For the past forty years, I have been doing marketing research surveys, initially in an advertising agency and since 1960 in my own company. This book is an account of that work. It presents a theory of choice behavior, supported by major, previously unpublished studies, drawing upon experience and data based on cumulative sample sizes of millions of interviews.
From the outset, I was as interested in the general as in the specific. I pondered such questions as: How many really different variables do we need to measure? How many really different problems are there? What do studies of automobiles and baby foods have in common? What is the connection between trying to improve the taste of an orange juice and trying to set the right price for an opera ticket? This led to some surprising conclusions. We need to measure only a few basic variables. There are only three basic problems, and these problems have a single common denominator. One way or another, marketing research boils down to measuring and predicting the choices people make. In spite of the fact that this is totally obvious, I did not understand it right away. The variables are treacherous. Unless one takes firm hold of them, they twist and turn, change form, and slip through one's grasp. But if one persists, one is led to a theory. That is what happened to me. I did not start with this theory in mind. The theory was born and grew up in the real world. The data that gave it sustenance came from the real world. I was driven to it inexorably by my desire to solve practical business problems.
Measurement played an indispensable role in the development of the theory. Without it, there would have been no theory, only a melange of words — something that happens now and then in the social sciences. In my own mind, measurement always came first. The words — the labels attached to the measurements — came later. The theory came last, an after-the-fact synthesis in the domain of ideas of what had already been accomplished in the real world.
This duality is mirrored in the pages of this book. There are frequent changes of scale. On the one hand, there are sweeping generalizations that aspire to cover conceptual ground far beyond the commercial exchange of goods. On the other hand, there are painstaking accounts of specific techniques. And there are shifts from the forest to the trees, and back to the forest again. This is not happenstance. For better or worse, it reflects the fact that abstract theory and concrete measurement have gone hand in hand in my work. So I have presented them together.
At the core of the theory are three laws of choice behavior: the Law of Congruence, the Law of Primacy, and the Law of Persistence. These laws are both general and self-evident. At first glance, they are so self-evident that you might say, "I have known this all along." My reply is, "Of course you have known it all along. It probably wouldn't be true if you hadn't known it all along. But there is a difference between knowing and knowing, between the passive knowing that allows us to persist in actions that are inconsistent with what we know and the active knowing that helps us change the way we do things." I believe that knowing the laws of choice behavior — really knowing them — changes the way we do things. My crass but stringent criterion has been that a good theory should enable someone equipped with it to make more money than someone who isn't. I believe my theory has passed this test.
What kind of book is it? Don't expect it to fall too readily into predefined categories. It is certainly a technical book in that it deals seriously with technical issues. But it is not written like a technical book. Theory, methods, and data are not presented in disembodied form. They are located in space and time, positioned in the institutional environment in which they arose, supplemented by anecdotes when appropriate. I simply do not believe that science needs to be treated as an abstract exercise, ethereally removed from human concerns. This is particularly true for an embryonic discipline. In such a discipline, the ideas one rejected, the path by which one got to where one is, may be as important as where one has arrived. For that reason, I have also provided the personal context, which makes parts of the book read like an autobiography.
The book is addressed to both the business and the academic communities, to businessmen and businesswomen who pose marketing questions, to my colleagues in marketing research who try to answer these questions, to the advertising community which is charged with implementing some of these answers, to teachers and students in business schools, to survey research practitioners, and to social scientists of all stripes. It may also be of interest to a broader audience that is curious about anything that has to do with human behavior and wants to get an inside glimpse into a small but intriguing corner of the business world.
Through the years, I refrained from publishing. I simply did not want to put in print work I knew to be incomplete. But my business interests led me to promulgate every new insight far and wide, carrying it from company to company, from one prospective client to the next, talking to anyone who would listen and to many who wouldn't. In this process, I won acceptance of my point of view from the major companies that became my clients, and exercised indirect influence on the thinking of a wider circle of people in the field. In addition, I resolved privately to publish my work when it had progressed far enough to be presented as a coherent synthesis of theory and empirical evidence. Here it is: a technical book that freely straddles the technical and the personal, full of original data from massive real-life experiments, a summary of the thoughts I had and the results I obtained in many years of practicing my craft.
Eric Marder
New York
1996
These four chapters introduce Choice Research.
