Using Mind Genomics To Understand The Specifics Of A Customer’s Mind

Abstract

Approaching customers with the right message has always been a key objective of companies selling products and services. Achieving this target demands understanding what customers think about specific ideas and messages, determining whether there are different mind-sets for the same topic, and if there are, assigning people to the right mind-set. The advent of Internet has increased the opportunities for the advertising component. Companies need to have the tools to conceptualize these mind-sets ‘on-demand,’ for any topic, collect the relevant data for this specific topic, followed by a way to uncover mind-sets and assign new people to the mind-set. The issue is to achieve the foregoing with low cost, on demand, in areas where there is no behavioral or psychographic data, and then apply the results in actionable way. Several approaches have been created and used to address the problem, but all require either behavioral or expensive psychographic data, and time to do the research. This paper presents the emerging new technology of Mind Genomics, a marriage of experimental psychology and data analytics. Mind Genomics reveals, through simple, affordable, iterative experiments the specifics of what features of a service or product appeals to an individual through the notion of mind-sets, and then a method for assigning any new person to the most appropriate mind-set. A simple set of questions, embodied in the PVI (personal viewpoint identifier), assigns the new person to the mind-set, doing so in ‘real-time.’ The results are both better knowledge of the topic, and better sales performance.

Keywords: Digital marketingmind genomicspersonal viewpoint identifier

Introduction

This paper presents a different approach to uncover consumer’s mind. The approach is referred to as Mind Genomics (MG), a new technology which tries to understand the minds of consumers by their reaction to messages (Moskowitz, Gofman, Beckley, & Ashman, 2006; Moskowitz & Gofman, 2007). The objective is to create a scalable, cost-effective, rapid, archival, and scientifically powerful system, having application for the creation of NEW KNOWLEDGE and for the profitable use of this knowledge for social, educational and commercial use.

What it is discussed in this paper is the possibility to create a science-based system which creates a fundamental as well as action-oriented base knowledge in virtually any topic where human judgment is relevant. The objective is to use the system to populate the data-based of any topic, leading one to rapidly and profoundly understand the mind of the consumer regarding the specific topic. The objective is to create a ‘Big Data of Relevant Information’.

Literature Review and Theoretical Framework

The standard for understanding subjective reactions to service or products has been to directly ask people in focus groups or other qualitative methods about a particular topic (Esmerino et al., 2017). The more advanced methods are reserved for the quantitatively oriented practitioners, who create surveys, present them to many consumers, and analyze the results using a simple average, or multivariate statistics. Often surveys begin with topics about what one does in general, such as food preferences and food habits (Lusk, 2017), now evolving down to a momentary survey after a relevant experience to ask ‘ How did we do ?, or ‘ Would you recommend us to someone with whom you do business? ’ the now-ubiquitous NPS, (Net Promoter Score), analyzed by (Artz, 2017).

There are other approaches which have tried to understand the mind of consumers and the role of different marketing techniques. Efforts to understand the mind of consumers have been around since the early days of consumerism. Rindell, Korkman, and Gummerus (2011) analyzed the relevance and the role of brand images, in terms of how they impact consumer practices. They reported that that in order to identify and categorize consumer practice, consumers develop idiosyncratic systems, novel and personal practices related to brands.

Kniazeva (2008) analyzed the consumer perceptions of genetically modified food, in order to understand and to explore messaging to bring consumers closer to accepting the idea of this GMO food. The research used detailed interviews. Yarrow (2014), provides interesting insights about the new motivations and behaviors of shoppers, taking marketers where they need to be today. He focuses on what has come to be accepted as deeply psychological, often unconscious relationships, which people form with products, retailers, marketing communications, and brands, respectively (Zemel, Gere, Papajorgji, Zemel, & Moskowitz, 2018a). If Yarrow’s suggestion is to be followed, Yarrow (2014), this will require that each problem eventually being solved by expensive, extensive, long, and tortuous experimentation. During the last two decades, the notion of mind-set segmentation has become increasingly clearer and used with success in different areas of business activities. Segmentation refers to the approach of dividing people in clusters by the pattern of their responses. Traditional segmentation gathers people in clusters based on individual attributes such as (e.g., age, gender, social class, income). This clustering approach reflects “ Who they are ”. Another clustering approach is the so-called psychographic segmentation that is based on “ What they believe ” (Vij, 2012; Mitchell, 1994). With the advent of Internet, a new clustering approach started appearing widely and it is based on the pattern of “ What they do ” electronically when people use the internet say for shopping purposes (Gere, Zemel, Papajorgji, & Moskowitz, 2018).

