Life after death? Analyzing post-defection consumer brand equity

Journal of Business Research 63 (2010) 1135–1141

Contents lists available at ScienceDirect

Journal of Business Research

Life after death? Analyzing post-defection consumer brand equity Svetlana Bogomolova ⁎ Ehrenberg-Bass Institute for Marketing Science, University of South Australia, GPO Box 2471, Adelaide SA 5000, Australia

a r t i c l e

i n f o

Article history: Received 1 March 2009 Received in revised form 1 June 2009 Accepted 1 September 2009 Keywords: Reasons for defection Lapsed customers' winback Customer brand equity

a b s t r a c t The industry literature is full of the idea of winning back lapsed customers. Yet marketing practitioners and academics know very little about what happens to customers after they stop buying the brand. This research investigates the brand equity of lapsed customers of five major financial institutions. The analysis compares the propensity for positive and/or negative brand associations, overall brand evaluation, and the propensity to consider the brand in the future across the main segments of lapsed customers. The results show that the group of lapsed customers is not homogenous, but consists of distinct segments. Customers who defected for different reasons also differ in their post-defection brand equity. The paper concludes with implications for winback strategies for lapsed customers and brand equity measurement and management. © 2009 Elsevier Inc. All rights reserved.

1. Introduction Marketing managers have a consistent goal of winning back lapsed customers (Griffin and Lowenstein, 2001). Industry consultants claim that lapsed customers, that is, customers who previously bought the brand but have defected away, are a fruitful source of customer acquisition (Reichheld, 1996; Reichheld and Sasser, 1990). The key rationale for this view is the belief that if brand management focuses on regaining lapsed customers, rather than attracting new ones, the return on investment will be higher. To date, however, very little empirical research on lapsed customers has taken place. (exceptions are Homburg et al., 2007 and Tokman et al., 2007) Whether researchers can treat lapsed customers as a homogenous group, or whether distinctive segments of lapsed customers exist, with varying propensities for future winback, is unclear. Consumers' past experiences influence their behavior and the way they perceive and interpret new experiences (Fazio and Zanna, 1981). Therefore, one way to assess the potential winback of lapsed customers is to examine their thoughts and feelings with regard to their former brand. Such cognitive information about a brand is an aspect of customer brand equity (Keller, 1993, 2003; Keller and Lehmann, 2006). The reasons for brand defection are a focus of many past studies (Keaveney, 1995). These studies show that customers defect from brands for a variety of reasons. Consumers might attribute some of these reasons to poor brand performance, for example, service failures or high fees. Other reasons might only partially relate to brand management's actions (e.g. when competitors attract customers), or not relate at all (e.g. when customers go out of business, leave the country, or change suppliers following a Head Office decision). The

⁎ Tel.: +61 8 83029170; fax: +61 8 83020442. E-mail address: [email protected]. 0148-2963/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.jbusres.2009.10.009

variability in the reasons for defection suggests that lapsed customers might have different judgments of their experience with their former brands. This difference might affect the lapsed customers' propensity to use their former brands in the future. This paper investigates three research questions: (1) what are the reasons for brand defection; (2) are lapsed customers more likely to retain associations about former brand that correspond with their reasons for defection; and (3) do lapsed customers who defect for different reasons differ in their post-defection customer brand equity. In particular, the last question focuses on the difference between customers whose reasons for defection relate to failures of the brand they switched from (e.g. service and price reasons) compared to customers whose reasons relate to the benefits of the brands they switch to (e.g. receipt of a better deal from a competitor). The research compares positive and negative brand associations, overall brand evaluations and the propensity to consider the former brand for future purchase as indicators of lapsed customers' brand equity after defection. Investigation of these research questions provides important implications for segmenting (or not) lapsed customers for future winback strategies. This paper uses a business-to-business (B2B) financial services industry context. Research into customer brand equity in B2B markets is important because people who rely on their own personal memories, associations and evaluations to make decisions about brands still make business decisions (Blombäck and Axelsson, 2007; Webster and Keller, 2004). Furthermore, effective brand management is crucial for services markets (e.g. financial services). Effective brand management instils trust and confidence in consumers when no physical differences in the product guide their choices (Berry, 2000). This paper consists of two parts, both of which use the same data set. The first part explores the reasons why customers defect from brands, and quantifies the proportion of lapsed customers stating each reason. This part provides discussion of the relevant literature,

