SUBMITTED BY
ADRĂŠ SCHREUDER Managing Director: Consulta Research In fulfilment of G-CEM Certification programme requirements
INDEX 1.
INTRODUCTION .............................................................................................................. 1
2.
OBJECTIVE ................................................................................................................... 1
3.
VOICE-OF-CUSTOMER AND CUSTOMER EXPERIENCE MEASUREMENT ................................................ 2
3.1.
Customer Satisfaction defined ........................................................................................... 3
3.1.1. CS viewed as an outcome of a consumption activity ................................................................. 3 3.1.2. CS viewed as a process .................................................................................................... 3 4.
DEFINING CUSTOMER EXPERIENCE ....................................................................................... 5
4.1.
Construct Definition of Customer Experience ......................................................................... 6
4.2.
Is Customer Experience the new Customer Satisfaction? ............................................................ 6
4.3.
Putting Customer Experience into Perspective ........................................................................ 7
5.
THE CONSULTA CUSTOMER EXPERIENCE MEASUREMENT MODEL AS A SOLUTION TO MEASURE ENTERPRISE WIDE CUSTOMER EXPERIENCE IN MULTI CHANNEL-PRODUCT-SEGMENT BUSINESSES ............................ 8
5.1.
Background & Overview of the Consulta CE Measurement Model .................................................. 9
5.2.
The Consulta Enterprise-wide 3D Model Solution for Multi Channel-Product-Segment Businesses .......... 14
5.3.
The Consulta CCEM as effective predictor of NPS ................................................................... 14
5.4.
Normality of modelled score ............................................................................................ 15
6.
CONCLUSION ................................................................................................................ 16
7.
BIBLIOGRAPHY .............................................................................................................. 17
LIST OF FIGURES Figure 1 – Customer Experience Management & Customer Feedback ....................................................... 2 Figure 2 – Historic Timeline of the Customer Satisfaction Construct ....................................................... 5 Figure 3 - Putting Customer Experience into Perspective ..................................................................... 8 Figure 4 – Consulta Conceptual model for Integrated Customer Experience Measurement ............................. 9 Figure 5 – The Consulta Conceptual Model Flow ............................................................................... 10 Figure 6 – The Consulta Instrument Development Process ................................................................... 10 Figure 7 – The Standardised Model Development Process .................................................................... 11 Figure 8 – The Principle Calculation of Modelled Scores ..................................................................... 11 Figure 9 – Confirmation-disconfirmation scale ................................................................................. 12 Figure 10 – Enterprise-wide application of the Consulta CCEM Methodology: An example of Retail Banks ......... 14 Figure 11 – NPS as dependent variable with a range of independent variables .......................................... 15 Figure 12 – Normality tests for customer experience index score .......................................................... 16
HOW TO MEASURE ENTERPRISE WIDE CUSTOMER EXPERIENCE IN MULTI CHANNEL-PRODUCT-SEGMENT BUSINESSES 1. INTRODUCTION The rationale for choosing the title for this assignment is based on the second option of assignment topics that was given i.e. “How to Solve A Particular Customer Experience Related Problem in Your Company” and the guideline that delegates should write in-depth on how to apply one of the six modules of the G-CEM Programme. My choice of topic is primarily related to the Voice-of-Customer Module and to a lesser extent Module 1 on Emotions and Loyalty. The Customer Experience Related Problem can be described as:
There is wide chasm between an academic-scientific and practitioner understanding of the foundations, concepts and theories underpinning the Voice-of-Customer construct;
The introduction and current popularity of “Customer Experience” is seen as the new “Customer Satisfaction” especially emphasizing the role of emotions more than ever;
If one has a balanced focus between the cognitive and emotional aspects of CEM, the complexity is further complicated by the fact that large service businesses such as banks function in multi-productchannel-segment environments. The challenge is how should CEM-scientists and practitioners approach this complexity and objectively measure the Voice-of-Customer.
2. OBJECTIVE The assignment objective has been formulated as – “The purpose of this Certificate Assignment is to give you an opportunity to apply what you have learned from the program, to conduct an in-depth study on the topic, and ultimately, to enhance your knowledge of customer experience management, now, and in the future.” The objective of this assignment is thus to describe how we in Consulta have addressed the problems (I prefer to call it challenges) described above.
