ArticlePDF Available

'Mind Mining': Better Customer Understanding by Applying Big Data Analysis to Neuromarketing

Authors:

Abstract

The added value of advanced data mining techniques is their ability to identify hidden structures (unknown relations) in large bodies of data. In contrast, the measurement of hidden signals from the mind and body in order to illuminate the customer’s conscious and unconscious thinking is the expected benefit of applying neuromarketing tools. In the present article, a fruitful cooperation for a better customer understanding is suggested by applying data mining techniques to neuromarketing data. The result might be called something new ... ”mind mining.”
This ar ticle (tex tual and pict orial conten t) is subject to c opyright . For educati onal purpose o nly.
More information: Neuromarketing Science & Business Association (www.nmsba.com)
NEURO MARK ETI NG 15 / 201610
‘Mind Mining:
Better Customer Understanding by Applying
Big Data Analysis to Neuromarketing
By Steen Schmidt, Philipp Reiter
In a recent article in the Neuromarketing Theory &
Practice, Carla Nagel raised the highly interesting
question about “competing or completing for better
consumer understanding” (Nagel 2015, p.24) regarding
neuromarketing and big data, both commonly considered
as “the next big thing” in marketing. Beyond the buzz,
both elds can provide valuable insights for marketers
about customer choices and behavior. In detail, big data
applications and neuromarketing tools have in common
the generation, collection, and analysis of large amounts
of data, but more importantly, they share a need to
extract potentially useful information for supporting
management decision-making. However, that is the
bottleneck in both of these emerging elds. It is easy
to store and receive tons of data from various sources,
or to apply innovative neuromarketing methodologies
such as facial coding and implicit response measures.
By all means, those techniques are becoming more
common business practices every day. In addition,
the development of new hardware and software has
helped to increase the acceptance and utilization
of both elds. However, it takes more than simply
collecting data (e.g., number of tweets per hour) and
easy information extraction (e.g., level of joy when a key
visual is perceived) to create any competitive advantage.
Indeed, it is much more dicult to turn knowledge into
actionable insights for business decision- making than it
is to acquire and crunch it.
Given the unimaginable amount of rich, unstructured,
and unrelated data available today from unlimited and
heterogeneous sources, what is needed is an ecient
search method for detecting patterns and extracting
insights from that mountain of data. Conventional
analysis methods such as widely-used correlation or
linear regression techniques are stretched to their
limits fairly quickly by the demands of both big data
and neuromarketing data. Heftier data mining tools,
such as neural network analytics, are needed to process
all the retrieved data and uncover hidden patterns
of knowledge. That is the pitch point for a fruitful
cooperation between big data and neuromarketing: apply
data mining techniques from big data to neuromarketing
data in order to achieve eective “mind mining” as
illustrated in Figure 1.
Figure 1: Pitch point between big data and neuromarketing
The added value of advanced data mining techniques is their ability to identify hidden structures (unknown
relations) in large bodies of data. In contrast, the measurement of hidden signals from the mind and body in
order to illuminate the customer’s conscious and unconscious thinking is the expected benet of applying
neuromarketing tools. In the present article, a fruitful cooperation for a better customer understanding
is suggested by applying data mining techniques to neuromarketing data. The result might be called
something new ... ”mind mining.”
background
Mind
mining
Big data
Processing
Neuro-
marketing
Data
This ar ticle (tex tual and pict orial conten t) is subject to c opyright . For educati onal purpose o nly.
More information: Neuromarketing Science & Business Association (www.nmsba.com) 11NEUROMA RKE TING 1 5 / 2016
» Page 12
Data mining techniques for uncovering hidden
structures in huge datasets
Companies are collecting more and more data every
day, often just because they can, even though they may
not know how, why or when to use it to improve their
business outcomes. Ordinary and isolated data “silos”
will not automatically support decision making. Instead,
data need to be converted into meaningful knowledge
to reveal valuable insights for evidence-based
management, in particular the identication of consistent
patterns and relationships between variables. Therefore,
the challenge is not so much to get the data, but more
to analyze the (right) data for knowledge discoveries to
support real-time and future actions.
