Customer complaints and service policy in electronic commerce

Handling customer complaints is an important strategy to retain customers. Therefore, in the event of service failure, eretailers should concentrate on recovery policies. However, studies discussing prevention policies to avoid customer complaints are scant. This study collected 5933 real customer complaint data from an electronic commerce customerservice database and classified customer complaints into 6 types. The findings showed that a number of customers were dishonest and took advantage of recovery policies. After interviewing business managers and consultants, this research suggests that e-retailers have prevention policies to guarantee accuracy of packaging and delivery processes. Prevention policies can reduce customer complaints, and avoid extra costs for businesses conducting recovery policies.


Introduction
With the rapid growth of electronic commerce (EC), the Internet has become an important channel for business transactions. The number of e-retailers and e-shoppers has increased dramatically in recent years (Goode & Harris, 2007). Online shopping allows customers to purchase products at their convenience. However, e-shoppers do not have the opportunity to diagnose the product during the transaction process. Customer complaints from an unsatisfied experience of online shopping would also spread more quickly by electronic word-of-mouth (eWOM) than by offline communication. An incremental negative eWOM is more powerful in cutting product sales than an incremental positive eWOM is in increasing sales (Chevalier & Mayzlin, 2006). For this reason, handling customer complaints has become a critical strategy for electronic customer relationship management (eCRM) toward customer satisfaction (Cho, Im & Hiltz, 2003).
Previous studies (Semeijn, Van Riel, Van Birgelen & Streukens, 2005) have found that loyal customers are crucial to EC business survival because the competition is just a mouse click away. Unfortunately, there are always situations where service failure occurs during service delivery. The service recovery policy is considered a key factor to influence customer satisfaction (Grewal, Roggeveen & Tsiros, 2008). Empirical findings show that customer satisfaction ensures customer loyalty, repeat sales, and positive word-of-mouth communication (Bearden & Teel, 1983). Effective service recovery also leads to enhanced perceptions of firm competence, and a positive image of perceived service quality (Kelley & Davis, 1994). However, the cost of proper service recovery is very expensive, and firms are eager to develop business policies for customer complaints. Service recovery policy expends extra costs for firms. Occasionally, dishonest customers may take inappropriate advantage by cheating because of the lack of product diagnostics in EC. Therefore, an appropriate business policy to prevent service failure decreases customer complaints. A service failure prevention policy can also protect firms from being cheated by dishonest customers and guarantee service quality. This research has two purposes, based on the importance of proper EC policy development for customers: to empirically study and analyze customer complaints in EC and to study current business policies for preventing service failure in EC. This section introduces the motivation of this study. The next section reviews the literature background. The third section analyzes empirical customer complaint data from an EC customer service database. The final section presents the discussion and conclusion.

Literature review
Perceived service quality and service failure Perceived service quality can be defined as customer overall impression of the relative superiority of the firm and its services (Bitner, 1990). This means that customers compare their expectations with customer perceptions of actual service performance to confirm the service quality (Boshoff & Gray, 2004). Perceived service quality positively influences customer continuance (Naidoo & Leonard, 2007). However, service products are intangible, inseparable, variable, and perishable (Wolak, Kalafatis & Harris, 1998), making it difficult to totally avoid service failure. Customer service failure may cause possible negative consumption emotions and exert a negative effect on satisfaction judgments (Phillips & Baumgartner, 2002).
A service failure is defined as service performance that falls below customer expectations (Hess, Ganesan & Klein, 2003). Forbes, Kelley, and Hoffman (2005) used the critical incident technique (CIT) to classify e-retailer service failure into three types. The first is "response to service delivery system/product failure," including service failures that occur in packaging and delivery processes. The second is "response to customer needs and requests," which indicates a misunderstanding between customers and the firm. The last type is "unprompted and unsolicited actions," which involves inappropriate actions of employees to customers.