Chapter 1 (Marketing and Choice) provides a succinct summary of a particular way of looking at marketing and defines the terms used in the theory of choice.
Chapter 2 (Defining Choice Research) defines Choice Research by contrasting it with other approaches.
Chapter 3 (First Principles of Choice Research) examines some widely held ideas and demonstrates that these take on a dramatically different aspect when viewed from the perspective of Choice Research.
Chapter 4 (The First Law: The Law of Congruence) is the cornerstone of the theory and of the entire book. It is the foundation for everything that follows — addressed to specialists, business executives, and social scientists alike.
This part of the book deals with the market shares deserved by integrated offers — offers specified by brand name, description, and price.
Chapter 5 (Measuring Choice) introduces a method that puts the First Law to practical use and presents empirical support for it.
Chapter 6 (Price Testing) and Chapter 7 (Concept Testing) summarize substantive findings from studies conducted over the course of many years.
Chapter 8 (Product Testing) provides a generic classification scheme for comparing all possible product testing methods.
Chapter 9 (Paired Comparisons) demonstrates, by analysis and by empirical data, that the method of paired comparisons — which has been used by some of the most respected packaged goods manufacturers — is fundamentally flawed and capable of generating wrong results.
Chapter 10 (Deserved Share) deals with how product tests should be conducted. The section "What's in a Name" contains surprising results about the role of product and image.
Chapter 11 (What STEP Measures) is a nontechnical recap of the ideas (not the findings) in the previous chapters. Together with Chapter 5, this chapter provides an overview of the measurement of integrated offers.
Chapter 12 (Beyond Product Category Boundaries) examines the implications of the First Law for offers that don't appear to fit into definable product categories, and introduces a self-validating technique for dealing with such cases.
This part of the book is the centerpiece of the theory. It should be of particular interest to social scientists and survey research practitioners. Business executives may find it theoretical. But at least three sections in these chapters contain practical advice on business strategy ("Desirability" and "An Interjection about Improving Brands" in Chapter 13, and "Product Category Differentiation" in Chapter 14).
Chapter 13 (The Anatomy of Questions) presents a general taxonomy of survey questions. It introduces the unbounded write-in scale, an original instrument for measuring desirability. It makes distinctions among the basic variables of the theory. And it demonstrates the importance of these distinctions by means of methodological analyses of large bodies of data.
Chapter 14 (The Second Law: The Law of Primacy) explores in depth the relationship between desirability and choice. This leads to the Second Law, which in turn leads to the tie matrix, with major implications for measurement theory and practice.
This part of the book deals with the market shares of synthesized offers, that is, offers constructed by modeling.
Chapter 15 (Struggling to See the Obvious) describes two problems — the strategy planning problem and the diagnostic problem — that once seemed intractable, and tells how an effort to deal with these problems led to the development of SUMM (Single-Unit Marketing Model).
Chapter 16 (The Partitioning of Choice) describes SUMM, a choice model, and the view of marketing on which SUMM is based. This culminates in the definition of the universal marketing problem — the one problem every market must confront.
Chapter 17 (Constructing the Map) and Chapter 18 (Twists and Turns in Measuring the Desirability of Attributes) describe the measurement tools used in SUMM.
Chapter 19 (Dynamic Assessments) and Chapter 20 (The Synthesis of STEP and SUMM) demonstrate the predictive power of SUMM and its relation to STEP.
Chapter 21 (Brand Positioning) deals with the positioning of new brands, the repositioning of established brands, and the collection and interpretation of customer satisfaction data.
This part of the book deals with advertising.
Chapter 22 (Measuring Advertising) covers the objectives of advertising.
Chapter 23 (Measuring Print Ads) discusses the institutional aspects of advertising. It summarizes substantive findings from four years of measuring the effects produced by single print ads under real-life conditions of exposure, and addresses the question whether ads can produce negative effects.
Chapter 24 (Measuring Television Commercials) examines the problems of measuring the effects of single TV commercials, and methods for doing so.
Chapter 25 (Measuring Television Campaigns) deals with the ultimate problem of advertising research: how to measure the effects produced by different advertising weights over the course of time. It presents Ad-Weight, a method for accomplishing this, summarizes a large body of substantive findings, compares the effects produced by television, coupons, and free samples, and identifies the generic relationship between advertising weight and effectiveness.
