Does customer satisfaction predict customer demand

DOES CUSTOMER SATISFACTION PREDICT CUSTOMER DEMAND IN THE RAIL SECTOR? Rob Sheldon and Miranda Mayes Accent Shepley Orr Accent/UCL Ben Condry Association of Train Operating Companies EXECUTIVE SUMMARY The Association of Train Operating Companies (ATOC) commissioned Accent to undertake research to examine the evidence for a link between customer satisfaction and customer demand, with a view to applying this knowledge to the rail sector. This has been addressed through a literature review and eight tele-depth interviews with users of customer satisfaction data from across a range of key market sectors including transportation, energy, communications, retail and finance. In addition to determining what literature exists to suggest or confirm a link between customer satisfaction and demand, the literature review has also addressed four key questions: 

whether or not recency of experience impacts upon demand;



whether satisfaction in any particular area (say value for money) has greater influence than satisfaction in another;



whether a negative customer experience has a greater impact than a positive experience;



and what drives demand and to what extent that is affected by satisfaction with the service, e.g. to what extent the purchase is discretionary as opposed to having no choice over the service used (as is often the case with rail).

The review found that – although there is some “suggestion” of links between customer satisfaction and demand – there is nothing conclusive. The findings suggested: 

with respect to recency of experience: that the view seems to be that satisfaction increases with repeated purchases; but that after satisfaction becomes increased with repeated purchases, it then plateaus, i.e. inertia sets in.

© Association for European Transport and Contributors 2011



with respect to types of satisfaction: that there does not seem to be any one type of satisfaction that matters most for demand; instead, different sectors are likely to have different characteristics where satisfaction matters more



with respect to positive versus negative satisfaction: that there seemed to be evidence of an “S-shaped” relationship between positive and negative satisfaction. However, though not necessarily linked to the customer satisfaction literature, the widely observed phenomenon of loss aversion suggests that negative satisfaction would weigh more heavily than positive satisfaction



with respect to discretion and substitutability: that, according to standard consumer theory, where there are available substitutes and high discretion, the link between customer satisfaction and consumer demand is likely to be weaker.

The depth interviews support the findings of the literature review by demonstrating that – whilst customer satisfaction plays a very important role within their organisations – few are focusing upon, or have identified, a relationship between satisfaction and demand. This is despite there being a clear interest in the idea, and a perception that there “ought” to be a link between satisfaction and demand. The majority are using the Net Promoter Score (NPS), but whilst it is seen as a valuable measure for assessing performance levels (and, in particular, an easy way of communicating performance levels internally), only one organisation has undertaken any analysis to see whether or not changes in the number of promoters or detractors are being matched in any way by changes in customer numbers or revenue. However, the depth interviews did identify a small minority who have looked for – and established – links between satisfaction and demand, and through so doing have shown that – for their sectors at least – there is a greater link between low satisfaction and attrition than between high satisfaction and growth. In most sectors however, rather than satisfaction, the key factors driving demand are felt to be:   

brand strength competitive pricing high profile marketing.

Customer satisfaction is important, but as a tool for driving performance rather than predicting demand.


The study suggests that ATOC has done more than many to determine whether or not there is a link between customer satisfaction and demand and that these explorations have identified instances where this has seemingly been proved to be the case. However, perhaps more strongly, it has suggested that there are other factors which are of much greater importance in determining demand and which, in many cases, override satisfaction. This, to us, suggests that there would be more value in looking at how customer satisfaction fits within the development of specific strategies of the different organisations and in continuing to use it to drive and monitor performance, rather than in undertaking further work to determine the link in the rail sector. That said, it would be of value to maintain a „watching brief‟ on this and to test the water in this area further in the future should its measurement and value become widely practised and accepted across organisations as a whole. 1.

INTRODUCTION

1.1

Background Accent was commissioned by the Association of Train Operating Companies (ATOC) to examine to what extent a relationship between customer satisfaction and customer demand can be determined, both generally to set out a broad industrial context and specifically within the rail sector. One of the most popular areas of market research is in customer satisfaction. Of course, any organisation wishes to know if its customers are satisfied. But it is to be expected that the organisation will also care about the potential impact of satisfaction on the bottom line: will those satisfied customers continue to buy that product? Will dissatisfied customers move elsewhere? How will this impact upon potential new customer acquisition? Despite this “bottom line” being a clear potential motivation for customer satisfaction studies, there is not a large subfield in market research which focuses solely on the correlation between satisfaction and demand. The research programme comprised a literature review coupled with a small number of depth interviews with practitioners to draw on their expertise. Our key findings were reported and discussed extensively within the industry.


