Effective market research is able to predict consumer behavior. However, relying only on the conscious statements of consumers may lead to false conclusions.
Consumers often do not know why they prefer one brand or product. Implicit market research methods are not affected by this deficit because they do not ask for customers’ opinions but assess underlying attitudes.
Market researchers can use implicit methods from cognitive psychology to asses subconscious preferences. Priming experiments or the IAT (Implicit Association Test) are widely used.
Newer methods include AI association prediction, a form of subconscious bias prediction. Both can help to understand your brands’ subconscious perception and helps to select the best marketing content, eliciting the desired implicit effects.
Overall, these methods therefore often give a better understanding of the unconscious functioning of brands and content.
In this article, we will illustrate with practical applications the boundaries of explicit questioning and the advantages of concrete, implicit methods.
We would like to emphasize that explicit and implicit market research methods do not exclude one another but can work hand in hand.
Practice-example NIVEA make-up: The attempt to position the successful brand NIVEA in the market of decorative cosmetics failed. The implicit meaning “care” and “security” of the brand NIVEA is problematic in this market since consumers use products such as make-up in the first place to look better which is contrary to what NIVEA implicitly stands for (Scheier & Held, 2012).
NIVEA’s positioning and communication conflict would ideally have been revealed in advance by using implicit market research methods. Consequently, customers ascribed fewer competencies to make-up products from NIVEA compared to products of the competition (Bielefeld, 2012).
The Psychology of the Implicit
The meaning of a brand reveals itself in the explicit and implicit system of a consumer (Kahneman, 2012). For a better understanding, these terms are also named pilot and autopilot.
The explicit system (pilot) can be understood as the perceivable “thinking”. Using this system we process information consciously and carefully but very slowly.
The explicit system deals for example with the sentence “the sun is shining,” which creates a cost-benefit analysis and is planning the future. In consumer surveys, the explicit system yields the answer “I compared prices and took the best offer.”
The implicit system, however, can be considered as the impulsive, automatic part of decision-making. As this system is mainly based on automatic responses, it allows for quick and efficient decisions.
The implicit part of us is not thinking twice about 2×2, whether we like to drink Coke or quickly buy the shampoo in the SUPER SPECIAL deal.
The Boundaries of Explicit Market Research
Contemporary market research is based on questioning consumers with explicit methods such as qualitative interviews and quantitative opinion research. This type of explicit consumer-questioning is an important part of market research but only examines the explicit system of the pilot.
The implicit system of the autopilot, which perceives the brand L’Oréal as more valuable than NIVEA in the make-up market, is not reached by this method of market research.
Implicit Bias in Explicit Market Research
The central weakness of conventional market research is to rely solely on the self-determined answer of the study participant. Thus, it is assumed that participants always answer honestly and beyond that also understand why they prefer a particular brand. In short: explicit methods assume that subjects can explicitly state their consumption-relevant or implicit preferences. The majority of the decision-making process, however, lies in the subconscious.
Implicit processes are a central element in psychological research to explain human behavior. This often leads to consumers making systematic failures when they explain their consumption in direct questioning. These systematic failures are summarized in psychology with the term “implicit bias” and are often not detected with conventional market research methods.
The so-called “social desirability bias” and the “narrative fallacy” are particularly relevant as examples of such implicit biases.
Social Desirability Bias
Consumers tend to see themselves consciously and unconsciously in an overestimated positive light. This serves to create a positive view on oneself. Conversely, this cognitive bias serves to produce a positive image towards the interviewer, institution or the general public.
Consumers could falsely state the usage of brands like Prada, Porsche or Rolex to be associated with higher social class. If questions prone for this bias are asked in the screener, the part at the beginning of the survey that is supposed to filter non-relevant consumers, the real target group is easily missed.
The social desirability bias is also observable in election research. The forecast for right-wing parties is often underestimated since voting for them is perceived as non-compliant and respondents want the interviewer to have a positive image of them.
The narrative fallacy describes the general tendency of humans to believe in something that makes (only) sense in the context of a story. Thus many consumers (including me) will answer the question of why they choose a certain beer brand with that they prefer the taste of this brand over others.
Numerous studies show that consumers often are unable to identify their preferred drink in a blind test or even prefer the taste of another brand. But why is this?
Humans generally understand their own behavior and their decisions in the context of their life story. If consumers buy a brand regularly, they implicitly grant that brand a certain status in their life. This status is then explicitly justified with the “obviously” better taste.
In reality, however, we can hardly differentiate the tastes of beer brands and base our choices considerably on the associated meaning of a certain brand. Explicit market research methods are very susceptible to this form of cognitive bias since consumers are fully convinced to use a certain brand for a reason that entirely differs from the actual one.
Implicit Market Research Methods
Implicit methods, in contrast to explicit methods, do not rely on the explicit statement of the consumer. Instead, they rely on reactions to certain stimuli.
The consumers reaction must be performed so quickly that the participants do not have time to make a conscious decision. The reaction to the stimuli is rather done by the unconscious, automatic system and represents the true decision-making of implicit processes.
Priming in Semantic Networks
Neuroscientists at the University of California, Berkeley have shown that different people have very similar patterns of brain activity when perceiving certain words and concepts.
The study shows not only that every concept appeals to a specific semantic network, but also that the semantic networks seem to be distributed so that different brain areas are involved in the decision-making process.
Each concept and thus each brand has a unique profile in the semantic network, dependent on all experiences one has with the brand (advertising, recommendations, private use, etc.).
One method to take advantage of semantic networks is priming. If a consumer sees a specific concept, such as the word “spaghetti”, a specific neuronal network gets activated representing this concept. Neuronal networks that represent associated concepts (for example, “noodles” and “Italy”) are activated as well.
