Digital Marketing Analytics 2nd Edition Kevin

Hartman pdf download

https://ebookgate.com/product/digital-marketing-analytics-2nd-edition-kevin-hartman/

★★★★★ 4.9/5.0 (36 reviews) ✓ 213 downloads ■ TOP RATED

"Fantastic PDF quality, very satisfied with download!" - Emma W.

DOWNLOAD EBOOK
 Digital Marketing Analytics 2nd Edition Kevin Hartman pdf
download

TEXTBOOK EBOOK EBOOK GATE

Available Formats

■ PDF eBook Study Guide TextBook

EXCLUSIVE 2025 EDUCATIONAL COLLECTION - LIMITED TIME

INSTANT DOWNLOAD VIEW LIBRARY

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Commonsense Direct Digital Marketing 5th Edition Drayton

Bird

https://ebookgate.com/product/commonsense-direct-digital-
marketing-5th-edition-drayton-bird/

ebookgate.com

Digital Nature Photography and Adobe Photoshop 1st Edition

Kevin Moss

https://ebookgate.com/product/digital-nature-photography-and-adobe-
photoshop-1st-edition-kevin-moss/

ebookgate.com

Introduction to Data Analytics for Accounting 2nd Edition

--

https://ebookgate.com/product/introduction-to-data-analytics-for-
accounting-2nd-edition/

ebookgate.com

Mastering Predictive Analytics with R 2nd edition Edition

Forte

https://ebookgate.com/product/mastering-predictive-analytics-
with-r-2nd-edition-edition-forte/

ebookgate.com
The Economics of Data Analytics and Digital Transformation
The theorems laws and empowerments to guide your
organization s digital transformation Bill Schmarzo
https://ebookgate.com/product/the-economics-of-data-analytics-and-
digital-transformation-the-theorems-laws-and-empowerments-to-guide-
your-organization-s-digital-transformation-bill-schmarzo/
ebookgate.com

Emarketing Excellence Planning and Optimizing your Digital

Marketing 4th Edition Dave Chaffey

https://ebookgate.com/product/emarketing-excellence-planning-and-
optimizing-your-digital-marketing-4th-edition-dave-chaffey/

ebookgate.com

Niche Envy Marketing Discrimination in the Digital Age 1st

Edition Joseph Turow

https://ebookgate.com/product/niche-envy-marketing-discrimination-in-
the-digital-age-1st-edition-joseph-turow/

ebookgate.com

Facebook Marketing Designing Your Next Marketing Campaign

2nd Edition Justin Levy

https://ebookgate.com/product/facebook-marketing-designing-your-next-
marketing-campaign-2nd-edition-justin-levy/

ebookgate.com

International Marketing 2nd edition Donald Baack

https://ebookgate.com/product/international-marketing-2nd-edition-
donald-baack/

ebookgate.com
 DIGITAL
MARKETING
ANALYTICS
In Theory And In Practice

Second Edition

KEVIN HARTMAN
 Copyright © 2020
Kevin Hartman
Digital Marketing Analytics
In Theory And In Practice
Second Edition
All rights reserved.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means,
including photocopying, recording, or other electronic or mechanical methods, without the prior written
permission of the publisher, except in the case of brief quotations embodied in critical reviews and
certain other non-commercial uses permitted by copyright law.
Kevin Hartman
Printed in the United States of America
First Printing April 2020
First Edition April 2020
Second Printing September2020
Second Edition September2020
ISBN: 979-8638631321
 To my family who has loved me,
my friends who have supported me,
and my colleagues who have inspired me.
 PRAISE FOR DIGITAL MARKETING
ANALYTICS
“Kevin has made science accessible. With the immense potential Digital
holds, I’m confident this book will allow you to unlock new ways to deepen
customer relationships and fatten your profits!”
Avinash Kaushik Digital Marketing Evangelist - Google, Author –
Web Analytics: An Hour A Day and Web Analytics 2.0

“Kevin combines the experience of a skilled analytics practitioner with the

expertise of an academic. In this book, he presents a fresh look at the field
through the lens of history: from its earliest beginnings to its modern
incarnation.”
Mike Chapple Director, M.S. in Business Analytics – Mendoza
College of Business, The University of Notre Dame

“Digital Marketing Analytics is an important resource for all marketers,

whether you are early in your career and learning or a seasoned practitioner
who wants to remain sharp. This book will help you understand the changing
dynamics and apply that thinking to maximize your business results.”
Karen Sauder Vice President – Google

“This book is an excellent examination of data analytics, providing both a

solid theoretical foundation and a deep dive into practical application.
Readers of all levels of digital savvy will benefit from Kevin’s expertise
across platforms, tools, and approaches.”
W. Brooke Elliott Associate Dean and EY Professor –
Gies College of Business, University of Illinois

