Customers Enhancing Purchase Intentions and Positive Word-of-mouth in the Banking Industry
Jasmin Bergeron Professor of Marketing University of Quebec in Montreal School of Business and Management C.P. 6192 Succ.Centre-Ville Montreal, QC CANADA Phone: [phone] X 1445# Email: [e-mail] H3C 4R2
Submission Type: Competitive Paper Option 1: To be published in its entirety in the ACR Proceeding
Acknowledgments: The authors gratefully acknowledge the financial and material assistance of the Institute of Canadian Bankers and the Social Sciences and Humanities Research Council of Canada.
Enhancing Customers Purchase Intentions and Positive Word-of-mouth in the Banking Industry
Abstract
A plethora of research have shown that to solidify long term relationships with customers, financial institutions must focus on key factors such as client knowledge, high quality service, expertise, trust, and satisfaction. This research examines the complex interactions between these key elements and their impact on customers purchase intentions and positive word-ofmouth recommendations. A sample of customers in a North American city was surveyed and structural equation modeling analyses were conducted. The results confirmed the predominant role of customer knowledge, service quality, account managers expertise, trust, and satisfaction, which accounted for 68% of the variance in the purchase intention construct and 84% in the word-of-mouth construct. Several strategic recommendations and interesting research opportunities are discussed.
1
Enhancing Customers Purchase Intentions and Positive Word-of-mouth in the Banking Industry
The only constant in today s banking environment is change. New technologies, the deregulation of financial industries, a plethora of new products and growing cultural diversity are some examples. Of course, clients have changed too. They are more sophisticated, more informed and, above all, more demanding. In the overcompetitive banking industry, the biggest challenge is perhaps to establish long-term relationships with customers and that these patrons talk positively about their institution. In order to obtain these cherished outcomes, a bank must focus on some key factors, such as client knowledge (Wilson, 1995; Ricard and Perrien, 1999), quality service (Rust and Zahorik, 1993), expertise (Ganesan, 1994), trust (Morgan and Hunt, 1994) and satisfaction (Crosby, Evans and Cowles, 1990). Any research that can help to address these key issues is likely to meet with a great deal of interest.
CONCEPTUAL FRAMEWORK A conceptual framework (Figure 1) was developed, based on previous studies conducted by the author and on an exhaustive review of the relevant literature. An explanation supporting the relationships posited in Figure 1 follows.
2
FIGURE 1 CONCEPTUAL MODEL
Relationship Antecedents
Affective Responses
Behavioral Intentions
Positive Word-ofmouth Intentions
Service Quality
Satisfaction
Customer Knowledge Purchase Intentions
Account Manager s Expertise
Trust
Customer Knowledge Teas (1988) refers to customer knowledge as familiarity and sensitivity concerning customers needs, specific problems, and characteristics. In the academic literature, customer knowledge has emerged in the past two decades in research on the quality of services and on relationship marketing. In the service quality field, Parasuraman, Zeithaml, and Berry (1985) claim that quality service is highly dependent on customer knowledge. In relationship marketing, knowledge of the customer has often been considered a significant dimension of the force of the relationship between a service provider and its clients (Paulin, Ferguson, & Payaud, 2000). According to Wilson (1995), an in-depth understanding of customers needs, wants and demands represents an imperative condition to provide high quality and personalized service.
3
Account Manager s Expertise In most banks, the account manager is responsible for providing expertise to analyze the client s needs and goals, recommend innovative and effective solutions, resolve complex financial problems, and generally build an aura of trust and warmth that should ensure a permanent bond between him/herself and the client. On one hand, Ganesan (1994) argued that an organization and its employees must demonstrate a high level of expertise in order to establish perceptions of credibility and trust in consumers minds. On the other hand, Brand, Cronin and Routledge (1997) observed that the expertise of the service firm, and particularly of its employees, influence to a great extent the customer s level of satisfaction. As indicated in Figure 1, we suggest that the client s trust and satisfaction, generated from the financial expertise of the account manager, will positively affect his/her purchase intentions and propensity to recommend the financial institution.
