9

INFORMATION SYSTEMS SOURCING

After 13 years, Kellwood, an American apparel maker, ended its soups-to-nuts IS outsourcing arrangement with EDS. The primary focus of the original outsourcing contract was to integrate 12 individual acquired units with different systems into one system. Kellwood had been satisfied enough with EDS' performance to renegotiate the contract in 2002 and 2008, even though at each renegotiation point it had considered bringing the IS operations back in-house, or backsourcing. The 2008 contract iteration resulted in a more flexible $105 million contract that EDS estimated would save Kellwood $2 million in the first year and $9 million over the remaining contract years. But the situation at Kellwood had changed drastically. In 2008, Kellwood had been purchased by Sun Capital Partners and taken private. The Chief Operating Officer, who was facing a mountain of debt and possibly bankruptcy, wanted to consolidate and bring the operations back in-house in 2012 to give some order to the current situation and reduce costs. Kellwood suffered from a lack of IS standardization as a result of its many acquisitions. The Chief Information Officer (CIO) recognized the importance of IS standardization and costs, but she was concerned that the transition from outsourcing to insourcing would cause serious disruption to IS service levels and project deadlines if it went poorly. Kellwod hired a third-party consultant to help it explore the issues and decided that backsourcing would help it save money and respond to changes caused by both the market and internal forces. Kellwood decided to backsource and started the process in late 2009. It carefully planned for the transition and the implementation went smoothly. By performing streamlined operations in-house it was able to report an impressive $3.6 million savings, or about 17% of annual IS expenses after the first year.¹

The Kellwood case demonstrates a series of decisions made in relation to sourcing. Both the decision to outsource, and then the decision to bring IS operations back inhouse, were based on a series of factors. These factors are similar to those used by many companies in the sourcing decisions. The global outsourcing market has been growing steadily. Companies of all sizes pursue outsourcing arrangements, and many multimillion deals have been widely publicized. As more companies adopt outsourcing as a means of controlling IS costs and acquiring “best of breed” capabilities, managing these supplier relationships becomes increasingly important. IS departments must maximize the benefit of these relationships to the enterprise and preempt problems that might occur. Failure in this regard could result in deteriorating quality of service, loss of competitive advantage, costly contract disputes, low morale, and loss of key personnel.

How IS services are provided to a firm has become an important strategic and tactical discussion. As briefly mentioned in Chapter 6, there are numerous alternatives to sourcing computing power, applications, and infrastructure. This chapter looks at the sourcing cycle to consider the full range of decisions related to who should perform the information systems work of an organization. The cycle begins with a decision to make or buy information services and products. If the decision is to buy, a series of questions must be answered about where and how these services should be delivered or products developed. The discussion in this chapter is built around the Sourcing Decision Cycle Framework discussed in the next section. Considering the answers to sourcing decisions can help explain a number of terms associated with sourcing: insourcing, outsourcing, cloud computing, full outsourcing, selective outsourcing, onshoring, offshoring, nearshoring, farshoring, captive center, and backsourcing. For each type of sourcing decision the risks, or likelihood of something negative occurring as a result of the decision, are discussed and some steps that can be taken to manage the risks are proposed.

SOURCING DECISION CYCLE FRAMEWORK

Sourcing doesn't really just involve one decision. It involves many decisions. The rest of this chapter is built around critical sourcing decision making as shown in Figure 9.1. The chapter headings are tied to key decisions in Figure 9.1. Though it is a cycle, and we could start anywhere, we choose to start with the original make-or-buy decision. In cases where the “buy” option is selected and the company outsources, the client company must decide on “how” and “where.” The answers to the “how” question focus on the scope of the outsourcing and the steps that should be taken to ensure success. The answers to the “where” question focus on whether to work with an outsourcing provider in its own country, offshore, or in the cloud. If the company decides to go offshore because labor is cheaper or needed skills are more readily available, the client company is faced with another decision: It must decide if it wants the work done in a country that is relatively nearby or in a country that is quite distant. It may even decide to have one of its own companies, a captive center, perform the work. Finally, the client company settles in on an outsourcing provider (or decides to do its own IS work). After a while, it faces another decision. It periodically must evaluate the arrangement and see whether a change is in order. If the in-house work is unsatisfactory or if other opportunities have become available that are preferable to the current arrangement, then the company may turn to outsourcing. If, on the other hand, the outsourcing arrangement is unsatisfactory, the client company has several options to consider: correct any existing problems and continue outsourcing with its current provider, outsource with another provider, or backsource. If the company decides to make a change in its sourcing arrangements at this point, the sourcing decision cycle starts over again.

FIGURE 9.1 Sourcing Decision Cycle Framework.

Starting the Cycle: Make or Buy Decision

Managers decide whether to make or buy information services. A simple “make” decision often involves insourcing some or all of their infrastructure, and a simple “buy” decision is mostly involves outsourcing. In its simplest form, the make-or-buy decision hinges on whether to insource (“make”) or outsource (“buy”).

Insourcing

The most traditional approach to sourcing is insourcing, or the situation in which a firm provides IS services or develops IS in its own in-house IS organization. Several “Yes” answers to the questions in Figure 9.2 favor the decision to insource. Probably the most common is to keep core competencies in house. Managers are concerned that if they outsource a core competency, they risk losing control over that competency or losing contact with suppliers who can help them remain innovative in relation to that competency. Failing to control the competency or stay innovative is a sure way to forfeit the company's competitive advantage. On the other hand, by outsourcing commodity work, a firm can concentrate on its core competencies. Other factors that weigh in favor of insourcing are having an IS service or product that requires considerable security or confidentiality, or that requires resources that are adequately available in house (i.e., qualified personnel or IS professionals with the needed skills).

