Is your company doing its best to address customer complaints and learn from mistakes?

Keeping and developing relationships with current customers is a key business strategy.1 Yet problems and complaints are bound to occur over the lifetime of customer relationships. Handling these effectively is vital to maintaining customer satisfaction and loyalty, as the example of Saturn illustrates. When faced with the need to repair a problem on 350,000 of its vehicles, the company chose to do whatever was necessary to satisfy customers: it set up an 800 number, went directly to some owners’ homes, and opened its checkbook to dealers. According to Joseph Kennedy, Saturn’s vice president of sales, service, and marketing, the goal was “to fix both the car and the customer. We want Saturn to be a 100-year car company.” Saturn has achieved high levels of customer loyalty and in 1996 was rated number one in sales and dealer satisfaction in its class.

The impact of recovery strategies on a company’s revenue and profitability is dramatic. Hampton Inn hotels, for example, realized $11 million in additional revenue from the implementation of its service guarantee and scored the highest customer retention rate in the industry.2 General Electric has found that, on average, customers buy fifteen major appliances in their lifetimes. Committed to maintaining customer loyalty, the company has invested heavily in the GE Answer Center to manage customer relationships and ensure that problems are addressed efficiently and effectively.

The ability to deal effectively with customer problems is also closely related to employee satisfaction and loyalty, which are critical concerns in industries where customer relationships are more closely associated with the individual service provider than with the organization.3 American Express pays careful attention to the job satisfaction of its investment advisers, since it estimates that more than 30 percent of an adviser’s clients would defect if the adviser left the company. In addition, customer complaints provide valuable insights into root causes of operations failures. Many quality-award winners, including Federal Express, Xerox, and Ritz-Carlton, use failure data when making decisions on process improvements, coupling service recovery with initiatives to increase customer satisfaction in the future.

The majority of customers are dissatisfied with the way companies resolve their complaints.

Despite the benefits offered by effective service-recovery strategies, our research shows that the majority of customers are dissatisfied with the way companies resolve their complaints. This result is consistent with other findings indicating that most customers have more negative feelings about an organization after they go through the service-recovery process.4 Furthermore, the vast majority of companies do not take advantage of the learning opportunities afforded by service failures. Why is this? How can firms develop a process that supports both effective recovery and learning?

To guide managers in designing an effective strategy, we provide a four-stage approach to service recovery see Figure 1), based on findings from a study we conducted to understand how customers evaluate service-recovery efforts. Our approach includes the implications of a successful recovery strategy on customer loyalty, employee satisfaction, and, ultimately, firm profitability. The first two stages focus on identifying and resolving individual customer problems. Stages 3 and 4 examine how recovery data can be classified and integrated with other firm data to identify profitable service-improvement investments. In presenting the approach, we share the best practices of companies that employ service-recovery strategies successfully.

Stage 1: Identifying Service Failures

The greatest barrier to effective service recovery and organizational learning is the fact that only 5 percent to 10 percent of dissatisfied customers choose to complain following a service failure.5 Instead, most silently switch providers or attempt to get even with the firm by making negative comments to others.6 Why are customers reluctant to complain? Our research and other studies have uncovered four key reasons: customers believe that the organization will not be responsive; they do not wish to confront the individual responsible for the failure; they are uncertain about their rights and the firm’s obligations; and they are concerned about the high cost in time and effort of complaining.7 In addition, some customers anticipate negative ramifications. Our work in the health care industry, for example, indicates that some patients are reluctant to complain because they fear that it may result in lower service quality when the need for care arises again. What can firms do to remove these barriers and encourage dissatisfied customers to complain?

Best Practices in Identifying Failures

Examining the recovery practices of leading firms highlights several approaches that have been effective in identifying service failures: setting performance standards; communicating the importance of service recovery; training customers in how to complain; and using technological support offered through customer call centers and the Internet.

Setting Performance Standards. Comparing services with tangible goods, Levitt observed that with services, “You don’t know what you aren’t going to get until you don’t get it.”8 His comment reflects customers’ unclear expectations about many services and helps explain why they do not complain when they are dissatisfied. One way to overcome the problem of unclear expectations is to implement service standards, which are often communicated through service guarantees.

By promising free delivery if a customer does not receive a package by 10:30 a.m., FedEx makes it clear when the service has failed. With the help of scanner technology, the data on package pickup and delivery are transmitted directly to FedEx, removing the need for service reps to report many types of failures. Several restaurant chains, after determining that customers value lunch service within fifteen minutes of ordering, have established delivery time as the critical service standard. One restaurant has even placed alarm clocks on tables so that customers are fully aware of its compliance to standards. Service standards can even enable recovery to take place without the customer having to lodge a complaint. An operator of several Seattle restaurants gives all employees the authority to provide complimentary wine, desserts, or entire meals to compensate for errors and delays in service.9 Customers who wait for a table for more than ten minutes beyond the reservation time are offered free drinks. The restitution is tied to service standards that management developed.

Communicating the Importance of Service Recovery. Companies that are committed to developing lifetime customers, such as Promus Corp. and Nordstrom, recognize the significance of service recovery in achieving customer satisfaction and enhancing customer relationships. Through their values (e.g., “The customer is always right!”), these companies signal to employees the importance of taking responsibility for service failures and resolving problems. Employees become important listening posts, discovering customer concerns and facilitating recovery.