Service quality After more than three decades of extensive service research, service quality is still an abstruse and abstract construct that is arduous to define. Harrison-Walker (2001) even reported that this construct have given from one to seven different dimensions. Hence, a global definition of service quality might be more appropriate since its inherent dimensions are likely to be dependent on service settings (Brown, Churchill, & Peter, 1993; Carman, 1990). Zeithaml (1988), for instance, defines service quality as the consumer's judgment about the overall excellence or superiority of a service. Likewise, Anderson and Fornell (1994) refer to service quality as a global evaluation of a firm's service delivery system. As a boundary spanner, the account manager plays a critical role in the service delivery process. They are normally the primary, if not only, contact between the organization and its customers (Sharma & Sarel, 1995). Each of these customer contacts has the potential of 4
positively or negatively impacting the customer's view of the service provided by the salesperson's company (Shepherd, 1999).
Satisfaction Although the research literature on satisfaction has swelled enormously over the past three decades, a universally agreed theoretical or operational definition still eludes marketing academics. A widely used paradigm of customer satisfaction is the disconfirmation-ofexpectations theory (Oliver, 1980). It defines satisfaction as a mental comparison of the quality that a customer expected to receive from an exchange and the level of quality the customer perceives actually receiving from the exchange. Even if this theory is commonly employed, it still lacks some coherence and common sense. Suppose, for example, that a man is going to dinner with his mother-in-law. If he expects the event to be boring, and that it is indeed tiresome, the man is most likely to be unsatisfied (even if his expectations were met). Like service quality, satisfaction was also attributed all-inclusive definitions. For example, Crosby, Evans, and Cowles (1990) refer to customer satisfaction with the salesperson reflects as an emotional state that occurs in response to an evaluation of the interaction experience that the customer has with the salesperson. Despite the obvious importance of customer satisfaction for most businesses today, Szymanski and Henard (2001) observed in their meta-analysis of 50 relevant papers that few studies have investigated the outcomes of satisfaction and only a few outcomes of satisfaction have been investigated in these studies. We will attempt to remedy this paucity in the present investigation.
5
Trust Trust is generally thought to be a key determinant of the quality of buyer-seller relationships (Swan, Bowers, & Richardson, 1999). Morgan and Hunt (1994) conceptualize trust as existing when one party has confidence in an exchange partner s reliability and integrity. Over the years, many researchers uncovered that trust is positively related to several variables in consumer research. For example, trust was found to be significantly correlated to cooperation (Morgan & Hunt, 1994), relationship commitment (Achrol, 1991; Morgan & Hunt, 1994), positive word-of-mouth intentions (Bergeron, Ricard, & Perrien, 2003), future interactions (Crosby, Evans, & Cowles, 1990; Ramsey & Sohi, 1997), and purchase intentions (Pilling & Eroglu, 1994).
Positive Word-of-mouth Intentions Word-of-mouth (WOM) is probably the oldest mechanism by which opinions on products, brands, and services are developed, expressed, and spread (Lau & Ng, 2001). File, Judd, and Prince (1992) define WOM as recommending the firm and the service to others. Specifically, Anderson (1998) refers to positive WOM as relating pleasant, vivid, or novel experiences, recommendations to others, and even conspicuous display. Numerous
research reinforced the contention that positive WOM is a cherished outcome in marketing. In 1955, Katz and Lazarsfeld found positive word-of-mouth seven times more effective than newspaper and magazine advertising, four times more effective than personal selling, and twice as effective as radio advertising in influencing consumers to switch brands. In 1983, Morin uncovered that referrals from others accounted for three times as many purchases as did advertising when consumers were asked what factors influenced their purchases of 60 different products.
6
Purchase Intentions Purchase intentions refer to the degree of perceptual conviction of a customer to repurchase a particular product (or service) or to repurchase any product (or service) at a particular organization. The implicit assumption made by those who gather and analyze consumer intentions data is that they will reflect future sales behaviors. Research in social psychology suggests that intentions are the best predictor of an individual s behavior because they allow each individual to independently incorporate all relevant factors that may influence the actual behavior. Armstrong, Morwitz, & Kumar (2000) compared buying intentions with other sales forecasting tools. They found that purchase intentions represent an accurate measure of future sales and that it provides better forecasts than an extrapolation of past sales trends. The benefits of maintaining a base of long-term customers are widely recognized by marketers. Given that the cost of retaining an existing customer is less expensive than prospecting for a new customer (Spreng, Harrell, & Mackoy, 1995), purchase intention is a very important consideration for financial institutions.