Page 1 of 22
3. VOICE-OF-CUSTOMER AND CUSTOMER EXPERIENCE MEASUREMENT In Module 2 of the G-CEM programme the following illustration appears that describes CEM and Customer Feedback:
Figure 1 – Customer Experience Management & Customer Feedback
Source: Redrawn from G-CEM Module 2 – Voice-of-Customer The relevance of this graphic illustration for this assignment is taken from the understanding of what we are really talking about when we refer to Customer Experience Measurement and whether it is something totally different from the measurement of Customer Satisfaction. If one look at the slide that explains the history of behind VoC the impression is created that Service Quality and Customer Satisfaction was primarily measured in the 1970’s and that measurement domain progressed to Customer Experience Management is now seen as an improvement and replacement of the former. In the following short paragraphs I will do a short literature review of the construct of customer satisfaction measurement and how it relates to customer experience. The problem, which is still not sufficiently addressed, is how to objectively measure customer satisfaction and remedy those shortfalls quicker and better than your competitors. In an article from Parker & Matthews (2001) the authors provide an excellent overview of the construct of customer satisfaction. The following paragraphs are partly referenced from this article:
Page 2 of 22
3.1. Customer Satisfaction defined Satisfaction is derived from Latin where “satis” refers to “enough” and “facere” (faction) means “to do/to make”. Early interpretation and use of the word mostly focused on “some sort of release from wrong doing” and later found more acceptance as a “release from uncertainty”. The modern understanding of the term points to an evaluation between what was received and what was expected. Most scientific literature basically views Customer Satisfaction through from at least one of the two following broad perspectives, i.e.:
Customer Satisfaction viewed as an outcome of a consumption activity, and
Customer Satisfaction viewed as a process
3.1.1. CS viewed as an outcome of a consumption activity In this view the focus is on the nature of satisfaction rather than the cause: •
Emotion - satisfaction is the affective (emotional) element of product acquisition and/or consumption experiences, or an affective response to a specific consumption experience
•
Fulfilment - motivation theories state that either people are driven by the desire to satisfy their needs or achieving specific goals.
•
State - Oliver’s (1989) framework of four satisfaction states, where satisfaction is related to reinforcement and arousal: –
Low arousal = “satisfaction-as-contentment”
–
High arousal = “satisfaction as surprise” (positive / delight or negative / shock)
–
Positive reinforcement = “satisfaction-as-pleasure”
–
Negative reinforcement = “satisfaction-as-relief”
3.1.2. CS viewed as a process •
Concentrate on the antecedents to satisfaction rather than satisfaction itself. (This view has its origins in the discrepancy theory (Porter, 1961) and Contrast Theory from Cardozo (1965);
•
The most common interpretation of customer satisfaction is “a feeling which results from a process of evaluating what was received against that expected, the purchase decision itself and/or the fulfilment of needs/wants.”
•
The most “well-known’’ descendent of the discrepancy theories is the expectation-disconfirmation paradigm (Oliver, 1977, 1981).
In the mid 80's renewed attention was created in customer service, service quality and services marketing with the introduction of the SERVQUAL-instrument of Parasuraman, Zeithaml & Berry (1988, 1991 & 1994). Since its introduction in 1988 there has been a service revolution and a virtual turn around in the way business manager's approach and measure customer satisfaction and service quality.
Page 3 of 22
Apart from various other approaches, the main schools of thought on customer satisfaction measurement involve a range of different paradigms (adapted from the original summary by Erevelles & Leavitt, 1992): Table 1 – Customer Satisfaction Models & Paradigms Models of Customer Satisfaction
Basis assumptions & underlying principles
Expectations Disconfirmation model
Consumers pre-purchase expectations are positively or negatively disconfirmed resulting in satisfaction or dissatisfaction judgements respectively
Perceived Performance model
For some products, consumers' satisfaction judgements are primarily determined by the perceived product performance and are independent of initial expectations
Norms in models
Multiple Process models
Norms serve as reference points for evaluating brands and satisfaction judgements are based on the resulting confirmation/disconfirmation relative to these norms Consumers sometimes use multiple standards or multiple comparison processes which may take place either sequentially or simultaneously to arrive at satisfaction judgements
Attribution models
Consumers tend to search for causes for purchase successes or failures and attribute these successes or failures using a multidimensional schema. Consumers post-purchase responses depends on the attributions made
Affective models
In addition to cognitive factors, satisfaction is a function of consumers' post-purchase affective responses. Positive and negative emotions directly affect satisfaction judgements, complaint and Word-of-Mouth behaviour
Equity models
Net Measure Models
Multi-attribute National Models Hybrid models (Multi-attribute & net measures)