For example, while UPS conducts a variety of customer surveys and has programs in which executives meet directly with a national customer advisory group, it claims that the most important source of customer information about service issues and problems is the UPS drivers. UPS has built into their schedules an extra half-hour each week to spend time with customers, answering questions and receiving feedback about procedures, packages, and new services. This sends a strong message about the importance of recovery and the drive toward zero defects. UPS uses its quality management training efforts to communicate the importance of both resolving customer problems as well as learning from the failure.10

In another example, USAA, a large insurance and financial services company, has created its ECHO (“Every Contact Has Opportunity”) approach to use its telephone reps as listening posts. The firm’s 6,000 sales-service reps gather market trend and competitor data, generate new product ideas, and deal with customer complaints. When a complaint is discovered, a service rep assigns it to an “action agent” for quick resolution. As part of their work routine, telephone reps also report service failures to the appropriate unit inside USAA that uses this data to change faulty processes. The emphasis on identifying and resolving customer problems has led to an annual customer renewal rate of 98 percent.11

Companies also use physical symbols to communicate the importance of service recovery. The Ritz-Carlton provides every employee from the CEO to the busboys with a wallet-sized card that lists the organization’s core values along with twenty statements referred to as “The Ritz-Carlton Basics.” The statements contain several references to service recovery. For example, Statement 8 reads, “Any employee who receives a complaint ‘owns’ the complaint”; Statement 9 declares, “Instant guest pacification will be ensured by all. React quickly to correct the problem immediately. Follow up with a telephone call within twenty minutes to verify the problem has been resolved to the customer’s satisfaction. Do everything you possibly can to never lose a guest.” Communicating the value of service recovery creates an atmosphere conducive to identifying failures and achieving effective recovery.

Some companies explicitly tell customers how to lodge a complaint and what to expect from the process.

Training Customers in How to Complain. Some companies explicitly tell customers how to lodge a complaint and what to expect from the process. Scotiabank, one of Canada’s large chartered banks, has prominently placed pamphlets in all its branches that explain the five-step process designed to help customers get their complaint heard and resolved. The pamphlet stresses the importance the bank places on maintaining customer relationships and promises quick action to resolve problems. It specifies with whom customers should initially lodge complaints and describes how to make appeals. The pamphlet provides phone numbers of a bank vice president and the bank’s ombudsperson for customers who are dissatisfied with the initial resolution. The strategy is designed to both encourage use of the complaint option and enhance the responsiveness of the process. Furthermore, the pamphlet reinforces to employees the value that Scotiabank places on service recovery.

Using Technological Support. The use of toll-free telephone call centers to handle customer contacts, including complaints, is a growing trend.12 Since the introduction of the 1-800 service in the United States thirty years ago, the number of calls has grown from 7 million a year to more than 10 billion a year. Call centers offer several advantages over written correspondence in providing customers with convenient, low-cost access to service recovery. For example, research has found that oral communication is better suited to conveying compassion and empathy to irate customers than written communication. In addition, customers may feel more comfortable complaining over the phone than face-to-face. Call centers are even widely used where there is a small or minor tangible component to the service (e.g., AT&T) and when customers are present and have the option of complaining directly to a service provider. Holiday Inn, for example, has placed a sign in rooms that encourages customers who have concerns or comments about their stay to call the 800 number for the Holiday Inn Worldwide Guest Relations office.

General Electric’s Answer Center serves as the core component of its customer relations and recovery strategy. The center is open twenty-four hours a day, 365 days a year to service the approximately 3 million customers who call annually. GE places its 800 number directly on all its products and encourages customers experiencing problems to inform the company directly. By maintaining easily accessible and thorough customer and product records, GE service reps can quickly diagnose and solve problems, enhancing customer perceptions of responsiveness and lowering the cost of complaining.

Some firms are now using Internet websites to facilitate service recovery. Cisco Systems has created a database that provides a key word search of questions and answers provided to other customers. As new problems arise, they are added to the database. For more complex problems, Cisco has developed a “troubleshooting engine,” an expert system that walks users through problem-identification and -resolution processes. The company uses a series of questions created by service experts to lead customers to the solution. Problems that cannot be resolved through on-line technology are referred to the telephone support system. In addition, Cisco offers a “bug toolkit,” a collection of interactive tools for identifying, tracking, and resolving software bugs. The company’s online service-recovery program has reaped high customer satisfaction scores and significant cost savings. Based on that success, Ethan Thurman, Cisco’s director of service marketing, concludes, “You don’t have to touch customers personally to make them feel good about your service.”

Stage 2: Resolving Customer Problems

We found that most complaints are lodged when customers experience what they perceive to be a serious problem, a result consistent with other research. Thus, once customers have voiced complaints, they expect action. More specifically, they want justice or fairness.13 Customers form perceptions of fairness by assessing three aspects of service recovery: outcomes, procedural features, and interactional treatment. Outcome fairness concerns the results customers receive from complaints. Procedural fairness refers to the policies, rules, and timeliness of the complaint process. Interactional fairness focuses on the interpersonal treatment received during the complaint process. Each fairness dimension reflects several themes that emerged from our study, in which we asked a sample of 320 consumers in a large metropolitan area to recall a time when they complained to a service firm and describe what they found to be fair or unfair about how their complaint was handled (see the sidebar). (We describe the methodology and additional results of the study in the appendix).

Being Fair with Customers

The study results indicate the critical need for firms to devote far more attention to recovery. For each of the three aspects of service recovery, most customers believed that they had been treated unfairly and overall were dissatisfied with the way their complaint was handled (see Figure 2). In this section, we look at the survey results and the fairness of outcomes, processes, and interactions in more detail.

