For decades, companies have struggled with the seemingly competing forces of excellent customer service and best-in-class operational cost. The prevailing view dictated that customer satisfaction required a high price and aftermarket service was a necessary evil.

Only recently have companies realized that they can run their service business as a profit center, making both their customers and their shareholders happy. The advantages of achieving excellence in service supply chain management—which includes those repair, planning, sourcing, and logistics activities associated with servicing the equipment-installed base—can be compelling. They include increased customer satisfaction and retention, product sales, and service contract revenue as well as higher margins.

A recent PRTM service supply chain benchmarking study (2003 Service Supply Chain Scorecard) revealed some of the performance benefits achieved by best-in-class companies. These leaders enjoyed many advantages over the median performers, including 75 percent lower inventory per dollar of installed base and 15 percent higher fill rates and on-time spare-parts deliveries.

Given the scope of the potential benefits, why haven't more companies capitalized on the opportunity to dramatically improve their service supply chain? There are six main reasons:

Service supply chain strategies and practices across and within industries are not standardized. As an example, consider the differences within the wireless industry. One major wireless carrier's returns strategy seeks to maximize the velocity of returned inventory and get the repaired product back on the shelf quickly. The company maintains swap stock (used product) at local stores and will provide a customer with a substitute before trying to fix the problematic phone. By comparison, another major carrier attempts to minimize returns and does not allow customers to obtain a substitute phone from a store. The company asks the customer to work through the problem over the phone or on the Web with a live customer-service agent and uses an advanced exchange model for support. (The customer sends the phone directly to the carrier for repair and receives a temporary replacement phone. Once the phone is repaired and returned, the customer is expected to send back the "loaner" phone; if not, he or she is charged for it.) Although these strategies may be viewed as only subtly different, they result in significantly different operating processes, philosophies, and systems. The net outcome is a lack of common language or framework for comparing performance.

Operations management traditionally has concentrated on optimizing the performance of the company's forward supply chain (planning, sourcing, making, and delivering new products) and not of its service supply chain (also known as the reverse supply chain). This focus is understandable given the level of goods and cost in the forward supply chain. But because operations managers have paid scant attention to the service supply chain, they have only limited understanding of this process and have accepted suboptimal performance To achieve breakthrough improvements, however, manufacturers need to transition from optimized factories to optimized supply chains. Accomplishing this requires a comprehensive understanding of the service supply chain among all operational managers.

The inventory accounting practices for service parts vary significantly from company to company. Some expense their parts when they are moved into service inventory; others amortize parts over their useful life. Further, these practices ripple through how companies account for repaired parts. The lack of standardization of cost and inventory investment levels across companies, coupled with the absence of a common service supply chain framework, makes it difficult for CFOs to understand the real cost of supporting a product and how their costs compare with their competitors'.

Optimizing service requires a trade-off between service level and cost (inventory and operating expenses). Optimizing along only one of these dimensions, for example service, will not lead to optimal overall performance. As with most other core processes, a balanced scorecard approach is required to drive optimization.

Optimizing the return on investment in service parts is different from optimizing investment in new product inventories. The demand characteristics are less predictable and more conjectural, the sources of supply are greater (new, repaired, and refurbished), the support life extends well beyond component and subassembly manufacturing lifecycles, a repair can take multiple paths, the level of part substitutability is greater, and multiple tiers of inventory require optimization. All of these factors contribute to the difficulty of understanding what's really involved in service supply chain effectiveness.

The rapid sales growth of the 1990s hid many evils, prominent among which was inventory excesses. The '90s mantra was service level at all cost. Most companies scrambled to ensure that growth was supported by consistent levels of service. In the euphoria of rapidly growing revenues, companies often greatly understated the cost of that service as a percent of sales. But with the drop in revenue and no corresponding decline in installed base, service costs as a percent of sales have increased significantly.

Fortunately, there's evidence to suggest that many companies now are addressing these failures of the past and are paying more attention to the service supply chain. More and more CFOs, for example, have become aware that they don't fully understand their service cost drivers or the total costs of providing different service levels. Vice presidents of sales and marketing are discovering that they can make money on service and that parts and repair management is a critical component of developing customer-differentiated and profitable offerings. Chief operating officers and vice presidents of operations are realizing that they need to demystify the service supply chain and enhance its performance. Academics, tool providers, and consultants have developed a deeper understanding of service supply chain management and are creating solutions to many service problems.