“The power of Digital Marketing Analytics is that it provides an insightful

approach to the art of analytics while demystifying its science. Whether
you’re a big brand or small company, Kevin’s book is a fantastic resource for
any marketer who wants to do more with data.”
Mike Clarke Director, Research – Facebook

“Kevin has honed wisdom gathered over decades as a practitioner and

lecturer to develop a strong foundation of analytics. As the evolution of
marketing analytics continues to accelerate, this foundation will prepare you
to successfully tackle any new challenge, approach, or data source you will
encounter throughout your career!”
Stan Balanovskiy Head of Ads Measurement and Insights – Quibi
 TABLE OF CONTENTS
INTRODUCTION
PART 1: The Day the Geeks Took Over
PART 1 | LESSON 1
The ART+SCIENCE Mind
The Three Roles of the Analyst
The Evolution of the Analyst in the Firm
Characteristics of the Successful Analyst
In Summary: The ART+SCIENCE Mind
PART 1 | LESSON 2
The Early History of Data Analytics
Epoch I: ‘Early Maps and Diagrams’
Epoch II: ‘Measurement & Theory’
Epoch III: ‘New Graphical Forms’
Epoch IV: ‘The Beginnings of Modern Graphics’
Epoch V: ‘The Golden Age of Data Analytics’
Epoch VI: ‘The Modern Dark Ages’ of Data Analytics
Epoch VII: ‘The Rebirth of Data Analytics’
Epoch VIII: ‘High-Definition Data Analytics’ (1975-1994)
In Summary: The Early History of Data Analytics
PART 1 | LESSON 3
The Contemporary History of Data Analytics
Epoch I: ‘When Anything Was Possible’
Epoch II: ‘The Bubble and the Burst’
Epoch III: ‘The Seeds of Prosperity’
Epoch IV: ‘The Age of Unicorns’
The Significance of 2014
In Summary: The Contemporary History of Data Analytics
PART 1 | LESSON 4
The Rise of Modern Data Analytics
P&G’s Three-Step Model of Marketing
The Introduction of the ‘Zero Moment of Truth’ (ZMOT)
Technological Influences
Market Influences
 Placing Today’s Data Creation in Context
Rise of the Machines: Machine Learning and Analytics
In Summary: The Rise of Modern Data Analytics
PART 2: Consumer/Brand Relationships
PART 2 | LESSON 1
Online Video
Online Video (OLV) Market Share
Consumer Behavior and OLV
Key OLV Measures
In Summary: Online Video
PART 2 | LESSON 2
Online Search
Search Engine Market Share
Search Engine Management Versus Search Engine Optimization
The Importance of Search Rank
Consumer Behavior and Search
Key Search Measures
In Summary: Online Search
PART 2 | LESSON 3
Display Media
Display Media Market Share
Evolution of Display Ad Formats
Key Display Media Measures
In Summary: Display Media
PART 2 | LESSON 4
Social Media
Social Media Market Share
Consumer Behavior and Social Media
How Brands Use Social Media
Key Social Media Measures
In Summary: Social Media
PART 2 | LESSON 5
The Consumer Decision Journey
The Introduction of McKinsey’s CDJ
The Steps of the CDJ
Applicability of the CDJ
 How Ad Blockers Affect Digital Measurement
In Summary: The Consumer Decision Journey
PART 3: The Science of Analytics
PART 3 | LESSON 1
Digital Data Infrastructure
Categories of Data
Digital Data Collection: Cookies, Tags and Sign-Ins
Data Availability and Value
Data Collection and the Effect of Privacy Concerns
In Summary: Digital Data Infrastructure
PART 3 | LESSON 2
Digital Measurement
The Four Categories of Digital Measurement
Making the Intangible Tangible: Brand Impact
Measuring Sales Levers: Consumer Outcomes
Understanding What Is Important: Customer Value
Assessing Marketing Effectiveness: Attribution
Getting to True Lift: Incrementality Testing
A Word on Clarity through Measurement Multiplicity
A Word on Digital Measurement Challenges
In Summary: Digital Measurement
PART 3 | LESSON 3
Analytics and Dataviz Tools
Key Categories of Digital Analysis Tools
Evaluation Criteria 1: Data Flexibility
Evaluation Criteria 2: Ease of Use
A Word on SQL
A Word on R and Python
A Word on Tools < The Analyst
In Summary: Analytics and Dataviz Tools
PART 3 | LESSON 4
Digital Marketing Maturity
BCG’s View of Digital Marketing Maturity
Bain’s View of Digital Marketing Maturity
Deloitte and MIT Sloan’s View of Digital Marketing Maturity
Benefits of Digital Marketing Maturity
 In Summary: Digital Marketing Maturity
PART 4: The Art of Analytics
PART 4 | LESSON 1
Navigating to Your Big Idea
The Four Steps of the Marketing Analytics Process (MAP)
First Step: Plan
Second Step: Collect
Third Step: Analyze
Step Four: Report
‘The Ask Behind the Ask’
In Summary: Navigating to Your Big Idea
PART 4 | LESSON 2
Planning for Your Analytics Expedition
When Objective-Setting Goes Awry
The Six Marketing Objectives
Determining Your Marketing Objective
Prioritizing Objectives
Building an Analysis Plan around Your Objective
A Word on Anchoring Bias
In Summary: Planning for Your Analytics Expedition
PART 4 | LESSON 3
Collecting Data, Data Everywhere
Unstructured Versus Structured Data
Collecting Data
Ensuring Data Quality
A Word on Data Tidying
A Word on Managing Collected Data
In Summary: Collecting Data, Data Everywhere
PART 4 | LESSON 4
Analyzing for Insights
The Five Categories of Marketing Data Analysis
A Word on the Importance of Context
Finding Patterns in Data
Countering Bias in Analysis
In Summary: Analyzing for Insights
PART 5: Storytelling with Data
PART 5 | LESSON 1
Pictures You See with Your Brain
Visual Perception and the Door Study
Preattentive Attributes
A Word on the Effect of Neuroscience
In Summary: Pictures You See with Your Brain
PART 5 | LESSON 2
Evaluation Framework for the Visual Form
Evaluating the Effectiveness of Dataviz
Connecting Data Visual Elements to the Analysis Journey
Understanding the Components of Visual Form
In Summary: Evaluation Framework for the Visual Form
PART 5 | LESSON 3
Sophisticated Use of Contrast
Size Contrast
Color Contrast
Shape Contrast
Contrived Contrast
Bringing Contrast to Numbers through Context
A Word on the Importance of Sketching Visuals
In Summary: Sophisticated Use of Contrast
PART 5 | LESSON 4
Ensuring Clear Meaning
Properly Titling Your Visual
Highlighting Messages Visually
A Word on Leveraging Emotions
In Summary: Ensuring Clear Meaning
PART 5 | LESSON 5
Refined Execution through Visual Polish
Color Theory Application in Charts
Harmonic Color Themes
Additional Color Considerations
Elements of Refined Design
Testing to Improve the Visual
A Word on Combining Tools to Refine Visuals
In Summary: Refined Execution through Visual Polish
PART 5 | LESSON 6
On Your Feet and Getting Your Story Across
‘The McCandless Method’ of Data Presentation
The McCandless Method in Practice
A Word on Presentation Style
The Storyteller’s Dash: Dashboarding Data Effectively
A Word on Building Habits over Time
In Summary: On Your Feet and Getting Your Story Across
CONCLUSION
GLOSSARY
ABOUT THE AUTHOR
 INTRODUCTION