Providing Fair Outcomes. When services fail, customers expect to be compensated. The typical forms of compensation are refunds, credits, correction of charges, repairs, and replacements, either singly or in combination. We found that apologies also contribute to customers’ compensation for being inconvenienced or treated rudely.

Most customers surveyed judged the outcomes they received to be unfair.

Most customers surveyed judged the outcomes they received to be unfair. The driving force behind unfair assessments was the failure of companies to compensate adequately for the harm done or to recognize the costs incurred by customers in getting their complaint resolved (“I had to come back to the store three times to get a 20 percent refund!”). The most negative reactions were in response to complaints that were never resolved (“We wrote them and then called three times, and the problem has not been looked at or fixed”). Clearly, firms need to better understand the outcomes that customers expect from a service failure and to construct compensation packages that better acknowledge costs of the failure to the customer. The minority of customers who were favorably impressed typically pointed to compensation that included reimbursement for the inconvenience associated with the failure as well as basic exchanges or repairs. Customers also reacted more positively when the firm gave them options than when it unilaterally prescribed the outcome (“They offered me the choice of a refund or a free upgrade to a better room”).

Providing Fair Processes. Study findings indicate that fair procedures begin with the firm assuming responsibility for the failure. Next, the complaint must be handled quickly, preferably by the first person who is contacted.14 Other aspects of procedural fairness include having a flexible system that takes individual circumstances into account and obtaining input from the customer on what the final outcome should be. Less than half the customers surveyed found the recovery procedures they experienced to be fair. When the procedures were rated as fair, it was mostly because they were clear to the customer, and the problem was dealt with quickly and without hassles (“All I had to do was explain what happened. They had my address and other information. The check arrived three days later”). In the more frequent cases where customers judged procedures to be unfair, those surveyed were frustrated by a prolonged, inconvenient process that required them to repeat their complaint to several firm representatives who seemed unconcerned about the situation. These employee responses prompted one respondent to comment that the firm operated on the principle that “the customer is always wrong.” According to respondents, the employees’ indifferent behavior was typically driven by an attitude that the firm was not responsible for the problem and that they were not going to resolve it (“He told me that I should have been aware of their return policy, and he wasn’t going to change it just for me”).

Providing Fair Interactions. We found that fair interpersonal behavior involves demonstrating politeness, concern, and honesty; providing an explanation for the failure; and making a genuine effort to resolve the problem (“The ticket agent was very concerned with the lost reservation. She called a supervisor, and the two of them resolved the problem to my satisfaction”). Unfair interpersonal treatment reflected the opposite behavior (“It didn’t matter to her that I would be late. She just continued talking with her friend and ignoring me”). Uncaring attitudes and little effort to resolve the problem caused half of the respondents to rate the treatment they received as unfair.

At first glance, the uncaring behavior of these employees may seem odd; is it so difficult for employees to be polite and to try to resolve the problem? Examining customers’ responses suggests two reasons for why it might not be so easy. First, customers often enter the complaint process in an angry mood. Considerable skill is required to defuse the anger and address the problem. Second, employees often lack the authority to take care of the problem. The situation is frustrating for customers and employees alike, particularly when the customer expects immediate action.

Our results indicate that the three fairness dimensions strongly affect customers’ evaluations of service recovery, together explaining 85 percent of the variation in satisfaction with the complaint handling. Further, performing poorly on even one fairness dimension severely limits the potential for customer satisfaction. This result indicates that firms fully compensating customers do not get a return on the investment if the restitution is provided by rude, unsympathetic employees or if the process is a hassle for the customer. It also appears that being fair helps in the court of public opinion when a customer’s request is clearly unreasonable, as in a case that Starbucks faced. A customer claimed that Starbucks sold him two defective coffee makers (total value of $500) and forgot to include a free half pound of coffee with the purchase and that he was treated rudely by a clerk. The customer stated that unless the company took out a full-page ad in the Wall Street Journal apologizing, gave him a $2,500 machine, and funded a shelter for runaways, he would take out ads in the paper denouncing the company. Starbucks responded, although late, with repeated apologies and an offer to replace both machines and provide a refund as well as other gifts. The company’s response and its history of treating customers fairly limited the potential damage of the episode.15

Strategies for Recovering Successfully

What can firms do to improve recovery performance? We observed four practices that together dramatically improve service-recovery effectiveness: hiring, training, and empowerment; establishing service-recovery guidelines and standards; providing easy access and effective responses through call centers; and maintaining customer and product databases.

Hiring, Training, and Empowerment. Our research indicates that successful service recovery is highly influenced by the effectiveness of the front-line employee who receives the complaint.16 This is not surprising since 65 percent of the complaints were initiated with front-line workers. What it means is that the design of a recovery system must focus on the initial contact and on developing policies that enable employees to resolve the complaint efficiently. Servicerecovery performance must therefore be incorporated into human resource management practices.