In some companies, the IS function is underappreciated by top management. As long as everything is running smoothly, top managers may not appreciate the services and products provided by the IS organization. Often, IS departments that insource have found they have to compete for resources differently than if they outsource. It is necessary for them to have the respect and support from top management needed to acquire resources and get the department's job done. A major risk of insourcing is that the complexities of running IS in-house requires management attention and resources that might better serve the company if focused on other value-added activities.

Outsourcing

Outsourcing means the purchase of a good or service that was previously provided internally or that could be provided internally, but is now provided by outside vendors. In the early days of outsourcing, providers often took over entire IS departments, including people, equipment, and management responsibility. The primary motivation for outsourcing then was on reducing costs. This classic approach prevailed through most of the 1970s and 1980s, but then experienced a decline in popularity. In 1989, Eastman Kodak Company's multivendor approach to meeting its IS needs created the “Kodak effect.” Kodak outsourced its data center operations to IBM, its network to Digital Equipment Company, and its desktop supply and support operations to Businessland.² Kodak managed these relationships through strategic alliances.³ Kodak retained a skeleton IS staff to act on behalf of its business personnel with outsource providers. Its approach to supplier management became a model emulated by Continental Bank, General Dynamics, Continental Airlines, and National Car Rental and many more.⁴

FIGURE 9.2 Make or buy? Questions and risks.

Kodak's watershed outsourcing arrangement ushered in changes to outsourcing practices in the 1990s that put all IS activities up for grabs, including aspects that provide competitive advantage. As relationships with outsourcing providers become more sophisticated, companies realize that even such essential functions as customer service are sometimes better managed by experts on the outside. Over the years, motives for outsourcing broadened beyond cost control. The next section examines factors and risks to be considered in making the outsourcing decision. The sourcing strategy suggested by the answers to key “Make/Buy” questions and associated risks are listed in Figure 9.2.

Factors in the Outsourcing Decision

Under what conditions is the answer to the “Insource/Outsource” question “Outsource”? That is, what factors drive companies to decide to outsource? There are three factors that are likely to lead to the decision to outsource: cheaper costs due to economies of scale, ability to handle peaks in processing, and the client company's need to consolidate data centers.

One of the most common reasons given for outsourcing is the desire to reduce costs. Outsourcing providers derive savings from economies of scale that client companies often can't realize. Providers achieve these economies through centralized (often “greener”) data centers, preferential contracts with suppliers, and large pools of technical expertise. Most often, enterprises lack such resources on a sufficient scale within their own IS departments. A single company may need only 5,000 PCs, but an outsourcing provider can negotiate a contract for 50,000 and achieve a much lower unit cost.

Second, as long as contract terms effectively address contingencies, the larger resources of an outsourcing provider make available greater capacity on demand. For instance, at year-end, outsourcing providers potentially can allocate additional mainframe capacity to ensure timely completion of nightly processing in a manner that would be impossible for an enterprise running its own bare-bones data center.

Third, an outsourcing provider may help a client company overcome inertia to consolidate data centers that could not be consolidated by an internal group or following a merger or acquisition. Outsourcing may also offer an infusion of cash as a company sells its equipment to the outsourcing vendor.

Above it was noted that if the service or product under consideration involved a core competency, then the company should strongly consider insourcing. However, if the product or service is considered to be a commodity instead of a core competency then there are some distinct advantages to outsourcing. By bringing in outside expertise, client company management often can focus less attention on IS operations and more on core activities. Although they must still manage the relationships with outsourcing providers, using outsourcing providers frees up their time so that they can devote their energies, and that of their employees, to areas that reflect core competencies for the business. It is important to remember, however, that client company managers are ultimately still responsible for IS services provided to their firm.

It was noted in discussing insourcing that if a client company does not have employees with the training, experience, or skills in-house to successfully implement the new technologies it should consider outsourcing. Outsourcing can be employed to help a company transition to new technologies. This is because outsourcing providers generally offer access to larger pools of talent with more current knowledge of advancing technologies. For example, many outsourcing providers gain vast experience solving business intelligence problems, whereas IS staff within a single company only have limited experience, if any. The provider's experienced consultants are more readily available to the marketplace advances and best practices than any comparably trained and experienced in-house IT professionals. Client companies turn to outsourcing providers to help them implement such technologies as Enterprise 2.0, Web 2.0 tools, and ERP systems.

Outsourcing providers also have an added advantage when it comes to having employees with the necessary skills, training, and experience because they can specialize in IS services. Outsourcing providers' extensive experience in dealing with IS professions likely helps them to understand how to hire, manage, and retain IS staff effectively. Outsourcing providers often can offer IS personnel a professional environment that a typical company cannot afford to build. For example, a Web designer would have responsibility for one Web site within a company, but for multiple sites when working for an outsourcing provider. It becomes the outsourcing provider's responsibility to find, train, and retain highly marketable IS talent. An outsourcing provider often opens greater opportunity for training and advancement in IT than can a single IS organization. Outsourcing relieves a client of costly investments in continuous training so that IS staff can keep current with marketplace technologies and the headaches of hiring and retaining a staff that easily can change jobs with more pay or other lures. These factors are listed in Figure 9.2.

Outsourcing Risks

Opponents of outsourcing cite a considerable number of risks (see Figure 9.2). A manager should consider each of these before making a decision about outsourcing. Each can be mitigated with effective planning and ongoing management.

First, outsourcing requires that a client company surrender a degree of control over critical aspects of the enterprise. The potential loss of control could extend to several areas: control of the project, scope creep, the technologies, the costs, financial controls, accuracy and clarity of financial reports, and the company's IS direction. By turning over data center operations, for example, a company puts itself at the mercy of an outsourcing provider's ability to manage this function effectively. A manager must choose an outsourcing provider carefully and negotiate terms that encourage an effective working relationship.