In this respect, using predictive analytics based on data
mining techniques is receiving increasing popularity.
A famous example of predictive data mining is Barak
Obama’s election campaign of 2012 (Issenberg, 2012).
Obama’s analytics team identied the interests of
individual voters and predicted which voters would be
positively inuenced by various campaign touchpoints
such as door knock, social media or TV ad. Specically,
the Obama campaign found the greatest value of
advanced data mining techniques to be in their ability to
identify hidden structures and unknown relations, which
would be beyond the capacity of any one human mind to
comprehend, or even recognize.
Data mining techniques are not limited by the restrictions
of conventional (multivariate) analysis methods. They can
reveal complex, nonlinear and dynamic relationships -
such as those found in living systems - in a practical way.
The history of the last 25 years – from the birth of the
World Wide Web to the thoroughgoing digitalization and
automation of virtually every domain of daily life –has
paved the way to the current era of big data. Today, the
role of the data scientist may very well be “the sexiest
job of the 21st century” as recently proclaimed in the
Harvard Business Review by Davenport & Patil (2012).
Neuromarketing tools for uncovering hidden
structures in the mind
As an applied science, neuromarketing uses tools
from brain research, cognitive neuroscience,
neuropsychology, and social psychology, and other
emerging disciplines. The motivation behind using tools
such as electroencephalography (EEG), electrodermal
activity (EDA), or latency-based measures (e.g. implicit
association test) is to resolve marketing issues, especially
with regard to the classic “four Ps” of marketing
– Product, Price, Promotion and Place. Among all
these source disciplines, neuroscience has provided
considerable insights for marketing science and business
practice by exploring human decision-making from a
consumer research perspective, e.g. the winner-take-all
/ rst choice brand eect indicating customer’s favor
for a brand or product by a reduced activation in brain
areas related to working memory and reasoning, and
an increased activity of areas associated with emotion
processing and self-reections (Deppe et al., 2005). It
is not surprising that expectations for neuromarketing
have been raised in the recent past with the promise
of tapping into customers’ “black box” of unconscious
and automatic, so-called implicit processes (Ariely &
Berns, 2010). Marketers have expressed great hope that
neuroscience-based techniques can reveal knowledge
about consumer preferences that are unobtainable
through conventional methods such as focus groups,
interviews, or self-reports. The measurement of that
hidden information in order to illuminate the customer’s
mind is the expected benet of applying neuromarketing
tools.
A use case: applying articial neural networks
to uncover hidden patterns in implicit and
explicit user experience data
In order to assess how an integrative analysis approach
might work, a case study was conducted that examined
people’s implicit and explicit user experiences (UX)
while they interacted with a smartphone interface. While
participants engaged in various tasks (e.g. writing a text
message, installing an app), two implicit responses were
recorded – task-related average cognitive workload
(using EEG) and average arousal (using EDA). Also
collected were participants’ explicit ratings of two
aspects of the UX, task-related usefulness and ease
of use (both self-reported on a 5-point Likert scale). In
addition, task outcome (fail or success), task satisfaction
(on a 5-point Likert scale) and product recommendation
(11-point semantic dierential) were captured as UX key
performance indicators (see Figure 2).
This ar ticle (tex tual and pict orial conten t) is subject to c opyright . For educati onal purpose o nly.
More information: Neuromarketing Science & Business Association (www.nmsba.com)
NEURO MARK ETI NG 15 / 201612
» continued reading from page 11
Nineteen subjects participated in the lab study. Each
subject performed  ve UX tasks. Thus, 95 UX cases
were included in the  nal dataset. A Universal Structure
Modeling (USM) analysis was applied to the dataset
using the statistical software package NEUSREL (Neusrel,
2015). USM uses a Bayesian neural network approach to
test structural equation models (SEM). This data mining
technique is able to quantify and visualize nonlinear
and interactive e ects among model constructs. USM
represents a more exploratory approach to  nding and
testing hidden model structures, in contrast to more
conventional SEM approaches (such as LISREL or Partial
Least Squares) that can only estimate models with
hypothesized paths pre-speci ed (Buckler & Hennig-
Thurau, 2008).