E ven if you know nothing about digital marketing analytics, digital

marketing analytics knows plenty about you. It’s a fundamental,
inescapable, and permanent cornerstone of modern business that affects
the lives of analytics professionals and consumers in equal measure. This
book is an attempt to provide the context, perspective, and information
needed to make analytics accessible to people who understand its reach and
relevance and want to learn more.
The contents of this book reflect my decades of work in the digital analytics
space, with most of that time spent leading large analytics teams at a major
advertising agency and a global technology company. I’ve also had the
privilege of teaching analytics at the graduate level for nearly 10 years. This
textbook represents the synthesis of those two worlds: the academic and the
actionable, the history and the hands-on, and the analysis and the application.
I hope readers find this book to be a valuable, enlightening, and engaging
review of the ever-evolving analytics field. I’m profoundly grateful for the
opportunity to share my experience, knowledge, and insights with you.
Kevin Hartman, September 2020
 PART 1: The Day the Geeks Took
Over
Lesson 1 The ART+SCIENCE Mind
Lesson 2 The Early History of Data Analytics
Lesson 3 The Contemporary History of Data Analytics
Lesson 4 The Rise of Modern Data Analytics

The ubiquity of data analytics today isn’t just a product of the past half-
century’s transformative and revolutionary changes in commerce and
technology. Humanity has been developing, analyzing, and using data for
millennia. Understanding where digital marketing analytics is now and where
it will be in five, 10, or 50 years requires a holistic and historical view of our
relationship and interaction with data.
Part 1 looks at modern analysts and analytics in the context of its distinct
historical epochs, each one containing major inflection points and laying a
foundation for future advancements in the ART + SCIENCE that is modern
data analytics.
 PART 1 | LESSON 1
The ART+SCIENCE Mind

Five things discussed in this lesson:

Data analytics affects every aspect of business
Opportunities for data analysts have never been greater, and
expectations of data analytics to influence business decisions have
never been higher
Analysts should focus on skill development in three
important roles: the “Data Strategist,” the “Techie,” and the “Data
Designer”
Analysts’ roles have evolved over time, moving from
focused areas of expertise to broad and balanced skills
Successful analysts develop deep skills but, just as
importantly, develop a mindset that helps set them up for success