1136

S. Bogomolova / Journal of Business Research 63 (2010) 1135–1141

followed by a brief description of the data set and a descriptive analysis of the reasons for defection. The second part examines lapsed customers post-defection customer brand equity in the form of positive and negative associations, overall evaluations and brand consideration. This part presents theoretical arguments that lead to hypotheses, research, and discussion of the results. The paper concludes with the discussion of managerial implications, the limitations of the study and implications for future research. 2. Part 1. Reasons for brand defection Lapsed customers are those who used to buy a brand but have stopped doing so. Various authors refer to this group of (non) customers as lost customers (Hogan et al., 2003; Stauss and Friege, 1999; Thomas et al., 2004), switchers (Hogan et al., 2003), former users (Bird and Channon, 1970; Romaniuk, 2001), or defectors (Thomas et al., 2004). Exploring the reasons for brand defection greatly benefits brand managers highlighting aspects of a product or service that might potentially cause further customer loss if managers do not rectify the problem (Griffin and Lowenstein, 2001; Reichheld, 1996). The customer defection literature uses three main categories of reasons for defection. The first category highlights the negative qualities of the brand they switch from. Various authors refer to this category of reasons for defection as push-away (Bansal et al., 2005; Stauss and Friege, 1999) or expectation disconfirmation (Lees et al., 2007). Common examples are service-related issues (poor core service, service encounter failure, poor mistake recovery, unethical or illegal behavior of staff members, etc.); price-related issues (dissatisfaction with fees and charges, interest rates, fees for termination, etc. of the current brand); product/service-related issues (poor quality of offering, limited amount or range of services/options); inconvenience (poor location of branches and hours of operation); or flexibility (inability of providers to meet specific requirements) (Colgate and Hedge, 2001; Gerrard and Cunningham, 2004; Keaveney, 1995; Lees et al., 2007). The second category includes reasons that prompt people to switch to competitors, highlighting the positive qualities of other brands, rather than negative information about the former brand. Prior literature calls this category pull-away (Bansal et al., 2005; Stauss and Friege, 1999) or utility maximization reasons (Lees et al., 2007). These reasons include receipt of a better deal in the form of monetary gain (lower fees, better interest rates, and price discounts) or non-monetary incentives (additional services, bundling products, better quality, etc.) (Colgate and Hedge, 2001; Gerrard and Cunningham, 2004; Keaveney, 1995; Lees et al., 2007). The final category contains reasons that usually do not relate to former or new brands, but are largely beyond the control of any brand management. Labels for this category include move-away (Stauss and Friege, 1999), stochastic (Lees et al., 2007) or reasons beyond the control of brand management (Bogomolova and Romaniuk, 2009). Examples include changes in personal life (moved away, got married, divorced, or died) or professional circumstances (sold business, joined a franchise network, or Head Office directed change of suppliers) (Bogomolova and Romaniuk, 2009; Gerrard and Cunningham, 2004; Keaveney, 1995; Lees et al., 2007). The first part of this paper examines the incidence rate of various reasons for brand defection in the business-to-business financial services context, with particular focus on categories that describe reasons to switch from the former brand and switch to new brands. 3. The data Researchers randomly select potential respondents using the electronic telephone directory (business section). The businesses in the sample represent a wide range of industries, including retail, hospitality, building, agriculture, electrical, consultancy, transport, church, real estate, and medical practitioners. More than 75% of the

sample has been in business for more than 10 years. The respondents' business turnover varies evenly from less than $100 K up to $20 million, however most are small businesses with three to five employees. This profile of business customers is representative of typical business customers in this market. Trained market research interviewers conduct 213 telephone interviews with the owners of the businesses or managers responsible for financial matters in their companies who have stopped using a brand of financial provider sometime in the past. In most cases, the respondent is the sole decision maker. This type of simple buying center (Johnston and Lewin, 1996) is typical for small businesses that have informal decision-making process based on the judgment, experience, associations and evaluation of the sole decision maker. Interviewers ask respondents which brands they stopped using in the past for business banking and what were the main reasons. Interviewers then record and classify responses into one of the precoded categories. The full list of categories includes more than 15 options. Respondents could provide more than one response; 29% did so. The time elapsed since defection is as follows: 8% defected in the past 12 months, 30% more than a year, but less than five years, 59% more than five years ago, and 3% could not remember. Because many contracts have terms of several years, the opportunity to change provider (or ‘come back’ to a former brand) might occur a number of years after the initial defection. Therefore, from this perspective, and bearing in mind the relatively low churn rate in financial services (about 5% per annum), the current distribution of time elapsed since defection is representative of all lapsed customers in the market. While such a long time lapse allows for memory loss and opinion modification (as a result of new experiences), lapsed customers use these modified perceptions when estimating their likelihood of coming back to a former brand (not the one they had right after the defection). The analysis includes responses about the five brands with the largest market share (from 8% to 20% of the market). Only 8% of lapsed customers indicate that they have switched from more than one financial institution. These respondents evaluate the brand they most recently defected from. Due to restricted sample sizes for each of the five individual brands, the analysis uses aggregate responses from all lapsed customers in regard to their respective former brands. Individual brand-level analysis shows consistent results, but lacks statistical power due to the small sample sizes. 4. Results — reasons for brand defection The first, most popular category cites service issues as the key reason for brand defection (36% of responses). This category includes reasons such as core service failure, poor mistake recovery, and unethical behavior by service provider employees. A further 28% of responses claim that receipt of a better offer from competitors prompted them to switch brands. Another 24% give pricing issues as the main reason for defection. A follow-up open-ended question (What was it about the [insert reason] that prompted you to switch?) verifies the validity of switch from vs. switch to classifications between price and competition reasons. Finally, 18% of responses quote reasons that are beyond brand manager's control, such as lack of need for the product or service; customers moving or selling a business; involuntary or corporate decision to change providers; and other personal reasons. As these reasons present no opportunities for brand managers to address customer loss, further analysis does not include this category of reasons (Table 1). Other reasons, which attract even lower number of responses, are inconvenience (11%) and inflexibility (9%, mainly related to the inability of brand managers to adapt the product and service to the needs of small/medium businesses). Further analysis does not include these groups due to their restricted sample sizes. In line with prior literature (Colgate and Hedge, 2001; Gerrard and Cunningham, 2004; Keaveney, 1995; Lees et al., 2007), these results