Second, outsourcing clients may not adequately anticipate new technological capabilities when negotiating outsourcing contracts. Outsourcing providers may not recommend so-called bleeding-edge technologies for fear of losing money in the process of implementation and support, even if implementation would best serve the client. Thus, poorly planned outsourcing can result in a loss in IS flexibility. For example, some outsourcing providers were slow to adopt social technologies for their clients because they feared the benefits would not be as tangible as the costs of entering the market. This reluctance impinged on clients' ability to realize social business strategies. To avoid this problem, outsourcing clients should have a chief technology officer (CTO) or technology group that is charged with learning about and assessing emerging technologies for their ability to support their company's business strategy.

Third, by surrendering IS functions, a company gives up any real potential to develop them for competitive advantage—unless, of course, the outsourcing agreement is sophisticated enough to comprehend developing such advantage in tandem with the outsourcing company. However, even these partnerships potentially compromise the advantage when ownership is shared with the outsourcing provider, and the advantage may become available to the outsourcing provider's other clients. Under many circumstances, the outsourcing provider becomes the primary owner of any technological solutions developed, which allows the outsourcing provider to leverage the knowledge to benefit other clients, possibly even competitors of the initial client.

Fourth, contract terms may leave clients highly dependent on their outsourcing provider, with little recourse in terms of terminating troublesome provider relationships. That is, the clients may be locked-in to an arrangement that they no longer want. The more reliant the client company is on the provider the harder and more expensive it is to switch to another outsourcing provider when the contract sours. Despite doing due diligence and background checks, the outsourcing provider may be unreliable or go out of business before the end of the contract. The risk of over-reliance for any number of reasons typically increases as the size of the outsourcing contract increases. DHL Worldwide Express entrusted 90% of its IT development and maintenance projects to a large Indian-based company, Infosys. “There's a lot of money wrapped up in a contract this size, so it's not something you take lightly or hurry with,” said Ron Kifer, DHL's Vice President of Program Solutions and Management.⁵ Clearly DHL faced considerable risk in offshoring with Infosys because of its reliance on them.

Fifth, when a company turns to an outsourcing provider, it must realize that its competitive secrets might be harder to keep. Its databases are no longer kept in house, and the outsourcing provider's other customers may have easier access to sensitive information. Although all outsourcing agreements contain clauses to keep customer data and system secure, managers still voice concern about data security and process skills when it's managed by a third party. This risk is mitigated by thinking through the security issues carefully and implementing controls where possible. Often, the outsourcing provider has more secure processes and practices in place simply because their business depends on it—it's a competitive necessity and often a core competency of the outsourcing provider.

Sixth, the outsourcing provider's culture or operations may be incompatible with that of the client company, making it difficult to deliver the contracted service or system. Conflicts between the client's staff and the staff of the outsourcing provider may delay progress or hurt the quality of the service or product delivered by the outsourcing provider.

Seventh, although many companies turn to outsourcing because of perceived cost savings, these savings may never be realized. Typically, the cost savings are premised on activities that were performed by the company. However, implementation of new technologies may fail to generate any savings because the old processes on which they were premised are no longer performed. Further, the outsourcing client is, to some extent, at the mercy of the outsourcing provider. Increased volumes due to unspecified growth, software upgrades, or new technologies not anticipated in the contract may end up costing a firm considerably more than it anticipated when it signed the contract. Also, some savings, although real, may be hard to measure.

Finally, there may be challenges that come from working with multiple vendors. While multiple vendors allow client companies to distribute work to the “best of breed,” it comes with its downsides. Having more vendors requires more coordination than with working with a single outsourcing provider. Further, when a major problem occurs, there may be a tendency to “finger-point.” That is, each vendor may claim that the problem is caused by, or can only be corrected by, another vendor. And as vendors expand their service offerings, an unexpected competition among vendors can result.

Decisions about How to Outsource Successfully

Clearly the decision about whether or not to outsource must be made with adequate care and deliberation. It must be followed with numerous other decisions about how to mitigate outsourcing risks and make the outsourcing arrangement work. Three major decision areas are selection, contracting, and scope.

Selection-related decisions focus on finding compatible outsourcing providers whose capabilities, managers, internal operations, and culture complement those of the client company. This means that compatibility and cultural fit might trump price, especially when long-term partnerships are envisioned.

Many “how” decisions center around the outsourcing contract. In particular, client companies must ensure that contract terms allow them the flexibility they require to manage and, if necessary, sever supplier relationships. The ten-year contracts that were so popular in the early 1990s are being replaced with shorter-duration contracts lasting three to five years and full life-cycle service contracts that are broken up into stages. In fact, the average contract for outsourcing today, is nearly a third of the 2000 average of $360 million.⁶ Often, outsourcing arrangements have formal service contracts between clients and outsourcing providers, called Service Level Agreements (SLAs), where the level of service is defined. SLAs often describe the contracted delivery time and expected performance of the service. Contracts are tightened by adding clauses describing actions to be taken in the event of a deterioration in quality of service or non-compliance to service-level agreements. Service levels, baseline period measurements, growth rates, and service volume fluctuations are specified to reduce opportunistic behavior on the part of the outsourcing vendor. Research demonstrates that tighter contracts tend to lead to more successful outsourcing arrangements.⁷ Unfortunately, a tight contract does not provide much solace to an outsourcing client when an outsourcing provider goes out of business. It also does not replace having a good relationship with the outsourcing provider that allows the client to work out problems when something unanticipated occurs.

Most enterprises outsource at least some IS functions. This is where scope questions come into play. Defining on the scope of outsourcing means that the client must decide on whether to pursue outsourcing fully or selectively.

