1 Introduction

In today’s world, firms face the challenge of increasingly competitive environments, making it essential to establish successful, long-lasting relationships with customers. Bonnemaizon et al. (2007) establish that relationship marketing is still in a process of development. Many studies show that a company’s customer portfolio is its greatest asset, determining the firm’s capacity to generate present and future wealth (Petersen et al. 2009). To build a stable, long-term customer base, firms employ relationship marketing strategies (De Wulf et al. 2001); in other words, a set of specific investments are made in each customer aimed at establishing, developing, and maintaining a strong company-customer relationship in order to maximize profits and gain a competitive edge (Morgan and Hunt 1994). Customer and market trends are forcing companies to develop and to manage more participatory approaches and, therefore, aspects such as co-creation of value with consumers and interaction with communities are becoming key aspects for understanding the success of relational approaches (Bonnemaizon et al. 2007).

The wealth of literature on customer management in recent years is remarkable (Verhoef et al. 2010), one of the main topics being the analysis of “customer engagement” (Kumar et al. 2010; van Doorn et al. 2010; Verhoef and Lemon 2013). The literature has recognized customer engagement as the evolution of relational concepts and represents one of the key research priorities for the Marketing Science Institute, customer engagement has been conceptualized as “customers’ behavioral manifestations that have a brand or firm focus, beyond purchase, resulting from motivational drivers” (van Doorn et al. 2010, p. 254) with a special focus on non-transactional customer behaviors. The concept goes deeper than simple customer satisfaction, individual loyalty, and repurchase behaviors (van Doorn 2011); it refers to a wide array of non-transactional behaviors, including personal recommendations (word-of-mouth, referrals, and blogging) and co-creation that have a significant, long-term impact on the profitability of both current and prospective customers (Bijmolt et al. 2010; Kumar et al. 2010; Verhoef et al. 2010; Verhoef and Lemon 2013). Firms need to understand that customers’ experiences are critical and that they can easily interchange opinions among themselves so they must ensure excellent experiences whenever the customers interact with the company (Bonnemaizon et al. 2007).

However, in the process of building strong relationships and customer engagement, companies may make mistakes (service failures) which can jeopardize customer satisfaction and, hence, the success of the company-customer relationship (DeWitt et al. 2008; Michel and Meuter 2008; Mattila 2001). Once a service failure has occurred and the customer has filed the complaint, firms tend to launch a complaint-handling process aimed at providing a solution to the problem and regaining customer satisfaction (Bitner et al. 1990; Davidow 2000; Ro 2014; Cambra-Fierro et al. 2015a). Complaint handling refers to a “formal organizational procedure for registering and processing customer complaints that is consistent with complainants’ needs” (Homburg and Fürst, 2005, p. 96). The quality of a complaint-handling system depends on the degree to which a firm has clear, simple, and customer-oriented guidelines for dealing with complaints to ensure appropriate actions by employees (Homburg et al. 2010). If these efforts are successful, new-found customer satisfaction can rekindle and enhance the company-customer relationship (Smith and Bolton 2002; Nikbin et al. 2013), so an effective response from the company is essential (Cambra-Fierro et al. 2015a). A review of the literature reveals that much of the research to date has focused on antecedents of complaint handling, particularly on the variables perceived justice (Chang and Chang 2010; Kau and Loh 2006) and perceived effort (Huang 2008). With regard to the consequences of complaint handling, the literature analyzes the impact of complaint handling on customer satisfaction (Buttle and Burton 2002; Chang and Chang 2010; Ro 2014), dissatisfaction after a service failure (Buttle and Burton 2002), customer loyalty (Sajtos et al. 2010), and repurchase intentions (De Matos et al. 2007). However, an extensive review of the complaint-handling literature reveals a surprising lack of research into customer engagement as a consequence of complaint-handling efforts, perhaps because customer engagement is still an emerging topic.

The present study, therefore, aims to integrate the concept of customer engagement and traditional notions of complaint handling. Our research proposition is that, by offering fair solutions in service failure scenarios, customer satisfaction can be restored, which ultimately will lead to an increase in customer engagement. From a business standpoint, then, customer value grows after successful complaint-handling initiatives and is reflected in enhanced customer engagement, since highly engaged customers are more valuable for firms both in the short- and long term (Kumar et al. 2010).