Selected results are presented in Figure 2. With regard to
the coe cient of determination (R-squared value) results,
the integrated approach (incorporating both implicit and
explicit measures) and the USM approach performed best
for predicting the dependent variable ‘task satisfaction.’
Indeed, the R-squared value is roughly twice as high as
that for the conventional UX assessment (including explicit
measures only) and the analytic approach (using modest
linear structural equation modeling) – two approaches
still often applied in contemporary marketing research.
The superior performance by the integrative measurement
approach appears to be based largely on the detection
of nonlinearity and interaction e ects (cf. Figure 2).
However, the impact of the key performance drivers (KDIs)
varies depending on the chosen KPIs. With regard to task
success and task satisfaction, results indicate highest
e ect strength (here: average simulated e ect, ASE)
by cognitive workload. In accordance to the research in
the  eld of human-computer interaction, the impact of
workload is negative, meaning high workload increases
the chances of errors (failed task outcomes) and decreases
the perceived satisfaction (since subjects feel frustrated).
For product recommendations, the explicit measurement
of usefulness shows highest in uence. The higher the
perceived usefulness, the higher the rated intention to
recommend the product. Finally, the USM prediction
accuracy regarding the estimated task outcome is highest
when integrated implicit and explicit measures are used.
This is especially true when estimating a succeeded task
outcome. Taken together, these results demonstrate that
performance can most accurately be predicted using an
integrated, implicit-explicit analysis approach.
Predictive Mind Mining Analysis:
What belongs together comes together
Returning to Carla Nagel’s question at the beginning
of this article about the competing or complementary
relationship between big data and neuromarketing, we
see in this case that integrating the best elements of both
approaches provides the greatest potential enhancing
customer understanding and improving marketing
business results. Applying sophisticated tools from
neuromarketing for data collection in cooperation with
advanced data mining techniques from big data for data
analysis enables an e cient knowledge and decision-
support platform for optimizing future management
actions. In conclusion, one could state in reference to
the HBR quote, “Predictive Mind Mining Analysis” is the
hottest job in marketing.”
Dr. Ste en Schmidt is Assistant Professor at the Institute
of Marketing and Management, Leibniz University of
Hannover. Phillipp Reiter is partner and COO of Eye Square
References
Ariely, D., Berns, G.S. (2010): Neuromarketing: the hope and hype of
neuroimaging business. Nature Reviews Neuroscience, 11(4), 284–292.
Buckler, F., Hennig-Thurau, T. (2008). Identifying Hidden Structures in
Marketing’s Structural Models Through Universal Structure Modeling:
An Explorative Bayesian Neural Network Complement to LISREL and
PLS. Marketing - Journal of Research and Management, 4(2), 47–66.
Davenport, T.H., Patil, D.J. (2012). Data Scientist: The Sexiest Job of the
21st Century. Harvard Business Review, October 2012, 70–76.
Deppe, M., Schwindt, W./Kugel, H., Plassmann, H., Kenning. P. (2005.
Nonlinear responses within the medial prefrontal cortex reveal when
speci c implicit information in uences economic decision making.
Journal of Neuroimaging, 15(2), 171–182.
Issenberg, S. (2012). How President Obama’s campaign used big
data to rally individual voters. MIT Technology Review, www.
technologyreview.com/featuredstory/509026/how-obamas-team-
used-big-data-to-rally-voters. Accessed 2015/12/11.
Nagel, C. (2015). The Battle: Big Data vs. Neuromarketing.
Competing or complementing for better consumer understanding?