S. Bogomolova / Journal of Business Research 63 (2010) 1135–1141

1137

Table 1 Distribution of responses about the reasons for brand defection. Reasons for defection

N

Service (switch from):

% of cases

% of responses

76

28

36

59

21

28

51

18

24

38

14

18

23

8

11

19 10 276

7 4 100

Service failure, poor customer service, poor mistake recovery, ethical, immoral, illegal actions

Competition (switch to): Getting a better deal from competition either in monetary or non-monetary deals

Price (switch from): Dissatisfaction with fees, charges, interest rates

Beyond brand's control: Corporate or group decision, lack of need for the product, business sold or moved, other personal reasons

Inconvenience: Inconvenient location, working hours

Inflexibility Don't know/refused Total a

9 3 129a

Multiple responses were possible.

show that the most common reasons for defection, among those within brand manager's control, relate to service, price and competition. The service and price reasons represent the switch from category, while the competition segment represents the switch to category. Customers who defect due to these reasons represent the majority of customer loss (more than 65% of those who still use the product category and have the ability to start using their former brand again, as opposed to those who have left the category altogether i.e. the beyond control category). Therefore, further analysis focuses on the most popular and managerially actionable groups of reasons — service, price and competition. 5. Part 2. Post-defection brand equity Consumer brand equity is a term describing a brand's assets in the minds of consumers that have a differential impact on their future brand response (Keller, 1993, 2003; Keller and Lehmann, 2006). Two important elements are: (1) Examination of cognitive information that resides in consumers' memories which allows estimating a brand's equity (Keller, 1993); and (2) this equity's ability to impact future behavior in regard to this brand (Agarwal and Rao, 1996). Therefore, examination of lapsed customers' cognitive responses with regard to their former brand can help in estimating the former brand's equity. Brand equity can indicate lapsed customers' winback potential and possible heterogeneity within the group. This section of the paper examines whether lapsed customers who defect for different reasons (service, price and competition-related) also differ in post-defection customer brand equity in their propensity to give positive and negative brand associations, overall brand evaluations, and brand consideration. Brand experience has a significant impact on consumer cognition and likelihood of future behavior (e.g., Fazio and Zanna, 1981). Therefore, different experiences with the former brand could prompt the formation of different brand associations or evaluations that lapsed customers retain in their memories. Brand associations are links in consumer memory between a brand name and various pieces of information including brand attributes (Holden and Lutz, 1992; Keller, 1993). This view of how memory works draws on the Associate Network Theory of Memory (Anderson and Bower, 1979). Lapsed customers are likely to retain both positive and negative associations in regard to their former brands (Winchester et al., 2008). Customers' first-hand experience prompts formation of a range of positive brand associations (for example, what the brand does and stands for, and its basic functional and consumption characteristics). On the other hand, poor brand performance might prompt development of negative brand associations. Alternatively, negative associations could form after the defection to justify the decision to stop using the brand, as per cognitive dissonance reduction (Festinger, 1957). Earlier work in fast-moving consumer

goods (Bird et al., 1970) and more recent replications in other areas, including subscription markets (Winchester and Romaniuk, 2003; Winchester et al., 2008), provide support for both these assumptions. Yet, no research examined the correspondence between specific reasons for defection and positive and negative associations that lapsed customers retain about their former brands. Another element in the customer brand equity model is the overall evaluation of the brand (Solomon, 1992; Wilkie, 1986). This evaluation is partly a function of salient brand attribute associations (Fishbein and Ajzen, 1975) and an affective or emotional component (Eagly et al., 1999). Consumers store overall evaluations and brand associations separately in their memory and their retrieval is also independent (e.g., Lichtenstein and Srull, 1985; Lynch et al., 1988). Since brand evaluation forms under the influence of past experience, various experiences that lead to defection and consumer's judgements could prompt the difference in the post-defection evaluation of the former brand. Support for this proposition comes from a study by Richins (1983), who reports that consumers' beliefs about how much a brand is responsible for unsatisfactory experience moderate brand evaluations. Further support comes from the study in B2B electronic funds transfer services. The authors found differences in the lost customers' evaluations of former providers depending on whether the reasons for defection were within or beyond the service provider's control (Bogomolova and Romaniuk, 2009). The final brand equity element under examination in this paper is brand consideration. Brand consideration is a process by which consumers narrow down their options to a subset of brands to simplify their decision-making process (Howard and Sheth, 1969). The inclusion of a brand in a consideration set is a necessary step towards purchase. This inclusion depends on the customer's ability to retrieve the brand from memory. This process of memory retrieval relates to the amount and valence of brand-related information that consumers retain in their memory (Nedungadi, 1990; Ratneshwar and Shocker, 1991). The higher the amount of positive information, the greater the propensity to consider the brand for purchase (Romaniuk and Sharp, 2004). On the other hand, consumers might reject brands after retrieval if their evaluation of the brand is negative (Laroche et al., 1986; Narayana and Markin, 1975; Urban et al., 1996). Hence, brand consideration would correspond with both propensity to retrieve positive and negative information about the former brand and the overall evaluation of the former brand. If associations and evaluations vary with the reasons for defection, so should brand consideration. Prior literature shows support for this notion. A study in consumer banking (Lees et al., 2007) reports differences in the propensity to consider former brands in consumer financial services, depending on the reason for defection. A study by Bogomolova and Romaniuk (2009) also shows that variations in lapsed customers' propensity to consider the former brand relate to whether the reasons for defection were within or outside the service provider's control.

