Measuring Respondent Information : Attitudes, Satisfaction, Loyalty And Behavior

 Information From Respondents

All marketing decisions involve recognizing alternatives to, and making predictions of, the behavior of market participants. Marketing decisions ultimately hinge, in whole or in part, on a prediction of the behavior of consumers, industrial users, marketing intermediaries, competitors, and, at times, the government. Whether the decision is to introduce a particular new product, raise the price of an existing product, change distribution channels, or determine an advertising budget, the solution involves forecasting the behavior of one or more groups of market participants.

We now consider the most frequently studied variables used to predict behavior: demographics, attitudes, intentions, satisfaction and loyalty. The information used to predict behavior is grouped into behavioral correlates (past or intended behavior) and nonbehavioral correlates (demographics, attitudes, lifestyles, values).

Behavioral Correlates

Behavioral correlates address the question of what past behaviors or intended behaviors predict future behavior.

Past Behavior

Past behavior is widely used as a predictor of future behavior. Each of us relies heavily upon our knowledge of others’ past behavior in our everyday relationships with our family, friends, and associates. When we state that we “know” someone well, we are implicitly saying that we believe we are able to predict that person’s behavior in a wide range of social and interpersonal situations. In economics applications, we examine past trends, seasonal averages, and cyclical patterns for forecasting.

Regardless of the nature of the variable or variables to be forecasted, a basic premise of using past behavior to predict future behavior is that there is a relationship between the two that, to some extent, is stable. Recognizing that the degree of stability is sometimes difficult to determine, and that we do not always understand the underlying causal relationships, we nonetheless must believe that there is some continuity and stability in the behavior patterns of people.

A typical consumer brand purchase study would concern itself in part with determining such facts as what brands have been used, the last brand bought, where and with what frequency purchases are made, what the exposure to company advertising has been, and similar aspects of past behavior.

Information about past behavior toward products is often classified into three categories: acquisition, use, and possession. Within each of these behavioral areas, we collect detailed information about who, what, when, where, how much, and in what situation the behavior occurs. This data produces detail useful in understanding product consumption patterns. The particular study’s requirements will dictate which of these types of information will be most useful. Table 9.1 shows the requirements for a study on tomato juice to determine, among other things, whether a new type of container should be developed. Often, such information comes from secondary sources, as previously discussed.

Intended Behavior

Intentions may be defined as presently planned actions to be taken in a specified future period of time. What more logical method of predicting the future behavior of respondents could be used than determining their intentions? After all, intentions are selfpredictions of behavior, and thus, if obtained from people whose behavior we want to predict, would seemingly be the most direct and reliable method of prediction

Intentions are relevant and commonly sought as predictors of behavior. However, consideration of our own experiences in terms of what we have planned to do vis-à-vis what we have actually done later should raise some questions concerning the reliability of intentions as a predictive tool. The question “What will you do?” must always be answered conditionally.

Table 9.1 Past Behaviors: A Study of Tomato Juice Usage Patterns

	Acquisition	Use	Possession
Who	Who in your family usually does theshopping	Who in your family drinks tomato juice
What	What brand of tomato juice did you buy last time?	What dishes do you cook or prepare with tomato juice?	What brands of tomato juice do you now have on hand?
When	How long has it been since you last bought tomato juice?
Where	Do you usually do your shopping at a particular storeor supermarket? (if Yes, where?)
sssHow Much	What size can of tomato juice do you usually buy? About how often do you buy tomato juice? About how many cans do you buy at a time?	About how much tomato juice does your family drink in a week? For which purpose does your family use more juice?	Do you now have any unopened cans of tomato juice on hand? (If Yes, about how many cans?)
Usage Situation		How does your family use tomato juice? Beverage         􀆑 Cooking          􀆑 Both                􀆑 Beverage         􀆑 with friends	How do you store tomato juice after it is opened? Can         􀆑 Bottle      􀆑 Plastic     􀆑 Other       􀆑

The degree of assurance that planned actions will be translated into actual actions varies widely depending on circumstances and future happenings, many of which are outside the respondent’s control.

The results of a study of expected and purchase rates of a few products and services are shown in Table 9.2. Researchers collected intentions data from a consumer panel sample using a 0 to 10 scale to measure purchase probabilities. Verbal definitions were assigned to each point on the scale. A 10 was defined as “absolutely certain of buying” and a 0 as “absolutely no chance of buying.” The definition of a 5 was given as “five chances out of ten of buying,” and the other points between 1 and 9 inclusively were similarly defined.

Table 9.2 Intentions Fulfilled and Not Fulfilled During a 60 - Day Period

Product/Service	Intentions-Based Expected Purchase Rate %	Purchase %	Difference
Ride local public transportation Purchase tax-sheltered investment Purchase stereo system Take a trip on cruise ship Purchase new automobile	22.5 11.4 17.6 4.2 14.3	21.7 7.2 15.6 3.7 14.1	-0.8 -4.2 -2.0 -0.5 -0.2

Expected purchase rates were calculated as the average purchase probability for each item. The actual rate was determined by reinterviewing the panel members 60 days later to find out what purchases they had actually made.

Intentions to buy are often conditioned by judgments and expectations of future events or future situations, as well as by our past experiences. Such variables as expected change in financial status, price expectations, general business forecasts, and predictions of need all contribute to the final intention decision. Since each of these is subject to variation, it seems plausible to suppose that the intender views them as such and that his or her stated intention is based on a subjective probability of purchase. This supposition is supported by the fact that intentions data with assigned probabilities have generally proven to be more accurate than those expressed in “either/or” form. 

