Information About Consumer Problems By the Policy-Delphi Procedure: a New Method For Evaluating the Marketing System

Solveig Wikstrom, Stockholm University
ABSTRACT - The basic premise of this paper is that the marketing system does not perform adequately as seen from a consumer's perspective. For various different reasons the sellers' offerings do not sufficiently match the consumers' needs and wants. The performance of the socioeconomic system to which the marketing system belong depends on adequate feedback. Existing information mechanisms are inadequate. In this paper a new method id presented for gathering information about the consumer situation for use as feedback. By this method a panel of informed persons assess the consumer situation in several rounds according to the so called Delphi procedure. By this method the gathering of subjective and objective information is made through one approach. Experience reported in this paper indicates that further development of the method should be carried out.
[ to cite ]:
Solveig Wikstrom (1978) ,"Information About Consumer Problems By the Policy-Delphi Procedure: a New Method For Evaluating the Marketing System", in NA - Advances in Consumer Research Volume 05, eds. Kent Hunt, Ann Abor, MI : Association for Consumer Research, Pages: 596-602.

Advances in Consumer Research Volume 5, 1978      Pages 596-602


Solveig Wikstrom, Stockholm University


The basic premise of this paper is that the marketing system does not perform adequately as seen from a consumer's perspective. For various different reasons the sellers' offerings do not sufficiently match the consumers' needs and wants. The performance of the socioeconomic system to which the marketing system belong depends on adequate feedback. Existing information mechanisms are inadequate. In this paper a new method id presented for gathering information about the consumer situation for use as feedback. By this method a panel of informed persons assess the consumer situation in several rounds according to the so called Delphi procedure. By this method the gathering of subjective and objective information is made through one approach. Experience reported in this paper indicates that further development of the method should be carried out.


In Sweden as in most industrialized countries intense activities are going on in the consumer policy area to strengthen the consumer's position on the market. Examples of important policy actions in Sweden in the 70's are:

1971: The Consumer Ombudsman, the market Court Act, the Marketing Practices Act, the Act Prohibiting Improper Contract Terms, the Door-to-Door Sales Act.

1972: The Food Office, the Food Act.

1973: The National Board for Consumer Policy, the Act on Products Hazardous to Man and Environment.

1974: The Consumer Sales Act.

1975: Decision about expanding the local consumer advisor activities.

1976: The New Marketing Practices Act.

There are more activities under consideration as a result of recent government commissions.

If we take these new laws and norms to be well motivated, they can be seen as indicators of deficiencies in the marketing system. Consumers as a whole or categories of consumers seem to have problems which consumer policy activities are supposed to counteract. What are the effects of these activities for the performance of the marketing system? What are the consequences for the consumers?

As a matter of fact there are no answers to such fundamental questions about the consumers' situation as:

--How often do consumers meet problems in their planning of purchases, buying and consumption of goods and services?

--What has been the trend over time?

--How serious are these problems?

--Are There product and service areas which have more problems than average?

--How much of the problems are the consumers themselves aware of?

In reality very little effort has been made to get this kind of feedback. To some extent this may be due to the lack of workable information mechanisms.

This paper presents some of the work carried out last year developing and implementing a new method for evaluating some aspects of the marketing system from a consumer's perspective (Wikstr÷m, Eliasson 1976) This was done by having a panel of informed people assess the extent to which consumers meet problems and what implications these problems have for the consumers. The assessments were carried out in several rounds following the so called Delphi method.

In this case the Delphi assessment of the consumer situation was used as a method for evaluating what has been done up to date to improve the marketing system. I particularly want to emphasize that this new approach by no means fills the information gap pointed at above. It however does reveal new aspects of the problem not obtainable by existing methods and the experience gained indicates that further development of the method should be carried out. As a frame of reference for the presentation of the new approach a short overview is given of methods used at present.


From the literature discussing criteria variables for evaluating the marketing system from a consumer perspective the following matrix can be arranged (Renoux, 1973, Lingoes, Pfaff, 1972, +lander, 1976, Andreasen, 1975)


Subjective data basis means statements based on individual consumers' evaluations of their situation and the state of their environment. Objective data basis implies that data used are not directly dependent on the consumers' evaluations of the situation. Information about the micro marketing system includes aspects such as individual products and firms, shopping possibilities, the buyer situation in the outlets, the consumption and repair of products etc. The macro marketing system covers the total system. The evaluation concerns its functions from a holistic view.