In order to address the gaps in the research, we propose a conceptual model designed to explain the impact of customer satisfaction with complaint handling on customer engagement. Specifically, we aim to answer the question: can complain handling initiatives translate into increased customer engagement? Based on the former research question, we propose a model that links complaint-handling investment (customers’ perceived effort and perceived justice), complaint-handling satisfaction, and customer engagement.

The rest of the paper is organized as follows. The following section reviews the link in the literature between complaint handling and customer engagement. The section entitled Conceptual model takes a closer look at recent customer engagement literature, along with the most relevant complaint-handling research to date. Grounding our work in two well-established theories, relationship marketing and the principle of reciprocity, we propose a conceptual model for understanding the relationships that link complaint-handling investment (customers’ perceived effort and perceived justice), satisfaction with the complaint-handling process, and the impact of this satisfaction on engagement. This model serves as a framework for deriving a set of hypotheses which we explore empirically in the context of the Spanish financial services industry in the Empirical study section. After that, we provide an in-depth analysis of our findings and present their implications, both for the literature and for best business practice. Finally, we draw our conclusions.

2 Review of the literature

As stated earlier, the present study aims to integrate the concept of customer engagement and traditional notions of complaint handling. Customer engagement has received little attention in the literature to date—even less so in relation to complaint-handling processes—despite the fact that highly engaged customers are more valuable customers as they may contribute to improving business performance both directly (through increased spending, cross-buying, service usage, and up-buying) and indirectly (through positive word-of-mouth and co-creation) (Kumar et al. 2010). In this line, van Doorn et al. (2010) define engagement as a set of non-transactional behaviors with future impact in both transactional actions and profitability. This concept is of interest because, as authors such as Bonnemaizon et al. (2007) highlight, market trends are forcing firms to involve consumers in the value creation process as well to assure satisfaction in this experience, both as individuals and as part of a community. Therefore, firms need to extend their focus from the standard relationships with customers to a set of relationships with a network of easily connected players.

Likewise, customer lifetime value (CLV), or the value that the relationship with a customer provides for the company over the entire duration of the relationship, is underestimated when non-transactional engagement behaviors (e.g., word-of-mouth and feedback) are not taken into account (Kumar et al. 2010; Verhoef and Lemon 2013). Yet, engaged customers are not an easy commodity to manage and any company can make mistakes that damage the customer-company relationship and lead to a loss of customer value. This is, precisely, why research into the nature of both engagement and complaint handling is so essential.

Maxham (2001), Teng et al. (2014), and Cambra-Fierro et al. (2015b), among others, consider service failure to be a real or perceived problem occurring at some point during the customer-company interaction. In order to solve these problems and address customer complaints, companies implement a complaint-handling process (Davidow 2000). For authors such as Bijmolt et al. (2010) and Kumar et al. (2010), among others, customer complaints and how firms manage them are inexorably linked to customer engagement. Engagement has an impact on customer–customer interactions and may affect transactions down the line (Kumar et al. 2010; Verhoef et al. 2010; Verhoef and Lemon 2013); hence, when customers experience service failures, we can expect negative attitudes toward the firm and potential negative word-of-mouth (Kumar et al. 2010; Verhoef et al. 2010). Complaint-handling processes allow firms to restore customer satisfaction (Bitner et al. 1990; Nikbin et al. 2013) as well as to provide a platform for company-customer interactions that are essential for spawning and cultivating customer engagement (Bijmolt et al. 2010). This interaction allows feedback from customers and the chance to co-create and improve service delivery quality, a valuable non-transactional opportunity for firms (Kumar et al. 2010). Authors such as van Doorn et al. (2010) and Verhoef et al. (2010) suggest that firm-led initiatives have a positive impact on customer engagement. In sum, with the specificities of complaint-handling interactions in mind, it seems that firm-driven initiatives are associated with investment and honest solution-seeking.

The following sections provide a more in-depth look at the state of the research into the antecedents and consequences of complaint-handling processes, respectively.