The net result of all of this is a rapid transformation in how leading organizations view and execute their service supply chain. This transformation is not unlike the earlier transformation in operations from functional to supply chain optimization. This time, however, the rate of transformation will be much faster because of the maturity of overall operations management thinking and the capabilities of service supply chain planning and execution systems. Moreover, the people driving this change will have participated in the earlier operations transformation. Those companies that move quickly on service supply chain excellence can expect to gain higher profits and shareholder value. The slow movers, by contrast, will see only higher costs and less revenue.

We have identified a five-step process to help companies swiftly get on that road to service supply chain excellence:

1. Understand your service supply chain.
2. Determine how to measure your service supply chain.
3. Develop the business case for improvement.
4. Develop and execute service supply chain strategy.
5. Sustain performance improvements.

STEP 1: Understand Your Service Supply Chain

The first step toward excellence is to understand your company's core service supply chain processes and strategies and determine how the functional areas align with these strategies and processes. We define the service supply chain as all processes and activities involved in the planning, movement, and repair of materials to enable after-sales support of the company's products. (Exhibit 1 shows a comprehensive service supply chain framework.)

Most companies don't yet view the service supply chain as a cross-functional core process, as depicted in the framework. They tend to have the service supply chain subprocess elements report into different organizational functions. Unfortunately, the resulting lack of overall process understanding typically suboptimizes the performance of the process. To illustrate, one network equipment maker had no single person responsible for the service supply chain. The vice president of customer advocacy set the service policies but was not responsible for service performance. The return material authorization (RMA) manager was responsible for planning and replenishing spare parts but did so on an ad hoc basis. A buyer reported to the RMA manager but bought spares only when no part could be found in the warehouse. All customer RMA requests were approved, but no one followed up with customers for "exchange-owed" parts—that is, customers were required to return the defective part when an exchange part was shipped. The result was suboptimized performance and general finger-pointing throughout the service network.

Another, more enlightened, company put nearly all of the service supply chain performance levers—spare parts planning and purchasing, spares logistics and distribution, and returns management and dispositioning—under the control of one vice president. The only element not under his control was repair, which he closely monitored. Because of his position within the organization, this executive was responsible for the most important service performance measures, such as inventory, cost, and service levels. This company is recognized in its industry as the thought leader in service supply chain. In fact, many of its competitors and companies in other industries are trying to emulate this company's model.

Although alignment of the organizational reporting structure can help solve some of the problems associated with process understanding, it is neither required nor recommended for all situations. The higher level principle that applies is that a company must have a common understanding of the core service supply chain processes, their overall objectives, and performance expectations.

STEP 2: Determine How to Measure Your Service Supply Chain

The adage "you can't manage what you can't measure" holds true here. The challenge is determining what you should measure and how you should measure it. There is no single metric that defines the service supply chain's performance. Simply measuring delivery performance against service-level agreements (SLAs), for example, does not appropriately measure overall asset performance. The challenge is further complicated by the multitude of operational models typically in play. For instance, measuring a guaranteed uptime contract is different from measuring a response-time-based contract. Similarly, the different models make comparisons of service supply chain performance from one enterprise to another extremely difficult. In most cases, the result is a paucity of available benchmarking data that measure performance in a similar way.

A decade ago, we saw this behavior in the supply chain arena. Back then, companies were suboptimizing supply chain performance and struggling with how to measure it. To establish a common framework for analyzing and measuring the supply chain, PRTM co-developed the Supply Chain Operations Reference-model (SCOR) — which was later modified by the Supply-Chain Council. Responding to the comparable lack of comprehensive benchmarking data for the service supply chain, we developed a set of metrics similar in structure to the SCOR metrics (see Exhibit 2). These metrics represent a balanced scorecard of operational and financial measures, which is needed to measure performance appropriately.

At level 1, the highest level, these metrics measure customer service (fill rate, on-time delivery to SLA, and perfect-order fulfillment) and operational efficiency (response time, cost, and inventory). The level 2 metrics are the building blocks of the level 1 measures. It's essential to choose the right metrics and to measure them in the right way. If your choice of metrics is not aligned with your strategy and does not measure the service supply chain's overall performance, then you suboptimize performance. The following example illustrates the point.

For years, a capital equipment maker measured its service-parts inventory. This was one of the level 1 metrics used to measure the company's service business. A close look at the metrics used in level 2, however, revealed that parts in-transit between the company's international depots were not included in the service-parts inventory numbers. Including the in-transit inventory raised the company's overall inventory levels by 8 percent. Although this news was not good, it did force the company to focus on driving down its total inventory. It now includes the in-transit inventory as a source of supply, thereby decreasing the need for purchasing new service parts.