Full outsourcing implies that an enterprise outsources all its IS functions from desktop services to software development. An enterprise would outsource everything if it does not view IT as a strategic advantage that it needs to cultivate internally. Full outsourcing can free resources to be employed in areas that add greater value. It can also reduce overall cost per transaction due to size and economies of scale.⁸ Many companies outsource IS simply to allow their managers to focus attention on other business issues. Others outsource to accommodate growth and respond to their business environment. Kellwood, the case discussed at the beginning of the chapter, appeared to have used full outsourcing to improve operations.

With selective outsourcing, an enterprise chooses which IT capabilities to retain in house and which to give to an outsider. A “best-of-breed” approach is taken in which suppliers are chosen for their expertise in specific technology areas. Possible areas for selective sourcing include Web site hosting, Web 2.0 applications, business process application development, help desk support, networking and communications, social IT services, and data center operations. Although an enterprise can acquire top-level skills and experience through such relationships, the effort required to manage them grows tremendously with each new provider. Still, selective outsourcing, sometimes called “strategic sourcing,” reduces the reliance of the client company being locked-in to outsourcing with only one provider. It also provides greater flexibility and often better service due to the competitive market.⁹ To illustrate, an enterprise might retain a specialist firm to develop social business applications and at the same time select a large outsourcing provider, such as EDS, to assume mainframe maintenance.

Consider JetBlue, an airline that recently turned to Verizon to manage its IT infrastructure—its network, data center, and help desk. The six-year contract with Verizon will allow the data centers to scale as JetBlue grows and it will also help JetBlue “reduce the cycle time for delivery of those capabilities and allow the rest of IT to focus on other capabilities,” said JetBlue CIO, Joe Eng. Eng asserts that JetBlue will still have control over IT: “We own the decision paths, the service-level agreements and what direction we want to take, but Verizon will be key in the implementation.” Verizon was chosen over other providers for a number of reason, but especially because the operation of networks is its core business.¹⁰

Deciding Where—Onshore, Offshore or in the Cloud?

Until recently, outsourcing options were either to use services onshore (worked performed in the same country as the client) or offshore (work performed in a distant country). More recently a new sourcing option has become more available and more accepted by managers: cloud computing. Below we describe each of the three sourcing options. We also describe some answers to the “how” to make the arrangement successful. Many best practices were discussed in the previous subsection because they are common to all three outsourcing options. Some are unique to the various options, as indicated in Figure 9.3.

FIGURE 9.3 Trade-offs between outsourcing options.

Source: Partially adapted from J. Rottman and M. C. Lacity, “Twenty Practices for Offshore Sourcing,” MIS Quarterly Executive (2004), 3(3), 119, Table 1.

Cloud Computing

As discussed in Chapter 6, cloud computing is the dynamic provisioning of third-party-provided IT services over the Internet. Companies offering cloud computing make an entire data center's worth of servers, networking devices, systems management, security, storage, and other infrastructure available to their clients. In that way their clients can buy the exact amount of storage, computing power, security, or other IT functions that they need, when they need it, and pay only for what they use. Thus, the client can realize cost savings by sharing the provider's resources with other clients. They also provide 24/7 access from multiple mobile devices, high availability for large backup data storage, and ease of use.

Netflix realized the advantages of cloud computing to support its strategic initiative to stream movies to its customers instead of mailing them DVDs. To do so it needed so much more infrastructure that the cloud appeared to be their only option. “Netflix.com is nearly 100% in the cloud. . . We really couldn't build data centers fast enough,” says Jason Chan, Netflix's cloud security architect. The introduction of a Netflix application for iPhones will place even greater spikes in demand, at least temporarily. But Chan isn't concerned: “That's what cloud is really intended for.”¹¹

Cloud computing comes in many different forms. Options include on-premise or private clouds, community clouds, public clouds, and hybrid clouds that are combinations of two or more other clouds. In private clouds the data is managed by the company and remains within the company's existing infrastructure, or it is managed offsite by a third party for the company. In community clouds, the cloud infrastructure is shared by several organizational and supports the shared concerns of a specific community. In public clouds data is stored outside of the corporate data centers in the cloud provider's environment. Public clouds include:

• Infrastructure as a Service (IaaS)—provides infrastructure through grids or clusters or virtualized servers, networks, storage, and systems software designed to augment or replace the functions of an entire data center; the customer may have full control of the actual server configuration allowing more risk management control over the data and environment.
• Platform as a Service (PaaS)—provides services using virtualized servers on which clients can run existing applications or develop new ones without having to worry about maintaining the operating systems, server hardware, load balancing, or computing capacity; the cloud provider manages the hardware and underlying operating system, which limits their enterprise risk management capabilities.
• Software as a Service (SaaS)—provides software application functionality through a Web browser. Both the platform and the infrastructure are fully managed by the cloud provider with means that if the operating system or underlying service isn't configured correctly, the data at the higher application layer may be at risk. This is the most widely known and used form of cloud computing. SaaS is sometimes calls an ASP, or Application Service Provider.¹²

Despite its advantages, some managers shy away from cloud computing. They are concerned about security, specifically about external threats from remote hackers and security breaches as the data travels to and from the cloud. Tied to the concerns about security are concerns about data privacy. The standards, monitoring, and maintenance tools for cloud computing are still not mature. This makes security, interoperability, and data mobility difficult. Finally, the ability to tailor service-level requirements, such as uptime, response time, availability, performance, and network latency, to the specific needs of a client is far less than with insourcing, or even with many other outsourcing options. To manage this risk, an SLA needs to spell out these requirements.

Onshoring

Outsourcing does not necessarily mean that the IT services and software development is shipped abroad. Onshoring, also called inshoring, is performing outsourcing work domestically. Onshoring may be considered the “opposite” of offshoring. A growing trend in onshoring in the United States is rural sourcing, which is hiring outsourcing providers with operations in rural parts of America. Rural sourcing firms can be competitive because they take advantage of lower salaries and living costs when compared to firms in metropolitan areas. Dealing with a rural company can have advantages in terms of closer time zones, similar culture, and fewer hassles that crop up when dealing with foreign outsourcing providers. However, the rural sourcing firms are usually too small to handle large-scale projects and may not have the most technologically advanced employees. Rural sourcing is often viewed as more politically correct.¹³ See Figure 9.4 for a discussion of related political issues.