Neuromarketing Theory & Practice, 13, August 2015, 24–26.
Neusrel (2015). Neusrel Analytic Soft ware. www.neusrel.com/causal-
analytics. Accessed 2015/12/07.
Figure 2: Conceptual modeling and selected study results
theme consumer habits
... Der dritte und letzte Schritt widmet sich detaillierter der Interpretation der Wirkungsverläufe, um darüber beispielsweise z.V.g.31M A R K T-U N D S O Z I A L F O R S C H U N G S C H W E I Z 2 018konnte wie in vergleichbaren Studien (vgl.Schmidt et al., 2015;Schmidt & Reiter, 2016)nachgewiesen werden, dass der ent- wickelte Analyseansatz des «Mind Mining» für ein «Customer Insights» 4.0 (explizite und implizite Messung sowie Struktur- gleichungsanalyse) eine rund dreimal so hohe Erklärungsgüte mit Blick auf die Nutzungsabsicht als essenziellen Verhaltens- indikator erreicht im Vergleich zum klassischen Analyseansatz (explizite Messung sowie multiple lineare Regression). Konkret gewährleistet das «Customer Insights» 4.0 eine substanziell hohe Erklärungsgüte (R2=0.84), ...
Book
Full-text available
Disruptive Forschungsmethoden wie «Machine Learning» und implizite Erhebungsverfahren verändern nicht nur die Marketingforschung strukturell, sondern ermöglichen in der Symbiose auch inhaltlich einen präziseren Blick in die Köpfe der Kunden - dem «Mind Mining» für ein «Customer Insights» 4.0.
Article
Full-text available
In the modern era, customer retention is one of the visionary tasks for the firm to sustain the customer from the competitors. One of the major issues for the firm is to focus on the existing customer rather than searching for new customers. Hence to overcome these obstacles eye-tracking is a long-lasting technique that will be useful for the marketer to attract customers and retain them for the remote future. It is also cost-effective and the easiest way to admire the customer. Visual attention always creates awareness in the mind of the consumer when they are attracted it will make them for purchasing the product. This paper mainly focuses on using of Eye-tracking technique of neuromarketing to identify the factors influencing the consumer attitude and perception towards durable goods. For this purpose Factor Analysis and Principle Component Analysis is used.
Chapter
Powerful brands have created strong and deep associations in the minds of consumers. Adapting and enhancing brand knowledge to improve mental availability is an efficient approach for long-term brand growth. Indeed, brands exist in consumers’ memories as stored experiences for future consumption opportunities. Therefore, a brand that is not memorized in consumers’ minds is an empty shell, interchangeable, invisible, and de facto nonexistent from a marketing perspective. This chapter presents a case study to show how advanced methods and evidence-based insights from neuromarketing and causal analysis can be used to transform a strong brand into an even stronger brand.
Article
Full-text available
The transformation of societies is the core business of innovative social enterprises. These hybrid organizations require deep customer insights to address the manifold needs of the public and business. The present work shows how marketing research enhanced the marketing intelligence of the social enterprise known as Little Sun, enabling the company to better promote its latest innovative product that uses “the power of sustainable light and energy to transform lives”.
Article
Full-text available
Universal Structure Modeling (USM), introduced herein, offers an exploratory complement to confirmatory structural equation modeling methods such as covariance-based structural equation modeling (e.g., LISREL) and partial least squares. Because USM combines the iterative methodology of partial least squares with a Bayesian neural network approach involving a multilayer perceptron architecture, it enables researchers to identify “hidden” structures within their models and highlights theoretically unproposed model paths, nonlinear relations among model variables, and interactive effects. Using synthetic data, the authors demonstrate USM’s ability to identify linear and nonlinear relationships and provide evidence that the method does not overfit the original data. They also find hidden paths, nonlinearity, and interactions in two structural models published in the Journal of Marketing, which illustrates the practical relevance of USM. They provide recommendations for researchers regarding in which conditions and how USM should be used.