STEP 3: Develop the Business Case for Improvement

In the services business, shareholder value comes either from driving higher top-line business growth or from lowering the cost to support the programs in the field. Enhanced service supply chain performance can do both. To illustrate, we have seen leading companies:

• Drive service-level improvements that stimulate increased customer loyalty and ultimately new product sales. One computer board manufacturer was proud of its low service supply chain cost and inventory. Even its service delivery seemed acceptable. What the company failed to understand, however, was that its service levels were no longer competitive. Although it shipped replacement parts within a 48-hour time-window, the industry standard had become 8 to 24 hours. Identifying this gap and the related financial impact on sales helped the company determine how to drive top-line growth.
• Create response-time-differentiated services that can be sold separately to increase revenue. The electronic-equipment industry has been driving shorter response times and charging for differentiated services. Some computer companies are even trying to take differentiated service to a commodity level.
• Develop delivery performance-based contracts that reward or punish suppliers for meeting or not meeting parts-availability thresholds. The U.S. Department of Defense, for example, is working with numerous suppliers to prompt adoption of performance-based logistics contracts.
• Devise uptime performance-based contracts that reward or punish suppliers for meeting or not meeting an equipment uptime threshold. The high-availability server manufacturers have offered this differentiated service for years.

Of course, taking service supply chain cost reductions to the bottom line also can be quite lucrative. One of the biggest sources of service supply chain waste (at least in parts-intensive industry segments) is materials. Companies regularly flood their service supply chain with spare parts. But rather than balancing service levels and cost, they tend to err on the side of overstocking. The optimal trade-off between cost, or inventory investment, and service level for a given service supply chain infrastructure is represented as the efficient frontier (see Exhibit 3), a risk-return concept originating in the financial space.

Determining where you are relative to the efficient frontier is a critical step in identifying the savings that can be brought to the bottom line. We have devised a relatively simple grid that identifies the different levels of cost/service performers (see Exhibit 4). The levels are defined as follows:

1. Money wasters have low customer-service levels and are investing large sums of money in inventory. The problem is that this investment is not contributing to service levels.
2. Over-Investors also are investing heavily in inventory but are achieving higher service levels than the money wasters. They are closer to the steep portion of the efficient frontier, where every percentage point in service level requires a disproportionate amount of inventory investment.
3. Companies in the no priority category have decided not to invest in achieving high customer-service levels. This is a dangerous strategy because customer service is increasingly becoming a competitive differentiator.
4. Best-in-class organizations achieve high service levels at low cost.

By benchmarking cost, inventory, and service performance, a company can determine its relative position in the service supply chain grid. The company's position on the grid determines its next step on the path to best-in-class service supply chain performance.

STEP 4: Develop and Execute Service Supply Chain Strategy

Even though different companies operate in different segments of the service supply chain grid, a generic service supply chain improvement strategy can be defined. Further, this strategy can be applied—at least in part—to every company's situation. The service supply chain improvement involves three executable stages:

• Stage 1: Improve efficiencies to get on the efficient frontier (Exhibit 5).
• Stage 2: Determine if further investments are required and move up the efficient frontier (Exhibit 6).
• Stage 3: Implement structural changes to actually move the efficient frontier (Exhibit 7).

Companies in the money wasters or over-investors categories need to get on the efficient frontier. They should stop investing in additional inventory and determine what current inventory is contributing to service levels and what is not. The several categories of noncontributing inventory are:

• Excess or obsolete inventory. The definition of excess inventory will depend on the remaining service life of the product the part is supporting.
• Parts in the wrong location. In practice, this inventory may be used to satisfy service demand but will either arrive late (and therefore not contribute to service as defined in the service-level agreement) or arrive on time at the expense of high express-freight charges.
• Field engineer's inventory, also called trunk stock. This inventory is usually not visible to the planners and therefore not available to satisfy service demand.

To move onto the efficient frontier, companies must reduce inventory levels until they eliminate most—or all—of the noncontributing inventory. Excess and obsolete inventory must be scrapped or liquidated, parts in the wrong locations must be redistributed, and processes must be established to move trunk stock back to the service depots.

The network equipment company discussed above that was using mostly new product to fulfill orders was clearly in the money-wasters category. We worked with this company to take immediate action on three fronts: (1) reduce noncontributing inventory by outsourcing product repairs to a contract manufacturer, (2) develop a planning process that sets target inventory levels for each of the regional depots, and (3) fill the depots with repaired products first.