1138

S. Bogomolova / Journal of Business Research 63 (2010) 1135–1141

The current study contributes to the literature by replicating prior findings in a new B2B financial services context and comparing different groups of lapsed customers — groups based on service and price, which represent the switch from group of reasons, and competition, representing switch to reasons. 6. Hypotheses development To develop hypotheses about the cognitive processes that occur in the minds of lapsed customers, this research draws on the psychology literature, specifically on two theories: the Theory of Cognitive Dissonance (Festinger, 1957) and Attribution Theory (Weiner, 2000). The next section provides rationales for the application of the above theories to the development of the hypotheses. 7. Correspondence between reasons for defection and respective brand associations The relevance of Cognitive Dissonance Theory to consumer behavior research is widely documented (Cummings and Venkatesan, 1976). The theory states that people have strong motivation to adjust their beliefs in line with their past and present behavior, in order to avoid emotional discomfort. Applying this theory to current research, lapsed customers would have a strong motivation to justify their act of abandoning the brand. This justification could come from adjusting their beliefs about the former brand to reflect the reason for their defection. For example, those who defect because of dissatisfaction with price would no longer believe that the brand has appropriate fees and charges. Hence, the brand associations that lapsed customers retain about their former brands would correspond with the reasons for defection. The valence of brand associations can be either positive (emphasizing desirable attributes for the product category) or negative (highlighting undesirable attributes) (Krishnan, 1996). Therefore, the reflection of the reason for defection in brand associations would be in the opposite direction: H1. Lapsed customers will be less likely to link their former brand to positive associations that correspond with their reasons for defection, compared to customers lapsed for other reasons (i.e. less likely to say the brand has appropriate fees and charges if the reason for defection is price-related). H2. Lapsed customers will be more likely to link their former brand to negative associations that correspond with their reasons for defection, compared to customers lapsed for other reasons (i.e. more likely to say the brand has high fees if the reason for defection is price-related). 8. Reasons for defection and post-defection brand equity responses Another theory this research draws on is Attribution Theory (Weiner, 2000). This theory explains how consumers perceive and make judgments about past events. The theory suggests that unusual or extraordinary events prompt consumers to search for the cause of such events. During this search consumers make judgments about whether such events are likely to occur again, who to blame for it, and whether this person/organization had the ability to control the event (Weiner, 2000). These attribution judgments then influence cognitive and affective responses concerning the event and the person/ organization responsible, which in turn influence future behavior. Current research uses the ideas of Attribution Theory to hypothesize about the cognitive process that lapsed customers go through to attribute the cause of their defection. The amount of blame for their defection that lapsed customers attribute to the former brand's management should influence how lapsed customers feel and think about their former brand. These post-defection thoughts, evaluations and considerations are aspects of customer brand equity.

Consider the three most populated groups of lapsed customers — those whose defection relates to service, price and competition reasons. The reasons for defection in the first two groups (service and price) belong to the switch from category and mainly relate to the former brand's (poor) performance. Hence, the formation of negative evaluations, feelings and thoughts should follow an unsatisfactory brand experience. On the other hand, attraction by the competition, which is the reason for defection in the switch to category, mainly relates to external factors (other brands), rather than the former brand per se. This means that evaluations and associations with the former brand should be more positive than in the case of the first two groups, where the former brand is the primary cause for defection. A study by Bansal et al. (2005) shows the applicability of the switch from/switch to paradigm to customer defection. Their study shows that customers who have different propensities to claim switch from or switch to reasons for considering changing brands differ in their intentions to switch from the brand. Current research takes the next step by examining switch from and switch to reasons for defection in the context of real defectors who have already switched brands. A further contribution of this research is in comparing post-defection brand equity responses between lapsed customers who claimed switch from (service and price) reasons and switch to (competition) reasons. This research can indicate the likelihood of lapsed customers in each of these groups coming back to their former brand. Following the discussion above, the research expectations are that: H3. Lapsed customers whose defection relates to competition (switch to) will have more positive evaluation of the former brand by comparison to lapsed customers whose defection is service and pricerelated (switch from). H4. Lapsed customers whose defection relates to competition (switch to) will have a higher propensity to consider the former brand again in comparison to lapsed customers whose defection is service and pricerelated (switch from).