Service recovery and customer satisfaction
In the event of a service failure, the firm's challenge of commitment to service quality and customer satisfaction depends on how it responds after the disconfirmation of customer expectations (Zemke & Bell, 1990;Boshoff & Staude, 2003). If a firm does not have recovery strategies, customers have higher switch intention (Forbes et al., 2005). Customers often switch to an alternative service provider not only because of service failure, but also because of the unacceptable response of customer attempts to redress failure (Keaveney, 1995). Customer switch cost is substantially lower in the online shopping environment because they can search and find an alternative provider or product easily.
Service recovery refers to the actions by a service firm to restore a customer to a state of satisfaction after a service failure (Boshoff, 2007) and is an effective strategy to solve service failure and change customer attitudes from dissatisfied to satisfied (Craighead, Karwan & Miller, 2004;Hoffman, Kelly& Rotalsky, 1995). Day and Landon (1976) indicated that customers have different satisfaction levels according to their previous purchasing experience. Therefore, firms should have different recovery strategies to resolve customer complaints. Miller, Craighead and Karwan (2000) indicated two typical types of recovery strategies. One is a psychological form, in which a service provider apologizes to and sympathizes with their customers to comfort them and decrease customer complaints. The other is a tangible form, in which a firm makes customers feel that they are fairly treated by giving them physical reparation, such as a discount or gift. Customer belief of a good purchase experience, including managing service failure well, positively influences customer satisfaction, loyalty, repurchase intention, and willingness to spread positive WOM (Simons & Kraus, 2005;McDugall & Levesque, 2000;Spreng, Harrell & Mackoy, 1995).

Methodology
The importance of service recovery is obvious. However, service recovery is a compensatory strategy following a customer complaint. A service provider with strategies in place to prevent service failure might be another approach to decrease customer complaints. When service providers focus more on prevention strategies, they can save recovery costs and give customers a more satisfying purchase experience.
Data were collected from an EC customer service database. This research gathered 5933 real customer complaint data from January 2008 to March 2011. Customer complaints were classified, and the business prevention policies were listed.

Phase 1: Classifying service failure of EC
Knowledge of the cause of customer complaints is necessary before designing prevention policies. The data findings showed that service failure occurred from different causes. First, a product producer must be responsible for product quality. Second, many products are not sold directly by the producer, but by a dealer, typically an e-retailer or a platform provider, which is considered the main service provider to the e-shopper. Third, sometimes service failure occurs during the delivery process because the logistics is separated from the dealers. Finally, the dishonest customer plays a part. Table 1 lists the types of customer complaints from the data by different causes.
Customer complaints were divided into six types according to the empirical data. First, Product defect, includes all types of broken products during different processes. Occasionally, defective items are not checked strictly by the producer; therefore, the product producer has responsibility of Defective new goods. Products delivered to the e-retailer might be damaged during the package to shipment process. Therefore, e-retailers have to manage the Defective packaging process. Products can also break during the delivery process in a traffic accident. Therefore, logistics should focus on the Defective delivery process. Finally, because of customer inappropriate use, products might be broken during the warranty, another cause of customer complaints.
Second, defects are occasionally unrelated to the product, but involve Packaging errors. One type of packaging error is the original package mistake, for instance, Packaging the wrong product in the wrong product box or Losing attachments. The product producer is responsible for these two situations. The e-retailer typically sells various products on a web site; thus, when a customer order includes several products, different items are in one delivery package. Dishonest logistics might Steal some of the items because they think e-retailers compensate customers to maintain their reputation. Finally, Dishonest customers lie about lost items, or Return incomplete items.
Third, Compatibility involves information service. For example, the producer attaches an incomplete manual or no manual with the product, resulting in a product Without compatibility information. Similar to the e-retailer, the service counter that has a shortage of product information and cannot answer customer questions well causes customer complaints. Fourth, Pricing is an important reason behind customer complaints that typically results from Unexpected price cutting by the producer or e-retailer after customer purchasing. Another situation is if customers find a Lower price in other channels, they ignore the online order and purchase the cheaper product.
Fifth is Customer cognitive differentiation. This situation originates with customers' incorrect images of the product, including Product not as expected and Buying the wrong product. A special situation is customer Dishonest warranty extension. For instance, a customer purchases a product whose warranty expires, and then buys a new one and returns the old one to extend the warranty. When the eretailer questions the customer, the customer denies returning the old one. Sometimes customers Cancel orders unilaterally without any reason, and the e-retailer who argues with customers causes customer complaints. Finally, Poor service quality of the e-retailer is the main reason for customer complaints.
Sixth, Delivery delay is a common problem in EC because products do not arrive directly from the e-retailer to customers. Wrong customer address occurs if the e-retailer records or the customer provides the wrong one. Sometimes package process error or delivery process error causes Dealer delay and Logistics delay. Dealer inventory shortage, for example, is when the e-retailer shows wrong inventory numbers on the web site, causing customers to not receive their order or having to wait a long time. Table 2 shows the percentage of customer actions following service failure. When customers receive defective products, wrong items, or have cognitive differentiation with the products, they typically choose to exchange or return the products. Fewer customer complaints on compatibility and delivery delay might be caused by customers concentrating more on the product. However, when customers face the price problem (typically a cheaper choice), 67 of 67 customers choose to return the products.