Developing hiring criteria and training programs that take into account employees’ service-recovery role directly affects customers’ fairness evaluations.17 Roger Dow, vice president of sales and marketing at Marriott, observes: “The guest-contact employee is the only one close enough to the customer to recognize and evaluate a problem and make it right for the customer and keep that customer. These point are brought up over and over again in the training of our employees.”18

FedEx gives customer service reps five weeks of training before assigning them to a service center and gives them an additional four hours of training per month. Because of this training, reps can resolve problems so effectively that few matters rise to higher organizational levels. According to one FedEx manager, “The system has become so efficient that nothing escapes the net. I take care of three customer problems a week because everything has been headed off closer to the line. My phone would be ringing off the hook if our front-line people were not legitimately trained and empowered to handle customer needs.”19

Another company that pays careful attention to hiring and training is Ford, which puts all job candidates through its Assessment Center to evaluate nine skill levels. Assessment areas that are particularly relevant to service recovery include written and oral communication, listening skills, problem analysis, organizing and following through, resilience, and stress management. Ford includes recovery skills in the training of employees at its Customer Assistance Center, the unit with primary responsibility for service recovery. Among the training topics covered are company policies, warranties, listening skills, defusing anger, and interpersonal skills. To provide new employees with a realistic preview of their jobs, the company uses simulations such as an irate customer meeting, a meeting with dealers, and a written report of a dealer meeting. Training programs that include learning how to strengthen interpersonal skills and that provide a detailed knowledge of company complaint procedures support the implementation of a fair recovery plan.

Using databases in conjunction with call centers and websites helps firms achieve fairness goals.

Providing employees with the authority to respond to service failures strongly affects all three fairness dimensions.20 First, employees’ attitudes improve and their efforts increase when they are given the power to resolve problems. Second, the speed and convenience of the process are enhanced when employees can act immediately rather than having to seek out a manager or another department to respond to the problem. Third, outcome fairness improves if employees have the flexibility to provide the appropriate compensation, based on the customer’s circumstances and requirements. In addition, empowerment complements the use of service guarantees. Hotels such as Hampton Inn and Holiday Inn rely on employees to deliver on the promise and provide restitution if the service fails. At Hampton Inn, any employee, including reservations staff, janitors, and maids, can invoke the guarantee and provide the customer with a complete refund.

While empowerment contributes to service recovery in several ways, if used inappropriately it can lead to problems. If, for example, a front-line staff member allows customers to check out late from a hotel room, it causes problems for housekeepers trying to ensure that rooms are ready for new guests. Or a restaurant hostess who allows customers to arrive late and finish late from the first seating may cause complications for the waiters and waitresses and those who prepare tables for the second seating. Companies can overcome the difficulties by setting boundaries. For example, Marriott managers clarify “safe zones” for employees. Safe zones represent decisions for which employees are allowed to take action. Employees spend a day reviewing the principles of empowerment and discussing how various situations were handled. Safe zones apply to both actions (e.g., letting a guest check into a room early) and dollar limits (e.g., setting a $2,500 limit on customer compensation), based on the nature of the problem and the value of the customer.21

Establishing Guidelines and Standards. Developing guidelines for service recovery that focus on achieving fairness and customer satisfaction represents a direct approach to improving performance. In the southwestern United States, Samaritan Health Services has developed a framework, known as the “AAAA” Action Plan for Service Recovery, that is designed to score highly on all three fairness dimensions. The As stand for Anticipate and correct problems before they occur; Acknowledge mistakes when they occur without placing blame or making excuses; sincerely Apologize for the mistake, even if you are not at fault; and make Amends for the mistake by taking corrective action and following up to ensure the problem has been resolved. These principles guide the specific policies for the recovery program. Similarly, the statements from the Ritz-Carlton Basics card give employees clear direction for providing fair service recovery. They mandate that employees be polite (interactional), take quick action (procedural), and correct the problem (outcome). Ford has also developed standards for service-recovery responsiveness and accessibility. The metric for responsiveness includes a five-day turnaround on correspondence and a twenty-day closure on executive complaints; accessibility standards include 95 percent accessibility to incoming callers and a maximum hold on calls of thirty seconds.

Most firms fail to document and categorize complaints adequately, making learning more difficult.

Providing Easy Access and Effective Responses. In addition to removing barriers to the customer’s decision to complain, call centers contribute to all three fairness dimensions. The hours of operation (twentyfour hours a day, 365 days a year) and the timely customer and product information at the GE Answer Center make for convenient access and fast resolution, enhancing procedural fairness. GE follows up with a letter of apology to all callers, which contributes to procedural, outcome, and interactional fairness. Customers receive a goodwill certificate, free home repair, and/or compensation for food spoilage, depending on the nature of the failure. These gestures further contribute to customer perceptions of fair outcomes. Extensive training of service representatives in recovery practices helps achieve high scores on interactional fairness. GE’s customer research has verified that effective complaint resolution enhances customer repurchase decisions, demonstrating that the company’s efforts yield significant financial rewards.

Maintaining Customer and Product Databases. Using databases in conjunction with call centers and websites helps firms achieve fairness goals. The “entire focus” of PCS Health Systems “is on solving the customer’s problem on the first call,” according to Kirby Bessant, vice president of service operations. This goal poses a challenge because service reps’ decision-making latitude is limited; PCS’s health plan and HMOs have stringent requirements for administering benefits. The company uses customer-knowledge databases as the key source for immediate problem solving. The databases capture historical trends of clients (HMOs, health plans) and patients. PCS analyzes these trends with its clients. Together, they develop appropriate treatment protocols and plans. Throughout, patient confidentiality is rigorously protected, and PCS strives for win-win-win outcomes for its clients, patients, and the company.