METHODOLOGY The research was carried out in the service sector because exchanges in many service contexts involve long-term commitments and a continual stream of interaction between buyers and sellers (Lovelock, 1983). As Normann (2000) argues, service exchanges tend to be inherently more relational since they are often personality intensive. Within the service realm, we selected the banking industry as our study domain because it is one of the first sector of activity to jump onto the bandwagon of relationship marketing. In other words, this is a industry in which long-term relationships have stimulated interest as financial institutions respond to omnipresent competition. 7
Sampling procedures and data collection A total of 1900 names were extracted at random from a file of customers provided by a professional marketing company. A first phone call was made to ask respondents whether they wanted to participate in the study. Then, 947 questionnaires were sent by mail only to respondents who agreed to participate in the research. Relying on mail surveys seemed like a valid choice given the reticence of people to grant personal or telephone interviews, especially in a context characterized by delicate financial and insurance matters. Two weeks later, a second phone call was made in order to verify if the respondents received the questionnaire and to encourage its completion. Three hundred and six (306) customers completed the questionnaire, equivalent to a satisfactory response rate of over 30%.
Measures Several steps were taken to ensure good measures of the constructs. First a literature review led to the identification and adaptation of the various scales. Then, a draft of the questionnaire was shown to five university researchers. After making certain modifications, the questionnaire was submitted to a banking sector expert for analysis and review in order to incorporate the respondents' point of view. Finally, a pre-test was conducted with 10 bank customers. Based on the respondents comments, slight modifications were accomplished to ensure a common understanding of the scales. A summary of the items used to measure each factor is presented in Table 1. Once the data had been collected and preliminary analyses completed, reliability tests were conducted. Results show that the Cronbach alphas varied from 0.66 (for the purchase intention construct) to 0.92 (for the expertise and word-of-mouth constructs), which is satisfactory (Peterson, 1994). Then, several factor analyses were run to evaluate the capacity of the questions to measure each factor (constructs). Results are presented in Table 1. 8
TABLE 1 MEASURES AND RELIABILITY ANALYSES
a b
MEASURES
FACTORS
/ CR
CUSTOMER KNOWLEDGE (CK) CK3 My account manager knows the industry and the environment in which my company works. CK2 My account manager is knowledgeable about the products and services of my company. CK1 My account manager knows the financial needs of my company. CK4 My account manager knows the abilities and skills of the administrators of my company. EXPERTISE (E) E1 E2 My account manager knows a lot about financial analyses. My account manager is highly knowledgeable in finance. .909 .883 .92 / .84 .866 .86 / .79 .860 .824 .810
SERVICE QUALITY (SQ) SQ4 SQ1 SQ2 SQ3 This bank is very reliable. This bank offers an excellent service. My account manager takes the time to adequately meet my needs. The discussion between my account manager and myself are usually very efficient. .822 .811 .719 .710 .75 / .74
SATISFACTION (S) S1 S3 S2 I am satisfied with the service offered at this bank. I am satisfied with the overall relation I have with this bank. I am satisfied with my account manager. .914 .907 .878 .88 / .81
TRUST (T) T2 T1 T3 I trust my account manager. I can trust my account manager when he gives me his word. I can rely on my bank when I need it. .875 .858 .783 .79 / .76
PURCHASE INTENTION (PI) PI1 The probability that my company will do business with this bank in two years is very high. PI2 In the next period of negotiation, the renewal of the financial services used by my company at this bank will be a simple formality. PI3 To the best of my knowledge, my company tried to dissolve its relation with this bank during the past year. WORD-OF-MOUTH INTENTION (WOM) W2 I would make positive recommendations about my account manager. W1 I would make positive recommendations about my bank.
a b
.797 .66 / .63 .557 .549
.974 .862
.92 / .85
Measures were calculated on seven-point Likert scales (strongly disagree to strongly agree) The first number represents the alpha of Cronbach. The second number represents the Composite Reliability index (Fornell and Larcker, 1981).