FIGURE 9.4 Government involvement with offshoring.

Source: Partially based on information from Erran Carmel and Paul Tjia, Offshoring Information Technology (Cambridge, UK: University Press, 2005).

Offshoring

Offshoring (which is short for outsourcing offshore) is when the IS organization uses contractor services, or even builds its own data center in a distant land. The functions sent offshore range from routine IT transactions to increasingly higher-end, knowledge-based business processes.

Programmer salaries can be a fraction of those in the home country in part because the cost of living and the standard of living in the distant country are much lower, maybe as much as 70% when only considering direct labor costs. However, these savings come at a price because other costs increase. Additional technology, telecommunications, travel, process changes, and management overhead are required to relocate and supervise operations overseas. For example, during the transition period, which can be rather lengthy, offshore workers must often be brought to the U.S. headquarters for extended periods to become familiar with the company's operations and technology. Because of the long transition period, it may often take several years for offshoring's labor savings to be fully realized. And even if they are realized, they may never reflect the true cost to a country. Many argue, especially those who have lost their jobs to offshore workers, that offshoring cuts into the very fiber of the society in the origin country where companies are laying off workers. Yet it helps the economies of the countries where offshoring is performed. For example, in December 2011, the IT services industry in India was a $76 billion industry, the largest source for IT services and offshoring, primarily from foreign companies offshoring.¹⁴

Even though the labor savings are often very attractive, companies sometimes turn to offshoring for other reasons. The employees in many offshore companies are typically well educated (often holding master's degrees) and proud to work for an international company. The offshore service providers are often “profit centers” that have established Six Sigma, ISO 9001, or another certification program. They usually are more willing to “throw more brainpower at a problem” to meet their performance goals than many companies in the United States or Western Europe. In offshore economies, technology know-how is a relatively cheap commodity in ample supply.¹⁵

Offshoring raises the fundamental question of what you send offshore, and what you keep within your enterprise IS organization when implementing the selective outsourcing model. Because communications are made difficult by differences in culture, time zones, and possibly language, outsourced tasks are usually those that can be well specified. They typically, but not always, are basic non-core transactional systems that require little in-depth knowledge of the users or customers. In contrast, early stage prototypes and pilot development are often kept in-house because this work is very dynamic and needs familiarity with business processes. Keeping the work at home allows CIOs to offer learning opportunities to in-house staff. In summary, the costs savings that lure many companies to turn to offshoring need to be assessed in relation to the increased risks and communication problems in working with offshore workers and relying on them to handle major projects.

Deciding Where Abroad—Nearshoring, Farshoring, or Captive Center?

As the outsourcing phenomenon has matured, the marketplace has differentiated across ways of offshoring. Offshoring can be either relatively proximate (nearshoring) or in a distant land (farshoring). An alternative to offshoring is a captive center. Each of these is described in more detail below.

Farshoring

Farshoring is a form of offshoring that involves sourcing service work to a foreign, lower-wage country that is relatively far away in distance or time zone (or both). For big outsourcing countries such as the United States and United Kingdom, India and China are the most popular farshoring destinations. Ironically, companies in India and China are now farshoring themselves to countries with lower labor costs.

Nearshoring

Nearshoring on the other hand, is when work is sourced to a foreign, lower-wage country that is relatively close in distance or time zones. With nearshoring, the client company hopes to benefit from one or more ways of being close: geographically, temporally, culturally, linguistically, economically, politically, or from historical linkages. Nearshoring basically challenges the assumption on which offshoring is premised: Distance doesn't matter. The advocates of nearshoring argue that distance does matter, and being closer on one or more of these dimensions, the client company faces less challenges in terms of communication, control, supervision, coordination, or bonding socially.

A recent analysis of the nearshoring literature found three major global clusters of countries focused on building a reputation as a home for nearshoring: a cluster of 20 nations around the United States and Canada, a cluster of 27 countries around Western Europe, and a smaller cluster of three countries in East Asia. This smaller cluster contains China, Malaysia, and Korea.¹⁶

The ways, or dimensions, of being close clearly extend beyond distance and time zone. For example, language makes a difference in nearshoring. That is why Latin American nearshoring destinations are appealing to Texas or Florida, where there is a large Spanish-speaking population and why French-speaking North African nations are appealing to France. These dimensions likely play a key role when companies are trying to decide between a nearshore or a farshore destination (particularly India). Ironically, India, which exports roughly five times the software of the strictly nearshoring nations, is responding to the competitive threat that these nations pose by offering clients nearshoring options. For example, India-based Tata Consulting Services (TCS) offers its British clients services that are nearshore (Budapest, Hungary), farshore (India), or onshore (London, United Kingdom). It is likely that the differentiation based on “distance” is likely to continue to be important in the outsourcing arena.

Captive Centers

With the cost advantages firms have found in less expensive economies, and to manage the risks associated with outsourcing vendors managing such an important resource as the information systems, a new type of sourcing arose: captive centers. A captive center is an overseas subsidiary that is set up to serve the parent company. Firms have set up these subsidiaries to operate like an outsourcing provider, but they are owned by the firm. Two common types of captive centers are hybrid and shared.¹⁷ The hybrid captive center typically performs the more expensive, higher-profile or mission-critical work for the parent company and outsources the more commoditized work that is more cheaply provided by an offshore provider. The shared captive center performs work for both a parent company and external customers. Captive centers can be nearshore or farshore. Their distinguishing characteristics are that they are in less expensive locations, usually away from the company's headquarters or major operating units, and that they are owned by the parent company rather than by an outsourcing provider.