Past experiences can often be inferred from a customer’s level of satisfaction. Increasingly, customer satisfaction surveys fill a critical role in many firms’ customer relationship management (CRM) systems (James, 2002). Customer satisfaction relates to intentions as it can significantly affect the repurchase decision of consumers.

Nonbehavioral Correlates

So far we have discussed how people’s past behaviors and their intentions are correlates of what they will do. We now need to examine the nonbehavioral correlates that are useful for predicting their future behavior.

Socioeconomic Characteristics

How is information on the social and economic characteristics of respondents useful for forecasting what people will do? The answer can be readily suggested by an illustration. The Radio Corporation of America (RCA), when introducing color television in the 1950s, was very much interested in the age, income, educational, and occupational composition of the market. They judged that the initial market willing to pay for a premium priced color television set would be families proportionally higher in income and educational levels and older, than either the black-and-white set owners or the population as a whole. These judgments were subsequently confirmed by a study of early purchasers of color sets. This information was useful for both pricing and promotional

Exhibit 9.1 Measuring Intentions

decisions, since an association was found to exist between families with these characteristics and the purchase of color televisions sets.

In studies of consumers where there is a basis for believing that such associations might exist, researchers obtain information on one or more socioeconomic characteristics; those most frequently obtained are income, occupation, level of education, age, sex, marital status, and size of family. While socioeconomic characteristics are by far the most widely used bases for classification of consumers, other bases exist. Among these are attitudes, preferences, personality traits, perceived risk, and such measures of actual buying behavior as amount purchased and brand loyalty. It may be interesting to know, for example, that owners of SUVs show different personality traits than owners of other vehicles; such knowledge will be useful in marketing automobiles, however, only if it can be used to develop and evaluate appeals for each type of buyer. Doing so can enhance segmentation, positioning, and market targeting.

In general, the identification of consumer segments is useful in marketing so long as the following four statements apply :

1.	Substantial : The value in terms of potentially increased sales makes it worthwhile to do so.
2.	Differentiable: There are practical means of differentiating purchase behavior among market segments. There is homogeneity within and heterogeneity between segments.
3.	Operational : There is a cost effective means of reaching the targeted market segment
4.	Responsive : The differentiated market segments respond differentially to marketing offerings tailored to meet their needs

Two commonly used and widely accepted classifications of consumers are by stage of the life cycle and by lifestyle. One classification identifies the household lifecycle groups as the following :

1.	Young unmarrieds
2.	Young marrieds, no children
3.	Young marrieds, with children, youngest child under six
4.	Older marrieds, with children, youngest child six or older
5.	Older marrieds, with children maintaining separate households
6.	Solitary survivors or older single people

Some writers have expanded the number of stages by distinguishing in the last two stages whether a person is in the labor force or retired. See Wells and Gubar (1966) and Wagner and Hanna (1983) for more detailed explanations of the life-cycle concept and marketing research 

The life-cycle stage has obvious implications with respect to purchases associated with family formation (furniture, appliances, household effects, and housing) and addition of children (food, clothing, toys, expanded housing). Other, less obvious relationships exist as well. New-car buying reaches its peak among the older married couples whose children have passed the age of six. A second stage of furniture buying takes place when children begin to date and have parties at home. Dental work, travel, and purchases of insurance are examples of service purchases associated with the life cycle.

Lifestyle has a close association with membership in a social class. It is a basis for segmenting customers by values, activities, interests and opinions, as well as by income. These differences tend to be expressed through the products bought and stores patronized, as well as the area in which one lives, club membership, religious affiliation, and other means. The media used for expression are often either consciously or subconsciously, symbolic representations of the class to which the person perceives he or she belongs (or would like to belong). When used with personality traits, lifestyle variables form the basis of psychographic classification, as illustrated in Exhibit 9.2. An illustration of psychographic questions is shown in Table 9.3.

As an example, let us consider the life styles of the Harley Owners Group (HOG). By examining a group of questions used in a segmentation study of values and motorcycle use, we find a divergent group of lifestyles that have embraced the mystique of owning a Harley.

Exhibit 9.2 Harley Owners Group (HOG) Classification by Psychographics

Psychographic research has suggested many different segmentation schemes. Such schemas represent interesting demographic and product markets, and provide a much more colorful description of the group as a whole as well as the diversity within.

Research by William Swinyard (1994a, 1994b) suggests that Harley-Davidson owners are a diverse group consisting of six distinct segments with very different motorcycling lifestyles:

-	Tour Glides find the appeal of motorcycling in long distance touring. They like riding long distances, use their bike both for touring and everyday transportation, are more interested in the comfort of their motorcycle than its speed, prefer riding with a passenger, and wear a helmet. More than the average Harley rider, Tour Glides are religiously traditional, have somewhat old-fashioned tastes and habits, are disciplinarians with their children, like reading, and feel they live a full and interesting life. They are less ambitious than others, and are distinctively unattracted by social gatherings and danger.
-	Vanilla Dream Riders. The Vanilla Dream Riders are more interested in the dream of motorcycling than in motorcycling itself, and are otherwise just plain vanilla a relatively undistinguished group. This is the largest, oldest, wealthiest, and among the best educated segment of Harley owners, who have the newest motorcycles yet ride them least, and spend little in accessorizing them. You see the Dream Riders taking riding on short trips around town (often by themselves), wearing a helmet, and riding a stock bike. They are distinctively unaffiliated with the “live to ride” ethic, and receive relatively little psychic satisfaction from riding. Their motorcycle is merely a possession, having no real place as a “family member.” They are conservative in their moral values, marital roles, and daily behavior.
-	The Hard Core segment is on the fringe of society, and identifies with the stereotypical biker subculture. They are the youngest, next to  least well-educated, and certainly the poorest, yet spend nearly 50 percent more than any other segment on accessorizing their motorcycles. Virtually all are blue-collar workers. In relative terms, Hard Core members are much more likely than others to feel like an outlaw, and believe  people would call them and their friends “dirty bikers.” Note, however, that they still only “slightly agree” that these lifestyles describe them well. More than others, the Hard Core likes to be outrageous, enjoys danger, favors legalizing marijuana, and embraces the ethic of “eat, drink, and be merry, for tomorrow we die.”
-	Hog Heaven finds great psychic and spiritual satisfaction in owning and riding a Harley. More than others, these riders feel like an “old wild west cowboy” and closer to nature when they ride. They have many motorcycle friends, and when group riding,  they feel the group becomes “one.” They do not like helmets, and feel cars are like a “cage.” This segment is distinctively mechanically inclined, likes to work on their motorcycles, but spends little on accessories. They have old-fashioned tastes and habits, and read relatively little. They are less likely than others to believe in a life after death, but often think about how short life really is.
-	Zen Riders too find solace and spiritual satisfaction, but find it in motorcycling itself, and escape life’s stresses in doing so. They include the highest percentage of married riders, but otherwise are typical of Harley owners in most demographic characteristics. More than others, Zen Riders find motorcycling fulfilling in many dimensions: their motorcycle seems alive and they like all motorcycles, from dirt bikes to four-cylinder Japanese motorcycles. Zen Riders are more impulsive and are willing to take chances and to run risks. They believe they are more ambitious than other segments. They like to party, and have trouble relaxing in everyday life, but they are “modern” husbands and are opposed to legalizing marijuana.
-	Live to Ride rides more than any other segment, and motorcycles represent a total lifestyle to them. They “ride to live and live to ride.” Members of this, the smallest segment, are most likely to have bought their motorcycle new, and ride it the most by a wide margin. They simply love riding. More than other Harley owners, they use their bike for everyday transportation, enjoy riding long distances, and use their bike for touring. They find riding to be a magical experience, and motorcycling is a total lifestyle to them. If they did not have family, members of this segment would quit their jobs and take off. They agree with an “eat, drink, and be merry” premise, like to create a stir, like danger, and get lots of satisfaction from their hobbies. They care little about their appearance and tend not to believe in a life after death.

Life style segmentation offers significant implications for marketing and advertising to these segments in terms of not only better communicating with each group, but in creating the product configuration and accessories that appeal to each group.

A common designation of social classes is the one originally used by Warner and others (1960) that designates the familiar upper, middle, and lower class designations, each divided into upper and lower segments. Thus, the Warnerian classification results in six classes, ranging from the UU (upper upper) down through the LL (lower lower). Somewhat newer is the value and lifestyles (VALS) schema, which classifies the American public into: survivors, sustainers, belongers, emulators, achievers, I-am-me’s, experientials, societally conscious, and integrateds (Mitchell, 1983).

Exhibiy 9.3. 

Psychographic Measurement Questions for Three Motorcycle Owner Segments: Most and Least Frequent Motorcycle Lifestyle Descriptors

Although less direct and more subtle than life-cycle stage in its effect on overt buying behavior, there can be little question that an upper-middle-class household will show more similarity in purchasing and consumption patterns of food, clothing, furniture, and housing to another upper-middle-class household than it will to a blue-collar, upper lower-class household. The media to which the managerial-professional, upper-middleclass family is exposed, and the appeals to which it responds, are also likely to be closer to those of other managerial-professional families than to those of the blue-collar family. Similarly, on the basis of VALS, a processor and packager of tofu may find that because “experientials” have a greater appreciation for natural things, they are heavier users of tofu. The marketer can then direct the product at this lifestyle group.

Another approach to values and lifestyles is the List of Values (LOV) developed by Kahle (1983). These values fit well into the qualitative research discussed in Chapter 6, (means-end laddering to produce hierarchical value maps and ZMET analysis) :

􀁸 Self-respect

􀁸 Security

􀁸 Warm relationships with others

􀁸 Sense of accomplishment

􀁸 Self-fulfillment

􀁸 Being well-respected

􀁸 Sense of belonging

􀁸 Fun and enjoyment in life

Classification of consumers is vital if we are to learn more about consumer behavior and utilize this information to develop more efficient marketing techniques. But caution is needed to ensure that managers do not segment too finely and use categorizations when they should not. Given the new software available that simplifies using such advanced techniques as neural networks, latent-class models, fuzzy or overlapping clustering, and even occasion based segmentation, there will be a tendency to use a backhoe when a shovel is more appropriate. The technology in this area is far ahead of other aspects of the marketing process, and ahead of most managers’ needs (Freeman, 2001). Advanced statistical techniques for consumer classification purposes are discussed in later chapters.