Page 4 of 22
Consumers' satisfaction judgements are based on equity interpretations derived from the costs an individual expends in the transaction and the anticipated rewards 1. NPS: Claimed to be the Ultimate Question that calculates the percentage of Detractors subtracted from the percentage of Promoters identified in a likelihood to recommend question 2. Customer Loyalty and the Secure Customer Index - Customers then could be grouped into subgroups or loyalty segments: Secure, Favourable, Vulnerable and At Risk Customers. Looks at concepts such as Earned Loyalty, Likelihood to Recommend, Likelihood to Repurchase, Overall Satisfaction and Preferred Company American Customer Satisfaction Index (ACSI), Swedish Customer Satisfaction Index (SCSI) The Consulta Customer Experience Measure – CCEM (discussed later in the paper)
The history of Customer Satisfaction Measurement can by illustrated by the following: Figure 2 – Historic Timeline of the Customer Satisfaction Construct
•
TQM of Edwards Deming, Zero Defect, Six Sigma
•
The Nordic approach (Grönroos 1984: Technical/Functional Model, Lethinen & Lethinen 1982 : Technical, Corporate, Interactive)
•
The North American Debate (PZB 1988, 1991, 1994: SERVQUAL (Gap-based measure, Familiar five quality dimensions, Cronin & Taylor 1992: SERVPERF - Performance only measure, Brown Churchill & Peter 1993: Better/worse than expected scale, Teas 1993: Evaluated Performance Model = gap between perceived performance & ideal amount of feature)
•
Sheth & Parvatiyar (1994) introduced Relationship Marketing Theory in the mid 90’s
•
Growth of CRM-systems and popularity from 1990-2002
•
NPS introduced by Reichheld in 2003
•
CEM era is born when Kirby & Wecksell & Janowski (2003) published a strategic analysis report for Gartner.
4. DEFINING CUSTOMER EXPERIENCE Starting with G-CEM publications, the “Defining CEM Whitepaper” with the foreword by Shaun Smith and coauthored by various G-CEM associates (2006) was use to select a relevant definition supplied by Paul Ward worth mentioning and giving context to this assignment. He writes: “So, what is the customer experience? My definition is that the customer experience is the provisional disposition a person has about your company based on all the information in his or her environment, and their interactions with you and your competitors, plus their reflections on what this means to them.” Second I found it instrumental to reference Sampson Lee in the G-CEM Whitepaper titled “Effective Experience Framework 2.0” (2010) where it is stated for customer experience to be effective it must be remembered, branded and contrasted. Colin Shaw (2008) from Beyond Philosophy defines CE as: “A Customer Experience is an interaction between an organisation and a customer. It is a blend of an organisations physical performance, the senses stimulated and emotions evoked, each intuitively measured against Customer Expectations across all moments of contact”
Page 5 of 22
The Harvard Business Review published an excellent article from Meyer & Schwager in 2007. The authors define customer experience as: “…the internal and subjective response customers have to any direct or indirect contact with a company. Direct contact generally occurs in the course of purchase, use, and service and is usually initiated by the customer. Indirect contact most often involves unplanned encounters with representatives of a company’s products, service or brands and takes the form of word-of-mouth recommendations or criticisms, advertising, news reports, reviews and so forth.” Verhoef, Lemon, Parasuraman, Roggeveen, Tsiros and Schlesinger (2009) provide a comprehensive overview of existing customer experience literature, including some recent definitions of customer experience. From this seminal piece of work by these renowned scientists I have taken the construct definition for this assignment:
4.1. Construct Definition of Customer Experience •
The customer experience construct is holistic in nature and involves the customer’s cognitive, affective, emotional, social and physical responses to the retailer.
•
•
This experience is created by: –
controllable elements - service interface, retail atmosphere, assortment, price,
–
uncontrollable elements - influence of others, purpose of shopping
Customer experience encompasses the total experience, including the search, purchase, consumption, and after-sale phases of the experience, and may involve multiple retail channels.
•
Three major focus areas: –
cognitive evaluations (i.e., functional values)
–
affective (emotional) responses
–
social and physical components
The next section in the building the background case for my proposed solution to the CEM Challenged addressed in the assignment deals with the relation between Customer Experience Measurement and Customer Satisfaction. Quite contrary articles and publications have appeared in various blogs and other internet sites, whilst I believe we should make a strong scientific case for answering this question.
4.2. Is Customer Experience the new Customer Satisfaction? Historically researchers have focused on measuring service quality and customer satisfaction (e.g., Parasuraman, Zeithaml & Berry, 1988; Verhoef, Langerak & Donkers, 2007), where satisfaction is a rating of the customer’s experience with the service outcome. Yet despite the recognition of the importance of customer experience by practitioners, the academic marketing literature investigating this topic has been limited.