2.1 Antecedents

Several studies have examined antecedents to complaint-handling processes and how these variables affect final customer satisfaction (Nikbin et al. 2013; Gelbrich and Roschk 2011; Smith et al. 1999). Maxham (2001) is a cornerstone in the analysis of complaint-handling antecedents, introducing perceived effort as a key variable and showing how greater perceived effort in complaint-handling processes has a positive impact on customer satisfaction. The company, therefore, should make it clear to the customer the amount of resources, time, and effort it has invested in solving the problem because, otherwise, the level of satisfaction will be lower (Mattila and Patterson 2004).

Another pioneering study, Tax et al. (1998), looks at perceived justice and demonstrates that customers are more satisfied with complaint handling when justice is perceived in the firm’s reaction to their complaint and in the final solution offered. Smith et al. (1999) define perceived justice as the individual subjective assessment of organizational responses. The three dimensions of perceived justice (distributive, procedural, and interactive) and their impact on satisfaction have also received much attention in prior research (Gelbrich and Roschk 2011). Karatepe (2006), for example, suggests that the three dimensions of perceived justice have a positive impact on customer satisfaction.

According to the recent literature on customer engagement (e.g., van Doorn et al. 2010; Verhoef et al. 2010), company initiatives not only affect customer satisfaction but also may influence engagement as well. In sum, both perceived effort and perceived justice are key antecedents for measuring customer satisfaction effectively in complaint-handling contexts.

2.2 Consequences

With regard to the effects of complaint handling, previous researchers have mainly focused on satisfaction (Maxham 2001), repurchase behaviors (Maxham 2001), loyalty (Karatepe 2006; Kau and Loh 2006; Nikbin et al. 2013), trust (Kau and Loh 2006), and word-of-mouth (De Matos et al. 2009) as key variables to measure the consequences of companies’ complaint-handling investments.

However, what companies are really interested in is seeing complaint-handling efforts translated into financial results, and Johnston (2001) determines that the effective management of complaint-handling processes has a strong positive impact on profitability. A growing body of literature on customer relationship management (e.g., van Doorn et al. 2010; Verhoef et al. 2010) suggests that repurchase is a transactional behavior, while customer engagement is a superior construct, with potential present and future economic value for firms, which simultaneously secures non-transactional behaviors (word-of-mouth, referrals, feedback, influencers, and co-creation). As Bijmolt et al. (2010) and van Doorn (2011) emphasize, customer engagement is not the sum of established relational constructs; it is a new, much stronger concept where the common variance of a set of relational behaviors is simultaneously displayed. Both the literature (e.g., Kumar et al. 2010) and business logic suggest that satisfied customers may become engaged customers, creating value for firms, while unhappy customers will tend to be disengaged and detract value.

Grounding our research in the literature, the present study proposes a conceptual model where customer engagement is a key product of successful complaint-handling processes. In the following section, we present our conceptual framework, drawing upon relationship marketing theory and the principle of reciprocity.

3 Conceptual model

One of the fundamental notions underpinning relationship marketing theory is that building, developing, and maintaining successful relationships are beneficial for both companies and customers (Morgan and Hunt 1994, p. 22). Both parties are willing to invest resources to foster cooperation and to reinforce a stable relationship which will prove mutually beneficial in the long term (Macneil 1980; Morgan and Hunt 1994); there is a mutual commitment for working together to make the partnership profitable (Anderson and Weitz 1992). The resulting relationship increases the value of the product or service and transforms indifferent consumers into loyal customers, while bolstering company-customer ties (Berry 1995). Yet all companies make mistakes that may lead customers to reconsider the relationship with their service provider. Adopting a relationship marketing approach—and aware of the potential benefits of remaining on good terms with the customer—firms will invest time and effort in providing solutions for problems that arise (Grönroos 1998).

If customers are aware of company efforts to solve their problems, a spirit of reciprocity may blossom (Palmatier et al. 2009). This is precisely what the principle of reciprocity postulates: that investments made by one party in a relationship generate the other party’s desire to reciprocate, which, in turn, motivates investment in the relationship in order to avoid a feeling of guilt for not having corresponded to the initial investment action (Bagozzi 1995). If this occurs, the customer may reciprocate in a number of ways (Palmatier et al. 2006). We aim to determine whether reciprocity generated by complaint-handling satisfaction may take the shape of greater customer engagement.