Now that the company has eliminated its noncontributing inventory, it is operating in the flat portion of the efficient frontier. Basically, a relatively modest investment in inventory yields a big improvement in service levels.

At this point, the challenge for the money wasters is to convince senior management that additional inventory is required to raise service levels. In companies with a history of throwing good money after bad, this can be a formidable task. It requires quantifying the new service opportunities available through increased inventory levels, as described in Step 3: Developing the Business Case for Improvement.

How do you determine what inventory to invest in, especially if you have a history of investing in the wrong parts? A critical look at your service-parts planning system should be the first step: What assumptions are made? How accurate are the planning data? Are the planning algorithms in line with expectations?

The next step is to assess how effectively the planning recommendations have been executed. Are parts ordered on time, delivered on time, and delivered to the right location? Is all inventory accounted for?

One international service company demonstrated how careful investments in inventory can improve service. The company had been experiencing low service levels from its local depots. Analysis of the inventory at these locations revealed a high level of excess and obsolete inventory and a shortage of parts contributing to service levels. The company's strategy was to reduce its excess and obsolete inventory and modestly invest in parts to move up the efficient frontier. To that end, the company scrapped 13 percent of its inventory classified as excess and obsolete, while simultaneously investing in an additional 11 percent of high-contributing parts. The result was a 30-percent improvement in local fill rate. The company clearly took advantage of the move up the flat portion of the efficient frontier. It is now considering its options to improve service levels further by shifting the efficient frontier, itself.

Best-in-class companies achieve the highest service levels at the lowest cost and inventory investment. Once a company has reached the steep portion of the efficient frontier, any further improvements in both service levels and cost/inventory can only be achieved by shifting the efficient frontier downward. And because the efficient frontier determines the best achievable combination of cost and service for a given service supply chain infrastructure, the only way to shift the frontier is to modify the infrastructure.

The service supply chain infrastructure means more than just the physical footprint of the service and repair depots. Rather it is comprised of the combination of network design, processes, organization, policies, and information systems. A change in any one of the elements affects the others.

The objective of any structural change in the service supply chain should be to move the efficient frontier downward by reducing cycle times, optimizing inventory allocations, and minimizing costs. This is the last stage in advancing to best-in-class service supply chain performance. It is also the most difficult stage to plan and implement. The following describes how organizations can make structural changes in each of the service supply chain elements of plan, source, deliver, return, disposition, and repair.

Plan. With regard to service parts, companies go through four planning stages. In the start-up stage, the focus is on growth, not on optimizing the planning and allocation of service parts. Parts are allocated to service centers using no more than gut feeling. Soon after this first stage, simple rules are put in place on where parts are allocated. Typically these are spreadsheet-type calculations of safety stock based on historic demand. In the third stage of maturity, companies start to use their enterprise resource planning/distribution requirements planning (ERP/DRP) systems to find the optimal forecasting method, but the forecast is still based on historic demand and individual depot allocation. The fourth and final stage uses spare-part optimizers. These software tools, which calculate target stock levels for each depot, base their recommendations on installed-base-related demand rather than part-number demand. The tools have sophisticated algorithms that optimize inventory across multiple echelons of depots. Low-usage, high-dollar parts will be allocated to the highest echelon (say, a central warehouse location), and high-usage, low-dollar parts will be allocated to the lowest echelon facilities. The tools also consider pooling of parts among depots to lower inventory levels. Most solutions are flexible enough to handle multiple scenarios—for example, maximum service levels for a given asset level or minimum inventory for a given minimum service level.

Source. The sourcing of service parts entails unique challenges not found in the sourcing of production parts. The key challenges relate to:

• Multiple sources of supply (new and refurbished parts). The price of refurbished parts can be half that of new parts. So maximizing the use of refurbished parts obviously will yield great returns. For this to happen, however, refurbished parts must be visible through the entire return and repair loop and tracked as a source of supply by the spares planning system. Planners should buy new parts only if no refurbished parts are available.
• End-of-life purchases. The service organization must be able to support the company's products long after the original equipment manufacturer and its vendors have stopped manufacturing them. Consequently, the service organization must make a best estimate on what quantities of parts will be needed to fulfill its service obligations. This is the classic newsboy problem: Underestimating demand leads to missed service opportunities, and overestimating demand leads to obsolete inventory. In our experience, few companies take a scientific approach to end-of-life purchases—that is, an approach that takes into account diminishing installed base, increasing product-failure rates, and the probability of out-of-warranty returns.
• Retrieving supplier warranties for returned products. Following the classic bathtub curve of product failures, most product returns will happen in the first year of use and then at the end of the product's lifecycle. Given that most material vendors provide a one-year warranty on their components, you can see the huge opportunity for retrieving vendor warranties. Yet few companies capitalize on this opportunity due to the complexities in tracking component serial numbers and vendor warranty terms. The return on investment for a simple database solution to do this can be compelling. If the company chooses not to track individual-part serial numbers and related vendor warranties, it should negotiate to eliminate warranties in return for lower component prices.