Although the discussion thus far has focused on consumers, similar classification requirements exist and are used in studies of industrial users and marketing intermediaries. Comparable characteristics of these firms include sales volume, number of employees, and the type of products manufactured or handled.

Extent of Knowledge

Prediction of what actions respondents will take is often aided by knowing “how much they know.” This is especially so when making advertising budget and media allocation decisions, where consumers’ choices are strongly affected by their levels of awareness and extent of knowledge of the product and its attributes. To illustrate questions measuring consumer awareness, a study of homeowners examined their knowledge of conventional and variable-rate mortgages. One question sought knowledge of interest rate differences :

Attitudes and Opinions

Investigators in the behavioral science fields of psychology, sociology, and political science have made extensive studies of attitudes and opinions over a wide range of subject areas. The study of people’s behavior in business and economic contexts is also a behavioral science. As such, it has been a natural consequence that marketing research has adopted, adapted and applied many techniques, including attitude-opinion studies to obtain information applicable to the solution of marketing problems.

The terms attitude and opinion have frequently been differentiated in psychological and sociological investigations. A commonly drawn distinction has been to view an attitude as a predisposition to act in a certain way, and an opinion as a verbalization of an attitude. Thus, a statement by a respondent that he or she prefers viewing Blue-Ray DVD’s to HDTV color television programs would be an opinion expressing (one aspect of) the respondent’s attitude toward high-definition television

When used to predict actions that the respondent will take, this distinction between attitude and opinion rapidly becomes blurred. Since the major purpose of attitude-opinion research in marketing is to predict behavior, this differentiation is, at best, of limited usefulness. We shall therefore use the terms interchangeably. Attitude research in marketing has been conducted with the use of both qualitative and quantitative techniques. In either form, researchers encounter problems that are more severe than those involved in obtaining any other type of descriptive information discussed. Despite these problems, which we will discuss in later chapters in some detail, attitude-opinion research has been widely used to provide information for choosing among alternatives. Its greatest use has been in the areas of product design (including packaging and branding) and advertising. Other uses have been in selecting store locations, developing service policies, and choosing company and trade names. In fact, attitudes and opinions are central in customer satisfaction studies.

Measuring Attitudes

Expectancy Value Measures of Behavioral Intention (BI) and Attitudes (A)

Expectancy value models were first developed in the 1960’s as a method of predicting behavioral intentions (a precursor of actual behavior). Expectancy value models use attitudes to predict behavioral intention (intention to try, purchase, recommend, or re-purchase a product or service). This methodology has become a mainstay of marketing research and is found to perform well in predicting both consumer behavior and consumer satisfaction/dissatisfaction.

The Expectancy value model uses attitudes and beliefs in a mathematical formulation that is read as follows :

-	Behavior (purchase of brand X) is approximated by Behavioral Intention (intention to purchase brand X), which in turn is approximated by the Overall Attitude toward brand X.
-	The Overall Attitude toward brand X equals the sum of all salient attitudes about brand X, weighted by how important each attitude is in the purchase decision process.
-	Normative beliefs are added to this compensatory evaluative process by considering the norms surrounding an attitude ( My friends insist I should buy an environmentally friendly car that gets 50 mpg), and applying weights reflecting the buyer’s motivation to comply with the norm (Will they praise me for buying a 50 mpg car? – or Do I really care about getting 50 mpg?).   where:
	B        =	Behavior
	BI       =	Behavioral Intention, which is measured using a question such as “how likely are you to buy a Toyota Prius sometime during the next year?" with a five or seven point Likert or semantic differential scale labeled "definitely will purchase" and "definitely will not purchase" at the endpoints.
	Ao           =	Overall Attitude toward the object (brand, product, service or company)
	w1, w2 =	Weights that indicate the relative influence of the overall attitude toward the object and the normative influence to purchase the product
	∑ai * bi =	the sum of individual attitudes toward the object, weighted by importance. The overall attitude is formed by the multiplicative product of ai (the person’s affective evaluation - liking of the attribute i), and bi (here defined as the importance of attribute i in the purchase decision). The sum is taken over the k attributes that are defined as salient in the purchase decision.
	∑nbi * mci =	The normative component of the decision process that asks “should the norms surrounding this attribute be important to me?” This is computed as the multiplicative product of nbi (the norms governing attitude i), and mci (the motivation of the respondent to comply with those norms).
	Attitude (ai*bi) ai is the affective (liking) component of the evaluation of attribute i. A representative measure of ai would be using the context "In terms of buying a Toyota Prius”, to evaluate “Gets 50 miles/gallon”. The evaluation would use a five or seven point scale with endpoints ranging from “Poor” to “Excellent”, or “Not at all Desirable” to “Very Desirable”. This is found on the right side of Figure 9.1.   bi - the importance of attribute I in the context of behavior B. This is sometimes measured as the probability of attribute i being associated with brand X. At other times it is measured as the importance of attribute i in achieving behavior B. Using the context of purchasing a Toyota Prius, the attribute would be “Gets 50 miles per gallon” could be rated on a seven point scale with endpoints labeled "Very Unlikely" and "Very Likely", or as in the example below, is measured as “Not at all Important” to “Very Important”. This is found on the left side of Figure 9.1.

Figure 9.1 Expectancy Value Importance-Performance Rating Scales

∑nbi * mci = The overall normative component of the decision process. This is computed as the multiplicative product of nbi (the norms governing belief i), and mci (the motivation of the respondent to comply with those norms) see Figure 9.2.