9
DATA ANALYSES AND RESULTS
Descriptive statistics We now present our results according to the methodological approach defined in the preceding section. First, descriptive statistics were computed on the demographic profile of the sample. Results indicate that our sample is composed of a majority of undergraduate (49.3%) male respondents (75.7%), aged between 35 and 54 years (64%), and working as president or owner of its business (52.5%).
Confirmatory Factor Analyses A confirmatory factor analysis was run on the model (Figure 1) using the EQS structural equation modeling software (Bentler, 1992). We hypothesized a priori that (a) each item has a non-zero loading on the factor it was designed to measure and zero loadings on all other factors, (b) factors that are not linked with structural paths are uncorrelated and (c) error terms associated with each item are also uncorrelated.
Assessment of Overall Fit and Parameter Estimates We relied on several statistics to evaluate the goodness-of-fit of the model. First, the normed fit index (NFI) of .93 and the comparative fit index (CFI) of .94 constitute a good indication that the hypothesized model represents an adequate fit to the data. The average off-diagonal value of the standardized residuals matrix is .039, which also reflects a fairly good fit. Although the
2
is significant (
2
(179)
= 380.2, p<.01), it is known to be sensitive to
large samples and trivial discrepancies (Fornell and Larcker, 1981) and thus, is a poor indicator of model fit (Singh, 2000). The goodness-of-fit statistics and the parameters of the measurement model are summarized in Table 2. The causal paths are graphically illustrated in their standardized format in Figure 2. 10
TABLE 2 MEASUREMENT MODEL PARAMETERS AND GOODNESS-OF-FIT STATISTICS
PATHS
Consumer knowledge (CK) CK CK1 CK CK2 CK CK3 CK CK4 Expertise (E) E E1 E E2 Service quality (SQ) SQ SQ1 SQ SQ2 SQ SQ3 SQ SQ4 Trust (T) T T1 T T2 T T3 Satisfaction (S) S S1 S S2 S S3 Purchase intention (PI) PI PI1 PI PI2 PI PI3 Word-of-mouth (WOM) WOM WOM1 WOM WOM2
b
UNSTANDARDIZEDa STANDARDIZED
1.00 .98 (12.88) .81 (11.04) .92 (12.35) 1.00 1.00 (22.18) 1.00 1.16 ( 9.55) 1.22 (11.72) 1.01 ( 9.77) 1.00 .99 (13.60) .91 (11.90) 1.00 .98 (19.67) .90 (15.10)
.80 .80 .70 .77 .91 .97 .77 .64 .78 .66 .80 .84 .75 .93 .85 .75 .80
1.00 .75 ( 8.24) .81 ( 7.50) 1.00 .93 (27.08)
.60 .55 .98 .91 380.2 .039 .93 .94
Fit Indices 2 Chi square - (179) Average off-diagonal value of the SR matrix Normed fit index (NFI) Comparative fit index (CFI)
a b
T-values are in parenthesis. Based on one-tailed t-tests: t-values>1.65, p<.05; t-values>2.33, p<.01. Estimates without t-values are fixed parameters. Good model fit is indicated when the average off-diagonal value of the standardized residuals matrix and the chi square value approach zero. Also, the closer to 1.0 the NFI and the CFI are, the better the model fit.
In reviewing information related to the measurement equations, we see in Table 2 that each measurement parameter is both reasonable and statistically significant. Relying sound on statistical and theoretical rationales, we feel confident that the model stands as an adequate representation of the sample data.