Publications on
customer experience are mainly found in practitioner-oriented journals or management books and tend to focus
Page 6 of 22
more on managerial actions and outcomes.
The literature in marketing, retailing and service management
historically has NOT considered customer experience as a separate construct. Instead researchers have focused on measuring customer satisfaction and service quality. (Verhoef, Lemon, Parasuraman, Roggeveen, Tsiros, & Schlesinger, 2009) One reason for the apparently weak observed link between satisfaction and future behaviour may lie in the role of emotions. Previous studies have mostly emphasised cognitive aspects of satisfaction although the original definitions of the customer satisfaction construct (refer to page 3 where this was discussed in detail) clearly provided for both cognitive and affective elements of customer satisfaction. Practitioner-lead publications and some scientific papers in the last two years (2007 to 2009) provide a growing body of evidence that affective measures of satisfaction (which incorporate emotions) may be a better predictor of behaviour
As a cognitive
measure, satisfaction is more likely to be distorted over time than a measure that incorporates an affective component (emotions are more deep-seated & more stable over time).
Satisfaction measures should thus
include a combination of an evaluative (cognitive) and emotion-based (affective) response to a service encounter (Koenig-Lewis & Palmer, 2008). With such strong titles like “The One Number You Need to Grow” and “The Ultimate Question, Driving Good Profits and Growth” (Reichheld, 2003; 2006), it comes as no surprise that Reichheld’s Net Promoter® score (NPS) has on the one hand found rapid application in a variety of industries while on the other hand have also been challenged; and it seems, could still be in an evolutionary phase. Furthermore with the more recent debate questioning whether Reichheld’s Net Promoter metric is in fact the best standalone predictor of a firm’s growth (Morgan & Rego, 2006; Keiningham, Cooil, Andreassen & Aksoy, 2007; Molenaar, 2007; Hayes, 2008), the debate has unintentionally caused some traditional customer satisfaction researchers to naively shy away from using the question. The popularity of the Net Promoter Score has highlighted the use of net measures in customer experience measurement. Similar to other net measures, the “recommend” question is an excellent research question and was used extensively before Reichheld (2003) popularised it. But as a standalone question, the NPS has limitations.
4.3. Putting Customer Experience into Perspective The term Customer Experience Management is used within the broader context of Customer Relationship Management (CRM) – clearly seen in the view of Kirkby, Wecksell & Janowski (2003) when they say: “CEM is part of customer relationship management (CRM) and the natural extension of building brand awareness. Where brand gives the promise, CEM is the physical delivery of that promise and is vital in an economy where a brand is increasingly built on value delivered rather than product features”. This perspective is graphically illustrated below:
Page 7 of 22
Figure 3 - Putting Customer Experience into Perspective
5. THE CONSULTA CUSTOMER EXPERIENCE MEASUREMENT MODEL AS A SOLUTION TO MEASURE ENTERPRISE WIDE CUSTOMER EXPERIENCE IN MULTI CHANNEL-PRODUCTSEGMENT BUSINESSES Considering the preceding literature review and discussion regarding different measures of customer experience, it is obvious that no single measure can be used successfully in measuring the complex constructs of customer experience, customer satisfaction and customer loyalty. This assignment will describe quantitative proof of a Customer Experience Measurement Model used by Consulta Research that is proposed to address the challenge of objectively measuring enterprise wide customer experience in multi-channel-product-segment businesses. The quantitative proof comes from an extensive statistical meta-analysis on data collected over a time frame of more than 5 years, covering more than 1.5 million customer experience interviews. Survey results have been consolidated from proprietary customer experience surveys across a range of clients. For the purpose of this paper (and reliability) the data is limited to results from surveys in the financial services industry. Survey data was collected via telephonic, web-based and face-to-face interviews. To ensure minimal sampling observation errors, all interviews were subject to strict quality assurance processes, and advanced technology was used to capture data.
Page 8 of 22
5.1. Background & Overview of the Consulta CE Measurement Model As founder and ex-academic my post-doctoral research focus was to develop a conceptual cause-and-effect model that is illustrated in Figure 4 as an integrated customer experience measurement model.
The
development is the result of years of academic research combined with extensive experience regarding Customer Experience measurement across multiple industries. In order to summarise all the literature discussed earlier and all other practical research experience into a conceptual model, we have devise the following illustration (Figure 4) of the Consulta Conceptual model for Integrated Customer Experience Measurement. Figure 4 – Consulta Conceptual model for Integrated Customer Experience Measurement
In a flow format the detail components of the Consulta CCEM is shown – particularly the Conceptual causal relations between different components and the enhancement of the model by using a combination of multiattribute and net measurement (delight & failure).