9. Method This section uses the same data set as Part 1, comprising a random sample of 213 telephone interviews completed by business customers in regard to five major financial institutions. To address the hypotheses, the research uses analysis of variance in a series of Chi-square and ANOVA tests. The independent variable in this analysis is a reason for defection. Researchers classify responses into mutually exclusive segments based on the reason for defection. As the data collection method allows for multiple reasons, and 29% of participants give more than one reason, analysis takes response, rather than respondent, level. This analysis compares the three most populated segments of lapsed customers — those whose reason for defection relates to service (36% of responses), price (24%) and competition (28%). The first two groups describe the reasons for defection as switching from the former brand; while the last represents reasons for switching to new brands. The dependent variables in this analysis are brand associations, brand evaluations, and brand consideration. A free choice, pick any approach (Barnard and Ehrenberg, 1990) measures lapsed customers' propensity to retrieve positive and negative brand associations with regard to their former brands. Respondents choose from a list of brands, including former providers, and state which, if any, of the brands they associate with each given attribute. This method produces similar outcomes to scales, but allows the collection of more information in a shorter period of time (Driesener and Romaniuk, 2006). Academic and industry brand image studies use the pick any approach extensively (Romaniuk, 2006).

S. Bogomolova / Journal of Business Research 63 (2010) 1135–1141

The individual attributes are part of an instrument from an ongoing commercial brand equity study for a financial services company. Past qualitative and quantitative research validated these associations as relevant to business consumers in the financial services market. The list includes 18 positive and five negative attributes in a mix of tangible and intangible attributes following a suggestion in McDowell-Mudambi et al. (1997). The difference in the number of positive and negative attributes in the questionnaire mimics the nature of consumer memory and information processing — people are likely to think of and retain higher numbers of positive attributes, than negative (Krishnan, 1996; Thelen and Woodside, 1997). Hence, the ratio of positive to negative attributes is representative of the pool of associations that a consumer may hold about a brand. For ease of analysis of the correspondence between the three most common reasons for defection (service, price and competition) and brand association, two independent raters code all associations into broad categories. Initial categorizations match in 90% of cases. Raters solve disputes about the attribute category membership in the other 10% through a discussion. In total, raters create six broad categories for positive associations: service, price, competition, business partnership, citizenship and flexibility/convenience. Negative associations represent four categories: service, price, business partner and flexibility. No negative associations represent the competition category, consequently the analysis tests H2 using only service and price associations. Table 2 presents the list of all brand associations and categories. The brand evaluation measure uses a verbally anchored seven-point Likert scale, where one is extremely negative and seven is extremely positive (Likert, 1967). Brand consideration is a prompted measure, which has the advantage of capturing mental rejection of a brand (i.e. I would not consider brand A) over unaided measures (Horowitz and Louviere, 1995). For each brand, respondents answer whether they Table 2 Propensity of lapsed customers to retrieve the former brand with regard to positive and negative brand associations (comparison of three groups: service, price and competition). Categories

Brand associations

Service

Price

Comp

χ2

p

1139

would or would not consider the brand for future purchase or if they are unsure. The Computer Assisted Telephone Interview (CATI) system randomly rotates the brands for all three dependent measures to minimize the order effect. Variables for brand equity are in separate parts of the questionnaire. Different scales and questioning techniques also minimize the possibility of common method variance when measuring cognitive responses in the same instrument (Avolio et al., 1991; Doty and Glick, 1998). 10. Results — propensity to give positive and/or negative associations H1 states that lapsed customers will be less likely to link their former brand to associations that correspond with their reasons for defection comparing to other groups of lapsed customers. For example, those who defected because of service will be less likely to state that the former brand always improves service. The analysis compares the propensity to link the former brand to corresponding associations across the three groups of reasons for defection (service, price and competition) and six groups of positive associations (service, price, competition, business partnership, citizenship and flexibility/convenience). Table 2 summarizes the results for the six groups of positive brand associations across the three groups of reasons for defection. The results show that none of the propensities to retrieve positive associations differ significantly from each other. The results reject H1. H2 states that groups with corresponding reasons for defection will be more likely to respond to their respective negative brand associations. The analysis for five negative associations (bottom half of Table 2) shows partial support for H2. Those who defect because of price reasons show slightly higher propensity (p < 0.05 and p < 0.1) to respond to the two price associations has high fees and is too profit-oriented, compared to other lapsed customers. Tests for negative service associations show no significant differences between the groups. 11. Results — overall brand evaluation and brand consideration

% associating Positive associations Service Is easily contactable Has helpful staff Improves service Responds quickly Price Appropriate fees and charges Good interest rates Competition Leading-edge technology Offers good value Business A business partner partnership Is a business bank Conducts business with integrity Citizenship Makes decisions locally Supports community Helps economy Is local Helps rural community Flexibility/ Conveniently located convenience branches Is flexible Negative associations Service Is impersonal Price Has high fees Too profit-oriented Business Does not care for partner businesses Flexibility Is inflexible

• •

16

16

12

0.49

0.78

8

4

12

2.33

0.31

13

16

15

0.20

0.91

20

24

25

0.65

0.72

24

26

25

0.76

0.96

28

31

34

0.63

0.73

26 22 29 29

24 35 37 26

17 15 19 15

1.70 6.20⁎⁎ 4.77⁎ 3.58

0.42 0.05 0.09 0.17

26

24

14

3.37

0.19

p < 0.1⁎; p < 0.05⁎⁎. Last two columns show results of the Pearson Chi-Square tests, df = 2, n = 186, split by groups n = 76, 51, 59, respectively.

