Sanders, Locke, Moore and Autry (2007) documented the existence of “total outsourcing,” where the buyer lets suppliers manage everything. This kind of disconnection between the buyer and its customers is in line with the argument of TCE (Williamson 1975, 2008; Jensen and Meckling 1976; Milgrom and Roberts 1992; Lacity and Willcocks 1995; Ang and Straub 1998). Unless the buyer lets the supplier take care of its customers, it would continue to incur the transaction cost or the “frictional cost” that occurs at the interface with another firm. Therefore, by design, the buyer wants its supplier to interface with its customer and handle the relationship; after all, the supplier is being compensated for taking care of the buyer's customer. The supplier understands this arrangement and will interact with customers directly. Besides setting up the direct contact with customers, the supplier also maintains its contact with the buyer, who is obligated to pay the supplier's invoice. The supplier now spans over two disconnected agents and enjoys the bridge position, as shown in the last network diagram in Figure 4.

Once this bridge transfer occurs, the supplier who used to be in a subservient position to the buyer now takes on a more powerful position. Some scholars (i.e., Stratman 2008) have cautioned services buyers of the power shift between buyers and suppliers. It is true that the buyer still writes the check to the supplier; however, as long as the bridge transfer holds wherein the buyer stays disconnected from the end customer, the supplier enjoys its position of the tertius gaudens. In other words, the supplier now accesses information from disconnected entities, the buyer and its customer, and these two now depend on the services supplier to control information. By accessing information as the new bridge, the services supplier also emerges as the gatekeeper of the information and this role grants it more power in dealing with the buyer. For instance, because the supplier is now in the position of having to provide services, the buyer may be forced to hand off proprietary knowledge to the supplier. Lenox and King (2004) caution how key knowledge resources could be handed off to a services supplier and the buyer would then become beholden to the supplier to transfer the knowledge back when needed. Similarly, Ray Perry, vice president at Avon Products, cautioned similar power shift after the information services outsourcing arrangements. He pointed out the difficulty a services buyer faced during the contract renewal stage when the services supplier became dominant in power (O'Leary 1990). We offer the following proposition as an explanation for such power shift.

Proposition 4: In services outsourcing, the triad structure of bridge decay would evolve into the triad structure of bridge transfer, where the leverage position would be assumed by the supplier.

The new network arrangement increases dependency of the buyer on the services supplier and this shifts the power position (Emerson 1962; Brass and Burkhardt 1993). The question at this point becomes how the supplier would react to this new-found power as the bridge. After all, the supplier is a rational actor that seeks to maximize its utility. As discussed earlier, the extent to which the supplier might abuse its bridge position by acting opportunistically would again largely depend on its past relationship with the buyer. A collaborative past relationship would tend to minimize opportunistic behaviors, but an adversarial relationship would amplify such behaviors. From the supplier's perspective, a collaborative relationship would offer long-term gains while an adversarial relationship would not. The prospective for long-term gains can act as a containment mechanism for suppliers' opportunistic behaviors. When suppliers under a collaborative relationship with its buyers are in the newly found bridge position, they would tend to act less opportunistically. Conversely, suppliers aimed solely at making “quick money” would show more tendencies to take advantage of their bridge position and act opportunistically.

Kodak outsourced its data center operations to IBM in the 1980s and it was viewed as a landmark success (Lacity and Hirschheim 1993). Kodak's director of data center services, Vaughn Hovey, pointed out that the secret to outsourcing success was in the way Kodak saw its suppliers as partners. By entering into strategic alliances with IBM, Kodak effectively contained the opportunistic behavior of IBM. Although the outsourcing contract was only a few pages long and the majority was under gentlemen's agreement (Lacity and Hirschheim 1993), which might be considered a perfect setting for opportunistic behavior, IBM did not behave opportunistically and instead provided superior services to Kodak's customers. However, the reverse can also be true. Unless care is taken, a supplier can act opportunistically and the outsourcing arrangement can fail. For instance, Hirschheim and Lacity (2000) studied a company called Chem2 and its failed outsourcing attempt. Chem2 selected its supplier based on price without paying attention to relationship building. Once the knowledge was transferred, the supplier started to exhibit opportunistic behaviors such as providing low quality service, charging high costs, and reassigning talented employees to work on other company projects. Eventually, Chem2 had to reverse its outsourcing decision.

The underlying reason behind Kodak's services outsourcing success and Chem2's outsourcing failure lies in the nature of the past buyer–supplier relationship or lack thereof. The buyer and supplier in a strategic partnership typically align their goals closer compared with the buyer and supplier in an adversarial relationship, leading to less goal incongruence between them. Additionally, research has shown that a prior history of cooperation between firms has been found to reduce the opportunistic behavior (Parkhe 1993) and decrease perceptions of exchange hazards (Deeds and Hill 1999). Further, in order to build a deeper supplier relationship, the buying company would typically invest in knowing the supplier firm's background, including past performance. Thus we state Proposition 5 below. Note that this proposition differs from Proposition 3. Proposition 5 is concerned with the ties between the buyer and its supplier, whereas Proposition 3 is concerned with the ties between the supplier and the buyer's customers.

