A toolkit for balanced outsourcing
By Tarikere T. Niranjan and Shashank S. Rao
Industry opinion is divided about the benefits of outsourcing, but a holistic, long-term view of this tool’s performance would necessitate looking beyond financial measures to market and consumer-centric measures. A research study that used the online retail industry as an example provides guidelines on how organizations can evaluate themselves and develop insights into whether they are outsourcing too much, too little or just right.
Outsourcing is not a new phenomenon. In the early 19th century, Britain imported cotton from the United States, spun the cotton, wove the fabric in England and then exported the finished textiles abroad, mostly to India, then its colony. Britain kept almost 75 percent of the monetary gains from these exports, thus creating substantial wealth for the crown. Outsourcing as we understand it today is best exemplified by Eastman Kodak outsourcing its entire IT services to IBM, Businessland and Digital Equipment Corp. in the late 1980s.
Outsourcing has grown phenomenally over the years, especially with the advent of the Internet and other technological developments that allow quality communication between geographically separated business entities to an extent that was unthinkable two decades ago. Today the size of IT outsourcing business from the U.S. alone is approximately $77 billion. Industry surveys indicate that worldwide this number would be almost $500 billion. According to a study by Gartner research, nearly 65 percent of all companies have a formalized approach for selecting outsourced vendors for almost every business process. And yet, despite this tremendous adoption of outsourcing and a long history of use, outsourcing remains a contentious matter, with opinions being divided on whether firms should outsource in the first place.
Doing vs. not doing
There are several sound reasons for outsourcing. A vendor, which in this article refers to the company doing the outsourced work, typically services multiple clients. This allows the vendor to enjoy lower asset specificity that a single outsourcer may find hard to match. Therefore, a company can lower its asset base and reduce costs by outsourcing. Further, vendors often enjoy greater economies of scale and lower operating costs. When the difference between the cost structure of the client, which in this article refers to the company that is outsourcing work to a third party, and the vendor exceeds the transaction costs, the vendor can pass on some of the residual value to the client. Other things being equal, these lead to improved returns on investment and assets for the client, which in turn would lead to better market performance. This makes outsourcing an attractive proposition.
Of course, cost savings may not always happen. So some argue that cost saving is hardly the main reason for a corporation to outsource. Instead, proponents of core competency argue that organizations ought to be outsourcing for strategic reasons, not cost savings. The advantage to be gained by directing the company’s efforts exclusively on activities closely related to its own core competencies, while outsourcing all peripheral activities, is of far greater reach and impact than mere cost savings. Studies also have shown that a company’s overall quality can improve due to outsourcing because the vendor specializes in its work. That work may be the vendor’s core competency, which allows the contractor far greater control and higher quality on its narrow range of activities than a client could ever hope to achieve.
Apart from these reasons, perhaps the single most important benefit of outsourcing is the flexibility and risk management it offers. When a market changes drastically in any product line, a client that has outsourced its activities to a high degree can pass on a proportionately large share of the shock to the vendors. The vendor, by virtue of having a broader client base, is much better equipped to absorb these shocks than one company that has to bear the whole brunt alone. As a corollary to this type of risk management, outsourcing allows organizations to be more flexible to changes in both product demand and variety.
Clearly, there are many reasons that drive companies to outsource. However, outsourcing has its share of opponents as well. Several pitfalls of outsourcing have been debated in companies and industry circles. A major threat associated with outsourcing is the opportunistic behavior of the vendor. Once the outsourcing relationship is established and begins to run smoothly, the client would tend to divest from the now redundant physical and human assets associated with the outsourced function, thereby losing organizational learning and knowledge capital. These changes, although associated with some cost savings, could make the client increasingly dependent on the vendor and vulnerable to arm-twisting.
Moreover, there is the risk that the vendor, after gaining a foothold in the value chain, begins to compete in the very market that the client operates. Further, some of the very reasons that are put forth in favor of outsourcing can be used against it. For example, cost savings may be exaggerated; if one considers the whole spectrum of transaction costs associated with outsourcing, such as communication, control and additional complexities and risks associated with offshore outsourcing, the claimed cost savings may evaporate. While, in theory, quality can improve through outsourcing, reality might include many factors that make outsourcing to a vendor a more risky quality proposition. Such factors include lack of visibility into and control over the vendor processes.
For these reasons, outsourcing may not always yield the rosy picture painted by its proponents, and this is reflected in industry surveys that show mixed success. For example, three-quarters of company managers responding to an outsourcing survey conducted by the American Management Association confided that outsourcing outcomes have fallen far short of expectations, and more than half reported that they have brought at least one outsourced activity back in house. Outsourcing may or may not benefit the client, depending on the situation. So the debate over the sensitive issue of outsourcing continues as organizations attempt to answer the following questions: Does outsourcing create value? Should companies outsource?