Steps of Mind Genomics Research Process

MG relies on the concepts of inter-market variability and horizontal segmentation, ideas pioneered by American market researcher, psychophysicist, Moskowitz (Moskowitz et al., 2006; Milutinovich & Jacob, 2016), who suggested that in any area where “opinion holds sway,” one can discover different segments of consumer needs. In other words, there are always different consumer mind-sets whenever there are topics calling for an opinion (Zemel, Gere, Papajorgji, Zemel, & Moskowitz, 2018a).

Step 1 identifies a problem. As mentioned before ahead, the problem could be of any nature, be it business, social or political nature. Step 2 uses a tool empowering rapid collection of scientifically valid and robust information. Although this instantiation of MG appears to be nothing more than a survey, the reality is that the MG interview is really an experiment. The respondent is presented with a set of varied vignettes. By systematically varying the composition of the vignettes, and having respondents rate these vignettes as single test concepts, it becomes possible through statistical (regression analysis such as OLS) to identify the contribution of every element to the response. The regression analysis approach was originally created to understand how people make decisions, but has been applied extensively to many aspects of consumer behavior. In its original, the method was called “conjoint measurement”, a name which remains today, a half century later (Luce & Tukey, 1964).

The experiment design embodied in a MG study dovetails well with the topics being explored. The topic is made explicit and concrete by asking four relevant questions, driven by the requirement that the sequence of the four questions ‘tell a story.’ Each of the four relevant questions, is answered by four separate statements for each question, i.e. a total of 16 answers or ‘elements.’ The selection of the topic, the four questions, and the four answers to each question remains in the purview of the researcher.

An experimental design creates the 24 combinations of elements, with the property that the 24 combinations comprise 2-tuples of elements 3-tuples of elements, and finally 4-tuples of elements. Each answer appears equally often. No vignette ever comprises more than one answer from the same question. The 16 elements, embedded in the 24 vignettes, are combined in a way which makes the 16 elements statistically independent of each other, permitting regression analysis to be performed on the ratings, to relate the rating to the presence/absence of each answer.

The ratings are transformed to produce a binary scale. Even though a 9-point scale is used, managers have a difficult time with this type of Likert or category scale. By changing the scale to a binary scale, things are made easier for the manager, and for the interpretation. Ratings of 1-6 are converted to 0, and a small random number added to the converted value. Ratings of 7-9 are converted to 100, and again a small random added to the converted value. The experimental design ensures that the 24 elements are statistically independent of each other so that the coefficients, the impact values of the elements, have absolute value. The inputs are of 0/1 nature, 0 when the element is absent from a vignette, 1 when the elements is present in the vignette. The dependent variable is either 0 or 100.

The equation is defined as: Rating = k₀ + k₁(A1) + k₂(A2) …k₃₆(D9). The equation can be estimated for each respondent, and the corresponding coefficients averaged (individual models averaged to create a group model), or all the data can be aggregated and then one regression analysis run (grand model.) The results will ‘tell the same story,’ even though somewhat different numerically. The equation says that the rating is the combination of an additive constant, k₀, and weights on the answers. The answers appear either as 0 (absent) or as 1 (present), so the weights, k₁ – k₃₆, showing the driving force of the different answers.

The next step divides people into subgroups, mind-sets and then creates a PVI (personal viewpoint identifier), to assign a new person to the proper mind-set. MG results are generated based on the original research data and are used as a description of the sample. The pattern of coefficients is, in a sense, the ‘signature’ of a person’s mind with respect to a topic. MG clusters people based upon the patterns of their individual coefficients, putting people with similar patterns of coefficients into the same group, called a mind-set. The approach uses a combination of regression to obtain the coefficients, as explicated above, together with k-means clustering to divide the people into a parsimonious set of groups, with the nature of the group, the mind-set, defined by the meaning of the answers which score highest in that group. This criterion is referred to as the criterion of interpretability. The discovery of the mind-set is of prime importance, a major contribution of the science. Yet, the mind-set is limited to the respondents who participated.

To extend the power of the discovered mind-sets, MG projects these mind-sets into the general population, allowing the researcher to assign a hitherto unknown, new person, to one of the mind-sets. The method is a classification task, through which new participants are assigned to mind sets based on their response pattern to a limited set of questions emerging from the study (the PVI or Personal Viewpoint Identifier.) In order to create the PVI and determine the response pattern, the most discriminating elements from the original study are extracted. These elements are those which show the highest distant between the existing mind sets. These calculations can be done using different distance metrics, the metric being left up to the discretion of the researcher. The chosen answers are ordered, and the most discriminating are extracted to form the PVI.

From practical considerations, such as e-commerce and sequencing a person on many different topics with many PVI’s, it is vital to make the PVI as simple, robust and fast as possible. For each PVI, participants are asked to rate six elements separately on a binary scale, a 15-20 second task. The results from the PVI are analyzed immediately in a cloud server, and the mind-set membership stored, as well as ‘reward feedback’ given to the respondent. The content of the feedback might vary from a simple text describing the mind sets to links to mind set dependent web pages or designs of the same page. By recording the occurrence of the mind sets, a more detailed picture can be achieved from a broader sample within a short period of time.