Proposition 5: Once the bridge transfer is complete, a collaborative relationship between the buyer and its supplier would mitigate the potential opportunistic behavior by the supplier, while an adversarial relationship would increase it.

Desirable Strategy: Acceptance of Bridge Decay

We have discussed how the services buyer loses its bridge position after the outsourcing occurs. However, if the buyer must give up the bridge position, the last thing it wants to do is to set up its supplier in a new bridge position, because the supplier will act opportunistically. After all, opportunism is a fundamental assumption about firm behavior (e.g., Williamson 1975). We propose that a better alternative would be to fill in the structural hole completely, creating a permanent state of bridge decay. This outcome requires the buyer to maintain close communication with its customer and obtain feedback regarding the performance of the supplier.

Simply because services have been outsourced, this does not mean the buyer should leave everything up to the supplier (Sanders et al. 2007). The buyer should still maintain contact with its customer and monitor how well the supplier is doing its job. This is the hallmark of Honda and Toyota — they keep a close watch on their suppliers, work very closely with them, and ensure a superior level of customer satisfaction (Liker and Choi 2004). Given the real-time delivery of services requiring close interaction, the buyer should monitor the process of the service delivery by the supplier. The buyer should focus on three loci of control: the supplier, its customers, and the relationship between the supplier and the customers. This arrangement is shown in Figure 5.

Figure 5.

Potential State of Permanent Bridge Decay

To address the issues of bridge transfer and the associated transfer of power to a services supplier, the buyer should maintain its information exchange with its customers as well as information exchange with its suppliers. In this scenario, all three players shown in Figure 5 will be connected with one another and thus effectively eliminate structural holes or advantages enjoyed by a bridge. It may appear to the buyer that continued monitoring and involvement seems like a poor option, but the alternative of bridge transfer is worse, because it would mean that the services supplier becomes the full-fledged bridge in the new relationship structure. To maintain control, or more precisely, to guard against the increasing leverage of its supplier, the services buyer must constantly monitor its customers, suppliers and the relationship between its trading partners.

Proposition 6: In a steady state, bridge decay is a much more strategically desirable outcome for the buyer compared to the bridge transfer.

Here the steady state occurs after the implementation of services outsourcing and after the relationship structures for postoutsourcing take place as shown in Figure 4. The concept behind this proposition is exemplified in the creation of the “solution integrator” role at HP (Parker and Anderson 2002). HP created a solution integrator role after the decision to outsource its internal IT support function to an external software support center located in low-cost regions. Employed by HP, these solution integrators served as go-betweens for the external support center and the end customers of HP. Their roles included coordination, monitoring, issue escalation, and issue resolution to ensure a high level of service quality. In this case, the services buyer would know what the supplier knew and engaged in constant information exchange with its customers as well.


Services outsourcing has been laden with stories of failed attempts. We posit that an important root cause of these failures is the lack of understanding of the dynamic nature of the triadic relationships among the services buyer, services supplier, and the buyer's customer. By extending the social network theory into the services outsourcing context, we were able to reveal the shifting relationship structures within the triad of services supply network at different stages of the services outsourcing.

During the contract negotiation stage and before outsourcing arrangements, the services buyer acts as a bridge between the supplier and the buyer's customer and enjoys information and control benefits associated with that bridge position. Once the service outsourcing arrangements are in place and a services supplier begins to interface with the customer, a situation unique to services outsourcing emerges. The services buyer invariably loses its bridge position to the supplier. Loss of the bridge position means loss of leverage. Unless intervened, the services supplier would end up gaining the advantage that comes with being a tertius gaudens. To mitigate this risk, we propose the use of a bridge decay arrangement as the permanent state. This means that the services buyer should continue to monitor the supplier, the buyer's customers, and the relationships between the services supplier and the buyer's customers after the outsourcing arrangements.

This approach may appear to be counter-intuitive. One of the major incentives behind outsourcing in a business setting is cost reduction, and extra monitoring costs extra money. What a buyer typically would prefer to do is only monitor its relationship with the services supplier after outsourcing because the buyer lets its supplier handle the relationship with its customers. After all, the supplier is being compensated for taking care of the buyer's customer. However, the buyer must realize the consequences of such bridge transfer. The field is littered with failed services outsourcing with the supplier as the new bridge, as illustrated earlier, and this outcome can have long-term, negative consequence to the buyer. A common negative outcome is the reduction in customer service quality (Anderson, Fornell and Lehmann 1994; Ittner and Larcker 1998). In this regard, the state of the bridge decay as a permanent state should be considered as a viable option.