Figure 9.2 Normative Rating Scales

The normative component is often considered to be absorbed in the overall attitude component and depending on the application, is often ignored.

The expectancy value model is an aggregation of attitudes and beliefs about the most important attributes that predict behavior. This compensatory type of decision model results in a powerful approach to measuring customer attitudes and predicting customer behavior, and is well used in research. It is not surprising that many popular methodologies, including conjoint analysis, are theoretically based on this measurement model.

Measuring Satisfaction

Customer satisfaction is the most common of all marketing surveys and is part of the “big three” research studies in marketing that include market segmentation and concept testing. Customer satisfaction has been defined as the state of mind that customers have about a company and its products/services when their expectations have been met or exceeded over the lifetime of product or service use (Cacioppo, 2000, p. 51). The positive attitudes that lead to customer satisfaction usually, in turn, lead to customer loyalty and product repurchase. But measuring satisfaction is not measuring loyalty. The following are typical measures of overall satisfaction :

1. Overall, how satisfied are you with (brand name)?

2. Would you recommend (brand name)?

3. Do you intend to repurchase (brand name)?

According to William Neal (2000), these questions are usually measuring the same thing—satisfaction with the product or service. Satisfaction is a condition that is necessary, but not sufficient to predict loyalty. Customer satisfaction can be measured by traditional surveys, by using comment cards, and for business-to-business situations by field or call reports, to name just some of the methods (Cochran, 2001). Many companies routinely measure satisfaction in consumer tracking studies (see Exhibit 9.4).

The attitudes and opinions of prospective buyers will clearly affect expectations and the resulting satisfaction in a purchase decision. Consequently, the marketing manager should be as well informed as possible about both the nature of their relevant attitudes and opinions and the intensity with which they are held. Subaru of America, for example, has a program that includes a Purchase Experience Survey and a Service Experience Survey that goes out to all customers who have purchased a Subaru or had it serviced. Findings from these surveys are reported back to the dealer, who then acts on them. In addition, Subaru sends out a Product Survey to a sample of new Subaru owners every year. This survey examines the quality of the product and whether new owners are satisfied with the performance, fit, and finish of their new vehicle.

Measuring satisfaction and building a satisfaction survey requires at least a basic knowledge of the satisfaction measurement literature, combined with knowledge and experiences with the company’s customers. We will first introduce the theoretical and methodological underpinnings of satisfaction research by first defining the concept of customer satisfaction and how satisfaction is used in business. Next, different satisfaction survey measures are discussed and presented. Finally, the components of a satisfaction survey are presented, along with sample satisfaction survey questions. The Qualtrics.com survey library contains sample satisfaction questionnaires, questions and scales. This discussion of satisfaction research provides the basis for understanding how to measure satisfaction and why the suggested should be used.

Exhibit 9.4 Common Ingredients of a Customer Satisfaction Survey

Product Use

Frequency of product use

Primary use location

Primary precipitating events or situations for product use or need

Usage rates and trends

Product Familiarity

Degree of actual product use familiarity

Knowledge (read product information, read product label, etc.)

Knowledge and Involvement with product and the purchase process

Awareness of other brands

Reasons for original product purchase (selection reasons)

Primary benefits sought from the product

Product Evaluation

Attribute evaluation matrix: (quality, price, trust, importance, performance, value)

Perceived benefit associations matrix

Importance, performance

Identification of primary benefits sought

Comparison to other brands (better, worse)

What is the best thing about the brand, what could be done better

Message and Package Evaluation

Packaging size, design

Advertising Promise, message fulfillment evaluation

Value Analysis

Expectation of price

Expectation of relative price (full price, on sale)

Current price paid

Satisfaction Measurements

Overall Satisfaction

Reasons for Satisfaction Evaluation

Satisfaction with attributes, features, benefits

Satisfaction with use

Expected and Ideal Satisfaction-Performance Measures

Likelihood of recommending

Likelihood of repurchasing

What Is Customer Satisfaction?

Customer satisfaction measures indicate how well a company’s products or services meet or exceed customer expectations. These expectations will reflect many aspects of the company’s business activities including the actual product, service and company. Customer satisfaction measures will tap the customer’s lifetime of product and service experience.

Why is Customer Satisfaction So Important?

Effective marketing focuses on two activities: retaining existing customers and adding new customers. Customer satisfaction measures are critical to any company, be it consumer or B2B, because customer satisfaction is a strong predictor of customer retention, customer loyalty and product repurchase.

When to Conduct Customer Satisfaction Surveys

When to measure customer satisfaction depends on the kind of product or service provided, the kinds of customers served, the number of customers served, the longevity and frequency of customer/supplier interactions, and the intended use of the results.

Three very different approaches may each produce meaningful and useful results that are appropriate for specific situations, uses and needs :

-	Post Purchase Evaluations reflect the satisfaction of the individual customer at the time of product or service delivery (or shortly afterwards). This type of satisfaction survey is typically used as part of a CRM (Customer Relationship Management System) and focuses on securing a long term relationship with the individual customer
-	Periodic Satisfaction Surveys provide an occasional snapshot of customer experiences and expectations, and are conducted for specific groups of consumers on a periodic basis.
-	Continuous Satisfaction Tracking is often part of a management initiative to assure quality is at high levels over time, and can involve post purchase evaluations or a succession of periodic satisfaction surveys conducted on a regular basis (daily, quarterly or monthly basis). Satisfaction feedback is obtained from the individual customer at the time of product or service delivery (or shortly afterwards). Satisfaction tracking surveys are.