The right question
While it may be one of the most engaging controversies that industry is grappling with at the moment, the above questions are framed incorrectly. The simple insight that often is missed in the frenzy surrounding these debates is that outsourcing most likely is useful, but the critical question is, “How much to outsource?” and not whether to outsource. Accordingly, our research goal is to provide a toolkit that allows companies to evaluate the extent of their outsourcing, helping them decide whether they need to outsource more or less than their current levels.
Numerous studies, including many in reputable academic journals, have attempted to find out the impact of outsourcing on corporate bottom lines. The focal measures of outsourcing success invariably are financial metrics such as return on assets, return on equity, return on sales or some variant thereof. This is unsurprising because, after all, this is what most managers primarily would focus on. This reflects one of the most fundamental management problems, which is that managers seldom have perfectly designed incentives; they therefore tend to focus on short-term and immediately observable successes to keep their jobs instead of taking a long-term view. Moreover, the findings of these studies have been mixed, and perhaps add to the existing confusion.
But financial measures do not capture the full impact of outsourcing, making them an unreliable indicator of outsourcing’s full benefits. This partly explains why earlier studies haven’t answered all the questions surrounding outsourcing. For starters, this approach is too indirect. Several variables intervene between outsourcing, which is the cause, and company financial performance variables, which is the proposed effect. This makes the link between them tenuous and, therefore, hard to measure reliably. And intangible factors such as learning have a delayed effect and do not reflect immediately in financial figures.
The right measurements
For these and other reasons, we argue that a more direct and forward-looking implication of outsourcing would be on end-customer based measures such as sales, sales growth, lead conversion, customer satisfaction and customer loyalty. Arguably, these are the key factors that lead to long-term financial success and profitability for corporations. They look forward instead of only focusing on past performance, and it is useful to know how they are affected by outsourcing. Unfortunately, no one knows. This brings us to a secondary goal: Lay out the various dimensions of outcomes that, when taken together, give a more reliable indicator of the success of outsourcing, demonstrating how organizations can adopt them to measure their own position.
To answer these questions, we specifically chose to focus our study on a single industry — online retailing — in order to keep industry-specific variables from confusing the results. We discuss below the results from this study carried out during the past couple of years. The overview of the study is provided in the sidebar on Page 24. Extending the issues raised previously, what would a holistic performance outcome measure look like? The key business processes of an online retailer are marketing, logistics, customer support and other functions. Therefore, how many of these processes are outsourced gives a fair idea of the extent of outsourcing: A client that outsources all or most of these functions is likely to have an inordinately high number of vendors; a client that outsources few of these key areas likely will have few vendors.
Each of these processes can be broken down further into its components for increased precision of measurement. For example, the marketing function consists of affiliate marketing, email marketing, e-commerce platforms and search engine marketing. Similarly, the logistics function consists of content delivery, order fulfillment, order management and payment collection. The customer support function typically includes customer relationship management, website design, website search and web analytics. Anything that is not captured under those three business functions can be classified under a fourth category, call it “others,” as depicted in Figure 1.
Next, a binary coding can be used to quantify the extent of outsourcing by using a mark of zero if the company is not outsourcing a particular function, while using a mark of one if it is. The sum of these scores gives a comprehensive picture of the current state of the organization’s “extent of outsourcing.” Using the binary coding as weights, the maximum score possible for the marketing function is four and the minimum is zero. Likewise, scores for the other three functions range from zero to four.
The fifth dimension
In light of the previous arguments that relying on financial performance measures alone is myopic and can mislead the findings, companies should measure the comprehensive success of outsourcing by using five distinct, but related, dimensions that predict long-term success. They are:
- Sales prospect conversion rate: This is the percentage of visitors to an online store who actually end up buying from that store.
- Sales and sales growth: Self-explanatory.
- Customer satisfaction: The satisfaction level of customers from the online retailer in question. Any of the customer satisfaction tools that the retailer may already be using can be used.
- Repurchase intention: This can be obtained through self-stated likelihood of the customers buying from the retailer or any other means the retailer may have.
- Customer retention: The percent of customers of the retailer who actually did come back and buy again in the same calendar year.
These five dimensions capture the impact, or success, of outsourcing.
This study analyzed data for 260 retailers. This enabled the study to organize the data statistically into naturally forming clusters by ranked comparison of performance, or “high outsourcers,” “moderate outsourcers” and “low outsourcers,” as shown in Figure 2. High outsourcers are those with an average composite score exceeding 12 on the extent of outsourcing. Low outsourcers on average score below three. Those that score around seven rank between these thresholds and may be considered moderate outsourcers.