H3 states that the competition group will have a more positive overall brand evaluation of their former brand compared to price and service groups. The results support H4. The ANOVA test shows that the groups have significantly different mean evaluations (F2, 183 = 8.08; p < 0.0001). The Tukey post-hoc test shows that significant differences exist between the competition (mean 3.7 out of 7) and price (M = 2.9) groups (p < 0.014), and the competition and service groups (M = 2.7) (p < 0.001). H4 states that the competition group will have a higher propensity to consider their former brand than the price and service groups. The competition group has a slightly higher proportion of those who would consider their former brand (27%), compared to the price (18%) and service (13%) groups. The Chi-square test shows a marginally significant difference between the groups (χ22, n = 161 = 5.34; p = 0.07). The analysis includes proportions of those who said yes they would consider their former brand in the future, and those who said no they would not, but leaves out those who are unsure, due to their lack of contribution to hypothesis testing. The marginal significance of the difference in the results for H4, and the fact that the results are in line with the previously tested dependent variables, allow the tentative acceptance of H4. 12. Conclusions This paper explores the heterogeneity of lapsed customers — those who used to buy a brand, but stopped doing so for various reasons, in the B2B financial services market. The first part of the study finds that the most common reasons for defection relate to (1) service; (2) price; and (3) attraction by competition. The first two groups

1140

S. Bogomolova / Journal of Business Research 63 (2010) 1135–1141

represent the switch from category of reasons, which highlights negative qualities of the former brand. By contrast, the third group represents the switch to category, which highlights positive qualities of other brands that customers can switch to. The second part of the study hypothesizes that the different reasons for defection will translate into a difference in the former brand's equity in the form of brand associations, brand evaluations and brand consideration. Table 3 presents the summary of hypotheses and results. The results show that lapsed customers (even after more than five years) still remember that they had lapsed from a brand of financial service provider. Furthermore, the reasons for that defection have a long-lasting impact on lapsed customers' judgment of their former brand. The main effects exist in negative price-related brand associations, brand evaluations and propensity to consider the brand for future purchase. A possible explanation for the lack of variability in positive and some negative brand associations could lie in uneven retention of information in consumer memory. Because of the time elapsed since defection, some perceptions may have been lost or modified (Bogomolova et al., 2009), while others remain in consumer memory and continue to influence lapsed customers' judgments. Future research should include investigation of lapsed customers' brand equity at different points in time after defection. Such research could provide valuable information about the timing of the impact that reasons for defection have on different components of consumer brand equity, shedding light on the durability of different types of perceptual brand information. From a managerial perspective, replication of this research at different points in time would suggest how long brand management should wait after defection before approaching their lapsed customers with a winback offer.

their former brand again. Following suggestions in the literature that brand equity can have a differential impact on future behavior (Keller, 1993), likely this group would have the highest winback potential. On the other hand, these customers might also be more likely to switch away again, as their preference might lie with whoever gives the best offer at the time (Stauss and Friege, 1999). So, while this group of lapsed customers might look like the easier catch, managers should balance their efforts to acquire these customers with an estimation of their (secondary) Customer Lifetime Value and potential return on investment. The two other groups (those who lapsed because of price and service reasons) are more negative about their former brands than the first (competition) group. This result supports the idea of Attribution Theory (Weiner, 2000), that in conditions when consumers are more likely to blame the brand (when the former brand's actions pushed customers away from the brand), consumer brand equity is more negative. Conversely, when the reasons for defection relate to consumers themselves or other external (non-brand related) factors, consumer brand equity is more positive. This finding supports the applicability of Attribution Theory to brand defection, and reveals the potential of this theory to provide explanations for lapsed customers' behavior after defection. Comparison between the price and service groups shows a mixture of similarities and differences across these variables. The propensity to give positive and negative associations is about the same in these two groups. However, the overall evaluation and brand consideration of former brands are slightly more favorable in the price group. These variations in scores are not statistically significant in the current analysis, however the general pattern supports the findings of a previous research (Bogomolova and Romaniuk, 2009), which shows that the service group tends to have the most negative brand equity of all lapsed customers.

13. Research implications The major implication of this research is that lapsed customers are not homogenous. The reasons for defection do influence postdefection brand equity (even some years after the event), creating groups that have varying propensities to come back to the former brand. These results support suggestions in the literature that the reason for defection is an important segmentation variable for lapsed customer winback strategies (Stauss and Friege, 1999). This conclusion suggests that brand managers must maintain accurate records of interactions with consumers prior to and during contract termination to gain in-depth understanding of the reasons for defection. This research also shows that the group of lapsed customers who defected from a brand because of attraction by the competition has the most positive former brand equity, compared to other lapsed customers. This group has the lowest propensity to respond to negative brand associations, the most positive overall evaluation and the highest proportion of customers who would consider buying from

14. Limitations and future research A major limitation of this research relates to its scope — only one category of financial business-to-business services. Further extension of research beyond financial services and into personal markets would show whether the findings are generalizable across industries. Another limitation is the classification of the reasons for defection. Because respondents could provide more than one answer (and about 1/3 did so), interviewers instructed respondents to think about their major reason. Hence, in about 1/3 of cases, possibly more than one reason influenced responses. Future research could explore the impact of multiple reasons for defection on post-defection consumer brand equity. One of the potential confounding factors in cognitive responses is the time elapsed between the last brand experience (or defection) and evaluation. This paper does not examine the impact of time on responses. The vast majority of lapsed customers who participated in

Table 3 Summary of hypothesis and results. Hypotheses

Results

H1: Lapsed customers will be less likely to link their former brand to positive associations that correspond with their reasons for defection. H2: Lapsed customers will be more likely to link their former brand to negative associations that correspond with their reasons for defection.