Selecting an Offshore Destination: Answering the “Where Abroad?” Question

A difficult decision that many companies face is selecting an offshoring destination. To answer the Where Abroad? question client companies must consider attractiveness, level of development, and cultural differences.

Attractiveness

Approximately 100 countries are now exporting software services and products. For various reasons, some countries are more attractive than others as hosts of offshoring business because of the firm's geographic orientation. With English as the predominant language of outsourcing countries (i.e., United States and United Kingdom), countries with a high English proficiency are more attractive than those where different languages are spoken. Geopolitical risk is another factor that affects the use of offshore firms in a country. Countries on the verge of war, countries with high rates of crime, and countries without friendly relationships with the home country are typically not suitable candidates for this business. Regulatory restrictions, trade issues, data security, and intellectual property also affect the attractiveness of a country for an offshoring arrangement. Hiring legal experts who know the laws of the provider's company can mitigate legal risks. Nonetheless, some countries are more attractive than others because of their legal systems. The level of technical infrastructure available in some countries also can add to or detract from the attractiveness of a country. Although a company may decide that a certain country is attractive overall for offshoring, it still must assess city differences when selecting an offshore outsourcing provider or creating wholly owned subsidiaries (“captives”). For example, Chennai is a better location in India for finance and accounting, but Delhi has better call-center capabilities.¹⁸

Some countries created an entire industry of providing IT services through offshoring. India, for example, took an early mover advantage in the industry. With a large, low-cost English-speaking labor pool, many entrepreneurs set up programming factories that produce high-quality software to meet even the toughest standards. One measure of the level of proficiency of the development process within an IS organization is the Software Engineering Institute's Capability Maturity Model (CMM)¹⁹ Level 1 means that the software development processes are immature, bordering on chaotic. Few processes are formally defined, and output is highly inconsistent. At the other end of the model is level 5, where processes are predictable, repeatable, highly refined, and consistently innovating, growing, and incorporating feedback. The software factories in many Indian enterprises are well known for their CMM level 5 software development processes, making them extremely reliable, and, thus, desirable as vendors. However, if the client company is not at the same CMM level as the provider, it may want to specify which CMM processes it will pay for to avoid wasting money. Further, it may seek to elevate its own CMM certification to close the process gap between what it can do and what the outsourcing provider can do.

Development Tiers

A very important factor in selecting an offshore destination is the level of development of the country, which often subsume a variety of other factors. For example in the highest tier, the countries have an advanced technological foundation and a broad base of institutions of higher learning. Carmel and Tjia suggest that there are three tiers of software exporting nations:²⁰

• Tier 1: Mature Software Exporting Nations—These include such highly industrialized nations as United Kingdom, United States, Japan, Germany, France, Canada, the Netherlands, Sweden, and Finland. It also includes the three “I's” (i.e., India, Ireland, and Israel) that became very prominent software exporters in the 1990s, as well as China and Russia, which entered the tier in the 2000s.
• Tier 2: Emerging Software Exporting Nations—These nations are the up-and-comers. They tend to have small population bases or unfavorable conditions such as political instability or an immature state of economic development. Countries in this tier include Brazil, Costa Rica, South Korea, and many Eastern European countries.
• Tier 3: Infant Stage Software Exporting Nations—These nations have not significantly affected the global software market, and their software industry is mostly a “cottage industry” with smaller, isolated firms. Some Tier 3 countries are Cuba, Vietnam, Jordan, and 15 to 25 others.

The tiers were determined on the basis of industrial maturity, the extent of clustering of some critical mass of software enterprises, and export revenues. The higher tiered countries tend to offer higher levels of skills, but also higher costs.

Cultural Differences

Often misunderstandings arise because of differences in culture and, sometimes, language. For example, GE Real Estate's CIO quickly learned that American programmers have a greater tendency to speak up and offer suggestions, whereas Indian programmers might think something does not make sense, but they go ahead and do what they were asked, assuming that this is what the client wants.²¹ Thus, a project that is common sense for an American worker—like creating an automation system for consumer credit cards—may be harder to understand and take longer when undertaken by an offshore worker. The end result may be a more expensive system that responds poorly to situations unanticipated by its offshore developers. It is important to be aware of and to manage the risks due to cultural differences.

Sometimes cultural and other differences are so great that companies take back in-house operations that were previously outsourced offshore. Carmel and Tjia outlined some examples of communication failures with Indian developers due to differences in language, culture, and perceptions about time:²²

• Indians are less likely than Westerners, especially the British, to engage in small talk.
• Indians often are not concerned with deadlines. When they are, they are likely to be overly optimistic about their ability to meet the deadlines of a project. One cultural trainer was heard to say, “When an Indian programmer says the work will be finished tomorrow, it only means it will not be ready today.”²³
• Indians, like Malaysians and other cultures, are hesitant about saying “no.” Questions where one option for response is “no” are extremely difficult to interpret.
• What is funny in one culture is not necessarily funny in another culture.

Reevaluation—Status Quo or Change?

The final decision in the Sourcing Decision Cycle requires an assessment as to whether the sourcing arrangement is working as it should be. If everything is satisfactory then the arrangement can continue. If everything is not totally satisfactory, adjustments may need to be made to the arrangement. If the arrangement is unsatisfactory, another outsourcing provider may be selected or backsourcing may occur. Kellwood, the company described at the start of this chapter, frequently re-evaluated its outsourcing arrangements. Eventually it backsourced.

Backsourcing is a business practice in which a company takes back in-house assets, activities, and skills that are part of its information systems operations and were previously outsourced to one or more outside IS providers.²⁴ It may be partial or complete reversal of an outsourcing contract. A growing number of companies around the globe have brought their outsourced IS functions back in-house after terminating, renegotiating, or letting their contracts expire. Some companies, such as Continental Airlines, Cable and Wireless, Halifax Bank of Scotland, Sears, Bank One, and Xerox, have backsourced contracts worth over a billion dollars or more.