The advantage of information based on subjective indicators is that these reveal what the consumers themselves perceive as their problems and as deficiencies in their marketing environment. These types of indicators can not, however, be used as a decision basis either by public policy regulators or by business planners without analysis and evaluation. One has to clarify what, if any, are the alternatives available within the present technology and the resulting costs to consumers. The consumer should be informed as to whether or not there are alternatives. There are serious "consumer problems" which can not be dealt with within the marketing system. In this case the government must step in and provide some solution outside the marketing system, e.g., in public transportation, in housing, or in education.

The main weaknesses of the subjective indicators are their dependence on consumer aspiration levels and their inability to generate information about those problems which the consumers may not be conscious of.

Evaluations based on objective indicators, on the contrary, have the advantage of recognizing problems which the consumers are unaware of. Trends in this kind of data can also be used as indicators for future problems.

There are however validity problems in using objective indicators. You really do not know how close a relationship there is between the indicators and the state you want to measure. Besides that you do not know if the set of indicators used is sufficient for measuring a certain consumer state.

The conclusion is that either subjective or objective indicators separately are sufficient as information instruments. Both types of indicators are needed, which implies that the information gathering becomes extensive and costly. Besides that there are still unfilled gaps.

The new method of evaluating the consumer situation by using assessments by a panel of "experts" combines the subjective and objective approach. By treating the gathering of subjective and objective information through one approach we are able to put them on a relative or comparative basis in terms of what we believe is now an efficient process for producing relevant information. First the panels' professional background implies good insights into what is called objective indicators. Second these people are also well informed and knowledgeable consumers themselves. Therefore, they should have qualifications to integrate subjective and objective aspects in their assessments.

The panel assessment in the actual method study has been directed towards micro aspects, but in a systematic way so that the marketing system in large has been covered. This way micro and macro aspects have been integrated.

Referring to the matrix in Figure l, the position of the Delphi assessment is between subjective and objective and between micro and macro and thus integrating these aspects.


In Russell Ackoffs terms (1962) the concept problem implies a tension between what a person should achieve and in fact does achieve, i.e., a deficiency in goal fulfillment. What a person should achieve has however to be kept within what is technologically possible and attainable. With the restriction to what is attainable the utopias are out of question. This definition is illustrated in Figure 2.



Defining consumer problems in this manner presupposes knowledge about consumers' achieved states, knowledge about attainable states and values about what is desirable. Differences in these components lead to differing views about how widespread and urgent the consumer problems are.

When defining consumer problems it is possible to distinguish between a number of concepts related to this problem concept in the way indicated in Figure 3.

(1) Problems from consumer complaints. Spontaneous dissatisfaction measured by number of consumer complaints.

(2) Problems from surveys. All problems which the consumers are aware of and admit. This concept is used in surveys of consumer satisfaction/ dissatisfaction.

(3) Unconscious problems as well. Relevant subsets of problem areas (1) and (2) as well as some specific problems not admitted or perceived by the consumers. We know that is some situations the consumer denies his problems. He rationalizes his poor decisions in order to reach internal harmony, acknowledging information which supports his decision and avoiding stimuli which are in conflict with it. These mechanisms are explained by the theory of dissonance (see e.g. Festinger, 1964). The consumer would not admit these kind of problems.

Then there are problems which the consumer is unaware of for another reason. He lacks information about his possibilities to attain better states. Better alternatives already exist, but the consumer does not know about them.

(4) Problems caused by lack of dynamics. Better Alternatives are attainable with present technology including better organization of the marketing systems. Because of lack of dynamics these alternatives are not realized.

(5) Problems from deliberate behavior. Certain kinds of consumer behavior give rise to a state which, according to established norms, is poor. An inadequate diet and consumption of alcohol and drugs are illustrative examples. Buying groceries very frequently is another kind of behavior which consumer policy planners regard as a problem. The consumers themselves however behave as they are used to. According to their own present values their state apparently is better than the other state which the planners consider highly preferable.



In Figure 3 the consumer concept is seen as a vector. The subconcepts - the numbers refer to the explanatory text above - are arranged along the vector after degree of verbalization, consciousness etc.

It is to be noticed that all consumer dissatisfaction can not be regarded as consumer problems. If there are no better alternatives attainable, the consumer dissatisfaction does not indicate consumer problems, according to the definition used. This is illustrated in Figure 3. Parts of the subconcept (1) in (2) are located outside the consumer problem vector. Seen from an individual consumer perspective the type (4) problems are not consumer problems either. The better state is not desirable for the consumer.

It is also to be noticed that we do not know the relations between the subconcepts. The size of these vectors in Figure 3 are arbitrary.

All the above problem areas have been discussed in the literature; however, the distinction between the problem types is hardly ever made explicit. When policy actions are considered these distinctions should be noticed.