Page 9 of 22
Figure 5 – The Consulta Conceptual Model Flow
The basis for the measurement is a structural model of customer satisfaction that incorporates the important antecedents (drivers) of satisfaction that will identify underlying service or product deficiencies (or strengths) that would otherwise not be identified and a proprietary algorithm for integrating net measures into this multiattribute model. It is important to note that the conceptual models shown above merely provide the conceptual platform when we are commissioned to measure a client’s customer experience.
The process of model
development follows a strict standardised methodology of instrument and model development – shown in two separate graphic illustrations below. Figure 6 – The Consulta Instrument Development Process
Page 10 of 22
Figure 7 – The Standardised Model Development Process
During the instrument development our Client Directors (project managers) will ensure that the customer experience measure will cover both the cognitive and affective components of customer experience. Through these processes Consulta always ensure that we develop a custom design instrument and CE model for our clients – addressing the challenge to objectively measure customer experience in complex enterprise wide environments. Since the calculation of the Customer Experience Index score incorporates the net effect of “failure” and “delight” ratings, it can be classified as a combined multi-attribute and net measure approach. The proprietary calculation algorithm is shown in Figure 8 below. Figure 8 – The Principle Calculation of Modelled Scores
Page 11 of 22
The integrated customer experience measurement, although resulting in a final index score, acknowledges the fact that a single value for an index might hide more that it reveals. It is important to be able to delve deeper into the results to enable the receiver to delve deeper than satisfaction. For this reason the customer experience index score is not reported in isolation as a single measure, but merely as the net result of multiple items, each of which contains detail results and offers valuable strategic information into the management of customer delight, loyalty, propensity to shift, service recovery, corrective improvement measures and consequence management. All the research instruments used for data collection followed the same basic layout and included sections corresponding to the components contained in the conceptual model for customer satisfaction measurement (Figure 4). However, in conjunction with the clients (several large corporations in the financial services industry) the first section of each questionnaire was developed to measure a specific channel’s value proposition to the client by means of a range of custom designed service attributes. The measurement of these attributes incorporates both customer perception and customer expectation, extracted by using the well-known confirmation-disconfirmation scale (Danaher & Haddrell, 1996), illustrated in Figure 9 this incorporates into customer satisfaction the important aspect of expectations, as discussed earlier by Anderson et al. (1994), and overcomes the longer and taxing application of double administration of perception and expectations measurements as earlier proposed by Parasuraman, Zeithaml & Berry (1985) in the famous SERVQUAL-approach. Figure 9 – Confirmation-disconfirmation scale
Much better than expected
Much worse than expected
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
Further, the research instruments include specific questions relating to the components of product quality, service quality, relationship quality and pricing, as contributing factors of customer satisfaction. The instruments were tested to ensure the two most critical cornerstones of good research:
Reliability, i.e. the extent to which a research instrument produces consistent results if repeated measures are made, by using Cronbach alpha reliability analysis. The international benchmark is 70%.
Validity, i.e. the extent to which a research instrument measures what it intends to measure, by using cumulative explained variance in principle factor analysis. The international benchmark is 60%.
Page 12 of 22
The final data set used for the meta-analysis contained each of the components, defined below in Table 2 for 704 separate customer experience studies forming part of the enterprise wide measurement of customer experience, for each of the financial institutions - each with a sample of at least 100 respondents or more. Table 2 – Customer Satisfaction & Loyalty Metrics for meta-data analysis Metric
Description
Weighted service attribute average score
A weighted average of the (unique channel) service attributes measured in terms of customer expectation
Service problems %
The proportion of respondents who indicated that they experienced a service problem within a certain time period. This is different from the proportion of respondents complaining (formally or informally) as measured in ACSI
Problem recovery %
The proportion of respondents who indicated that their service problem was recovered to their satisfaction
Overall delight %
The proportion of respondents who gave a 9 or 10 rating out of 10 for overall satisfaction. This is much more strict than the typical Top 2 Box metric calculated on a 5 point verbal scale or the “equivalent” “top four” boxes on the ten-point ACSI scale
Overall failure %
The proportion of respondents who gave a 0 or 1 rating out of 10 for overall satisfaction
Average overall satisfaction score
A simple average of overall satisfaction rated on a scale from 0 to 10
Customer satisfaction index score
The index score (out of 100) is a function of the following key elements: The underlying structural model – incorporating the weighted effect (estimated contribution) of the components product and/or service and/or relationship quality The basic calculation principle of being “rewarded” for positive ratings (e.g. overall delight ratings and problem recovery) and being “penalised” for negative ratings (e.g. service problem occurrence and overall failure ratings) – corresponding to the concept of a net measure
Net Promoter Score
Calculated according to the original definition of Reichheld (2003) the Net Promoter Score equals the % of promoters minus the % of detractors: * Promoters = respondents indicating a 9 or 10 out of 10 likelihood to recommend; * Detractors = respondents indicating a 0 to 6 out of a 0 - 10 likelihood to recommend
Page 13 of 22
5.2. The Consulta Enterprise-wide 3D Model Solution for Multi-Channel-Product-Segment Businesses
Figure 10 – Enterprise-wide application of the Consulta CCEM Methodology: An example of Retail Banks
Through our unique 3D Modelling capabilities (as shown in Figure 10 above) Consulta will develop an Enterprisewide Customer Experience Measurement solution for corporate clients that deal with customers through the multiple interactions of channels x product x segment (we call this the CPS-triangulation). The end result of this is a very realistic representation of the totality of customer experience in the business that will enable the rollup of the customer experience index scores to an executive corporate level – one score for the organisation as a whole, whilst having all the detail of any other CPS-triangulation in the roll-down.