Rejected. All groups have equal propensities. Partially supported. Price group has higher propensity to retrieve negative price associations than other groups. All groups have about equal propensity to give negative service association. The survey did not test for competition association. Supported.

H3: Competition group will have a more positive evaluation of the former brand compared to price and service groups. H4: Competition group will have a higher propensity to consider the former brand again compared to price and service groups.

Supported.

S. Bogomolova / Journal of Business Research 63 (2010) 1135–1141

this study defected from brands more than 12 months ago. Hence forgetting and post-hoc rationalization (as per the Cognitive Dissonance literature) could potentially affect the way lapsed customers view their former brands. Further analysis should ideally include lapsed customers with different time lapsed from the defection. Future research should also have sufficient sample sizes to allow for detailed analysis of other lapsed customers groups, beyond the three most common ones. An important extension of this research is to examine the brand equity of lapsed customers who defect for reasons that were beyond the brand management's influence (for example, who left the category altogether). The current research does not examine this group, but earlier exploratory study suggested that this group forms a substantial proportion of defectors (Bogomolova and Romaniuk, 2009). Finally, this research compares brand equity and the potential for future brand uptake (winback) only among lapsed customers. Yet, whether this group would have a higher purchase potential compared to the remaining non-users — those who have never used the brand — is unclear. After all, lapsed customers have already rejected the brand, and this experience affects their propensity to have more negative than positive associations and lower than neutral overall evaluations. Possibly, new customers who have never had such (negative) brand experience would be a more fruitful source of customer acquisition. This is an agenda for future research. References Agarwal Manoj K, Rao Vithala R. An empirical comparison of consumer-based measures of brand equity. Mark Lett 1996;7(3):237–47. Anderson John R, Bower Gordon H. Human Associative Memory. Hillsdale, NJ: Lawrence Erlbaum; 1979. Avolio Bruce J, Yammarino Francis J, Bass Bernard M. Identifying common methods variance with data collected from a single source: an unresolved sticky issue. J Manage 1991;17(3):571–87. Bansal Harvir S, Taylor Shirley F, St. James Yannik. “Migrating” to new service providers: toward a unifying framework of consumers' switching behaviors. J Acad Mark Sci 2005;33(1):96-115. Barnard Neil R, Ehrenberg Andrew SC. Robust measures of consumer brand beliefs. J Mark Res 1990;27(4):477–84. Berry Leonard L. Cultivating service brand equity. J Acad Mark Science 2000;28(1):128–37. Bird Michael, Channon Charles. Further analysis of former users brand image deviations — part II. Admap 1970:28–32 (January). Bird Michael, Channon Charles, Ehrenberg ASC. Brand image and brand usage. J Mark Res 1970;7(3):307–14. Blombäck Anna, Axelsson Björn. The role of corporate brand image in the selection of new subcontractors. J Bus Ind Mark 2007;22(6):418–30. Bogomolova Svetlana, Romaniuk Jenni. Brand defection in a business-to-business financial service. J Bus Res 2009;62(3):291–6. Bogomolova Svetlana, Romaniuk Jenni, Sharp Anne. Quantifying the extent of temporal decay in service quality ratings. Int J Mark Res 2009;51(1):71–91. Colgate Mark, Hedge Rachel. An investigation into the switching process in retail banking services. Int J Bank Mark 2001;19(5):201–12. Cummings William H, Venkatesan M. Cognitive dissonance and consumer behavior: a review of the evidence. J Mark Res 1976;13:303–8 (August). Doty DHarold, Glick William H. Common methods bias: does common methods variance really bias results? Organ Res Methods 1998;1(4):374–406. Driesener Carl, Romaniuk Jenni. Comparing methods of brand image measurement. Int J Mark Res 2006;48(6):681–98. Eagly Alice H, Chaiken Shelly, Chen Serena, Shaw-Barnes Kelly. The impact of attitudes on memory: an affair to remember. Psychol Bull 1999;125(1):64–89. Fazio Russel H, Zanna Mark P. In: Berkowitz Leonard, editor. Direct experience and attitude–behavior consistency. Advances in Experimental Social Psychology. New York: Academic Press; 1981. p. 161–202. Festinger L. A Theory of Cognitive Dissonance. Stanford: Stanford University Press; 1957. Fishbein Martin, Ajzen Icek. Belief, Attitude, Intention and Behaviour: An Introduction to Theory and Research. Reading, MA: Addison-Wesley Publishing Company; 1975. Gerrard Philip, Cunningham JBarton. Consumer switching behavior in the Asian banking market. J Serv Mark 2004;18(3):215–23. Griffin Jill, Lowenstein Michael W. Customer winback: how to recapture lost customers — and keep them loyal. San Francisco: Jossey-Bass, Inc. Publishers; 2001.