By recording service failures in a database, Cisco Systems also improves recovery, since identifying systemic problems leads to faster and more accurate diagnoses of customer complaints and to improvement of the company website’s expert systems and troubleshooting engine databases. Cisco’s approach gives the customer greater control and flexibility in resolving the problem, enhancing service fairness. GE maintains extensive databases on customer purchases and service incidents that help guide recovery decisions. For example, recognizing that customers become particularly upset when products fail just after a warranty expires, GE includes warranty information in the customer database, and service reps can adjust compensation accordingly (e.g., waive the cost of the repair call).

Stage 3: Communicating and Classifying Service Failures

J. Willard Marriott, Jr., offers the following advice on service recovery: “Do whatever is necessary to take care of guests. Also track, measure, and follow up on how to handle it better next time, the first time.”22 Marriott’s approach reflects the Malcolm Baldrige National Quality Award scoring system, which provides service-recovery points for “complaint resolution that leads to quality improvement.” The system also requires organizational learning, defined as “the capacity or processes within an organization to maintain or improve performance based on experience.”23

Our study results show, however, that most firms fail to document and categorize complaints adequately, making learning more difficult. We found four explanations for this failure. First, in several cases, employees showed little interest in hearing the customer describe the details of the problem. They treated the complaint as an isolated incident needing resolution but not requiring a report to management. Second, many employees and managers devoted their energies to avoiding responsibility for the problem, instead blaming the customer for the failure. Third, numerous complaints were never resolved. Customers left telephone messages, complained to several employees, and wrote letters — and still no action was taken. Fourth, firms appeared to have no systematic way to collect and distribute complaint information to the individuals responsible for the process that failed.

What can firms do to ensure that complaints are properly recorded, disseminated, and classified? Many of the practices identified in Stages 1 and 2 can also be used during Stage 3. For example, call centers record the details of each complaint and ensure that the information is made accessible or is distributed to all appropriate managers and employees. Emphasizing service recovery in training front-line employees and cultivating a “learn from failure” attitude are central to getting complaints reported internally. The Disney Orientation Program, in which all new hires participate regardless of position, gives considerable attention to handling guest inquiries and complaints. An important element of the Disney orientation in service recovery is stressing the importance of reporting service failures to supervisors. Disney tracks “moment of truth” failures in its effort to eliminate sources of customer problems.

In addition to the influence of training on the reporting of service failures, the design of the recovery process can make the reporting of problems a seamless component. For example, Ritz Carlton employees who receive complaints “own” the complaint. It is their job to report the incident to the operations unit responsible for fixing the problem, complete a guest incident form that goes to the manager, and respond to the customer on the status of the situation. The design of the process requires that failures be reported.

A customer who complains frequently or is never satisfied with recovery efforts may be the “wrong customer.”

Similarly, the effectiveness of service guarantees has been attributed in large measure to the extensive training employees receive in recording details of the failure and implementing the guarantee. In addition, firms can facilitate the effective classification of service-failure data by creating internal complaint forms, accessing complaints made to front-line employees, and categorizing customers who complain.

Creating Internal Complaint Forms. Complaint forms are internal documents used to record service failures (we are not referring to paperwork filled out by customers to register their complaints), including documentation of service guarantees that have been invoked. The purpose of the complaint form is to facilitate organizational learning and ensure that the complaint is brought to a fair conclusion. Xerox’s “Customer Action Request Form” provides detailed information about customer problems. Complaints are identified as resulting from failures in one or more of thirteen business areas (e.g., equipment performance, service, mail order/delivery/installation, customer inquiries, sales). Within each category, more detailed coding is provided. For example, customer service problems have twelve possible codes (e.g., difficulty obtaining service, service rep unable to fix problem, service pricing, repair time). Three additional codes for root causes are identified. The first code deals with whether the problem primarily concerned process, people, product, or policy. Second, the problem is coded as primarily involving one of seven factors: attitude, communication, training, ethics, human error, technical, or invoice. The third code relates to business area (e.g., sales, service, supply, and logistics). Xerox also tracks the resolution, which includes who handled the problem, how it was handled, the financial implications of the recovery, and the details of customer contacts. When completed, the form is delivered to the customer relations administrator, who ensures that the information is provided to all relevant individuals. This comprehensive data collection and dissemination provides key information needed for service improvement.

Accessing Field Complaints. When they are present in the “service factory,” customers often lodge complaints with the nearest employee they can find (the individuals receiving the most inquiries at Disneyland and Disney World are the garbage collectors). Ensuring that complaints delivered in this way are communicated back to the organization is a major challenge that requires answering the question, “Why should an employee report a complaint?” Meeting the challenge requires involving front-line employees in the management of quality and customer satisfaction and rewarding rather than punishing complaint reporting.

At Motorola and Marriott, front-line employees are responsible for finding opportunities for service-quality improvement, which encourages them to identify complaints. For example, Marriott’s decision to invest in ironing boards in the rooms rather than small televisions for bathrooms (which managers thought advisable) was based on the housekeepers’ tracking of the number of requests for ironing boards (and the complaints when they were not immediately available) and the time they spent delivering the few ironing boards they had. By reporting and recording these complaints, the housekeeping department had the data necessary to justify its position.

When IBM Canada introduced an empowerment policy that allowed service reps to write checks to satisfy customer problems, it was a failure. Years of experience in IBM’s culture had convinced most employees that they would be “punished” for spending this money, thus discouraging them from taking action to compensate customers or alerting management to problems and complaints. This resistance can be overcome by stressing and reinforcing values that promote the empowerment process.