Article
Full-text available
The application of neuroimaging methods to product marketing - neuromarketing - has recently gained considerable popularity. We propose that there are two main reasons for this trend. First, the possibility that neuroimaging will become cheaper and faster than other marketing methods; and second, the hope that neuroimaging will provide marketers with information that is not obtainable through conventional marketing methods. Although neuroimaging is unlikely to be cheaper than other tools in the near future, there is growing evidence that it may provide hidden information about the consumer experience. The most promising application of neuroimaging methods to marketing may come before a product is even released - when it is just an idea being developed.
Article
Back in the 1990s, computer engineer and Wall Street "quant" were the hot occupations in business. Today data scientists are the hires firms are competing to make. As companies wrestle with unprecedented volumes and types of information, demand for these experts has raced well ahead of supply. Indeed, Greylock Partners, the VC firm that backed Facebook and LinkedIn, is so worried about the shortage of data scientists that it has a recruiting team dedicated to channeling them to the businesses in its portfolio. Data scientists are the key to realizing the opportunities presented by big data. They bring structure to it, find compelling patterns in it, and advise executives on the implications for products, processes, and decisions. They find the story buried in the data and communicate it. And they don't just deliver reports: They get at the questions at the heart of problems and devise creative approaches to them. One data scientist who was studying a fraud problem, for example, realized it was analogous to a type of DNA sequencing problem. Bringing those disparate worlds together, he crafted a solution that dramatically reduced fraud losses. In this article, Harvard Business School's Davenport and Greylock's Patil take a deep dive on what organizations need to know about data scientists: where to look for them, how to attract and develop them, and how to spot a great one.
Article
The authors used functional magnetic resonance imaging (fMRI) to investigate how individual economic decisions are influenced by implicit memory contributions. Twenty-two participants were asked to make binary decisions between different brands of sensorily nearly undistinguishable consumer goods. Changes of brain activity comparing decisions in the presence or absence of a specific target brand were detected by fMRI. Only when the tar get brand was the participant's favorite one did the authors find reduced activation in the dorsolateral prefrontal, posterior parietal, and occipital cortices and the left premotor area (Brodmann areas [BA] 9, 46, 7/19, and 6). Simultaneously, activity was increased in the inferior precuneus and posterior cingulate (BA 7), right superior frontal gyrus (BA 10), right supramarginal gyrus (BA 40), and, most pronounced, in the ventromedial prefrontal cortex (BA 10). For products mainly distinguishable by brand information, the authors revealed a nonlinear winner-take-all effect for a participant's favorite brand characterized, on one hand, by reduced activation in brain areas associated with working memory and reasoning and, on the other hand, increased activation in areas involved in processing of emotions and self-reflections during decision making.
How President Obama's campaign used big data to rally individual voters. MIT Technology Review, www. technologyreview.com/featuredstory/509026/how-obamas-teamused-big-data-to-rally-voters
  • S Issenberg
Issenberg, S. (2012). How President Obama's campaign used big data to rally individual voters. MIT Technology Review, www. technologyreview.com/featuredstory/509026/how-obamas-teamused-big-data-to-rally-voters. Accessed 2015/12/11. Nagel, C. (2015). The Battle: Big Data vs. Neuromarketing. Competing or complementing for better consumer understanding? Neuromarketing Theory & Practice, 13, August 2015, 24-26.
The Battle: Big Data vs. Neuromarketing. Competing or complementing for better consumer understanding? Neuromarketing Theory & Practice
  • C Nagel
Nagel, C. (2015). The Battle: Big Data vs. Neuromarketing. Competing or complementing for better consumer understanding? Neuromarketing Theory & Practice, 13, August 2015, 24–26.
Neusrel Analytic Software. www.neusrel.com/causalanalytics
  • Neusrel
Neusrel (2015). Neusrel Analytic Software. www.neusrel.com/causalanalytics. Accessed 2015/12/07.