As we have already commented, customer engagement is considered to be a set of customer behaviors beyond purchase which result from individual motivational factors (Kumar et al. 2010; van Doorn et al. 2010). These behaviors—which may not have an immediate impact on company revenues—conceptually increase CLV by generating future profitability for the firm (Kumar et al. 2010; Verhoef and Lemon 2013) (e.g., an increased number of acquired customers, a better brand image, a more loyal and less price-sensitive customer base, etc.). Drawing upon relationship marketing theory and the principle of reciprocity, we propose a conceptual framework (Fig. 1) designed to analyze and explain how customers, on perceiving investment in satisfaction and relationship consolidation, respond with a greater degree of customer engagement.

Fig. 1
figure 1

Causal model

Full size image

3.1 Antecedents of complaint-handling satisfaction: perceived effort and perceived justice

Based on research by Huang (2008), perceived effort has been defined as customer perception of the resources a company invests—and the degree of interest shown—when providing solutions to service-related problems. When consumers feel that a company has made a concerted effort to resolve their problems, this can have a positive impact on the level of customer satisfaction. The literature suggests that complaint-handling efforts strengthen company-customer interactions and relationships (Maxham and Netemeyer 2002). Furthermore, company personnel clearly play a key role in conflict resolution (De Matos et al. 2007; Huang 2008). Authors such as Baker and Meyer (2014), Robinson et al. (2011), and McColl-Kennedy and Sparks (2003) find a positive link between perceived employee effort and complaint-handling satisfaction. Hence, we can formulate the following hypothesis:

H1 1

The greater the perceived effort by the customer regarding the company’s complaint handling, the greater the level of the customer’s complaint-handling satisfaction.

The second antecedent of complaint-handling satisfaction is perceived justice. The correlation between the two has been amply analyzed in the literature to date (Gelbrich and Roschk 2011; Maxham and Netemeyer 2002). Customers tend to make evaluative judgments about how a company manages their complaints. They value the justice received throughout the complaint-handling process and the fairness of the proposed solution. Whether customers are satisfied or dissatisfied is often a function of whether they feel they were treated fairly or unfairly (DeWitt et al. 2008; Hocutt et al. 2006). If we bear in mind that customers generally evaluate the solutions firms propose, there is no doubt that perceived justice is a key variable (Sparks and McColl-Kennedy 2001).

Authors who have studied perceived justice point out the need to break the variable down into more specific units for measurement purposes. The most widely accepted breakdown is distributive, procedural, and interactive justice (Smith et al. 1999). Distributive justice refers to the compensation customers receive as a result of the complaint-handling process; procedural justice assesses the justice in the complaint-handling process itself; finally, interactive justice is the desire of the staff to solve the problem, as expressed in their treatment of customers throughout the complaint-handling process (Chang and Chang 2010). Therefore, in line with the literature, we can formulate the following hypothesis:

H 2

The greater the degree of perceived justice by the customer regarding the company’s complaint handling, the greater the level of the customer’s complaint-handling satisfaction.

3.2 Impact of customers’ complaint-handling satisfaction: customer engagement

One of the underlying principles of relationship marketing is the importance of building and nurturing long-term company-customer relationships (e.g., Cambra-Fierro and Polo-Redondo 2011). The satisfaction-engagement correlation, while highly relevant for marketing theory and practice, has yet to be measured in complaint-handling contexts. The present study examines potentially deeper links between complaint-handling satisfaction and customer engagement. To that end, we draw on recent conceptual work on customer engagement to propose that customer satisfaction with complaint-handling processes may be a key driver of customer engagement.