Deliver. Most companies have already outsourced their spare-parts distribution to a third-party logistics provider (3PL) and have benefited from the 3PL's lower cost structure, experience, and flexibility. The number and location of the 3PL or company-owned warehouses used to handle these parts are dictated by several factors: location of customers and installed base, required customer-service levels, and total logistics cost.

Any changes in the location of customers and installed base or in the required customer-service levels wil affect the design of this warehouse network infrastructure. Many companies still operate with the same network they used when they started their service operations. Several network analysis tools now are becoming available to validate a company's existing service network and to develop an optimal network design—that is, the one that minimizes the total-cost and inventory requirements for a given service level.

Return. The two largest cost-reduction opportunities center on reducing the overall returns volume and improving returns cycle time. Companies can reduce returns by:

• Educating customers during the buying process. This should result in the customer buying the right product, which reduces the likelihood that it will be returned for credit.
• Screening returned products at the point of return. This activity will reduce the levels of "no trouble found" returns. Information gathered at the return point can be incorporated into scripts and trouble-shooting guides to help call centers, store personnel, and field engineers reduce the number of good products returned.
• Designing better products by incorporating inputs from product failures into new product designs.

While the company is reducing its overall flow of returns, it also should work to minimize the time it takes to transport returned products. The reason: The sooner products reach repair centers and are transformed into working products, the less inventory is tied up in the service supply chain and the closer the company gets to best-in-class performance.

Disposition. The product disposition process is an important, though often overlooked, element in the service supply chain. Decisions regarding the disposition cycle time must be based on clear, measurable, and executable criteria. These include such considerations as product revision level, date of original manufacture, physical condition, test performance, product lifecycle, competitors' offerings, manufacturers' policies, and repair vs. new buy cost trade-offs. One key objective here is to minimize the cycle time.

In addition, products should be designed with an eye toward facilitating the disposition of returns. For example, a smartphone manufacturer incorporated a water-sensitive sticker inside the product. The sticker changes color when exposed to water. By simply removing the cover, the repair operators can quickly determine if the phone has water damage and is, therefore, beyond repair economically.

Repair. Repair operations are similar to manufacturing operations in many respects. To reduce repair cost and cycle time, the repair process can be organized in cells using these best-practice lean-manufacturing techniques:

• Create self-contained work cells responsible for repair and reconfiguration of units.
• Perform component and module repairs offline.
• Use just-in-time inventory mechanisms to source and stage parts.
• Locate cells close to receiving and disposition operations.
• Maintain physical visibility of all materials.
• Develop a repair engineering function whose primary focus is to reduce repair cycle time.

STEP 5: Sustain Performance Improvement

Without sustainment, the path to service supply chain excellence will reverse itself. Process cycle time starts to creep up, noncontributing inventory starts to increase, and service levels begin to slip. To prevent this reversal of fortune, every service supply chain organization should be involved in three levels of activities, as shown in Exhibit 8 and explained below:

• Breakthrough activities, which will result in step-function improvements in service level, inventory, and cost.
• Continuous improvement activities, which will result in incremental performance improvements.
• Maintenance activities, which will prevent the deterioration of performance.

All three activities are important; leaving one out will result in piecemeal improvement or no improvement at all. The service supply chain organization needs to lay the groundwork for a continuous-improvement culture in which ideas for improvement are stimulated, rewarded, and quickly implemented. The progress achieved must be maintained through rigorous performance measurement and rapid corrective actions. Only then can the company achieve—and sustain—service supply chain excellence.

Many companies realize the importance of great customer service. Some are starting to understand the costs associated with providing this level of service. And a few have embarked on a path to provide excellent customer service at best-in-class operational costs.

These leaders have a clear understanding of their service supply chain performance, and they continue to make strategic changes to their service infrastructure and processes to improve that performance. Importantly, they also understand that sustaining these performance levels is just as important as achieving them.

By following the five steps to service supply chain excellence outlined in this article, your company can achieve what previously seemed impossible: It can become a champion both in customer service and in shareholder performance.