Conclusion and Discussions

After using Mind Genomics in a number of cases of different nature the time is to make some important conclusions. Mind Genomics is a powerful tool that allows analyzing and understanding business, social and political issues. The Mind Genomics strategy dimensionalizes the issue (be it of social, business or political nature), dividing a complicated issue into a set of questions, and then providing alternative answers to these questions. Analyzing the response to combination of these analyses allows the deconstruction of the responses into the contribution of the individual answers on a person-by-person basis, and the discovery of different modal patterns, the aforementioned mind-sets. It is often the case that respondents are not be aware of their proclivities, but the Mind Genomics approach reveals their mind-sets, confirming what they often say they ‘suspected was the case’. The Personal Viewpoint Identifier (PVI) is of scientific and commercial value as it classifies new people who have not participated in the study into one of the revealed mind-sets. In a typical application, e.g., for e-commerce, the PVI creates a personalized database for each person for a number of different products. For each product, the respondent’s mind-set for that product, obtained through PVI, reveals the mind-set segment for the individual, and suggests the most compelling message for that prospective customer.

References

1. Artz, M. (2017). NPS—The One Measure You Really Need to Grow?. Controlling & Management Review, 61(1), 32-38. Retrieved from
2. Esmerino, E. A., Ferraz, J. P., Tavares Filho, E. R., Pinto, L. P., Freitas, M. Q., Cruz, A. G., & Bolini, H. M. (2017). Consumers' perceptions toward 3 different fermented dairy products: Insights from focus groups, word association, and projective mapping. Journal of dairy science, 100(11), 8849-8860. Retrieved from http://www.sciencedirect.com/science/article/pii/S0022030217308159
3. Gabay, G., Zemel, G., Gere, A., Zemel, R., Papajorgji, P., & Moskowitz, H. (2018). On the Threshold: What Concerns Healthy People about the Prospect of Cancer? Cancer Studies and Therapeutics, Research, 3(4).
4. Gere, A., Zemel, R., Papajorgji, P., & Moskowitz, H. (2018). Weak Signals and Mind-Sets of Consumers: The Case of Milk. Journal of Food Science and Engineering 8, 125-136. https://doi.org/
5. Kniazeva, M. (2008) Marketing “Frankenfood”: Appealing to Hearts or Minds? Journal of Food Products Marketing, 11, 21-39.
6. Luce, R. D., & Tukey, J. W. (1964). “Simultaneous Conjoint Measurement: A New Type of Fundamental Measurement.” Journal of Mathematical Psychology 1(1), 1-27.
7. Lusk, J. L. (2017) .Consumer Research with Big Data: Applications from the Food Demand Survey (FooDS). Am J Agric Econ, 99(2), 303–20. Retrieved from: DOI:
8. Milutinovich, V., & Jacob, S. (2016) Mind Genomics A Guide to Data-Driven Marketing Strategy. Berlin: Springer.
9. Mitchell, V‐W. (1994), "How to Identify Psychographic Segments: Part 1", Marketing Intelligence & Planning, 12(7), 4 – 10. https://doi.org/10.1108/EUM0000000003905
10. Moskowitz, H. R., Gofman, A., Beckley, J., & Ashman, H. (2006). Founding a New Science: Mind Genomics. J Sens Stud, 21(3), 266–307. Available from: DOI:
11. Moskowitz, H. R., & Gofman, A. (2007). Selling blue elephants: How to make great products that people want before they even know they want them. Pearson Education.
12. Rindell, A., Korkman, O., & Gummerus, J. (2011). The Role of Brand Images in Consumer Practices: Uncovering Embedded Brand Strength. Journal of Product & Brand Management, 20, 440-446. 
13. Vij, S. (2012). Psychographic Segmentation Based on Belief Factors Underlying Attitude toward Advertising in General. International Journal for Management Research, 1(4), 17-28.
14. Yarrow, K. (2014). Decoding the new consumer mind: how and why we shop and buy. John Wiley & Sons.
15. Zemel, R., Gere, A., Papajorgji, P., Zemel, G., & Moskowitz, H. (2018a). Uncovering Consumer Mindsets Regarding Raw Beverages. Food and Nutrition Sciences, 9, 259-267. Retrieved from http://www.scirp.org/journal/fns
16. Zemel, R., Gere, A., Papajorgji, P., & Moskowitz, H. (2018b). “Candy is dandy …Different strokes …An algebra of expected sexuality”. American Chemistry Society National Meeting & Expo, March 18 - 22, 2018, New Orleans, LA, USA.