This relation is especially important in the light of existing literature that discovers links between complaint-handling satisfaction and relational constructs such as loyalty (e.g., Chang and Chang 2010), word-of-mouth (e.g., Villanueva et al. 2008), and commitment (e.g., Won-Moo 2010). Remarkably few studies, however, have looked at customer engagement as a global concept driven by complaint-handling satisfaction. Higgins and Scholer (2009) analyze customer satisfaction as a key antecedent of engagement, considering it to be a requisite if a high degree of satisfaction is to be sustained over time. However, it was van Doorn et al. (2010) who confirmed this line of argumentation. These authors understand customer engagement as being clearly satisfaction-driven. In other words, customer satisfaction is a prerequisite for engagement. In the context of our research, this means that customers will require proper complaint-handling in order to be satisfied. Through the principle of reciprocity (De Wulf et al. 2001), complaint-handling satisfaction leads to gratitude, which may be expressed in a high degree of customer engagement.

Given that engagement has not been studied as an individual construct in complaint-handling contexts, we propose that

H 3

The greater the level of the customer’s complaint-handling satisfaction, the greater the degree of customer engagement.

4 Empirical study

To carry out this study, we designed a survey based on a set scales grounded in the specialized literature, with a view to measuring the following variables: perceived effort, perceived justice, customer satisfaction with complaint handling, and customer engagement. It was deemed necessary to adapt the scales in order to fine-tune the final survey content and structure (Appendices 1, 2).

The perceived effort variable is a unidimensional construct that was measured using three reflective indicators from the scale proposed by Karatepe (2006). Following the nomenclature proposed by Edwars (2001) and Polites et al. (2012), perceived justice is a superordinate multidimensional construct design (reflective first-order and reflective second-order) which encompasses three reflective first-order dimensions, in line with DeWitt et al. (2008): distributive justice, procedural justice, and interactive justice. The modeling of complaint-handling satisfaction, based on the proposal of Karatepe (2006), is a unidimensional construct that involves three reflective indicators.

Customer engagement is a superior concept that involves a set of relational behaviors (e.g., WOM and co-creation) under a common variance in such a way that they collectively represent a set of non-transactional behaviors. Customer engagement is therefore considered a superordinate multidimensional construct design (reflective second-order) which considers ideas proposed by Sprott et al. (2009) and van Doorn et al. (2010) and encompasses two reflective first-order dimensions: recommendations and co-creation.

After designing the survey, a pre-test was carried out with financial services users (marketing students and researchers from several university marketing departments) in order to check the comprehensibility and adequacy of all the indicators included. The survey was adapted for use in Spain through an iterative process of translation and back translation by a team of bilingual people to assure equivalency in meaning (Brislin et al. 1973). The version was pre-tested for validation purposes (Douglas and Craig 2007). Appendix 1 provides the descriptive statistics of the scales.

Recent research has recognized the inherent complexity of analyzing complaint-handling processes due to the fact that only a small percentage of customers actually file a formal complaint; hence, the difficulty is in finding a large sample population of customers who have filed a complaint and received a response from the company (Michel and Meuter 2008). To meet this challenge, we opted to hire a specialized information-gathering service. Members of the sample population are adult financial services users who had experienced a problem with their carrier, filed a complaint, and received an answer from the company. The geographical scope is Spain. The survey is personal, employing a quota sampling procedure depending on customer age and gender. All the fieldwork was carried out between October and December, 2013. The final sample size is 316 respondents (52 % male; 18–25 years old: 30 %, 26–50 years old 45 %, >50 years old 25 %). The preliminary analyses indicate that 51.2 % of our customers (162 customers) stated that the failure was serious and 48.7 % experienced a mild failure (154 customers).

A partial least squares (PLS) structural equations analysis was carried out using SmartPLS software (version 2.0.M3) in order to test the hypotheses. This methodology has recently been defended and employed in the literature (Barroso et al. 2009; Roldán and Sánchez-Franco 2012; Henseler et al. 2014).

A reliability analysis for each item in relation to its construct was carried out in order to assess data quality. Our results demonstrate that all values overcome the 0.707 threshold set by Carmines and Zeller (1979). Reliability was also tested for each of the variables using Composite Reliability—considered superior to Cronbach’s Alpha. All constructs were reliable given that they are above the 0.8 benchmark (Nunnally 1978) (Appendix 2). A convergent validity analysis was carried out using the average variance extracted (AVE) (Fornell and Larcker 1981). The fact that the results were above the 0.5 benchmark shows that more than 50 % of the variable is expressed through its indicators (Appendix 2). Afterward, results of discriminant validity were adequate via an AVE comparison of each construct (main diagonal) and the correlations between the variables. We observe that the square root of the AVE is higher than the correlations between constructs in each case (Fornell and Larcker 1981) (Appendix 3).