By contrast, Nordstrom, with its mission to “provide outstanding customer service,” celebrates extraordinary efforts by employees to resolve complaints. The company’s only rule is to “use your good judgment in all situations.” At Nordstrom, resolving and reporting service failures is an important part of the customersatisfaction promise. At company meetings and celebrations, Nordstrom uses stories to communicate instances in which employees went to exceptional lengths to provide excellent service including resolution of customer problems. As these examples illustrate, companies encourage the reporting of complaints when they focus on customer satisfaction and service quality and when they support and reward front-line employees for achieving those goals.

Categorizing the Customer. Tracking who has complained is helpful for two reasons. First, an effective service-recovery effort can generate high levels of satisfaction; however, subsequent service failures can undo the goodwill generated by a successful recovery. To avoid this trap, the J. Peterman Company puts all customers who have complained on a VIP list, and every subsequent purchase by those customers is hand-picked, inspected, and tested. Second, a customer who complains frequently or is never satisfied with recovery efforts may be the “wrong customer.” Lovelock has developed several strategies for discovering and removing wrong customers.24 These customers may be seeking benefits not provided by the company, may require more resources to serve than profit generated, or may simply be criminals. They are particularly troublesome for firms that offer a 100 percent unconditional service guarantee. To manage this problem, Hampton Inn maintains a database of customers who have invoked its guarantee; when they determine that a customer has violated their trust, rather than booking that customer a room they recommend another hotel.

Stage 4. Integrating Data and Improving Overall Service

Since customers complain about problems that they find important, complaints represent a valuable form of market information. However, because customers rarely complain when a service fails, companies seeking to improve service quality need to locate additional sources of information. The goal of data management is to ensure that the organization gathers relevant, credible, timely information and disseminates it to everyone involved in decisions on investments in service quality.

Generating Service-Quality Data. A market-driven approach to service research consists of many tools in addition to identifying and responding to complaints.25 Berry and Parasuraman have developed a “Service- Quality Information System,” a comprehensive approach to: service-improvement planning and resource allocation.26 The system includes, but is not limited to: customer, employee, and competitor surveys; mystery shopping; focus groups; customer and employee advisory panels; and service operating performance data. In identifying opportunities to improve services, many firms integrate complaint data with data gathered through one or more of these research methods. For example, Delta Hotels uses guest comment cards and surveys, employee opinion surveys, mystery shoppers, and quality self-assessments. The multiple methods and sources of data provide the hotel with a comprehensive view of service quality.

Information from call centers and customer databases is also valuable. GE monitors the lifetime performance of every appliance sold as well as the quality of every customer interaction. The company integrates the data collected (e.g., requests for servicing, product or service compliments and complaints, customer reactions to service calls, requests for product information) into its product and service design decisions. Maintaining databases of customer purchases is particularly helpful in cases where the only indication that a service has failed is that the customer has stopped purchasing from the firm. Foremost Insurance Group’s information system generates a list of insurance- policy cancellations monthly, which triggers a call to the customer to determine the cause of the defection. Staples has salespeople call accounts that have discontinued purchasing to assess the situation and try to win the customer back.

Disseminating the Data. While data may be collected by various sources within an organization, it must be made available to those responsible for implementing service improvements. Delta Hotels has a variety of procedures to ensure that information is effectively disseminated. Department meetings, which are held at least once a month, focus on the sharing of information, ideas, and plans for process improvement. A top priority is to ensure that the staff has the tools, training, and resources to “wow” the guest. The hotel has also created an “Employee Representative Team,” which includes one representative from each department and the general manager of a property. The team examines processes that involve more than one department.

Delta brings together front-line employees and managers at meetings to ensure both the sharing and the shared understanding of information. While Delta’s approach emphasizes face-to-face communication, other companies use technology to accomplish similar goals. For example, Ford electronically distributes complaint information gathered at its service center directly to the dealerships responsible for settling the dispute. The same information is also sent to the marketing research and engineering departments, which integrate the complaint data with additional research information. John Deere uses e-mail, bulletin-board postings, and specialized monthly reports to ensure that the information reaches those who need it to improve service quality.27

Investing in Quality Improvements. Firms determining investment priorities in service improvement must examine the impact of various options (e.g., increasing the speed of front-desk check-in versus expanding the room-service menu at a hotel) on customer satisfaction, repurchase intention, process cost, and market share. The goal is to identify those process improvements that will have the greatest impact on profitability.28 Investment decisions should also be driven by customer profitability assessments.

United Airlines, for example, has determined that business travelers account for 40 percent of its business by headcount but supply 72 percent of its revenue; “mile-collecting vacationers” constitute 60 percent of travelers but provide only 28 percent of revenue. The most frequent business travelers, dubbed “road warriors,” generate 37 percent of revenue even though they make up only 6 percent of the passengers. United learned through complaints and survey results that the road warriors were the customers least satisfied and most frustrated with air travel. Based on the data, the company is investing $400 million to provide business travelers with better seats, food, and lounges; a special predeparture service to help them avoid long lines; more frequent-flier benefits; and perks such as showers in terminals. According to John Edwardson, United’s president, “One of the things we are trying to do is pay a lot more attention to customers who pay us most.”

Linking Recovery to Profits

The relationship between service recovery and firm profitability can be clearly seen by examining the service-profit chain.29 The profit-chain concept argues that profit results from customer loyalty, which results from customer satisfaction with the service system; customer satisfaction (value) is generated by satisfied, loyal, and productive employees. The impact of service recovery can be traced through improvements in the service system (Stage 4) and through the direct effect on satisfaction of resolving a customer complaint (Stage 2) (see Figure 1).