2.

METHODOLOGY Our approach to this study consisted of two elements: 

a literature review in which we examined published research sources to understand current theoretical thinking in this area



a series of eight telephone depth interviews with consultancies and corporate representatives in both transport and other sectors (such as retail, finance, energy and communications) to elicit information on their experience and understanding of this area.

The literature reviewed for the first element is shown in the bibliography. 3.

LITERATURE REVIEW FINDINGS

3.1

Introduction This section uses a literature review to examine the relationship between customer satisfaction and customer demand. Some studies are beginning to emerge on this subject and we discuss many of these below. There are many sceptical voices emerging though which challenge the conventional wisdom that satisfaction and demand should be highly correlated (Dixon, Freeman and Toman, 2010).

3.2

Customer Satisfaction and Demand: Basic Concepts Customer satisfaction is the general use of survey instruments which will typically ask about satisfaction with (a) particular aspect(s) of a product/service and/or overall levels of satisfaction. Customer satisfaction is more often than not subdivided into domains or types of satisfaction with various aspects or attributes of the product. An overall satisfaction measure might be either an aggregation of those subdivisions of satisfaction, or a new questionnaire item asking how satisfied the respondent is with a product/service overall. Customer demand is variously defined in the literature. Now, of course, the most reliable measure of demand is whether there has been an actual purchase. But it can be difficult to link survey data to actual purchasing behaviour, so instead surveys look into various categories. These include terms such as “repurchase”, “loyalty” or “price acceptance”, and these can be further subdivided into the self-reported past behaviours, future behaviours, or future intentions. In what follows, for purposes of convenience we will simply refer to demand as an umbrella terminology for any of the above, as there is nothing in the literature that indicates any substantial diversion from one such term to another.


Of course, holding all else constant, customer satisfaction should be an excellent predictor of demand. However, there are a number of mediating factors which influence the relationship. This is discussed below, but a brief illustrative list includes the following:    

length of “relationship” (i.e. having purchased the same product/service for a long period of time); switching costs (i.e. being in a contract where there might be a surcharge for leaving); price of product/service; presence and price of available substitutes.

Note that all but the first mediating variables would be predicted by standard economic theory. That is, switching costs, the price of good A versus the relative price of a substitute good B, all simply refer to the standard axiom of consumer theory that consumers will typically wish to save money for an equivalent utility gain. After that, there will be trade offs between utility and cost. However, there will be other mediating variables, including a sense of „loyalty‟ to a brand (wherein the consumer might not take a cheaper substitute because of this loyalty), which would not be predicted by standard consumer theory. Before proceeding further we need to look at another key factor regarding customer satisfaction which has received much attention of late, which is the importance of the feedback loop in implementing the lessons of customer satisfaction (Markey, Reichheld and Dullweber, 2009). It calls attention to the fact that customer satisfaction information should lead to performance management which aims to improve upon those areas of satisfaction which are weak, whilst also maintaining quality levels where satisfaction is strong. This should then be monitored to see where service improvements lead to increases in satisfaction, a kind of return on investment analysis. Although this is intuitive, it is often overlooked that the objective of a finance department of return on investment is not always perfectly matched to investment in those areas which produce greater customer satisfaction (Srivastava et al, 2006). Of course, this relationship – if followed through well – should lead to a strong indirect relationship between customer satisfaction and demand through the mediating influence of targeted improvements to performance. In what follows we will discuss how these various factors play mediating roles.