5 Findings

5.1 Structural model

In this section, we will analyze the relationships between the constructs in the proposed model. We will analyze whether perceived effort and perceived justice condition the degree of complaint-handling satisfaction and whether the latter, in turn, determines the degree of customer engagement.

Table 1 shows the significance of the structural paths: all the three hypotheses have been explored and path coefficient values are over 0.2 (Chin 1998).

Table 1 Structural model results
Full size table

Firstly, if we analyze the antecedents of customer satisfaction in a complaint-handling context, hypothesis H1 is supported (β = 0.4374; p < 0.001). Our model also establishes a direct relationship between perceived justice and customer satisfaction (H2). Parameter values are both positive and significant for the total sample (β = 0.4166; p < 0.001), which supports our second hypothesis.

As far as the consequences of complaint-handling satisfaction are concerned, H3 (β = 0.7058, p < 0.001) allows us to conclude that higher satisfaction levels lead to a greater degree of customer engagement. This result provides us with a better understanding of the special bond that is created between companies and customers following successful complaint-handling processes.

To evaluate the predictive relevance of the model, we used the Stone-Geisser test. The Q 2 values of this test for complaint-handling satisfaction and customer engagement were positive (Q 2−CHS = 0.447; Q 2−CEB = 0.4157). Thus, it can be assumed that the dependent variables can be predicted by the independent variables. Additionally, we calculated the goodness of fit proposed by Tenenhaus et al. (2005) and obtained a value of 0.6529, which Cohen (1988) considers high.

5.2 Post-hoc assessment of the mediating effect

We have also examined whether customer satisfaction acts as a mediator in the model. Following the recommendations of authors such as Real et al. (2014), among others, a new causal model must be analyzed. Two direct links from perceived effort and perceived justice to customer engagement were drawn (see broken arrows in Fig. 1).

To get information about the significance of the indirect effects, we applied a bootstrapping method (Preacher and Hayes 2004, 2008). Figure 2 describes the total effects of perceived effort (d) and perceived justice (e) on customer engagement. These total effects may be arrived at via a variety of direct and indirect forces (Hayes 2009).

Fig. 2
figure 2

Structure of relationships for exploring the mediating effect. Model with total effects

Full size image

Specifically, in Fig. 3, the total effect of perceived effort on customer engagement can be expressed as the sum of the direct (d) and indirect (a × c) effects: d = d′ + a × c (Taylor et al. 2008). This approach has the advantage of being able to isolate the indirect effect (a × c). The relationships d and e, although non-hypothesized, are included in our analysis in order to check the presence of either full or partial mediation (Baron and Kenny, 1986). The same procedure is applicable to the total effect of perceived justice on customer engagement, e = e′ + b × c, where b × c is the indirect effect.

Fig. 3
figure 3

Structure of relationships for exploring the mediating effect. Model with the mediating effect

Full size image

We followed a bootstrapping process, i.e., a non-parametric resampling procedure that does not impose the assumption of normality on the sampling distribution. MacKinnon et al. (2004) and Real et al. (2014) defend that the performance of bootstrapping is better than the traditional Sobel test. Using the latent variables scores obtained from the PLS analysis, we have applied the SPSS routine developed by Preacher and Hayes (2004) to calculate the total, direct, and indirect effects, as well as the 95 % confidence interval (CI) for the mediator variable. Next, we have applied the method of percentiles based on the work of Williams and MacKinnon (2008) which calculates the mediating effect of a variable on a proposed relationship in the conceptual model using a bootstrapping process with 5000 subsamples. When an interval for a mediating effect does not contain zero, the indirect effect is significantly different from zero with a 95 % confidence level. As Table 2 shows, in the confidence intervals obtained, the value zero is not contained in the paths PERCEIVED EFFORT → COMPLAINT-HANDLING SATISFACTION → ENGAGEMENT and PERCEIVED JUSTICE → COMPLAINT-HANDLING SATISFACTION → ENGAGEMENT, so we see that the indirect effect is statistically significant.