Satisfaction surveys are developed to provide an understanding of customers' expectations and satisfaction. Satisfaction surveys typically require multiple questions that address satisfaction with different dimensions of the product or service concept. That is, satisfaction measurement includes measures of overall satisfaction, satisfaction with individual product and service attributes, and satisfaction with the benefits recorded as a result of purchase. Satisfaction measurement is like peeling away layers of an onion-each layer reveals yet another deeper layer, closer to the core.

Each of the three methods of conducting satisfaction surveys are helpful in obtaining customer feedback for assessing overall accomplishments, degree of success, and areas for improvement. Ulwick (2005) advocates the combination of satisfaction and importance to identify unfulfilled opportunities in the marketplace. In this case, performance gaps not being adequately served by the products in the market are identified, evaluated for feasibility, and targeted for development.

Customer Satisfaction Survey Measures

Customer satisfaction surveys will often include several different multiple measures of satisfaction, including :

􀁸 Overall measures of customer satisfaction

􀁸 Affective measures of customer satisfaction

􀁸 Cognitive measures of customer satisfaction

􀁸 Behavioral measures of customer satisfaction

􀁸 Expectancy value measures of customer satisfaction

We will now explain each in detail. Because general measures of customer satisfaction usually involve product fulfillment, we will discuss product use scenarios focusing on where and how the product is used

Satisfaction Measurement: Overall Measures of Satisfaction Like when we have a great food experience at a favorite local restaurant, elevated levels of customer satisfaction usually leads to customer loyalty and product repurchase. But measures of satisfaction are different than measures of loyalty. Satisfaction measurement questions typically include items like those found in Exhibit 9.5 and like the following :

1.	An overall satisfaction measure (emotional) : Overall, how satisfied are you with “Yoni fresh yogurt”? This question reflects the overall opinion of a consumer’s satisfaction experience. It is noteworthy that we can meaningfully measure attitudes towards a product that a consumer has never used, but we cannot measure satisfaction for a product or brand that has never been used. There is no experience base for such a measure of satisfaction.
2.	A loyalty measure (affective, behavioral) : Would you recommend “Yoni” to your family and friends?
3.	A series of attribute satisfaction measures (affective and cognitive): How satisfied are you with the “taste” of Yoni fresh yogurt? How important is “taste” in your decision to select Yoni fresh yogurt? Satisfaction and attitude are closely related concepts. The psychological concepts of attitude and satisfaction may both be defined as the evaluation of an object and the individual’s relationship to it. The distinction is that here, satisfaction is a "post experience" state representing the emotional affect produced by the product’s quality or value.
4.	Intentions to repurchase (behavioral measures) : Do you intend to repurchase Yoni fresh yogurt ? Satisfaction can influence post-purchase/post-experience actions other than usage (these other actions might include word of mouth communications and repeat purchase behavior). Additional post-experience actions might reflect a heightened level of product involvement that in turn results in increased product or information search activity, reduced trial of alternative products, and even changes in preferences for shopping locations and choice behavior.

As shown in Figure 9.3, customer satisfaction is diagrammatically shown to be influenced by perceived quality of product and service attributes, associated product features and benefits, and is moderated by the customer’s expectations regarding the product or service. The researcher may want to define and develop measures for each of these constructs that influence customer satisfaction.

Affective Measures of Customer Satisfaction

Attitudes toward a product can be developed as a result of product information or any experience with the product, whether perceived or real. Affect (liking/disliking) is best measured in the context of product attributes or benefits. Again, it may be meaningful to measure attitudes towards a product or service that a consumer has never used, but it is not meaningful to measure satisfaction when a product or service has not been used.

Figure 9.3 Building a Customer Satisfaction Survey

Exhibit 9.5 Sample Satisfaction Measures from the Qualtrics Question Library

Cognitive Measures of Customer Satisfaction

A cognitive element is defined as an appraisal or conclusion that the product was useful (or not useful), fit the situation (or did not fit), exceeded the requirements of the problem/situation (or did not exceed), or was an important part of the product experience (or was unimportant). Cognitive responses are often specific to the situation for which the product was purchased and specific to the consumer’s intended use of the product, regardless if that use is correct or incorrect

Behavioral Measures of Customer Satisfaction

It is sometimes believed that dissatisfaction is synonymous with regret or disappointment while satisfaction is linked to ideas such as, "it was a good choice" or "I am glad that I bought it." When phrased in behavioral response terms for a future or hypothetical behavior, consumers indicate that “purchasing this product would be a good choice” or “I would be glad to purchase this product.” Often, behavioral measures reflect the consumer’s past experience with individuals associated with the product (i.e. customer service representatives) and the intention to repeat that experience.

Expectations Measures of Customer Satisfaction

As might be expected, many different approaches to measuring satisfaction exist in the consumer behavior literature. Parasuraman, Ziethaml and Berry (1998) refined their earlier work to identify five generic dimensions of a service satisfaction scale called SERVQUAL: tangibles, reliability, responsiveness, assurance and empathy. SERVQUAL has become the standard for service quality measurement, but it should be recognized that satisfaction dimensions seem to vary depending on the application (high tech vs. health care vs. leisure services, etc.). None the less, this important work is often used for the measurement of customer satisfaction in a service environment.