3.3

The Net Promoter Score One dominant trend in research and thought about the measurement of customer satisfaction has been the “net promoter score” (NPS). The idea of a net promoter is that the best proxy for both customer satisfaction and customer demand can be found in one question: namely, whether a person would recommend the product/service to a friend. Indeed, the article launching the concept from the Harvard Business Review in 2003 was called “The One Number you Need to Grow” (Reichheld, 2003). The question is: “How likely is it that you would recommend our company to a friend or colleague?" with respondents given a score of 1-10 to self-rate their likelihood of recommending. Respondents are classified into three categories: “promoters” who rate their likelihood of recommending as 9-10; “passives” who report their likelihood of recommending as 7-8; and “detractors” who score anywhere from 1-6. It should be noted that the NPS was developed not because it was seen as having significant methodological superiority over other survey instruments, but because of its simplicity. It was thought that having a single questionnaire item would be much more capable of being explained to management and other employees in comparison to multi-dimensional scorings of satisfaction. The model of customer satisfaction implicit here certainly has some validity. For example, it is easy to see that a strong indicator of being satisfied with a product/service would be to recommend it to someone else. However, one can also imagine recommending a product to a friend because it is inexpensive, not because one was satisfied necessarily as a customer. As a proxy of actual demand, the NPS also faces criticism. For example, being willing to recommend a good might reflect taste heterogeneity: one might think that a good/service is worth using for others, but not for oneself. And, even going beyond these basic conceptual points, researchers have found drawbacks to the NPS even in methodological terms. The key ones are outlined below. First, Hayes (2008) finds that NPS had no greater predictive validity than other loyalty measures, and indeed fared worse than many other loyalty measures. Hayes‟ own research and that of others is discussed in Hayes (2008: ch. 6). Nevertheless, although worrying that in some instances other measures might predict loyalty and demand slightly better than the NPS, it must be acknowledged that the NPS certainly has the virtue of ease of administration. And perhaps the overwhelming argument in favour of the NPS by its proponents is that the measure provides a lot of useful and largely valid information with only one simple question. Given the costs of survey administration this is a powerful argument in favour of the NPS.


It is also worth noting that there are a lot of organisations that have strongly embraced the approach (including many that we spoke to in our depth interviews). In the following sections we will move on to four questions that Accent has raised as being of relevance in understanding the relationship between customer satisfaction and customer demand. 3.4

The Role of Recency of Experience 

Q1: Customer satisfaction inevitably measures performance at a point in the past. How important is the recency of the measurement when looking at future demand? And, how frequently does the information need to be collected?

This first question concerns the role of recency of experience. One might naturally think that a highly satisfied customer whose last experience was five years ago would have, all else equal, less associated potential demand correlation than an equally satisfied customer whose last experience was five days ago. Research has suggested that satisfaction is higher among repeat customers for a good/service than for a first time user of a good/service (Homber, Koschate, and Hoyer, 2005). That is, loyalty, unsurprisingly, builds up over time. Some have suggested that there is a non-linear relationship between satisfaction and repurchase intentions (loyalty) which is mediated by time. Satisfaction is necessary to create loyalty in the first place; but after loyalty “kicks in” there is more inertia and reluctance to switch regardless of satisfaction levels (Mittal and Kamakura, 2001). We might say that satisfaction is necessary but not sufficient to determine loyalty. Time lapse, which is distinct from – though related to – frequency of purchase, is also shown to decrease the correlation between satisfaction and demand (Mittal and Kamakura, 2001). Also, perhaps obviously, goods which are frequently purchased (such as food or household items purchased during a weekly purchase) appear to be more susceptible to strong relationships between satisfaction and demand than infrequent purchases (such as cars or houses) (Chandon, Morwitz, and Reinhartz, 2005). 3.5

Which Type of Satisfaction Matters Most? 

Q2: Are particular variables more important in predicting demand than others? So does the overall satisfaction score need to be considered or is a variable such as „value for money‟ actually more influential in this regard?


In answering this question, we should note two facts at the outset. First, that there does not appear to be any results that indicate that particular types of satisfaction are correlated with demand in general. Second, we should note that this is very natural, given the context-dependent nature of the relationship between satisfaction and demand. That is, different types of goods in different sectors will have different valued forms of satisfaction and these will be potentially correlated with different types of demand. To suppose that the relationship between satisfaction and demand could potentially be the same for all goods would be to assume that all goods have the same bundle of characteristics and that utility for that set of characteristics is uniform among the population, which is obviously false. So, to conclude this section, there is nothing in the literature linking particular types of satisfaction to demand and perhaps we should not expect there to be one. 3.6

Positive vs. Negative Customer Satisfaction 

Q3: Does a negative customer experience have a greater impact than a positive experience? In other words, is demand reduced by poor satisfaction to a greater extent than it is increased by high satisfaction or do both impact equally?