Our research indicates that resolving problems effectively has a strong impact on customer satisfaction and loyalty. Studies show that some customers are actually more satisfied with a firm that follows a service failure with a remarkable recovery than they would have been had the failure not occurred in the first place.30 Furthermore, data from Xerox and American Express have demonstrated that only customers scoring at the highest level on satisfaction measures tend to be both loyal purchasers and “apostles,” customers who encourage others to buy from the company.31 Since recovery is closely tied to satisfaction, the revenue and profitability impact of service recovery can be dramatic. Studies on the return on investment in complaint-handling units in several industries, including retailing, banking, and automotive service, indicate that service-recovery investments provide substantial returns, ranging from 30 percent to 150 percent.32

Poor recovery following a bad service experience can create “terrorists,” customers so dissatisfied that they actively pursue opportunities to criticize the company.33 Our research found that the most negative reactions to poor recovery were expressed by customers previously loyal to the company. Loyal customers expected problems to be dealt with effectively and were disappointed when they were not, making service recovery key to maintaining the loyalty of these customers.34 While resolving a problem helps achieve the loyalty of one customer, using complaint data as an input into process improvement contributes to current and prospective customers’ satisfaction. Sustaining satisfaction over many encounters builds equity with customers, strengthening loyalty and further driving profitability.35

Another important profitability question for companies to consider is how service recovery affects the employees involved. In examining the customer descriptions of service recovery, we found two interesting themes related to employees. First, customers who found the service recovery handled fairly commented that the employee was concerned about the problem, eager to help, and happy that the complaint was resolved to the customers’ satisfaction. Second, when customers indicated that they found the complaint handled unfairly, employees were frequently observed to be rude and defensive, indifferent to providing assistance, and increasingly angry as the dispute progressed. These findings suggest that individual service-recovery incidents affect the satisfaction of not only the customer but also the employees involved. Our research further suggests that employees faced with a large number of complaints to handle and no effective way to deal with them are likely to be very dissatisfied. Therefore, developing effective recovery programs and improving the service system should enhance service quality and increase employee satisfaction and loyalty, contributing to customer value and ultimately to improved profitability.


Compelling evidence that customer loyalty drives profitability in service industries has led to recommendations that firms shift their focus away from “offensive” strategies aimed at seeking new customers and toward “defensive” strategies aimed at satisfying and keeping current customers.36 Service recovery is a cornerstone of a customer satisfaction strategy. Over the lifetime of customer relationships, conflicts are inevitable. Our data confirm that managing those conflicts effectively is vital in maintaining customer loyalty and trust.37

Our study results suggest that firms need to develop a comprehensive service recovery system that encourages dissatisfied customers to voice their complaints and that provides a fair process and outcome. To maximize the impact of the system, service design and investment decisions should reflect an understanding of the vital role that recovery plays in contributing to improved performance, customer and employee satisfaction, and firm performance and profitability.


1. Focusing marketing efforts on current customers is referred to as “customer satisfaction” or “defensive” strategy. For insights into the benefits of such a strategy, see: C. Fornell, “A National Customer Satisfaction Barometer: The Swedish Experience,” Journal of Marketing, volume 56, number 1, 1992, pp. 6–21; A. Griffin, G. Gleason, R. Preiss, and D. Shevenaugh, “Best Practices for Customer Satisfaction in Manufacturing Firms,” Sloan Management Review, volume 36, Winter 1995, pp. 87–98; and F.F. Reichheld and W.E. Sasser, Jr., “Zero Defections: Quality Comes to Services,” Harvard Business Review, volume 68, September–October 1990, pp. 105–111.

2. B. Ettorre, “Phenomenal Promises that Mean Business,” Management Review, March 1994, pp. 18–23; and R.T. Rust, B. Subramanian, and W. Wells, “Making Complaints a Management Tool,” Marketing Management, volume 1, number 3, 1992, pp. 41–45.

3. The importance of frontline employees in delivering service quality is developed in: L.A. Schlesinger and J.L. Heskett, “Breaking the Cycle of Failure in Services,” Sloan Management Review, volume 32, Spring 1991, pp. 17–29.

4. C.W.L. Hart, J.L. Heskett, and W.E. Sasser, Jr., “The Profitable Art of Service Recovery,” Harvard Business Review, volume 68, July–August 1990, pp. 148–156.

5. L. Dube and M. Maute, “The Antecedents of Brand Switching, Brand Loyalty and Verbal Responses to Service Failures,” in T. Swartz, D. Bowen, and S. Brown, eds., Advances in Services Marketing and Management, volume 5 (Greenwich, Connecticut: JAI Press, 1996), pp. 127–151; and Technical Assistance Research Program, Consumer Complaint Handling in America: An Update Study (Washington, D.C.: Department of Consumer Affairs, 1986).

6. J. Singh, “A Typology of Consumer Dissatisfaction Response Styles,” Journal of Retailing, volume 66, number 1, 1990, pp. 57–99.

7. Dube and Maute (1996).

8. T. Levitt, “Marketing Success through Differentiation — of Anything,” Harvard Business Review, volume 58, January–February 1980, pp. 83–91.

9. T.W. Firnstahl, “My Employees Are My Service Guarantee,” Harvard Business Review, volume 67, July–August 1989, pp. 28–32.