Figure 9.4 Product and Service Questions

Scott M. Smith and Gerald S. Albaum, An Introduction to Marketing Research, © 2010

A slightly different diagnostic approach to satisfaction measurement is one that examines the gap between the customer's expectations of performance and the consumer’s actual experience. This “satisfaction gap" involves measuring both perception of performance and expectation of performance along specific product or service attributes dimensions. This approach can be applied to an individual product or service experience, a brand, or even a complete product category, as proposed by Ulwick (2005).

Again we see that customer satisfaction is largely a reflection of both expectations and experiences that the customer has with a product or service. Expectations may reflect past product experiences, but will also reflect the purchase evaluation process that occurs when shopping for a product or service. For example, when we make major purchases, we research the product or service, collecting information from the advertising, salespersons, and word-ofmouth from friends and associates. This information influences our expectations and ability to evaluate quality, value, and the ability of the product or service to meet our needs.

Customer Expectations that Influence Satisfaction

Customers hold both explicit and implicit performance expectations for attributes, features and benefits of products and services. The nature of these expectations will dictate the form and even the wording of a satisfaction questionnaire.

Explicit expectations are mental targets for product performance, such as well identified performance standards. For example, if expectations for a color printer were for 17 pages per minute and high quality color printing, but the product actually delivered 3 pages per minute and good quality color printing, then the cognitive evaluation comparing product performance and expectations would be 17 PPM – 3 PPM + High – Good, with each item weighted by their associated importance.

Implicit expectations represent the norms of performance that reflect accepted standards established by business in general, other companies, industries, and even cultures. An implicit reference might include wording such as “Compared with other companies…” , or “Compared to the leading brand…”

Static performance expectations address how performance and quality for a specific application are defined. Performance measures for each application are unique, though general expectations relate to quality of outcome and may include those researched by Parasuraman or others such as: accessibility, customization, dependability, timeliness, and accuracy, tangible cues which augment the application, options, cutting edge technology, flexibility, and user friendly interfaces. Static performance expectations are the visible part of the iceberg; they are the performance we see and -- often erroneously – are assumed to be the only dimensions of performance that exist.

Dynamic performance expectations are about how the product or service evolves over time and includes the changes in support and product or service enhancement needed to meet future business or use environments. Dynamic performance expectations may help to produce “static” performance expectations as new uses, integrations, or system requirements develop and become more stable.

Technological expectations focus on the evolving state of the product category. For example, mobile phones are continually evolving. Mobile service providers, in an effort to deal with the desire to switch to new technology phones, market rate plans with high cancellation penalties for switching providers, but with liberal upgrade plans for the phones they offer. The availability of low profile phones with email, camera, MP3, email, blue tooth technology, and increased storage will change technology expectations as well as the static and dynamic performance expectations of the product. These highly involving products are not just feature based, but have expectations that enhance perceptions of status, ego, self-image, and can even invoke emotions of isolation and fear when the product is not available.

Interpersonal expectations reflect the relationship between the customer and the product or service provider. Person to person relationships are increasingly important, especially where products require support for proper use and functioning. Expectations for interpersonal support include technical knowledge and ability to solve the problem, ability to communicate, time to problem resolution, courtesy, patience, enthusiasm, helpfulness, assurance that they understood my problem and my situation, communication skills, and customer perceptions regarding professionalism of conduct, often including image, appearance.

Situational Expectations

In building a customer satisfaction survey, it is also helpful to consider reasons why prepurchase expectations or post-purchase satisfaction may or may not be fulfilled or even measurable. In some situations the following conditions may occur :

1.	Expectations may not reflect unanticipated service attributes;
2.	Expectations may be quite vague, creating wide latitude of acceptable performance and expected satisfaction;
3.	Expectation and product performance evaluations may not be cognitive, but instead sensory, as in expectations of taste, style or image. Such expectations are not only difficult to evaluate and understand, but may change over time.
4.	The product use may attract so little attention as to produce no conscious affect or cognition (evaluation), and result in measures that are meaningless satisfaction or dissatisfaction measures;
5.	There may have been unanticipated benefits or consequences of purchasing or using the product (such as a use, usage situation, or feature not anticipated with purchase);
6.	The original expectations may have been unrealistically high or low;
7.	The product purchaser, influencer and user may have each been a different individual, each having different expectations.

When fulfilled, expectations result in customer satisfaction (or when expectations are not fulfilled, result in dissatisfaction and complaining behavior). Each of the above discussed types of expectation should be considered in the context of the unique research project to determine if special consideration is warranted during questionnaire development. The research study may also benefit from consideration of expectations related to perceived quality and value.

Satisfaction Measurement: Perceived Quality Measures

Perceived quality is often measured through three measures: overall quality, perceived reliability, and the extent to which a product or service is able to fulfill the customer’s needs. Customer experiences that result in attributions of quality are the single greatest predictor of customer satisfaction.

Satisfaction Measurement: Perceived Value Measures

Perceived value may conceptually be defined as the overall price divided by quality or the overall quality divided by price. Perceived value is measured in many ways including overall evaluation of value, expectations of price that would be paid, and more rigorous methodologies including the Van Westendorp pricing analysis and conjoint analysis. Note that these are Qualtrics advanced option question types, and that the Qualtrics online survey university contains white papers and tutorials about these topics

The consumer behavior literature has long shown that price is a primary indicator of quality when other attributes and benefits are relatively unknown. However when repeat purchases are made in some product categories, price may be reduced in importance. This may reflect a decision simplification strategy that makes shopping or purchase decisions easier, or may reflect increased customer loyalty.