Table 2 Path coefficients and indirect effects for the mediation model
Full size table

When complaint-handling satisfaction is introduced as a mediator, perceived effort reduces its direct effect on customer engagement, although it remains significant (d′ = 0.1673; t = 2.5415), whereas its indirect effect via complaint-handling satisfaction achieves a point estimate of 0.1894 (a × c). This means that complaint-handling satisfaction partially mediates the influence of perceived effort on customer engagement. The same process and same arguments indicate that complaint-handling satisfaction also partially mediates the influence of perceived justice on customer engagement. Although these direct relationships were not hypothesized, empirical data suggest significant relationships between the antecedents of complaint-handling satisfaction (perceived effort and perceived justice) and customer engagement. These interesting, but for the moment, exploratory, results constitute an interesting line for future research.

6 Discussion and implications for management

In accordance with the fundamental ideas of relationship marketing the premise of this research is that, if customers perceive that a company has invested resources in the complaint-handling process, it will be a key driver of satisfaction. This, in turn, will make customers more engaged as a means to reciprocate the investments made by the firm.

Our central hypothesis (H1, H2, and H3) were confirmed. This reveals the relevance of perceived effort and perceived justice as antecedents of complaint-handling satisfaction and demonstrates the impact of satisfaction on customer engagement. This finding is of interest because this research is the first empirical attempt to investigate customer engagement after complaint-handling processes, which was conceptually, but not empirically, proposed by authors such as Bijmolt et al. (2010) or Kumar et al. (2010).

Regarding the antecedents of complaint-handling satisfaction, earlier studies suggested that perceived effort is reflected in higher levels of customer satisfaction (De Matos 2007; Huang 2008). Our data are in line with former studies which suggested the relevance of factors such as money, time, and effort in solving the problem (Baker and Meyer 2014; Robinson et al. 2011). Otherwise, satisfaction may not be restored and extra problems could arise (Mattila and Patterson 2004). With respect to the perceived justice variable, we find that it has a direct impact on customer satisfaction, which is similar to ideas proposed by authors such as Gelbrich and Roschk (2011) or Karatepe (2006); customers will be happier in contexts where they perceive justice in the complaint-handling process (Chang and Chang 2010; DeWitt et al. 2008; Tax et al. 1998). Since customer satisfaction is a prime objective, administrators and personnel should strive to (i) manage complaints effectively and (ii) know what consumers expect. The objective here is that customers perceive justice in the solution provided.

With respect to the consequences of customer satisfaction with the complaint-handling process, hypothesis H3 was confirmed: if customers are satisfied with the complaint handling, they will be more likely to respond to the investment made by committing to a high degree of engagement. This is in line with the basic tenets of the principle of reciprocity proposed by authors such as Bagozzi (1995), DeWulf et al. (2001), and Palmatier et al. (2009) which allow us to propose that customer satisfaction fosters customer engagement. However, this research represents the first attempt to apply these ideas to complaint-handling contexts. This is an interesting contribution to the literature because there was a lack of research on this relationship in complaint-handling contexts. The fact that this relationship has been demonstrated to a high degree of significance provides evidence of a positive correlation between complaint-handling satisfaction and customer engagement. It is essential, then, for firms to know how to manage complaints and company-customer interactions effectively; if they succeed, they can count on a portfolio of engaged customers among their most prized assets. Engaged customers will attract new customers via positive recommendations and will be more disposed to co-create with the firm. As the literature on Relationship Marketing shows, a strong company-customer bond is forged, providing long-term benefits for both parties.