11
FIGURE 2 STANDARDIZED RESULTS
R =.63
2
R =.60
2
R =.84
2
.76
Service Quality Satisfaction
.84
Positive Word-ofmouth Intentions
.79
Client Knowledge
NS
.14
.50 .46
Account Manager s Expertise Trust
R =.74
2
.37 .55
Purchase Intentions
R =.68
2
As evidence in Figure 2, each causal path (except the impact of expertise on satisfaction) is statistically significant (t > 2.33, p < 0.1). Results indicate that an effective relationship, when measured by purchase intentions and positive word-of-mouth, is fostered by the customer s trust and satisfaction in his/her bank, which is partially depending on the account manager s expertise and knowledge of the customer, as well as by the quality of the service provided. These results are discussed more in-depth in the next section. DISCUSSION This study provides both a theoretical and practical contribution to understanding the key elements enhancing customers purchase intentions and positive word-of-mouth in the banking industry. Some of the findings we obtained are consistent with the literature. For instance, the positive impact of consumer knowledge on the perceived quality of service has been identified in earlier research (Dunkelberg and Scott, 1984). The successive effect of service quality on trust and purchase intentions, which was originally acknowledged by Parasuraman, Zeithaml, and Berry (1994), was also confirmed in this research. Even if the present study contributed to the relevant literature by confirming some of its 12
theories, it still provided new and interesting insights. Three clearly important findings emerged. The first one involves the non-significant relationship between expertise and satisfaction. Although surprising at first, this result is in accordance with Teas (1988), who argues that since most customers do not know much about finance, they tend to evaluate their bank with relational criteria, as opposed to financial standards. Nevertheless, expertise is an important factor since it impacts the client s trust, which in turn affects his/her purchase intentions and tendency to make positive recommendations. This is consistent with Ganesan s (1994) study, which revealed that a service provider must demonstrate his/her expertise in order to augment the customer s feeling of trust. A second interesting finding was obtained with respect to the combined effect of trust and satisfaction on purchase intentions. Surprisingly, it appears that trust, as opposed to satisfaction, has a greater influence on customers intentions to patronize their banks. In other words, trusting customers are more likely to purchase financial products in the future than satisfied consumers. Garbarino and Johnson (1999) also found similar results, but only for high relational customers. One possible explanation (to be tested) is that most consumers would buy from trustworthy account managers even if they are not completely satisfied with them, whereas few individuals would purchase from untrustworthy account managers, even if they were somehow satisfied by certain aspects of their offering. These results confirm the importance of trust in the development of a long-term relationship. Several researchers have previously determined how crucial this factor is, going so far to integrate it into the definition of relationship marketing (e.g., Crosby, Evans and Cowles, 1990). The third important finding involves the strong repercussion of the client s satisfaction on his/her propensity to generate positive word-of-mouth publicity. Satisfied customers, as opposed to trusting customers, are very inclined to recommend their bank and/or their account managers. According to Bagozzi, Gopinath and Nyer (1999), customer satisfaction is highly correlated to joy, happiness, and other 13 related positive emotions. This might
explain why satisfied customers tell persons external to the transaction of their pleasure with the bank and/or its account managers. Finally, it is important to underline the predominant role of customer knowledge, service quality and account manager expertise. The combined indirect effect of these three variables, added with the effect of trust and satisfaction, accounted for 68% of the variance in the purchase intention construct and 84% in the word-of-mouth construct. This results are deemed as excellent, given that Sawyer and Ball (1981) qualifies an effect with an R2 between .13 and .26 as average.