5.3. The Consulta CCEM as effective predictor of NPS The customer satisfaction index score is calculated from various metrics, including the service attribute average score, percentage service problems and percentage problem recovery, as well as delight and failure ratings on overall satisfaction (top & low box ratings). The customer satisfaction model shows a negative correlation between service problems and overall satisfaction. The higher the percentage of service problems, the more
Page 14 of 22
index points will be subtracted in the calculation of the index score. Problem recovery can offset the negative effect of service problems and would earn "bonus points" being added to the index score. The Net Promoter Score, based on “likelihood to recommend" as indicator of loyalty, is calculated according to the definition of Reichheld (2003). The relationship between customer satisfaction and loyalty has been discussed and documented in detail. Now consider each of the net metrics used in calculating the customer satisfaction index score. Individually, as independent variables in modelling the Net Promoter Score, the graphs and correlation coefficients (R²-values in Figure 11) clearly show that the integrated index score with an R2 of 0.73 seems to be a strong predictor of the NPS. Figure 11 – NPS as dependent variable with a range of independent variables
5.4. Normality of modelled score Due to the more complex nature of its calculation, efforts to examine the statistical properties of net measures using a mathematical approach can be tedious and difficult. In these situations, computer-intensive simulation methods such as the bootstrap provide a solution to address questions concerning the probability distribution of the measures under consideration. Using four different studies from the financial sector data under consideration, as a starting point, the bootstrap method was applied to replicate 1000 bootstrap samples for each of these studies – each bootstrap sample consisted of 380 respondents chosen randomly with replacement from the survey data. The customer satisfaction index score is calculated for each of the 1000 bootstrap samples, providing 1000 simulated index scores, which can be plotted as histograms and normal probability plots, as shown below. The accuracy of the simulations increase as the number of bootstrap replications increase, but according to Davison & Hinkley (1997), 500 or more simulations are sufficient to reduce variability and provide accurate results.
Page 15 of 22
For all four studies, both the chi-square test and Shapiro-Wilk test did NOT reject normality of the customer satisfaction index score, which creates infinite opportunities regarding statistical inference for the index score (e.g. calculating confidence intervals and performing hypothesis testing). Although these results are based on only four studies, representing a small portion of the wide range of underlying models used to describe the results of the various studies, we believe that with additional research we will be able to establish similar results for the whole range of studies under consideration, and consequently establish normality for the customer satisfaction index score in general. Figure 12 – Normality tests for customer experience index score Variable: VoC1, Distribution: Normal
Normal Probability Plot of VoC1 (4 VoCs for normality graphs 4v*1000c) 4
Chi-Square test = 8.67399, df = 9 (adjusted) , p = 0.46790 20
3 18 2 Expected Normal Value
Relative Frequency (%)
16 14 12 10 8
1 0 -1 -2
6 -3
4
-4
2 0
-5 50
50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 Category (upper limits)
52
54
56
58
60
62
64
66
68
VoC1: SW-W = 0.998084051, p = 0.3196 Observed Value
Variable: VoC2, Distribution: Normal
Normal Probability Plot of VoC2 (4 VoCs for normality graphs 4v*1000c) 4
Chi-Square test = 5.06307, df = 7 (adjusted) , p = 0.65227 25
6. CONCLUSION
3
2 Expected Normal Value
Relative Frequency (%)
20
The relationship between customer satisfaction and loyalty, customer retention and economic performance has 15 1
been discussed and proven extensively; leading to the conclusion that continuously measuring customer 10
0
-1
satisfaction has enormous benefits for a company, and that knowing how to improve service holds the promise of -2
dramatic bottom line results. This paper presented an overview of previous research and literature regarding 5 -3
different forms of net measures such as the Net Promoter Score, Secure Customer Index and American Customer 0