The structure of the model suggests a potential mediating effect of satisfaction with the complaint-handling process. As a result, although they have not been hypothesized, we examine the direct relationships between perceived effort-customer engagement and perceived justice-customer engagement. Results indicate that these direct links are positive and significant. Perceived effort seems to affect customer engagement directly, while perceived justice leads customers to become more engaged—regardless of whether they are satisfied or not. These ideas are relevant for both theory and practice because they suggest that customers can become engaged whether their problem is solved or not, which seems to contradict the assumptions of van Doorn et al. (2010) that customer engagement is an only satisfaction-drive construct. In other words, perceived effort and perceived justice could be sufficient for companies to engage their customers, even if the solution that the company gives after the customer complaint is not satisfactory. However, this result must be interpreted with caution because we did not find strong theoretical arguments in former studies to support it. This part of the research could be considered exploratory in nature and provides an important line for future research.

A final aspect worth noting is that, despite the initial investment, complaint-handling processes can actually be extremely profitable. Currently, the industry is suffering a restructuration process, forcing Spanish financial firms to invest in marketing in the hope of capturing and retaining customers. By employing complaint management strategies, among other tactics, firms can find themselves in a win–win situation. Firms can get feedback that provides a better understanding of customer needs and expectations, leading to enhanced service delivery quality. Existing customers may wish to remain in the relationship and new customers can be won over from other firms—which explains the inclusion of engagement in the model. If the company’s customers are themselves voluntarily attracting new customers, the cost of building its customer portfolio will be dramatically lower. In short, successful complaint-handling should be a top priority for firms today as it strengthens the company-customer bond and results in more engaged customers.

7 Conclusions

This study demonstrates the importance of perceived effort and perceived justice as antecedents of complaint-handling satisfaction among customers. Our findings show that high satisfaction levels in complaint-handling contexts make for more engaged customers. A service failure episode is enough to lead customers to reconsider the relationship and demand compensation. For this reason, companies must strive to avoid service failures. If failure does occur, the company should resolve it to the customer’s satisfaction. When managed properly and effectively, complaint-handling investments pave the way for customers to reciprocate by becoming engaged. Thus, complaint-handling investments bring long-term advantages for the company in terms of stronger, longer lasting, and more profitable customer relationships.

This research is of interest because relationship marketing is still in a process of development (Bonnemaizon et al. 2007) and customer engagement has recently been considered an emerging topic in the relationship marketing and customer management literature (e.g., van Doorn et al. 2010; Verhoef and Lemon 2013). This paper has presented, both conceptually and empirically, the interrelation between complaint-handling satisfaction and customer engagement and is the first attempt to analyze these interrelationships.

However, our study is not without its limitations. Firstly, the results are based on the financial services sector, so we must be cautious when extrapolating our findings to other industries. Secondly, we have employed questionnaires to survey consumer opinions and perceptions at a given point in time, but following recommendations made by Podsakoff et al. (2003), potential biases have been checked. The common method bias was tested by conducting Harman’s single factor test (Harman 1967). Using a factor analysis, no single factor that explains variance across all the items is identified. The main factor explains 46.65 % of the variance. Because no single factor is found to explain more than 50 % of the variance, the study’s data can be accepted as valid (Podsakoff and Organ 1986). Following Armstrong and Overton (1977), we also compared early and late respondents and we did not find any significant difference. Non-random sample biases were also checked. Both measurement scales and the causal model were explored in a smaller sample of a different industry (mobile phone users), obtaining similar results. Therefore, our research does not present non-random sample biases.

With regard to potential lines for future research, it would be interesting to investigate whether the impact of complaint handling on satisfaction and engagement varies with the relationship life-cycle (Palmatier et al. 2013). As customers evolve through different phases in the relationship with the company (exploration, expansion, maturity, and decline), different aspects are more/less important in their decisions, and the effectiveness of different complaint-handling strategies in improving customer engagement behaviors (such as co-creation or word-of-mouth) can differ significantly. Research that addresses these challenges is thus needed. It would also be interesting to consider consumer profile variables—e.g., age, gender, income, education bracket, employment status, etc.—as moderating the structural model would be especially relevant. Authors like Verhoef et al. (2010) and Verhoef and Lemon (2013) note that demographic factors could have a lot to contribute to the study of customer engagement and the findings could be invaluable for management. A longitudinal analysis of the entire complaint-handling process—from service failure to solution—would be of interest. Lastly, a study replicating our research with a larger sample population and in different contexts would be valuable in terms of extrapolating our findings to other arenas.