Chapter 26 (The Third Law: The Law of Persistence) presents substantive finding about the decay and persistence of the effects of messages. This culminates in the Third Law, which has practical implications for business readers and theoretical implications for social scientists.
Chapter 27 (Budget Allocation Across Brands) presents a strategic method for deciding which of a company's brands should be supported, and how.
STEP, SUMM, and Ad-Weight are registered trademarks of Eric Marder Associates, Inc.
From the outset, I was as interested in the general as in the specific. I pondered such questions as: How many really different variables do we need to measure? How many really different problems are there? What do studies of automobiles and baby foods have in common? What is the connection between trying to improve the taste of an orange juice and trying to set the right price for an opera ticket? This led to some surprising conclusions. We need to measure only a few basic variables. There are only three basic problems, and these problems have a single common denominator. One way or another, marketing research boils down to measuring and predicting the choices people make. In spite of the fact that this is totally obvious, I did not understand it right away. The variables are treacherous. Unless one takes firm hold of them, they twist and turn, change form, and slip through one's grasp. But if one persists, one is led to a theory. That is what happened to me. I did not start with this theory in mind. The theory was born and grew up in the real world. The data that gave it sustenance came from the real world. I was driven to it inexorably by my desire to solve practical business problems.
Measurement played an indispensable role in the development of the theory. Without it, there would have been no theory, only a melange of words — something that happens now and then in the social sciences. In my own mind, measurement always came first. The words — the labels attached to the measurements — came later. The theory came last, an after-the-fact synthesis in the domain of ideas of what had already been accomplished in the real world.
This duality is mirrored in the pages of this book. There are frequent changes of scale. On the one hand, there are sweeping generalizations that aspire to cover conceptual ground far beyond the commercial exchange of goods. On the other hand, there are painstaking accounts of specific techniques. And there are shifts from the forest to the trees, and back to the forest again. This is not happenstance. For better or worse, it reflects the fact that abstract theory and concrete measurement have gone hand in hand in my work. So I have presented them together.
At the core of the theory are three laws of choice behavior: the Law of Congruence, the Law of Primacy, and the Law of Persistence. These laws are both general and self-evident. At first glance, they are so self-evident that you might say, "I have known this all along." My reply is, "Of course you have known it all along. It probably wouldn't be true if you hadn't known it all along. But there is a difference between knowing and knowing, between the passive knowing that allows us to persist in actions that are inconsistent with what we know and the active knowing that helps us change the way we do things." I believe that knowing the laws of choice behavior — really knowing them — changes the way we do things. My crass but stringent criterion has been that a good theory should enable someone equipped with it to make more money than someone who isn't. I believe my theory has passed this test.
What kind of book is it? Don't expect it to fall too readily into predefined categories. It is certainly a technical book in that it deals seriously with technical issues. But it is not written like a technical book. Theory, methods, and data are not presented in disembodied form. They are located in space and time, positioned in the institutional environment in which they arose, supplemented by anecdotes when appropriate. I simply do not believe that science needs to be treated as an abstract exercise, ethereally removed from human concerns. This is particularly true for an embryonic discipline. In such a discipline, the ideas one rejected, the path by which one got to where one is, may be as important as where one has arrived. For that reason, I have also provided the personal context, which makes parts of the book read like an autobiography.
The book is addressed to both the business and the academic communities, to businessmen and businesswomen who pose marketing questions, to my colleagues in marketing research who try to answer these questions, to the advertising community which is charged with implementing some of these answers, to teachers and students in business schools, to survey research practitioners, and to social scientists of all stripes. It may also be of interest to a broader audience that is curious about anything that has to do with human behavior and wants to get an inside glimpse into a small but intriguing corner of the business world.
Through the years, I refrained from publishing. I simply did not want to put in print work I knew to be incomplete. But my business interests led me to promulgate every new insight far and wide, carrying it from company to company, from one prospective client to the next, talking to anyone who would listen and to many who wouldn't. In this process, I won acceptance of my point of view from the major companies that became my clients, and exercised indirect influence on the thinking of a wider circle of people in the field. In addition, I resolved privately to publish my work when it had progressed far enough to be presented as a coherent synthesis of theory and empirical evidence. Here it is: a technical book that freely straddles the technical and the personal, full of original data from massive real-life experiments, a summary of the thoughts I had and the results I obtained in many years of practicing my craft.
Eric Marder
New York
1996