Satisfaction Measurement: Customer Loyalty Measures

Customer loyalty reflects the likelihood of repurchasing products or services. Customer satisfaction is a major predictor of repurchase, but is strongly influenced by explicit performance evaluations of product performance, quality, and value.

Loyalty is often measured as a combination of measures including overall satisfaction, likelihood of repurchase, and likelihood of recommending the brand to a friend. For a common measure of loyalty might be the sum of the following three questions measured on an agreement scale (Strongly agree – Strongly disagree) :

1. Overall, how satisfied are you with [brand]?

2. How likely are you to continue to choose/repurchase [brand]?

3. How likely are you to recommend [brand] to a friend or family member?

Measures of loyalty are generally not perfect predictors of the brand switching that often occurs in an actual consumer purchase situation. Switching may occur because of out of stock situations, inconvenience of going to a store that has the preferred brand, or the lack of commitment to the brand. This latter condition has been addressed by Gallup in the identification of four emotional states that are indicative of the degree of commitment and brand loyalty (Table 9.3).

Table 9.3 Brand Loyalty Measures

Construct	Measure
Confidence in the brand	[Brand] is a name I can always trust. [Brand] always delivers on what they promise.
Integrity of the brand	[Brand] always treats me fairly. If a problem arises, I can always count on [brand] to reach a fair and satisfactory resolution
Pride in the brand	I feel The to be a [brand] customer. [Brand] always treats me with respect.
Passion for the brand	[Brand] is the perfect company for people like me. I can't imagine a world without [brand].

Figure 9.5 Gallup Customer Loyalty Example

The application of these loyalty scales enables the differentiation of brands, companies and even industries according to how well they are “engaging” their customers (Figure 9.5).hat

This approach to measuring customer loyalty opens the door for the creation of performance benchmarks and for management decision making. An improved customer experience can be tracked and improved as it instills confidence in the brand through an image of trust, confidence, and fairness. Customer loyalty is created by front-line employees who interact with the customer. It must always be remembered that all companies face the challenge of meeting increasing customer expectations. Dealing with expectations in a timely and effective manner increases customer loyalty and retention.

Measuring Customer Expectations and Satisfaction

Expectations are beliefs that are measured as the (likelihood or probability that a product or service (with certain attributes, features or characteristics) will produce certain outcomes (benefits-values). As we have discussed, these expectations are based on the consumer’s reservoir of affective, cognitive and behavioral experiences. Expectations are seen as related to satisfaction and can be measured in the following ways :

1.	As an Importance-Value relationship that fulfills the expectations for the product/service. This can include overall expectations or expectations for the individual product attributes most important in the purchase decision or experience.
2.	Affect-Satisfaction Expectations: The (liking/disliking) of the product/service or of the attributes that comprise the product/service.
3.	Fulfillment of Expectations: the expected level of performance vs. the desired expectations. This is “Predictive Fulfillment” and is a respondent specific index of the performance level necessary to satisfy.
4.	Expected Value from Use: Satisfaction is often determined by the frequency of use. If a product/service is not used as often as expected, the result may not be as satisfying as anticipated. For example a Harley Davidson motorcycle that is purchased with high expectations for a fun and adventurous lifestyle, but sits in the garage, an unused year subscription to the local fitness center/gym or a little used season pass to the local ski resort or amusement park may produce more dissatisfaction with the decision to purchase than with the actual product/service.

Summary

Customer satisfaction is the most important element of business success and is a key measure of fulfillment of business strategies, including those involving segmentation and concept development.

In this chapter our discussion has provided an introduction to satisfaction measurementthat illustrates the theoretical and methodological underpinnings of satisfaction research for business. The different satisfaction survey measures discussed were presented as components of a satisfaction surv ey, along with sample satisfaction survey questions. The Qualtrics survey library contains sample satisfaction questionnaires, and the Qualtrics question library contains sample individual questions. These templates will assist you in further understanding what measures should be included in your satisfaction research and why those measures are necessary and of value

References

Cochran, C. (2001, November). Customer satisfaction: The elusive quality. Quality Digest, 45–50.

Freeman, 2001 Freeman, L. (2001, September 24). Small, smaller, smallest: New analytical tools can slice markets too thin. Marketing News, 35, 1ff.

James, D. (2002, May 13). Better together. Marketing News, 36, 15–16.Kahle (1983)

Mitchell, A. (1983). The nine American lifestyles. New York: Warner Books

Neal, W. D. (2000, June 5). When measuring loyalty satisfactorily, don’t measure CS. Marketing News, 34, 19.

Parasuraman, A, V. Zeithaml, and L. Berry."SERVQUAL: A Multiple-Item Scale for Measuring Customer Perceptions of Service Quality," Journal of Retailing, Spring 1988, pp. 12-40

Ulwick, A. (2005) What Customers Want: Using Outcome-Driven Innovation to Create Breakthrough Products and Services, New York: McGraw Hill

Wagner, J., & Hanna, S. (1983, December). The effectiveness of family life cycle variables in consumer expenditures research. Journal of Consumer Research, 10, 281 – 291.

Wells, W. D., & Gubar, G. (1966, November). Life cycle concept in marketing research. Journal of Marketing Research, 3, 355–363.