-4
40 41 42 43 44 45 46 47 48 49 50 52 53 54 40 42 44 46 Satisfaction Index and their application in51customer experience measurement. Category (upper limits)
48
50
52
54
VoC2: SW-W = 0.998708772, p = 0.6945 Observed Value
Variable: VoC3, Distribution: Normal
Normal Probability Plot of VoC3 (4 VoCs for normality graphs 4v*1000c) 4
Chi-Square test = 8.15932, df = 7 (adjusted) , p = 0.31876
Without denying the fact that net measures has a role to play, the use of net measures as standalone questions 3 25
has been shown to have some disadvantages. Reporting net measures in context, supported by the multiple 20 2 Expected Normal Value
Relative Frequency (%)
items it contains, provides the opportunity to analyse the detail of all the different metrics constituting the net 1 15
measure. This will assist greatly in the need for root cause analyses and strategic/tactical direction, while the 0 net measure in itself can provide a top line measurement to track performance or even be effectively used as a 10 -1 “top-of-house” executive indicator. The quantitative data -2analysis of these measures can further be enriched by 5
qualitative questions, including verbatim descriptions of service problems that were experienced, suggestions on -3 improving service delivery, etc. 0 28
29
30
31
32
33
34
35
36
37
38
39
40
Category (upper limits)
41
42
-4 28
43
Variable: VoC4, Distribution: Normal 25 3 20
2 Normal Value
equency (%)
15
32
34
36
38
40
42
Normal Probability Plot of VoC4 (4 VoCs for normality graphs 4v*1000c) 4
Chi-Square test = 6.36535, df = 7 (adjusted) , p = 0.49779
Page 16 of 22
30
VoC3: SW-W = 0.998033823, p = 0.2971 Observed Value
1
0
44
Using longitudinal meta-data analysis of more than 1.5 million customer satisfaction measurement interviews, we have presented reliable correlations between the Net Promoter Score and an integrated customer satisfaction index score, as well as establishing statistical properties of these measures. The customer satisfaction index score can be classified as a combined multi-attribute and net measure approach, since it incorporates the net effect of “failure” and “delight” ratings, as well as service problems and the recovery thereof. Understanding that customers, as human beings, are complex by nature and accepting that the measurement of customer satisfaction involves the measurement of a complex construct, the use of multiple-item net measures has the advantage of providing insight into underlying drivers of customer satisfaction, while also offering a simple “top-of-house” dashboard metric that is simple to communicate.
7. BIBLIOGRAPHY Brown, T. J., Churchill, G. A., & Peter, J. P. (1993). Improving the measurement of service quality. Journal of Retailing, Vol. 69 (No. 1), 127-39. Cardozo, R. N. (1965). An experimental study of consumer effort, expectation and satisfaction’. Journal of Marketing Research, Vol 2, 244-249. Cronin, J. J., & Taylor, S. A. (1994). SERVPERF versus SERVQUAL: Reconciling performance based and perceptions-minus-expectations measurement of service quality. Journal of Marketing, Vol. 58, January, 125-31. Danaher, P. J., & Haddrell, V. (1996). A comparison of question scales used for measuring customer satisfaction. International Journal of Service Industry Management, 7(4), 4-26. Duboff, R., & Heaton, C. (1999). Employee loyalty a link to valuable growth. Strategy & Leadership, January/February. Erevelles, S., & Leavitt, C. (1992). A comparison of current models of consumer satisfaction/dissatisfaction. Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, Volume 5, 104-114. Gentile, C., Spiller, N., & Noci, G. (2007). How to Sustain the Customer Experience: An Overview of Experience Components that Co-create Value with the Customer. European Management Journal, 25 (5), 395–410. Grönroos, C. (1984). A service quality model and its marketing implications. European Journal of Marketing, Vol. 18(No. 4), 36-44. Kahneman, D., & Krueger, A. B. (2006, Winter). Developments in the Measurement of Subjective Well-Being. Journal of Economic Perspectives, Volume 20, (Number 1), 3-24. Keiningham, T., Cooil, B., Aksoy, L., Andreassen, T., & Weiner, J. (2007). The value of different customer satisfaction and loyalty metrics in predicting customer retention, recommendation, and share-of-wallet. Managing Service Quality, 17(4), 361-384. Kirkby, J., Wecksell, J., Janowski, W., & Berg, T. (2003). Strategic Analysis Report to Gartner: The Value of Customer Experience Management.