1141

Hogan John E, Lemon Katherine N, Libai Barak. What is the true value of a lost customer? J Serv Res 2003;5(3):196–208. Holden Stephen JS, Lutz Richard J. Ask not what the brand can evoke; ask what can evoke the brand? Adv Consum Res 1992;19(1):101–7. Homburg Christian, Hoyer Wayne D, Stock Ruth Maria. How to get lost customers back? A study of antecedents of relationship revival. J Acad Mark Sci 2007;35(4):461–74. Horowitz Joel L, Louviere Jordan J. What is the role of consideration sets in choice modelling? Int J Res Mark 1995;12(1):39–54. Howard John A, Sheth Jagdish N. The Theory of Buyer Behavior. New York: John & Wiley Sons, Inc; 1969. Johnston Wesley J, Lewin Jeffrey E. Organizational buying behavior: toward an integrative framework. J Bus Res 1996;35(1):1-15. Keaveney Susan M. Customer switching behavior in service industries: an exploratory study. J Mark 1995;59:71–82 (April). Keller Kevin Lane. Conceptualizing, measuring, and managing customer-based brand equity. J Mark 1993;57(1):1-22. Keller Kevin Lane. Brand synthesis: the multidimentionality of brand knowledge. J Consum Res 2003;29(4):595–601. Keller Kevin Lane, Lehmann Donald R. Brands and branding: research findings and future priorities. Mark Sci 2006;25(6):740–59. Krishnan HS. Characteristics of memory associations: a consumer-based brand equity perspective. Int J Res Mark 1996;13(4):389–405. Laroche Michel, Rosenblatt Jerry A, Brisoux Jacques E. Consumer brand categorisation: basic framework and managerial implications. Mark Intell Plann 1986;4(4):60–74. Lees Gavin, Garland Ron, Wright Malcolm. Switching banks: old bank gone but not forgotten. J Financ Serv Mark 2007;12(2):146–56. Lichtenstein Meryl, Srull Thomas K. Conceptual and methodological issues in examining the relationship between consumer memory and judgment. In: Alwitt Linda F, Mitchell Andrew A, editors. Psychological Processes and Advertising Effects. London: Lawrence Erlbaum Associates; 1985. p. 113–28. Likert Rensis. The method of constructing an attitude scale. In: Fishbein Martin, editor. Readings in Attitude Theory and Measurement. New York, NY: John Wiley & Sons Inc; 1967. p. 90–5. Lynch Jr John G, Marmorstein Howard, Weigold Michael F. Choices from sets including remembered brands: use of recalled attributes and prior overall evaluations. J Consum Res 1988;15:169–84 (September). McDowell-Mudambi Susan, Doyle Peter, Wong Veronica. An exploration of branding in industrial markets. Ind Mark Manage 1997;26(1):433–46. Narayana Chem L, Markin Rom T. Consumer behavior and product performance: an alternative conceptualization. J Mark 1975;39:1–6 (October). Nedungadi Prakash. Recall and consumer consideration sets: influencing choice without altering brand evaluations. J Consum Res 1990;17(3):263–76. Ratneshwar S, Shocker Allan D. Substitution in use and the role of usage context in product category structures. J Mark Res 1991;28:281–95 (August). Reichheld Frederick F. Learning from customer defections. Harvard Bus Rev 1996;74 (2):56–69. Reichheld Frederick F, Sasser Jr WEarl. Zero defections: quality comes to services. Harvard Bus Rev 1990;68(5):105–11. Richins ML. Negative word-of-mouth by dissatisfied consumers: a pilot study. J Mark 1983;47(Winter):68–78. Romaniuk Jenni. The brand perceptions of former users. Mark Bull 2001;12:1–6. Romaniuk Jenni. Comparing prompted and unprompted methods for measuring consumer brand associations. J Target Meas Anal Mark 2006;15(1):3-11. Romaniuk Jenni, Sharp Byron. Conceptualizing and measuring brand salience. Mark Theory 2004;4(4):327–42. Solomon Michael R. Consumer Behavior: Buying, Having, and Being. Needham Heights. MA: Allyn and Bacon; 1992. Stauss Bernd, Friege Christian. Regaining service customers: costs and benefits of regain management. J Serv Res 1999;1(4):347–61. Thelen Eva M, Woodside Arch G. What evokes the brand or store? Consumer research on accessibility theory applied to modeling primary choice. Int J Res Mark 1997;14 (2):125–45. Thomas Jacquelyn S, Blattberg Robert C, Fox Edward J. Recapturing lost customers. J Mark Res 2004;41(1):31–45. Tokman Mert, Davis Lenita M, Lemon Katherine N. The WOW factor: creating value through win-back offers to reacquire lost customers. J Retail 2007;83(1):47–64. Urban Glen L, Weinberg Bruce D, Hauser John R. Premarket forecasting of really-new products. J Mark 1996;60:47–60 (January). Webster Frederick E, Keller Kevin Lane. A roadmap for branding in industrial markets. J Brand Manag 2004;11(5):388–402. Weiner Bernard. Attributional thoughts about consumer behaviour. J Consum Res 2000;27:382–7 (December). Wilkie WL. Consumer Behavior. New York: John Wiley & Sons; 1986. Winchester Maxwell, Romaniuk Jenni. Evaluative and descriptive response patterns to negative image attributes. Int J Mark Res 2003;45(1):21–34. Winchester Maxwell, Romaniuk Jenni, Bogomolova Svetlana. Positive and negative brand beliefs and brand defection/uptake. Eur J Mark 2008;42(5/6):553–70.