As a consequence of such connections, semantically associated concepts are more easily accessible and can take influence on the (consumer) behavior – both on conscious and unconscious processes. This is a fundamental principle of the human brain and is repeatedly found in numerous studies and experiments over and over again.
In market research, this effect can be used to assess whether a particular product is perceived positively. In a typical priming test consumers have to make fast decisions after seeing a target concept such as “premium”. Before seeing the target concept, they are primed with another word, such as the name of a new product.
This priming is usually done only for a fraction of a second so that the word is not consciously perceived. The faster a consumer responds to the target concept “premium” afterward, the stronger the corresponding semantic association. In this way, complex semantic relations can be examined and portrayed.
Implicit Association Test (IAT)
The Implicit Association Test (IAT) was developed in 1998 by Greenwald, McGhee, and Schwartz and takes advantage of the principle of priming and semantic networks. This test can identify which properties are assigned to a specific concept.
The IAT has significantly advanced the focus on implicit measurement methods in psychology because, in contrast to conventional methods, it is able to reliably measure socially undesirable concepts like racism of study participants (Greenwald, Poehlman, Uhlmann, & Banaji, 2009).
Application of the IAT in Marketing
Similar to priming, the IAT assesses how strongly a person associates two concepts, meaning how closely they are semantically. If a consumer associates the brand NIVEA with “care” and “security” rather than with “autonomy” and “beautify”, he/she will react more rapidly to words in the IAT that are semantically close to “care” and “security” than to those related to “autonomy” and “beautify”.
The IAT finds application in marketing in two ways: The current image of a brand can be assessed by priming several study participants with the brand logo. According to the principle of semantic networks, only characteristics are activated which are associated with the brand. Consumers show a faster reaction time in the IAT to the associated characteristics than to non-associated characteristics. As a result, the implicit meaning of the brand can be depicted.
The second possible application is to test which implicit values a communication measure conveys. First, the participant is presented with the new advertising campaign and subsequently, an IAT is performed. Based on the results an assessment can be made of whether the new measure carries the desired implicit values.
For example, an advertising campaign from NIVEA can be tested if it conveys implicitly the aimed values “care” and “security”. If different marketing measures carry distinct implicit motives, the brand image is diluted and the effect of the message is weaker than desired.
AI Association Prediction
From our experience with implicit testing in our company NEURO FLASH we developed “AI association prediction”. Based on the data of implicit tests, such as the IAT, it was found in 2017 that large so called “word embedding models” include implicit association patterns (Caliskan et al. 2017). Therefore we build software which made these associations accessible to anyone, instantly.
The strength of this approach is to understand the connection between different concepts visible with colors and lines.
To illustrate, have you ever been in a restaurant and the waiter gave beer to the man and lemonade to the woman, no matter who ordered what? He did that, because his brain could more easily think of the connection between man and beer.
As established before, subconscious associations are a big part of the way we think and strongly influence our behavior. By using very large text datasets, we can compute the associations more common for millions of consumers.
The Neuro Flash Software enables content writers and Brand communicators to see the associations connected with their Brand or Message anywhere on the planet. Results are instant and the machine even suggests new content that is more able to achieve your desired subconscious associations.
Implicit market research and explicit market research should go hand in hand to help decision makers to understand consumers and choose the right content to reach and excite them.
Traditional methods have a number of implicit biases that can alter the accuracy of results. However, explicit consumer responses are often quite straight forward to interpret.
Implicit market research methods, like the Implicit association test, are more difficult to setup and reaction times are not always the easiest data to interpret. Nevertheless, implicit desires are often a better predictor for brand and communication success.
To bridge both worlds, we have developed an AI association prediction software which enables brands to get instant consumer insights and which gives concrete predictions on the best content to use.
- Bielefeld, K. W. (2012). Die theoretischen Grundlagen der Arbeit. In Consumer Neuroscience(pp. 9-332). Springer Fachmedien Wiesbaden.
- Scheier, C., & Held, D. (2012). Was Marken erfolgreich macht: Neuropsychologie in der Markenführung (Vol. 97). Haufe-Lexware.
- Caliskan, A., Bryson, J. J., & Narayanan, A. (2017). Semantics derived automatically from language corpora contain human-like biases. Science, 356(6334), 183-186.
- Greenwald, A. G., McGhee, D. E., & Schwartz, J. L. (1998). Measuring individual differences in implicit cognition: the implicit association test. Journal of personality and social psychology, 74(6), 1464.
- Implicit Association Test Wikipedia Link
- Kahneman, Daniel, Schnelles Denken, Langsames Denken, München 2012
- Gawronski, B. and De Houwer, J. (2014). Implicit Measures in Social and Personality Psychology. In H. T. Reis, & C. M. Judd (Eds.), Handbook of research methods in social and personality psychology (2nd edition). New York: Cambridge University Press.
- Greenwald, A. G., Poehlman, T. A., Uhlmann, E. L., & Banaji, M. R. (2009). Understanding and using the Implicit Association Test: III. Meta-analysis of predictive validity. Journal of personality and social psychology, 97(1), 17.
- Snelders, D. & Hekkert, P. (1999). Association measures as predictors of product originality. Advances in Consumer Research, 26, 588-592.
- Reber, R., Schwarz, N. & Winkielman, P. (2004). Processing fluency and asthetic pleasure: Is beauty in the perceiver’s processing experience? Personality and Social Psychology Review, 8, 364-382.
- Huth, A. G., de Heer, W. A., Griffiths, T. L., Theunissen, F. E., & Gallant, J. L. (2016). Natural speech reveals the semantic maps that tile human cerebral cortex. Nature, 532(7600), 453-458.