Our exercise was aimed at clarifying problems of type (3) and (4). Using a panel with knowledge about consumer behavior as well as present and potential supply gives an opportunity to take into account those problems which individual consumers are unaware of or are unwilling to admit.

In the operationalization we tried to exclude (4) thus measuring the amount of consumer problems perceived by fully informed but not necessarily rational consumers in a narrow sense. The working hypothesis was that people strive at using their economic and other resources in such a way that they get as good a life as possible. In this goal seeking behavior conflicts arise. In balancing these conflicts people sometimes consciously choose a poor alternative in fulfilling a certain sub-goal. In a narrow perspective the decision seems irrational, from a wider perspective quite rational.


The Panel Composition

In selecting the panel members two qualities were important. First, the members should have good knowledge about consumer behavior and about present and potential supply, i.e., about data earlier described as objective indicators. The knowledge areas aimed at were identified from the members' employment or professional position. Second, as has been emphasized above, values hold an important role in defining consumer problems. Considering this, efforts were made to include members viewing the consumers' situations from different perspectives and representing different value structures. As a consequence the panel was composed of the following categories.

(1) Local consumer advisors (LA)

(2) Representatives of public authorities (PA)

(3) Representatives of the business sector (BR)

The guidelines for the panel size was about 30 participants, which is considered sufficient for a Delphi-procedure. The final panel consisted of 9 from the (LA) category and 10 each from both others.

Method for Date Collection

Data were generated by a survey in different rounds by means of the so called Delphi method. With this method the participants give their answers individually on written questionnaires. The individual answers remain anonymous. The answers of the panel are reported back to the panel in statistical form as median and quartile answers. Questions and remarks are also fed back to the panel, together with explanations and clarifications from the panel director. On the basis of such information the participants reassess their previous statements in new rounds.

The Delphi inquiry utilizes well-informed people's subjective judgements. The method was originally developed as a forecasting tool. It has however been used more and more to generate approximations and judgements about other situations where "hard data" are difficult to obtain. The method has been used for instance to define different opinions of important social issues (see e.g. Jilson, 1975, and Dahl, 1974), as a basis for decision making in marketing (Jolson and Rossow, 1971) and as a general tool in evaluation research. For more details about the method, see e.g. Linstone, Turoff (1975) or Wikstr÷m (1973).

The Questioning

The panel assessment concerned two different aspects:

a) how frequent and how serious are consumer problems in planning of purchases, buying and consumption of goods and services.

b) what desirable and feasible policy actions are to be taken to counteract these problems.

The latter aspect was included not only to give a wide spectrum of ideas about possible action, but to highlight what ideas there are about causes to these problems.

The panel was provided with some information about the general idea behind this assessment and a presentation of the outline of the study. This was done by the conceptual model in Figure 4.


The first part of the assessment considered (1) and (2) in Figure 4. In this the panel was asked to use available data about (5). The outcome of this assessment gives (3). In the second part, (4) and (5) are dealt with. The panel was asked to suggest possible and feasible actions to counteract the stated problems. The panel was extorted to seek information and ask questions if anything seemed confusing.

The Operationalization of the Consumer Problem Concept

Consumer problems were seen as situations in the planning, buying and consumption process, where consumers unwittingly make "incorrect" decisions, i.e., behave in a way which affects their budget or need satisfaction negatively. In order for an action to be defined as a wrong decision, it should have clearly noticeable negative effects. "Unwittingly" means that the consumer would not have behaved in that way if he had foreseen the consequences and if there had been other alternatives at hand.

The frequency of problems was measured as the average number of wrong decisions the consumer makes per 100 decisions. How serious the wrong decisions were was measured according to a 3 point ordinal scale.

As a second part of the study the panel suggested steps to be taken to counteract the problems they had established. In the second round the panel evaluated all measures suggested.

To get a systematic view of the consumer situation the panel was asked to assess the amount of problems in the following situations. For durable goods and more expensive services such as cars, private homes, boats, dishwashers, television sets and package vacation tours, the question was how often consumers make "incorrect" decisions on an average when they:

1. plan and consider and ultimately decide to buy a certain product or service.

2. choose product class, size, price level, and retail outlet.

3. carry out the buying: choose how to pay, agree to purchase conditions, make complaints etc.

4. use the product or service.

5. buy repair services.

For convenience goods the incorrect decisions concern:

6. the shopping pattern including frequency of shopping choice of retail outlets, time, transportation etc.

7. the choice of brand, package, size, price level, etc.

These steps in the buying process are based on ideas from Gredal's typology (Gredal, 1971).