Please note that the results are shown in ranks and not absolute values of the performance variables. This means that lower ranks indicate better performance, i.e., first rank is better than second, etc. While one plausibly might expect moderate outsourcers to outperform low and high outsourcers, it was surprising that the results held for each of the performance indicators. The study allowed a lag of one year between the outsourcing and performance variables because impact is not instantaneous.
Moderation is the key
Put together, what these results mean is that moderate outsourcers outperform high and low outsourcers on each of the performance metrics, even in the short term. It goes without saying that in the long term each of these performance variables would help increase the corporations’ financial performance. Moreover, rather than simply being additive, it seems plausible that some of these benefits act synergistically and the financial performance difference between moderate outsourcers and others would be even more pronounced than in these results.
The study also revealed that moderate outsourcing across the board is better than heavy outsourcing on one function and no outsourcing on another. For example, consider two companies that scored an eight out of 16 on the outsourcing continuum. Company 1 outsources marketing and logistics completely, for two scores of four. It does not outsource the other two functions at all, for two more scores of zero. Thus, its total outsourcing score is eight out of 16. Company 2 outsources a little of marketing (two out of four), a little of logistics (two out of four), a little of customer support (two out of four) and a little of others (two out of four), matching Company 1’s total outsourcing score.
But our results revealed that the pattern of outsourcing used by Company 2 clearly yielded superior benefits. In fact, Company 1 barely fares better than the low and high outsourcers. From this, the key lesson is that companies should outsource moderately, and the moderation should be distributed across all outsourced functions. Note that those with high and low outsourcing scores are not fictitious companies outsourcing at unreasonable rates — these are real companies in the same industry. Therefore, outsourcing moderately means “moderate” in comparison to others in the industry.
In summary, there is a multitude of opinions, often not based on facts and analysis, about the benefits of outsourcing. The few studies that attempted systematic analysis also failed to add any clarity, as findings on the efficacy of outsourcing have been mixed at best. One source of this problem is industry fixation on financial performance measures. The arguments provided and the evidence from this study should encourage organizations to think beyond financial performance measures and short-term goals, which could explain the conflicting findings of previous studies.
After all, a strong positive impact on market performance and consumer happiness almost certainly will result in financial benefits over the long run. Using the guidelines provided, corporations can evaluate the extent of their outsourcing and get valuable clues on whether they are outsourcing too much or too little. And, if they wish, executives can measure how their organizations fare on market and consumer-centric performance.
Tarikere T. Niranjan is a post-doctoral researcher at the Chair of Logistics Management, Swiss Federal Institute of Technology, Zurich. He holds a mechanical engineering degree from the National Institute of Engineering, India, and a doctoral degree in business management from the Management Development Institute, India. Prior to earning his Ph.D., Niranjan served as a flight lieutenant in the aeronautical engineering branch of the Indian Air Force. His core research areas are behavioral supply chain management and outsourcing. His research has been published in leading outlets such as the Decision Sciences Journal and the International Journal of Production Research.
Shashank S. Rao is an assistant professor of operations and supply chain management at Nova Southeastern University. He is an engineer and an MBA with corporate experience in the banking and financial products industries. He received his Ph.D. in supply chain management from the University of Kentucky. Rao’s research focus is on logistics strategy, customer service and logistics issues in online retailing. He has published and continues to publish articles in academic and professional journals, including the Journal of Business Logistics, the International Journal of Logistics Management, the International Journal of Physical Distribution & Logistics Management and the Journal of Electronic Commerce Research.
About this research
The recommendations in this article were borne out of recent research carried out at the University of Kentucky and the Swiss Federal Institute of Technology. The companies analyzed in this study were drawn from the Internet Retailer Top 500 Guide (IR500) and Bizrate.com, a popular price comparison engine available on the Internet. The IR500 database lists the top 500 online retailers in the U.S., with total sales exceeding $166 billion, accounting for more than 60 percent of the total online retailing market.
Together, these two organizations provide most of the data required to establish the relationship between the extent of outsourcing and its impact on market and consumer-centric performance. For example, the databases have data on what functions and sub-functions have been outsourced, what the customer satisfaction and repurchase intentions are, etc. A few data components, such as sales and sales growth, were obtained directly from the retailers, usually from their websites.
After eliminating weak and dubious data points, the final dataset elicited for analysis was 260 companies out of the 500 retailers. Data thus obtained were analyzed through relevant statistical techniques. In summary, we allowed the data to fall into naturally forming clusters. This resulted in the three clusters of companies that outsource a little, a moderate amount or a lot. The results, as shown in Figure 2, clearly point to the fact that companies with moderate outsourcing scores outperform those with high and low outsourcing scores. The thresholds to identify the three clusters were elicited from these results.