10. J. Kelly, “From Lip Service to Real Service: Reversing America’s Downward Service Spiral,” Vital Speeches of the Day, volume 64, number 10, 1988, pp. 301–304.

11. B. Rossello, “Customer Service Superstars,” ABA Banking Journal, volume 89, number 10, 1997, pp. 96–104.

12. J.L. Heskett, W.E. Sasser, Jr., and C.W.L. Hart, Service Breakthroughs: Changing the Rules of the Game (New York: Free Press, 1990); and C.L. Martin and D.T. Smart, “Consumer Experiences Using Toll-Free Corporate Hotlines,“ Journal of Business Communications, volume 31, number 3, 1994, pp.195–212.

13. S.S. Tax, S.W. Brown, and M. Chandrashekaran, “Customer Evaluations of Service Complaint Experiences,” Journal of Marketing, volume 66, April 1988, pp.60–76; and E.C. Clemmer and B. Schneider, “Fair Service,” in Swartz et al. (1996), pp. 109–126.

14. C. Boshoff, “An Experimental Study of Service Recovery Options,” International Journal of Service Industry Management, volume 8, number 3, 1997, pp. 110–130.

15. “On Achieving Excellence,” December 1995, pp. 2–3.

16. J. Carlzon, Moments of Truth (New York: Balligen, 1987).

17. Schlesinger and Heskett (1991).

18. L.L. Berry and A. Parasuraman, Marketing Services: Competing through Quality (New York: Free Press, 1991).

19. “Federal Express Uses a Three-Level Recovery System,” The Service Edge, December 1990, p. 5.

20. For a review of the use of empowerment in service recovery and other aspects of service management, see: D.E. Bowen and E.E. Lawler, “Empowering Service Employees,” Sloan Management Review, volume 36, Summer 1995, pp. 73–84.

21. Ibid.

22. Ibid.

23. E.C. Nevis, A.J. Dibella, and J.M. Gould, “Understanding Organizations as Learning Systems,” Sloan Management Review, volume 36, Winter 1995, pp. 73–85.

24. C.H. Lovelock, Product Plus (New York: McGraw-Hill, 1994).

25. G.S. Day, “Continuous Learning about Markets,” California Management Review, volume 36, Summer 1994, pp. 9–31.

26. L.L. Berry and A. Parasuraman, “Listening to the Customer — The Concept of a Service-Quality Information System,” Sloan Management Review, volume 38, Spring 1997, pp. 65–76.

27. Ibid.

28. R.T. Rust, A.J. Zahorik, and T.L. Keiningham, “Return on Quality: Making Service Quality Financially Accountable,” Journal of Marketing, volume 59, number 2, 1995, pp. 58–70.

29. J.L. Heskett, T.O. Jones, G.W. Loveman, W.E. Sasser, Jr., and L.A. Schlesinger, “Putting the Service-Profit Chain to Work,” Harvard Business Review, volume 72, March–April 1994, pp. 164–174.

30. M.A. McCollough and S.G. Bharadwaj, “The Recovery Paradox: An Examination of Consumer Satisfaction in Relation to Disconfirmation, Service Quality, and Attribution-based Theories,” in C.T. Allen et al., eds., Marketing Theory and Application (Chicago: American Marketing Association, 1992), pp. 102–107.

31. Heskett et al. (1994).

32. Technical Assistance Research Program (1986).

33. Heskett et al. (1994).

34. V. Zeithaml, L.L. Berry, and A. Parasuraman, “The Nature and Determinants of Customer Expectations of Service,” Journal of the Academy of Marketing Science, volume 21, Winter 1993, pp. 1–12.

35. F.F. Reichheld. “Loyalty-Based Management,” Harvard Business Review, volume 71, March–April 1993, pp. 64–74; and T.A. Oliva, R.L. Oliver, and I.C. MacMillan, “A Catastrophe Model for Developing Service Satisfaction Strategies,” Journal of Marketing, volume 56, number 3, 1992, pp. 83–95.

36. Fornell (1992); Reichheld (1993); Heskett et al. (1994); Griffin et al. (1995); and Reichheld and Sasser (1990).

37. M.J. Bitner, B.H. Booms, and M.S. Tetreault, “The Service Encounter: Diagnosing Favorable and Unfavorable Incidents,” Journal of Marketing, volume 54, number 1, 1990, pp. 71–84.


The authors thank Michael Hutt, David Bowen, and Mary Jo Bitner (all from Arizona State University), Lance Bettencourt (Indiana University), and Rebecca Grant and David McCutcheon (University of Victoria) for helpful comments on an earlier version of the manuscript. The authors also thank the Center for Services Marketing and Management at Arizona State University for funding the research.

2 Comments On: Recovering and Learning from Service Failure

  • debra netherton | January 26, 2015

    I am not sure how to get help, but I was enjoying reading this paper but don’t think I can access the remainder of the paper. The last paragraph visible is ‘Being Fair with Customers.’ Any help would be appreciated.

    In respect to the paper and its valid recommendations, it is extremely disappointing that management in large, profitable organisations are still failing their customers and employees by not managing increasing levels of complaints, in 2015!.

  • Leslie Brokaw | October 26, 2015

    Hello Ms. Netherton — Thank you for letting us know about the missing section of this article. We have updated this page so that the article is now complete. — Leslie Brokaw, Digital Contributing Editor, MIT SMR

Add a comment