The panel was also to mention and assess consumer categories and product areas with particularly high frequency of wrong decisions.

The Delphi Process

The written questioning consisting of two rounds was carried out in spring 1976. The panel answered each questionnaire within 3 weeks. After the first round some questions had to be rewritten and new instructions given. Misunderstandings of the concept and the assessment procedure had appeared.

After a working report had been finished and sent to the panel members a telephone round was carried out in early 1977. The subject was the panel's experience of the assessment process and the opinions about the final results.


The Amount and Seriousness of Consumer Problems

The assessment of the frequency of problems for the average consumer is shown in the profile below.



According to the median assessment there are problems for the consumers in 15-35 percent of the decisions in the different steps in the buying and consumption process.

The frequency of problems is one dimension, the seriousness of the problems is another.

In analyzing the panel assessments a total score was used for the negative economic and need satisfaction consequences. As mentioned before the consequences were measured according to a 3 point scale. The most serious consequences were given 3 points, the next group 2 points and the least serious 1 point. As shown by Figure 6, the first two aspects, planning in a wide sense, has the highest scores for negative consequences. As a whole negative economic effects are bigger than the need satisfaction effects. For the choice of convenience goods the opposite holds good.



An important part of this study was to inquire into the differences in values and views between the three sub-groups of the panel. These differences are shown in Figure 7.



The diagram shows that the (LA) representatives tend to consider the consumer situation to be more problematic than the other categories. The differences are, however, not as clear as I had expected.

For two reasons I had anticipated that the (BR) category would have given the most positive view of the consumer situation. First, business people regard consumers as a powerful group. The consumers "vote" by buying or not buying. From the business people's view they are not very easy to convince. Second, for tactical reasons this category could be expected to be optimistic. A problematic consumer situation calls for more regulations, which are mostly directed against the business.

As the diagram shows, there are no clear differences between the (BR) and (PA) categories. The assessments by these categories by no means reflect the differing views between the same categories in the public debate. After having analyzed the assessments and discussed the results with the panel in the telephone round the small differences between the categories can be explained. In the assessment process individual values and personal experience of buying have had a much stronger impact than group values and professional experience. In other parts of the assessment, concerning possible policy actions to be taken to counteract the stated problems, there was much more explicit groups differences.

Suggested policy actions

The panel as a whole suggested about 60 policy actions of very different character. About half of them were business oriented, one third were market oriented and the rest were consumer oriented. The panel evaluated these actions remarkably positively according to desirability and feasibility.

In order to give an idea of what kind of actions the panel suggested, some examples are quoted. All the three panel categories rated these actions as definitely feasible and desirable. There is also a set of more radical actions suggested, about which there were disagreements.

Actions to improve the consumers' planning of purchases:

--Improved education of young people and adults in consumer issues, particularly in budgeting.

--Increased consumer information in present and new forms.

--Extensive use of the television for consumer information.

--Expanding the local consumer advisor and consumer information activities.

--Increased product information based on product tests from the National Board for Consumer Policies.

Actions to improve the buying decisions:

--Compulsory use of simple contract forms.

--Compulsory information about actual interest in all installment buying.

--Develop standard forms for consumer complaints.

Actions to improve the use of products:

--Claim for instructions easy to understand for normal use and when troubles occur.

--More consideration to an easy use in product design and product development.

--Seller responsibility for complete and correct information about product use.

Actions to improve the household shopping organization:

--Prevent the closing down of retail outlets in rural areas.

--Increased consumer influence in local retail planning.

--Prohibit further hypermarkets.

Actions to improve the choice of products in the convenience goods area.

--Compulsory price comparison.

--Educate the consumers to read the declarations of content.

--Standard packs.

The business representatives however opposed some actions directed towards their own sector, such as:

--Increased consumer insight into and influence on the product development.

--Reducing advertising volume.

--Brand differentiation prohibited.

--Product variety reduced in certain areas.

--Reduced "bargain" prices.

The professional background of the panel members seemed to have a stronger impact in the assessment of possible policy actions than in their stating the amount of consumer problems.


What do these panel assessments measure and how valid are the results? Compared with the subjective and objective indicators discussed in an introductory passage the panel assessment aims at measuring the total consumer problem concept. This measurement procedure covers the problems which the consumers are aware of as well as those they do not know about.

The subjective and objective indicators illustrated in Figure 1 are fairly accurate measures. They are, however, partial and we do not know how valid they are as measurements of the total consumer problem situation. This issue was discussed in connection with the different subconcept in Figure 3.