The biggest backsourcing of a contract to date was the one that JP Morgan Chase had signed with IBM for a whopping $5 billion dollars. JP Morgan Chase terminated its contract and brought information systems (IS) operations back in house only 21 months into a seven-year mega-contract. The CIO of JP Morgan Chase, Austin Adams, stated at that time, “We believe managing our own technology infrastructure is best for the long-term growth and success of our company, as well as our shareholders. Our new capabilities will give us competitive advantages, accelerate innovation, and enable us to become more streamlined and efficient.”²⁵ A number of factors appear to have played a role in the decision to bring the IS operations back in house. As stated in the press release, outsourcing appeared to stagnate IT at JP Morgan Chase under the outsourcing arrangement. Another factor that it did not mention in its press releases is that it had undergone a major change with its July 2004 merger with Bank One, which had gained a reputation for consolidating datacenters and eliminating thousands of computer applications. And the man who had played a big role in the consolidation was Bank One's CIO Austin Adams. Adams, in his new role at JP Morgan Chase, managed the switch from IBM to self-sufficiency by taking advantage of the cost-cutting know-how he had gained at Bank One. The underperforming JP Morgan Chase learned much from the efficient Bank One.²⁶

It isn't only large companies that are backsourcing. Many outsourcing clients report having negative experiences with outsourcing and a number of these have backsourced or are considering backsourcing. Given the size and number of the current outsourcing contracts and the difficulties of delivering high-quality information services and products, backsourcing is likely to remain an important option to be considered by many client companies.

Ironically, the reasons given for backsourcing often mirror the reasons for outsourcing in the first place. That is, companies often claim that they backsource to reduce costs and become more efficient. Based on reports in the popular press, the most common reasons given for backsourcing are a change in the way the IS is perceived by the organization, the need to regain control over critical activities that had been outsourced, a change in the executive team (where the new executives favored backsourcing), higher than expected costs, and poor service. The study found that backsourcing wasn't always due to problems. Sometime companies saw opportunities, such as mergers, acquisition, or new roles for IS, that required backsourcing to be realized.²⁷

Outsourcing decisions can be difficult and expensive to reverse because outsourcing requires the enterprise to acquire the necessary infrastructure and staff. Unless experienced IT staff can contribute elsewhere in the firm, outsourcing major IT functions means staff will be lost either to the outsourcing provider or to other companies. When IT staff gets news that their company is considering outsourcing, they often seek work elsewhere. Even when staff are hired by the outsourcing provider to handle the account, they may be transferred to other accounts, taking with them critical knowledge. Though backsourcing represents the final decision in one sourcing decision cycle, it is invariably followed by another cycle of decisions as the company seeks to respond to its dynamic environment.

OUTSOURCING AND STRATEGIC NETWORKS

Typically outsourcing relationships are couched in terms of an outsourcing provider and a client—just as we have done in this chapter. A different approach to viewing outsourcing arrangements: the strategic network.²⁸ A strategic network is a long-term, purposeful “arrangement by which companies set up a web of close relationships that form a veritable system geared to providing product or services in a coordinated way.”²⁹ The client firm becomes a hub and its suppliers, including its outsourcing providers, are part of its network. The advantage of the strategic network is that it lowers the costs of working with others in its network. In doing so, the company can become more efficient than its competitors, as well as flexible enough to respond to its rapidly changing environment. Perhaps the strategic network is the best way to think about outsourcing arrangements in today's world. An example of a strategic network is the keiretsu. Japanese keiretsu is similar to a strategic network in that it has a hub company, a policy that encourages specialization within the network, and investments (financial and otherwise) in long-term relationships.³⁰ The Japanese companies manage their outsourcing activities based on the types of inputs from different types of suppliers.³¹ The strategic suppliers (kankei kaisa) fall into the keiretsu category, whereas independent suppliers (dokuritsu kaisha) do not. Japanese companies work very closely with companies in the keiretsu. Another type of strategic network that increasingly affects outsourcing arrangements is a network with a parent organization or multinational and a number of their subsidiaries. Often one subsidiary performs outsourcing services for another subsidiary in the network. Given the increasingly complex structure of today's multinationals, the role of strategic networks in outsourcing arrangements is likely to grow.

SUMMARY

• Firms typically face a range of sourcing decisions. The Sourcing Decision Cycle Framework highlights decisions about where the work will be performed. Decisions include insourcing versus outsourcing, onshoring versus cloud computing versus offshoring, and selecting among offshoring options (nearshoring versus farshoring). Captive centers are an offshoring option with a subsidiary of the client firm. The relationships are shown below in Figure 9.5. The cycle involves an assessment of the adequacy of the IS service/product delivery. The assessment can trigger a new cycle.
• Cost savings or filling the gaps in the organization's IT skills are powerful reasons for outsourcing. Other reasons include the ability for the company to adopt a more strategic focus, manage IS staff better, better handle peaks, or consolidate data centers. The numerous risks involved in outsourcing arrangements must be carefully assessed by IS and general managers alike.
• Full or selective outsourcing offers organizations an alternative to keeping top-performing IS services in-house. Firms can meet their outsourcing needs by using a single-vendor or multiple-vendor models.
• Cloud computing allows client firms to buy the exact amount of storage, computing power, security or other IT functions that they need. . . when they need it. It includes Infrastructure as a Service (IaaS), Platforms as a Service (PaaS), and Software as a Service (SaaS).
• Offshoring may be performed in a country that is proximate along one or a number of dimensions (nearshoring) or that is distant (farshoring). Offshoring must be managed carefully and take into consideration functional differences.

FIGURE 9.5 Sourcing options.