This is certainly an interesting hypothesis given the focus on “loss aversion” in contemporary behavioural economic psychological research. This concerns the result that losses or negative experiences are weighed much more heavily than positive experiences. Translated into the relationship between satisfaction and demand, one might speculate that while very high levels of satisfaction do not necessarily ensure loyalty, any amount of dissatisfaction would be enough to ensure “dis-loyalty” i.e. failing to re-purchase. We have not found any such results in the literature, but we should note that this is certainly a plausible theoretical assertion. Somewhat related to this, Homberg, Koschate, and Hoyer (2005) found the result that the relationship of satisfaction to demand is non-linear, and is Sshaped. That is, decreases in satisfaction at the lower levels are linearly related to decreases in demand; increases in satisfaction at the higher level lead to increases in demand, but in the middle levels of satisfaction neither increases nor decreases of satisfaction appear to have much effect on demand (Homberg, Koschate, and Hoyer, 2005). However, again, it must be noted that this result is for a largely discretionary good, namely for clothing from a department store. Because of the availability of substitutes, this pattern might be very different from another satisfactiondemand relationship in which the choices are far more constrained.


3.7

Substitutability and Discretion in Transport Demand 

Q4: An important area to address will be the question of what drives demand and to what extent that is affected by satisfaction with the service e.g. to what extent is the purchase discretionary? The target market is likely to segment into those who have no alternative mode of transport or choice as to whether or not to travel and those whose rail travel is entirely discretionary. For this first group, customer satisfaction would be expected to be unlikely to impact on their demand. Conversely, those with a range of choices are likely to take into account their customer experience in deciding whether to travel by rail, so there could be a high correlation between customer satisfaction and demand.

To answer this question will require more than a “blanket” answer, and this is because the available alternatives for travel depend on what kind of traveller they are. So, in the rail industry there are well known customer segments such as daily commuters, regular but non-daily commuters, business-only travellers, leisure-only travellers, and so on. Now, given that transport is highly segmented in this way, perhaps because it is such a large and varied market, it makes speculation about the source of which types of satisfaction will be potentially consistently correlated with demand complex. That is, for someone who is primarily a business-only traveller, they may have a number of different substitutes when it comes to transport for leisure, as they have very little strong preference over forms of leisure travel, so a cheap substitute for a train to the Norfolk broads might be a trip to the local nature reserve which can be reached by bike i.e. no costly travel at all is necessarily required to define the substitutes. On the other hand, for what we might call flexible segments, the link between satisfaction and commuter demand will potentially be higher because they have a “take it or leave it” flexibility to their choices. However, for those segments which are more fixed in their market segment (e.g. commuters) the link between demand and satisfaction will be much lower. This is a natural outcome of having fewer choices because of one‟s market segment. Research does show that loyalty (i.e. demand) and satisfaction are more highly correlated in contractual or semi-contractual relationships versus cases where purchases are discrete and voluntarily recurring (Lemon, Barnett, and Winer, 2002) However, a contract limits the availability of alternatives by constraining the customer so this could perhaps be explained by the fact that the exit costs are often higher in contractual relationships and so a customer “doesn‟t see the point” in being dissatisfied if they cannot leave. So as an adaptive response the customer might just say that they are satisfied as a form of “dissonance reduction”.


4.

DEPTH INTERVIEWS

4.1

Introduction The aim of the depth interviews was to see what could be learned from the experience of other companies, in particular:  

whether in their business they had evidence of a satisfaction-demand link; and, if so, how they had used this information:  for example, was it used to predict demand, to influence demand or for some other use?

We spoke to senior contacts in eight organisations who had responsibility for producing or analysing satisfaction data. Two organisations in each of the following sectors were represented:    

Transport Communications Retail/Finance Energy

After explaining the objectives of the research to the respondents, and summarising the findings of the literature review, we asked each whether or not their organisations looked at the relationship between satisfaction and demand. If they did, additional questions (such as whether a negative customer experience has a greater impact than a positive experience) were also explored. The key findings are shown below. 4.2

Findings Prevalence of Means of Measuring/Existence of a Link Between Satisfaction & Demand All of the organisations we spoke to were measuring customer satisfaction and recognised the importance of doing so, but for the majority the primary role of their satisfaction scores was to drive better performance. However, there was a good deal of interest in the idea of satisfaction driving demand, albeit very little evidence of organisations having established – or looked to establish – this link from either an econometric or more simplistic perspective. Only one organisation was regularly monitoring the link between satisfaction and demand. Another was trying to correlate overall satisfaction with attrition (although not through econometric models). A third had attempted to link customer satisfaction with market share statistics and one other was intending to try and establish a link in the future.