MANAGERIAL IMPLICATIONS From the customer s perspective, it is of paramount importance that financial institutions have a good understanding of both the company and its activities, and use this knowledge to provide a personalized high quality service. This research has shown that by doing so, banks can increase significantly their clients trust and satisfaction. Hence, policies and decisions of financial institutions must favor an in-depth knowledge of customer needs, goals and expectations. Recent technological advances have helped to make this objective achievable. Rapid improvements in information technology allow banks and their front-line staff to track customer characteristics more easily and respond with appropriate marketing offers. Businesses are no longer restricted to marketing at segment level, but can now tailor offers to individual customers. Financial institutions can also benefit from new information technologies to develop an image of expertise, given that this factor was found to affect customers trust. For instance, McGaughey and Mason (1998) suggest service providers to develop a community of satisfied clients where the customers could have access to real testimonies from other customers via the Internet. This research revealed that a customer s level of satisfaction toward his bank is strongly related to his/her will to make positive 14 recommendations. In that respect,
account managers should seek references from clients who they think are very satisfied. This strategy could prove to be very beneficial because positive word-of-mouth has been identified as a powerful factor in the purchase of financial services (File and Prince, 1992). Our results also suggest that account managers should try to sell new products to their more trusting clients, since the purchase intentions of these customers are highly correlated to their level of trust. The present study raises some intriguing issues with respect to the management of human resources. For instance, we found that client satisfaction is more dependent on service quality than on financial expertise. Hence, relational abilities must be a major input in the selection criteria, and be at least as important as financial knowledge. One may suggest that learning the basics of finance may take some months for most individuals, whereas learning (and applying) relational skills may take a lifetime. Skills and aptitudes to understand the customer and provide the appropriate service must also be considered in the design of a salesforce formation program. The importance of service quality reinforces the contention that the relational skills of account managers can be decisive. If a financial institution assesses the performance of its account managers simply on the basis of sales volume, they send the message that it is the numbers and not the consumers that matter. Therefore, the performance appraisal of account managers should include relational criteria, such as client satisfaction and trust. In other words, evaluation and remuneration policies must be modified to ensure that customer-orientated individuals are rewarded. Our results indicate that relational standards are not necessarily conflicting with profitability because satisfied and trusting clients intend to purchase financial services in the future and generate positive word-of-mouth publicity.
15
LIMITATIONS, RESEARCH OPPORTUNITIES AND CONCLUSION Clearly, the present study raises more questions than it provides answers. First, the structural equation modeling solution accounted for 68% of the variance in the purchase intention construct. Thus, the choice of factors in our conceptual framework (Figure 1) is clearly not exhaustive. What additional variables might also influence the customers buying intentions? On one hand, the model could have considered other factors associated with the bank, such as its reputation. On the other hand, customer-related factors such as attitudes and experiences could have been added to the conceptual framework. These ideas give matter to reflection and it seems pertinent to address them in future research. Our study was based on a descriptive perspective, rather than on a longitudinal one, which limits the knowledge of the long term impact of the factors in our model. Would a longitudinal research produce different results from ours? Wilson (1995) argued that some determinants are very important at certain stages of the relationship development, but may become latent in others. In that respect, it could prove interesting to study the phenomenon under a long term horizon. Such a study would also increase our understanding of the development process of a relationship. Does the results of our study apply to other business sectors? Would we obtain the same results at the international level? To answer these questions, researchers could eventually test the model presented in this paper, but under a completely different angle. Studies could, for example, be conducted in another industry (ex: telecommunications), for a particular type of clientele (ex: the government) or in another region (ex: Asia). The results that we obtained, although interesting, constitute only a spring-board for future investigations in a domain growing in theoretical and practical importance.
16
REFERENCES
Achrol, Ravi (1991), Evolution of the Marketing Organization: New Forms for Turbulent Environments, Journal of Marketing, 55 (4), 77-93. Anderson, Erin W. and Cleadus Fornell (1994), A Customer Satisfaction Research Prospectus, In Roland T. Rust and Richard L. Oliver (Eds.), Service Quality: New Directions in Theory and Practice, Sage Publications. Anderson, Eugene W. (1998), Customer Satisfaction and Word-of-mouth, Journal of Service Research, 1 (1), 5-17. Armstrong, J. Scott, Vicki G. Morwitz, and Vineet Kumar (2000), Sales Forecasts for Existing Consumer Products and Services: Do Purchase Intentions Contribute to Accuracy? International Journal of Forecasting, 16 (3), 383-397. Bagozzi, Richard P., Mahesh Gopinath, and Prashanth U. Nyer (1999), The Role of Emotions in Marketing, Journal of the Academy of Marketing Science, 27 (2), 184-206. Bentler, Peter M. (1992), EQS: Structural Equations Program Manual. Los Angeles: BMDP statistical software. Bergeron, Jasmin, Line Ricard, and Jean Perrien (2003), Determinants of Commercial Clients Loyalty in the Banking Industry, Canadian Journal of Adm...