KEY TERMS

backsourcing (p. 281)

captive center (p. 278)

cloud computing (p. 272)

farshoring (p. 277)

full outsourcing (p. 270)

insourcing (p. 264)

nearshoring (p. 278)

offshoring (p. 275)

onshoring (p. 274)

outsourcing (p. 264)

selective outsourcing (p. 270)

service level agreement (SLA) (p. 269)

strategic network (p. 283)

DISCUSSION QUESTIONS

1. The make-versus-buy decision is important every time a new application is requested of the IS group. What, in your opinion, are the key reasons an IS organization should make its own systems? What are the key reasons it should buy an application?
2. Is offshoring a problem to your country? To the global economy? Please explain.
3. When does cloud computing make sense for a large corporation that already has an IS organization? Give an example of when cloud computing might make sense for a start-up company?
4. Does a captive center resolve the concerns managers have about outsourcing to a third party vendor? Why or why not?

CASE STUDY 9-1
CROWDSOURCING AT AOL

Where would you go if you needed to find hundreds of people each willing to take on a tiny portion of a large task for minimal pay? Projects like these include filling out surveys, verifying or entering data, writing articles, and transcribing audio files. They are increasingly common in the digital age, so you might turn to an online marketplace such as Crowdsourcing.com, CrowdFlower, or Amazon's Mechanical Turk where people around the globe go to find work.

Daniel Maloney, an AOL executive, recently turned to crowdsourcing for help inventorying AOL's vast video library. (Note: This definition of crowdsourcing differs from the one used in Chapter 5 to describe crowdsourcing as a way to spur innovation). He broke the large job into micro-tasks and described the tasks that he needed to be done on Mechanical Turk. In particular, each worker was asked to find Web pages containing a video and identify the video's source and location on those pages. The over half a million workers that were registered at Mechanical Turk could read about the tasks and decide if they wanted to perform them.

Using the crowdsourcing service, the AOL project took less than a week to get up and running, and only a couple of months to reach completion. The total cost was about as much as it would have been to hire two temp workers for the same period.

Mr. Maloney was pleased with the cost savings and added: “We had a very high number of pages we needed to process. Being able to tap into a scaled work force was massively helpful.”³² However, he really didn't know very much about the workers who did the work for AOL and he likely had to make sure that their work was done correctly.

Critics of crowdsourcing feel it can lead to “digital sweatshops,” where workers, many of whom may be underage, put in long hours to generate very little pay and no benefits. Some also feel crowdsourcing will eliminate full-time jobs. The crowdsourcing marketplace services counter that they are trying to register stay-at-home parents or college students with spare time.

Discussion Questions

1. Is crowdsourcing as used by AOL a form of outsourcing? Why or why not?
2. What steps do you think Maloney might have taken to ensure that the crowdsourcing would be a success for the inventory project?
3. What factors should be considered when deciding whether or not to crowdsource a particular part of a business?
4. Describe the advantages and disadvantages of crowdsourcing.

Sources: Amazon Web Services Web site, http://aws.amazon.com/mturk (accessed on April 17, 2012); Haydn Shaughnessy, “How to Cut Consulting Costs by 90% and Keep Your Talent Happy!” Forbes (April 16, 2012), www.forbes.com; Scott Kirsner, “My Life as a Micro-Laborer,” The Boston Globe (April 1, 2012), www.boston.com; and R. E. Silverman, “Big Firms Try Crowdsourcing,” Wall Street Journal (January 17, 2012), http://online.wsj.com/article/SB10001424052970204409004577157493201863200.html?mod=djem_jiewr_IT_domainid (accessed on November 2, 2011).

CASE STUDY 9-2
ALTIA BUSINESS PARK

The road to Altia Business Park in San Pedro Sula, Honduras is quite memorable. On one side of the road are gated communities with small, but neatly maintained stucco houses. On the other side of the road is a small river with clear running water. One bank of the river is covered with tightly-cramped shanties. Further down the road towers a thirteen story monolith in black glass. This is the home of Altia Business Park, a technological park developed by Grupo Karims, a multinational corporation with core businesses in textiles and real estate and operations in Asia, North America, Central America and the Caribbean. The building is anti-seismic and Leed Certified, which means that it follows green building practices. It is energy self-sufficient and connected to North and South America through three fiber optic submarine cables. The building is the first of two that will comprise the Business Park.

On a recent visit, the Corporate Marketing Director, Barbara Rivera, guided an American student group through the marbled halls of the building. She introduced Marcus, who was a manager in the call center in the building. Marcus explained that call center business, especially to North America, was picking up. He was born and raised in the U.S. and graduated from the University of Maryland. Since he couldn't find work in the U.S. upon graduation, he moved to Honduras where he has family. Barbara also introduced Lena, a 20-something professional, who spoke to the visiting group in perfect English, complete with current idioms. Lena had recently graduated from a university in Honduras with a Masters degree in graphical design. She said this degree was very helpful in managing the room full of graphic designers working for the company that maintains the Web site for Sandal's Resorts. Barbara told the visitors that the average salary of the workers in the companies in the Business Park was $4,800 a year³³ and people were eager to get the jobs because of the excellent pay in a country where 65% of the population lives below the poverty line. The country has a literacy rate of 84.3% and 47% of the employable workforce between the ages of 20 and 34, so the competition for good jobs can be fierce. Honduras actually has more English speakers as a proportion of population than the average Central American economy.

Discussion Questions

1. Discuss offshoring from the perspective of potential workers in your country. Discuss offshoring from the perspective of potential workers in Honduras.
2. Barbara Rivera is marketing Altia Business Park as a nearshoring site to companies in North America. What characteristics make it a desirable nearshoring site to companies in North America?
3. Is this a good idea to market Altia Business Park as a nearshoring site to people in North America? Why or why not?