Marketing Research

Twelfth Edition
Carl D. McDaniel and Roger Gates

Chapter #10

Marketing Analytics
Learning Objectives
1. Understand what’s included in marketing analytics.
2. Review techniques for analyzing data.
3. Gain a greater understanding of big data.
4. Explore data mining.
5. Understand differences in analytical for big and little data.
6. Define artificial intelligence, machine learning and deep learning.
7. Outline the key issues regarding consumer privacy.

Copyright ©2021 John Wiley & Sons, Inc. 10-2

Introduction
• Structured data is data that are highly organized and can be represented as
tables made up of rows and columns
o Rows are observations for a single unit of the population
o Columns are variables like age, gender, and income
o Can be shown in Excel
• Unstructured data is data or information that is not neatly structured
o Examples include emails, social media posts, online reviews, photos, and text
• Chapters 12-14 discuss analyzing structured data
• This chapter looks at tools to analyze all types of data

Copyright ©2021 John Wiley & Sons, Inc. 10-3

What Is Marketing Analytics?
• Marketing analytics is the discovery, interpretation, and communication of
meaningful patterns in data
• The insights gained enables marketing managers to better understand their
customers and markets
• The development of marketing analytics has rapidly accelerated based on the
emergence of big data, advances in analytical tools and improvements in
information technology
• Parts of the process
1. Get the data
2. Organize and merge the data
3. Analyze
4. Act
Copyright ©2021 John Wiley & Sons, Inc. 10-4
The Marketing Analytics Process
• Getting the data: we need to go through a process that first involves
identifying data sources and then obtaining and analyzing the data to
answer our questions
• Big data sources:
o Online retailers
o Search engines
o Social media
• Data from traditional sources:
o Internal company data
o Data from traditional primary sources

Copyright ©2021 John Wiley & Sons, Inc. 10-5

The Marketing Analytics Process Continued
• Traditional database / data warehouse systems offer no means of storing
and analyzing unstructured data
• We now have tools for handling these types of data
• Managers who do not act on the results of analytics tend to say the results
are “not actionable”
o Could the issue be understanding or trust?
• Companies such as Google, Twitter, Instagram and Facebook demonstrated
to those in the data analysis business that there is significant value in
storing and analyzing unstructured data
• One of the widely used tools for handling big, unstructured data is Hadoop
Copyright ©2021 John Wiley & Sons, Inc. 10-6
Background on Big Data Issues
• Big data is the accumulation and analysis of massive quantities of
information
• To be able to handle these huge data sets, new tools needed to be
developed
• Big data has the three Vs:
1. Volume refers to the amount of data
2. Variety is concerned with putting together data from different sources
3. Velocity concerns the fact that the data are arriving continuously in data streams
• Big data are more about “what” than “why”

Copyright ©2021 John Wiley & Sons, Inc. 10-7

How Does It Work?
• In 2005 an open-source framework called Hadoop was developed to
analyze big data
• Hadoop consists of four principal modules:
1. A distributed-file system: allows data to be stored in an easily accessible format
2. MapReduce: reads the data and puts it into a format suitable for analysis
3. Hadoop Common: reads the data
4. YARN: manages the resources of the data storage system
• One of the important innovations of Hadoop is its use of networked servers
o This distributes complex queries to many inexpensive servers

Copyright ©2021 John Wiley & Sons, Inc. 10-8

Analyzing Data
1. Descriptive analytics
2. Predictive analytics
3. Prescriptive analytics

Copyright ©2021 John Wiley & Sons, Inc. 10-9

Descriptive Analytics
• In this type of analytics, we summarize data in rudimentary forms to
discern patterns
• This is the essence of old school marketing research analytics
o Cross tabulation tables are common
• The next level is predictive analytics

Copyright ©2021 John Wiley & Sons, Inc. 10-10

Predictive Analytics
• Predictive analytics uses more powerful tools to provide more insight and
more robust decision-making tools
• We use data from primary, secondary and big data sources
• We apply analytical tools such as regression analysis of various types,
discriminant analysis, cluster analysis, neural networks and others
• Examples of problems to address:
o Predicting the impact of improving customer service through our call centers on overall
customer experience
o Estimate the likelihood that an individual will buy or not buy our new product or service
o Classify people into market segments
o Predict which product different consumers are most likely to buy
Copyright ©2021 John Wiley & Sons, Inc. 10-11
Prescriptive Analytics
• Prescriptive analytics find the best course of action in a given situation
• It use the results of descriptive and predictive analytics to guide decision-
making in more specific ways
• Steps:
o You determine what quantity you want to maximize
o You list the decision levers available to you
o You build and calibrate a model that is robust under a wide range of ways of adjusting
the decision levers
o You embed the simulator inside an optimization tool that evaluates a large number of
different ways of setting the decision levers and identifies which combination of lever
settings maximizes your objective
Copyright ©2021 John Wiley & Sons, Inc. 10-12
Example Uses of Prescriptive Analytics
• A company could automatically adjust pricing in response to various factors
• Doctors and hospitals could use prescriptive analytics to choose the best
treatments for patients
• Airlines can adjust fares for different destinations based on multiple factors
• Fire departments can evaluate where to order evacuations