However, in two of the instances where attempts had been made to establish a link, this had been with limited, or no, success. For example, one respondent had found that the attrition rate amongst those with the highest levels of satisfaction was no different to the attrition rate for those at the midpoint. In other words, increased satisfaction did not lead to increased loyalty. However, there was most definitely a correlation between losses and those with the lowest levels of satisfaction. The other had had no success at all – wherever a peak in satisfaction was found, there was typically a dip in market share! This was, however, partly put down to problems with the timing of satisfaction surveys and measures were going to be put in place to try and address this and see whether the figures became more aligned as a result. For most (including from the transport sector) it was their brand (or corporate reputation) which was perceived to be of greatest importance in driving demand. A strong brand and brand values were what were seen as key to retaining and gaining customers: if they improved perceptions of their brand then they expected to see improvements in customer loyalty. One respondent stated that they only saw 18% customer attrition where they had a good brand value, but 33% attrition where they had a poor brand value. Another noted that even when the media was being immensely critical of them, they retained their customers due to the strength of their brand. The fact that they still saw attrition where they had a good brand value leads into the other area which was seen as key to demand in most sectors – that of price. That is, a number commented that it didn‟t matter how good their satisfaction was, if someone came along with a cheaper price for a product or service which – on the surface at least – gave them the same as their current supplier, but for less money, it often didn‟t matter how good their customer service was, they would switch anyway. Another argued the same thing, but from the other direction, for one of their market areas where – because “price is king” – expending efforts on improving customer satisfaction was very much of secondary importance to their organisation. That said, in those areas where it was possible to build a relationship and purchases were less price-driven, satisfaction was seen as more important. There it was seen as “logical” that a happy customer was “good for business”. Price, in turn, was closely linked to marketing, with a couple noting that even if their – or a competitor‟s – satisfaction ratings were very/relatively low, if they hit the market with a really high profile marketing campaign, they inevitably grew their customer base. For energy companies in particular, the focus has just been upon getting as many customers as possible, rather than upon building a relationship with them. In transport another factor felt to be more important than satisfaction as a link to demand was whether or not customers have a choice of whether to travel or not. As there was not felt to be a strong link between satisfaction and demand for this transport respondent, it was not something that they were going to spend time on.


So, overridingly, other factors were seen – and being used – as predictors of demand rather than satisfaction. A couple of respondents were using their satisfaction scores in a very focused, targeted manner, testing out new products, messages or service strategies with key segments and then looking for changes in satisfaction as a result. If we relate this to one of the findings of the literature review, i.e. “Another key moderating factor between demand and satisfaction is “value anticipation”, namely, being able to predict what customers will value. Customer satisfaction only weakly predicts demand when firms do not correctly anticipate the changing tastes of its customers” this suggests that their use of customer satisfaction measures – whilst not directly related, by them, to demand – will be having an impact on demand. One energy respondent also felt that as regulation increasingly curtailed the use of door to door sales people, and as online sales became more prevalent, customer satisfaction would become more important in the battle to retain and gain customers. The Net Promoter Score (NPS) The majority of the organisations spoken to were using the NPS, and some had clear evidence that a good service or product led to customers saying “I‟m going to recommend that/you to my friends”. However, we noted in the literature review findings that the NPS was developed not because it was seen as having significant methodological superiority over other survey instruments, but because of its simplicity. And, indeed, the depths would seem to confirm this. Whilst it is seen a valuable way of highlighting a company‟s performance to senior management, it is not actually being linked to demand, i.e. only one of the organisations that we spoke to had undertaken any analysis to see whether or not changes in the number of promoters or detractors were being matched in any way by changes in customer numbers or revenue. Others described it as being used simply as a “temperature check” or as simply being “more actionable and easy to explain”. Relative Impact of Positive & Negative Satisfaction As previously stated, one respondent had noted a correlation between losses and those with the lowest levels of satisfaction, i.e. losses were higher amongst those who were least satisfied – but, external factors (such as low prices, high profile marketing campaigns etc) meant that losses were still seen amongst those who were very satisfied, but to a lesser degree. A second had also seen evidence that there was a bigger relationship between satisfaction and demand where a score dropped than where a score increased.