Validity includes, however, good reliability. In a panel assessment we know very little about the measurement procedure, about what model the individual members use and about what data they feed into the model when they do their assessments. We know however that their assessments are some sort of synthesis of a very complex reality. In a Delphi procedure using panel assessment as the data source, the reliability problem is a validity problem on the individual level. Do the individual members process the data they are supposed to (see Figure 4) and do they generate good information? We just do not know. With these questions one enters another theory of science. Keywords such as dialectical and phenomenological indicate what kind of theory is referred to. According to this theory the reality is too complex to apply any concrete validity criteria. One criterion suggested is to what extent the actors involved accept the results and the interpretations. This subject is however very difficult and more thinking is needed.

The quality of the information gathered was also discussed with the panel. Their reaction was that the assessments could be used as rough measures. I would say that the outcome in this case should be expressed by the interquartile rather than by the median figures (see Figure 5). This means that you can not from the figures say if the proportion of "wrong" decisions is 15 or 35%, you can however see that some steps are more problematic than others.

Another vital question is how the stated problem situation should be interpreted. Do the figures indicate a satisfactory consumer situation or not? According to the panel the consumer situation is unacceptable. Policy actions should be taken. The panel has suggested a long list of possible actions, most of them directed toward suppliers. You can also see from the results to what extent different panel members agree on the suggested actions.

The policy Delphi procedure is not only a possible information and evaluation instrument but is also a tool for generating ideas about how the stated problems should be cured.

In our study the Delphi assessment was used to gather information about the effects of the policy actions taken up to now on the consumer situation and to evaluate that situation. The results indicate that there are considerable problems. For this kind of statement there is no need for very accurate measurements. The results also show comparisons between different areas. For this kind of assessment there is a need for a somewhat higher accuracy. I think that the method is valid for this kind of assessment too. One has however to take into consideration that if a special aspect has been openly debated recently this may effect the results considerably.

In comparing of situations at two different points in time a rather high accuracy in the measurements is needed. The same holds true if the aim is to compare the situation between different countries. At present I would be hesitant to use the method for this kind of comparisons. More knowledge, experience and thinking are needed.

Finally when judging the value and the possibilities of this method one should keep in mind that this study was a test of the method. It is possible to improve the results considerably by a better design of the assessment procedure.

You should also, when judging the value of this method, ask how good results you could get for the same money using another method. In this aspect I think the panel assessment is quite competitive.


R. L. Ackoff, Scientific method. Optimizing applied research decisions. New York. (1962).

A. R. Andreasen, Consumer dissatisfaction and market performance. Paper presented at Fourth International research seminar in Marketing, Senanque Abbey, Gordes, France. May (1977).

A. W. Dahl, Delphic and Interactive Committee Processes in a Comprehensive Health Planning Advisory Council: A comparative Case Study, PHD Dissertation, The Johns Hopkins University. (1974).

G. Eliasson, Problemformulering och problemuppfattning i den konsumentpolitiska planeringen. Arbetspapper. Foretagsekonomiska inst. Stockholms Universitet. (1976).

K. Gredal, Moderne forbrugeres motiver og adferd (The motives and behaviour of modern consumers), Kopenhavn. (1971).

L. Festinger, Conflict decision and dissonance. Stanford, California. (1964)

I. A. Jilson, The National Drug-Abuse Policy Delphi, in Linstone & Turoff, The Delphi Method. (1975)

M. A. Jobson and G. L. Rosson, The Delphi Process in Marketing Decision Making, Journal of Marketing Research. November (1971)

J. S. Lingoes and M. Pfaff, The Index of Consumer Satisfaction: Methodology. In M. Venkatesan (ed.), Proceedings of the Third Annual Conference of the Association for Consumer Research. (1972)

H. Linstone and M. Turoff (ed), The Delphi Method. Techniques and applications. Addison-Wesley. (1975)

Y. Renoux, Consumer Dissatisfaction and Public Policy. In F. C. Allvine (ed.), Public Policy and Marketing Practices, Chicago: American Marketing Association. (1973)

S. Wikstr÷m, Vara dagligvarors distribution. En framtidsstudie. (The Food Marketing Process: A Future Study), Bonniers. (1973)

S. Wikstr÷m and G. Eliasson, Konsumenters problem vid planering, kop och konsumtion av varor och tjanster -En policy-delphi studie. (An assessment of consumer problems in planning purchases, buying and consumption. A policy-delphi study). The Department of Business Administration Stockholm University. Working paper. (1976)

F. +lander, Consumer satisfaction - a skeptic's view. Paper presented at NSF Workshop on Conceptualization and Measurement of Consumer Satisfaction and Dissatisfaction, Chicago, April 11-12, 1976. (1976)