Copyright ©2021 John Wiley & Sons, Inc. 10-13

Data Mining
• Data mining is an umbrella term for analytic techniques that facilitate fast
pattern discovery and model building, particularly with large datasets
• The objective is to identify patterns
• The most powerful data mining tools of today use a combination of
statistics, artificial intelligence, and machine learning
• Neural networks are analytical tools that mimic the processes of the human
brain
• Through a process called backpropagation, the neural network makes
adjustments to improve future predictions and is thus capable of learning
and adjusting
Copyright ©2021 John Wiley & Sons, Inc. 10-14
A Standardized Approach to Data Mining
• A well-known methodology for conducting data mining projects is the
CRISP-DM framework
o Cross-Industry Standard Process for Data Mining
• It is based on a sequence of six steps:
1. Business understanding
2. Data understanding
3. Data preparation
4. Data modeling
5. Evaluation of results
6. Deployment of a plan

Copyright ©2021 John Wiley & Sons, Inc. 10-15

Behavioral Targeting is a Big Deal
• A major reason why companies “mine data” is the ability of analytics tools to
do behavioral targeting
• Behavioral targeting is the use of online and offline data to understand a
consumer’s habits, demographics, and social networks in order to increase the
effectiveness of online advertising
• eXelate gathers online consumer data through deals with hundreds of
websites
• In January 2019, Amazon reported that more than 100 million Alexa powered
devices had been sold
• The addition of social networking data has been a huge boost to behavioral
targeting in the form of something called remarketing or retargeting
Copyright ©2021 John Wiley & Sons, Inc. 10-16
Machine and Deep Learning
• Machine learning is where machines can learn by experience and acquire
skills without human involvement
o It perform predictive analytics far faster than any human can
• Deep learning is where artificial neural networks, algorithms inspired by the
human brain, learn from large amounts of data as in machine learning but
now we add backpropagation where machines learn from their mistakes

Copyright ©2021 John Wiley & Sons, Inc. 10-17

Artificial Intelligence (AI)
• Roughly, AI is any device that perceives its environment and takes actions
that maximize its chance of successfully achieving its goals
o Also known as intelligent agents
• AI is the end payoff that comes from the use of big data, data mining,
predictive analytics, and deep learning
• All of this can be taking place without “supervision” with the machine
managing the whole process
• The uses for AI number in the millions

Copyright ©2021 John Wiley & Sons, Inc. 10-18

A Few Uses of AI
• Virtual assistants such as Alexa and Hey Google
• Translations apps
• Vision for cars and other vehicles
• Chatbots and service bots
• Image colorization
• Facial recognition
• Medicine and pharmaceuticals
• Personalized shopping and entertainment

Copyright ©2021 John Wiley & Sons, Inc. 10-19

Surge or Dynamic Pricing
• Surge pricing is a pricing strategy in which businesses vary prices for
products or service based on current level of demand or demand at the
specific moment
o Also referred to as dynamic pricing, demand pricing, or time-based pricing
• The ability to price goods and services dynamically based on real time
demand can already be found or is spreading to other industries
• There is a fine line between surge pricing and price gouging

Copyright ©2021 John Wiley & Sons, Inc. 10-20

Data Visualization
• A picture is worth a thousand words
• Pictures (and graphs) can communicate much more effectively and
efficiently than words
• Data visualization is graphic tools that make data understandable to a wider
audience than just analysts and data scientists
o It include charts, graphs, photos and videos, even tables, and maps if appropriate

Copyright ©2021 John Wiley & Sons, Inc. 10-21

Infographics
• An infographic is a collection of imagery, charts, and minimal text that gives
an easy-to-understand overview of a topic
• There are no particular rules for creating them
• To create good data visualization, you need to consider the type of data,
the audience, and the message you wish to convey

Copyright ©2021 John Wiley & Sons, Inc. 10-22

Sample Infographic

Copyright ©2021 John Wiley & Sons, Inc. 10-23

Marketing Dashboards
• Marketing dashboards are a reporting tool that provides a comprehensive
snapshot of performance-based analytics, Key Performance Indicators
(KPIs), and other marketing metrics
o They are a specialized form of infographic
• There are three types of marketing dashboards:
1. Strategic dashboards monitor performance towards high-level goals
2. Tactical dashboards use past data to identify trends that can impact future plans
3. Operational dashboards are the most specific and are often used to track the
performance of a department

Copyright ©2021 John Wiley & Sons, Inc. 10-24

Sample Marketing Dashboard

Copyright ©2021 John Wiley & Sons, Inc. 10-25

Privacy vs. Customization
• The downside to big data is the associated loss of privacy
o Big data almost assuredly includes detailed information about the behavior and activities of
specific, identifiable individuals
• An emerging technique that is upsetting many privacy advocates is called scraping
o This is where firms harvest online conversations and collect personal details from social
networking sites
• The Internet is in an arms race over control of personal data
• The “right to privacy” is the right to have information that was never made public
to remain private
• The “right to be forgotten” allows individuals to have information, videos, or
photos about themselves deleted from Internet records
Copyright ©2021 John Wiley & Sons, Inc. 10-26
Key Terms
• Artificial intelligence or AI • Descriptive analytics
• Backpropagation • Machine learning
• Behavioral targeting • Marketing analytics
• Big data • Marketing dashboard
• CRISP-DM Framework • Neural networks
• Data mining • Predictive analytics
• Data visualization • Prescriptive analytics
• Deep learning • Surge pricing

Copyright ©2021 John Wiley & Sons, Inc. 10-27