Another respondent felt (although didn‟t have specific evidence to support this) that negativity had a greater impact than positivity, in the sense that if customers were happy with a product or service they would not necessarily purchase more of it (or more from that supplier), but if they were unhappy with a service they were quite likely to go elsewhere. Specific Comments on the Relationship between Satisfaction & Demand in the Rail Sector Customer satisfaction is taken very seriously in the rail sector, with the National Passenger Survey (NPS) – not to be confused with the Net Promoter Score – interviewing over 50,000 passengers every year. However, the data gathered tends to be used as a continuous improvement tool – i.e. to determine what service aspects need to be improved to improve customer satisfaction. Recent research undertaken for the Department for Transport, Transport Scotland and the Passenger Demand Forecasting Council, aimed to mine the NPS data and generate indices of service quality to be included as potential drivers of rail demand in subsequent econometric analysis. Just as our respondents had had difficulties doing so (where they had attempted this link) so there were difficulties here, with the service quality indices remaining in the final econometric models for just two out of the 28 market segments. However, the authors (one of whom was interviewed for this study) point out that “this should not be interpreted as service quality being unimportant in explaining the demand for rail passenger travel, but may have arisen because of the relatively short time series of the data, and the relatively aggregate nature of the indices”. It was also the case that other independent variables proved to be powerful determinants so undermining the potential role that customer satisfaction could play. This may, indeed, be the case. However, respondents that we spoke to from non-transport sectors did struggle to find a link between satisfaction and demand in a sector such as rail. Most saw the rail sector as different to their sector – i.e. you have to have an energy supplier, you have to eat etc, but most people don‟t have to travel by train and then have a limited (or no) choice if they do. This could consequently be seen as making competitive edge, and aspects such as customer satisfaction, less important to the rail market, where it was felt that demand was more likely to be linked to the need to travel by the most convenient, fast or cost effective means than to satisfaction with the operator.


BIBLIOGRAPHY Chandon, P., Morwitz V.G. and Reinartz W. (2005) Do Intentions Really Predict Behavior? Self-Generated Validity Effects in Survey Research, Journal of Marketing, 69: 1-14. Dixon, M., Freeman, K. and Toman, M. (2010) Stop trying to delight your customers, Harvard Business Review, July – August: 116-122. Flint, D.J., Blocker, C.P. and Boutin Jr, P.J., (2010) Customer Value Anticipation, Customer Satisfaction and Loyalty: An Empirical Estimation, Industrial Marketing Management, (doi:10.1016/j.indarmarman.2010.06.34). Fornell, C., Rust, R.T. and Dekimpe M.G. (2010) The Effect of Customer Satisfaction on Consumer Spending Growth, Journal of Marketing Research, XLVII: 28-35. Hayes, B.E. (2008) Measuring Customer Satisfaction and Loyalty: Survey Design, Use, and Statistical Analysis Methods, WI: Quality Press, Milwaukee. Kandampully, J. and. Duddy R. (1999) Competitive Advantage Through Anticipation, Innovation and Relationships, Management Decision, 37(1): 516. Lemon, K.N., Barnet T.W. and Winer R.S. (2002) Dynamic Customer Relationship Management: Incorporating Future Considerations into the Service Retention Decision, Journal of Marketing, 66(1): 1-14. Markey, R., Reichheld, F. and Dullweber, A. (2009) Closing the Customer Feedback Loop, Harvard Business Review, pp. 43-47. Mittal, V. and Kamakura W.A. (2001). Satisfaction, Repurchase Intent, and Repurchase Behavior: Investigating the Moderating Effect of Customer Characteristics, Journal of Marketing Research, 38(1): 131-42. Homburg, C., Koschate N., and Hoyer W.D. (2005) Do Satisfied Customers Really Pay More? A Study of the Relationship Between Customer Satisfaction and Willingness to Pay, Journal of Marketing, 69 (April), 84- 96. Homburg, C., Hoyer W.D., and Koschate N. (2005) Customers‟ Reactions to Price Increases: Do Customer Satisfaction and Perceived Motive Fairness Matter? Journal of the Academy of Marketing Science, 33 (1), 36- 49. Luo, X. and Homburg C. (2007) Neglected Outcomes of Customer Satisfaction, Journal of Marketing, 71: 133-149. Reichheld, F. (2003) The One Number You Need to Grow, Harvard Business Review.


Seiders, K., Voss G.B., Grewal D., and Godfrey A.L. (2005) Do Satisfied Customers Buy More?: Examining Moderating Influences in a Retailing Context, Journal of Marketing, 69: 26-43. Slater, S.F. and Narver J.C. (2000) Intelligence Generation and Superior Customer Value, Journal of the Academy of Marketing Sciences, 28(1): 1207. Srivastava, R., Reibstein, D.J. and Joshi, Y.V. (2006) Linking Marketing Metrics to Financial Performance, Zyman Institute of Brand Science Technical Report, Emory University