Our study integrates social network theory and services outsourcing by focusing on triads and the role of tertius gaudins in the triad. There are ample additional opportunities to extend our research. Future research can expand beyond the triadic structures and examine how the embeddedness of a service firm impacts its structural choices. For example, if a services buyer is embedded in a sparse network or a services supplier relies on the buyer's extended network for future business opportunities, the services buyer may have more enduring leverage over the supplier beyond a single services outsourcing contract. Under this type of scenario, bridge transfer may be a plausible strategy for the buyer. The buyer could potentially rely on its extended network as a containment mechanism for the supplier's opportunistic behaviors.

When the services buyer is the bridge to extended networks that are valuable to the supplier, the buyer can also exert its role as the tertius iungens, the third who joins (Obstfeld 2005). Similar to the tertius gaudens, the tertius iungens also derives benefits from the bridge position. However, in contrast to the role of the teritus gaudens who derives benefits from playing off the nodes on each side of the structural hole, the tertius iungens derives benefits via mediating between two disconnected nodes. An intriguing question then arises as to how the buying firm should play the role of the tertius iungens as well as the tertius gaudens.

Besides social embeddedness, other areas of social network theory deserve further exploration. For example, Tsai and Ghoshal (1998) distinguish three dimensions of social capital — the cognitive dimension, the relational dimension, and the structural dimension. How could these dimensions compete or complement each other in a services outsourcing context? How could they influence the formation and strength of different types of ties among the triad in the service supply network? These are all important questions that have significant managerial relevance.

Our structural relationship-shifting model is geared toward high-contact, pure services outsourcing arrangements where the services provider is in direct and intensive contact with the end customers, such as the outsourcing of a call center, banks, health care services, training and beauty salons. Besides pure services, as we noted earlier, other types of services entail lesser degrees of customer contact. It would be worthwhile to examine the patterns of structural shifting that exist in mixed services and quasi-manufacturing settings.

One interesting scenario that can also be considered in future studies entails complete supply chain disintermediation. Here, the services supplier completely bypasses the services buyer and works directly with the end customers. One common example of such disintermediation happens in tax preparation firms. Once the outsourced services suppliers (accountants) work with its customers, they are then able to obtain the contact information of these customers. They would have the option of contacting those customers directly for future tax returns, completely bypassing the tax preparation firm. Another interesting example happens in Internet portals such as the travel site hosted by In this case, the travelers (customers) use the portal (buyer) to locate an airline (services supplier) that sells the least expensive ticket. Once they have found the ticket, they can go directly to the airline's website to make the purchase and take advantage of a small savings by purchasing directly from the airline's website. In the above two scenarios, no triadic relationships remain among the services supplier, services buyer and buyer's customer. At the end of the transaction, the services buyer is completely dropped out of the supply network. We believe this loss of intermediary entity deserves further investigation.

Our model is limited largely to the substitution type of outsourcing, as opposed to the abstention type (Gilley and Rasheed 2000). As such, we are able to focus on the conscious choice of replacing internal production with external purchases (i.e., substitution), which contrasts the regular “sourcing” decisions that occur because of the lack of internal production capabilities (i.e., abstention). In other words, our model does not apply to some of the Internet sellers who purchase logistics services from carriers such as UPS to deliver their products to their customers. In this example, these Internet sellers simply do not have the capability to deliver their products on their own and have to purchase services from UPS. It is a regular sourcing decision, not an outsourcing decision. Therefore, the network dynamics involved in abstention types of services sourcing is left to future studies.

Services outsourcing practices are indeed very complex. Social network theory offers a wide array of opportunities to tackle this complex task. Our study is only the first step toward analyzing this challenging issue. Nevertheless, we have been able to reveal insights omitted by other theories used in services outsourcing studies. The insights gained from integrating services outsourcing and social networks concepts are of great theoretical and managerial relevance that can help us move the field forward.

Mei Li (Ph.D. Candidate) is a Ph.D. student at Arizona State University in Tempe, AZ. She previously served as a global program manager in supply chain information technologies at the headquarters of the Hewlett-Packard Company and had a wealth of industry experience in supplier selection, integration, evaluation and management. Mei Li's research interests include services operations, services outsourcing, self-services, innovation and sustainability. Her latest project empirically tests the implications of the shifting of the triadic relationship structure in the services supply network. Her earlier work on the evolution of inductive case studies with Dr. Barratt and Dr. Choi has been recognized as the finalist for the Chan Hahn Best Paper Award by the Operations Management Division of the Academy of Management Conference 2007.

Thomas Y. Choi (Ph.D., University of Michigan) is a professor of supply chain management and John G. and Barbara A. Bebbling Professor in Business at Arizona State University in Tempe, AZ. He also holds the title of Outstanding Operations Management Professor at Yonsei University. He received his A.B. degree from the University of California at Berkeley. His research interests are: dyads and triads in supply networks, supply base rationalization, second- and third-tier suppliers, supplier–supplier relationships, and supply chain disintermediation. His articles have been published in the Academy of Management Executive, Decision Sciences, Harvard Business Review, Journal of Operations Management, Journal of Supply Chain Management, Production and Operations Management, Supply Chain Management Review, and others.