Page 17 of 22
Koenig-Lewis, N., & Palmer, A. (2008). Experiential values over time – a comparison of measures of satisfaction and emotion. Journal of Marketing Management, 24:1-2, 69-85. Kristensen, K., & Martensen, A. (1996). Linking satisfaction to loyalty and performance. ESOMAR Publication Series, 204, 159–169. Lehtinen, U., & Lehtinen, J. R. (1982). Service Quality: A Study of Quality Dimensions. Working Paper, Service Management Institute. Helsinki, Finland. Meyer, C., & Schwager, A. (2007, February). Understanding Customer Experience. Harvard Business Review, 117– 126. Morgan, N., & Rego, L. (2006, Sep-Oct). The Value of Different Customer Satisfaction and Loyalty Metrics in Predicting Business Performance. Marketing Science, 25(5). Oliver, R. L. (1977). Effects of expectation and disconfirmation on post-exposure product evaluations: an alternative interpretation. Journal of Applied Psychology, Vol. 62, 480-486. Oliver, R. L. (1981). Measurement and evaluation of satisfaction process in retail setting. Journal of Retailing, Vol. 57, 25-48. Oliver, R. L. (1989). Processing of the Satisfaction Response in Consumption: A Suggested Framework and Research Propositions. Journal of Satisfaction/Dissatisfaction and Complaining Behaviour, 1-16. Parasuraman, A., Zeithaml, V., & Berry, L. (1988). SERVQUAL: a multiple item scale for measuring consumer perceptions of service quality. Journal of Retailing, Vol. 64(Spring), 12-40. Parasuraman, A., Zeithaml, V., & Berry, L. (1991). Refinement and reassessment of the SERVQUAL scale. Journal of Retailing, Vol. 67 (No. 4, Winter), 420-450. Parasuraman, A., Zeithaml, V., & Berry, L. (1994). Alternative scales for measuring service quality: a comparative assessment based on psychometric and diagnostic criteria. Journal of Retailing, Vol. 70(No. 3, Fall), 201-230. Parker, C., & Matthews, B. P. (2001). Customer satisfaction: Contrasting Academic and Consumers' Interpretations. Marketing Intelligence & Planning, Vol 19/1, 38-44. Porter, L. W. (1961). A study of perceived need satisfaction in bottom and middle management jobs. Journal of Applied Psychology, Vol. 45, 1-10. Reichheld, F. (2003, December). The one number you need to grow. Harvard Business Review, 1–11. Reichheld, F. (2006). The Ultimate Question, Driving Good Profits and True Growth. Harvard Business School Press. Reichheld, F., & Sasser, W. (1990). Zero Defections: Quality Comes to Service. Harvard Business Review, 68(5), 105-111. Rucci, A. J., Kirn, S. P., & Quinn, R. T. (1998). The employee-customer-profit chain at Sears. Harvard Business Review, 76(1), 83-97. Schneider, D., Berent, M., Thomas, R., & Krosnick, J. (2008). Measuring Customer Satisfaction and Loyalty: Improving the Net-Promoter Score. . Poster presented at the Annual Meeting of the American Association for Public Opinion Research, . New Orleans. Shaw, C. (2008). The DNA of Customer Experience: How emotions drive value. Customer Experience Conference. London.
Page 18 of 22
Sheth, J. N., & Parvatiyar, A. (1994). Relationship Marketing: Theory, Methods, and Applications. Research Conference Proceedings, Center for Relationship Marketing, Emory University, , (p. 2). Atlanta, GA. Teas, R. K. (1993). Expectations, performance evaluation, and consumers’ perceptions of quality. Journal of Marketing, Vol. 57(No. 4), 18-34. Verhoef, P. C., Langerak, F. L., & Donkers, B. (2007). Understanding Brand and Dealer Retention in the New Car Market: The Moderating Role of Brand Tier. Journal of Retailing, 83(1), 97–113. Verhoef, P. C., Lemon, K. N., Parasuraman, A., Roggeveen, A., Tsiros, M., & Schlesinger, L. A. (2009). Customer Experience Creation: Determinants, Dynamics and Management Strategies. Journal of Retailing, 85 (1), 31-41. Zeithaml, V. A. (2000). Service quality, profitability, and the economic worth of customers: What we know and what we need to know. Journal of the Academy of Marketing Science, 28, 67-85.
Page 19 of 22