Phase 2: Interviewing business managers and consultants
Business managers and consultants were interviewed in this study to determine their policies in practice. After product selling, the system records the delivered date to distinguish whether the product is under warranty. When service failure occurs, if the product is under warranty, customer complaints go into the exchange or return process. However, dishonest customers take advantage of recovery policies. First, customers declare product or attachment shortages to obtain extra items. Second, dishonest customers return incomplete products or even empty boxes. Third, customers dishonestly extend the warranty and return a used product that cannot be resold, resulting in e-retailer loss. Table 3 shows the prevention policies to protect e-retailers from suffering loss.

Conclusions
This research offers firms the knowledge of how to simultaneously retain customers and reduce loss. Because of the lack of product diagnostics in EC, customer cognitive differentiation frequently occurs. Many previous researches have focused on recovery from customer complaints. Service failure recovery strategies are important to retain customers, but some dishonest customers use recovery strategies to take advantage of product providers. This study suggests that not only customers want to decrease risks, but also firms in the EC environment.
This research collected 5933 real customer complaint data from an EC customer service database and classified six types of service failures including product defect, packaging errors, compatibility, pricing, customer cognitive differentiation, and delivery delay. The result showed that the main reasons for customer complaints were customer cognitive differentiation (66,16%) and product defect (27,81%) because customers are mainly concerned with product quality. However, dishonest customers also used customer complaints as an excuse to request additional attachments or to return untruthful product packages, causing unnecessary loss to e-retailers.  Recording package process and the serial numbers of products. Tagging the package weight and confirming with the logistics.

Incomplete return
Recording package process and the serial numbers of products. Tagging the package weight and confirming with the logistics. Taking photos before enclosing the package as evidence. Printing a lot number and packaging with the products. Dishonest warranty extension Printing a lot number and packaging with the products. Refusing a product return with a broken warranty sticker.
An interview with business managers and consultants provided information about some business policies when facing customer complaints. A service provider first divides customer complaints into two groups: those under the warranty period or expired. If products are still under warranty, customer complaints go into the exchange or return process. To prevent dishonest customers, e-retailers take photos or videos to record the package process and the serial numbers of products. This issue reminds EC firms to not only focus on recovery policies, but also to prevention policies. These policies give fewer chances to dishonest customers to take advantage of e-retailers. Prevention policies ensure accuracy and save time to exchange or return products for normal customers. Business prevention policies might offer customers a more satisfactory online shopping experience and avoid EC firms bearing extra recovery costs.