Digital marketing analytics, in my humble opinion, is the most vibrant and

undiluted blend of art and science in today’s business world. No other pursuit
requires its practitioners to balance such a depth of scientific application with
a more poignant need to communicate with audiences who neither understand
nor have patience for the technical underpinnings of that science. The science
of digital marketing analytics presents as verbose and academic. A purely
artistic display of analytics output suggests the same lack of substance as a
charlatan’s medicine show. This pursuit only works through a harmony of the
two: “ART+SCIENCE.”
Data analytics – the broad science of applying data analysis to solve business
problems – affects every aspect of business. From finance to operations to
human resources to marketing (and everything in between), it’s understood
“the world’s most valuable resource is no longer oil, but data.”[1] As a result,
everyone in a business today becomes an analytics consumer. Moreover, the
stakes have never been higher for analysts as the opportunities for data
analysis – and the expectations for their benefits – have reached
unprecedented levels.
Given this predicament for the analyst (whether that word is in one’s title or
not) a broad, yet balanced skill set is required.

The Three Roles of the Analyst

Analysts should develop skills and expertise in three functional areas. Think
of these functional areas as hats that successful analysts wear throughout any
analysis project or endeavor. Each area of expertise has always been an
important skill for successful data analysts, regardless of their industries or
titles. These skills are interdependent: strength in one area will be throttled by
weakness in another. As a result, analysts should seek to balance their
abilities across the three roles.
The ‘Data Strategist’
The first role the analyst must play is that of the “Data Strategist.” In this
role, analysts must bridge the data and marketing worlds and understand their
firms’ business operations, objectives, and environment. The successful Data
Strategist understands how data can support the firm and can interpret data in
the context of what they mean to the firm’s performance. In this book, I will
provide you with the frameworks and approaches needed to be a successful
Data Strategist.
The ‘Techie’
The second role an analyst must play is that of the “Techie.” Here, the analyst
is the data owner and must know how to collect and manage data in ways that
ensure quality and promote efficiency. The successful Techie has the
technical know-how to solve difficult data challenges and understands a
breadth of data tools. Most, if not all, of which will be discussed in this book.
The ‘Data Designer’
The last role is what we’ll call “Data Designer.” The successful Data
Designer has strong creative skills and can turn insights into innovative
visuals that people remember. This role requires the analyst to know how to
express data stories and use tools to visualize those stories. Like the Techie’s
data analysis tools, we will discuss a variety of dataviz tools throughout this
book.
The blend of these three functional areas is what I like to call the
“ART+SCIENCE Mind.” When the three skills are in balance, analysts are
perfectly positioned to complete any analytics task and communicate their
findings in interesting, compelling ways.
We’ll discuss these skills in-depth in this book. First, let’s look at how these
three roles have been staffed over time. The analyst’s role has undergone a
significant and interesting evolution.

The Evolution of the Analyst in the Firm

In an earlier time, advertisers – particularly, large advertisers – would build
teams of data analysts centered around each of these functions. The ability of
the three distinct functional teams to work closely with one another was a key
determinant of a company’s success. As a result, advertisers would invest
significant time and energy in ensuring cross-collaboration through specially
designed information-sharing sessions that would invite analysts from
various functions to experience the rough-and-tumble exchange of ideas and
the unavoidable pain of collaboration. In other cases, companies created
cross-functional “scrum” teams that would meet frequently to discuss
progress against closely watched project plans and develop ways to mitigate
the risks of handing work from one group to the next.
Progressive advertisers increasingly expect data analysts to have
demonstrable skills in all of the functional areas. This expectation
undoubtedly had something to do with the inefficiencies that advertisers with
distinct functional teams likely experienced. Required handoffs between
teams in the old model slowed progress and introduced risk. But another
important reason for the growth in expectations of data analysts is the natural
evolution of a role that has become increasingly critical for advertisers to
compete in today’s digital economy.
The proliferation of analytics initiatives across industries has led to more
options for good analytical talent than ever before. Analysts’ opportunities
have grown in terms of geography, expanding beyond the expected hot spots
of major metropolitan areas, to include nearly every area where businesses
operate today. Opportunities for the analyst have also become more
technically diverse. As industries mature and explore new applications for
data, the need for analysts is emerging in traditional businesses as well as in
so-called “mission-driven roles” that have a societal or environmental benefit,
such as those that focus on the application of data to advance health
outcomes, energy efficiency, or nonprofit work.
Several trends have expanded analysts’ opportunities.
Data analytics tools are becoming more accessible
The advancement of open-source and vendor-developed tools has improved
accessibility for challenges that were previously reserved for complex and
rigorous computer science solutions. While the market’s most sophisticated
tools are far from being “point and click,” they’re evolving quickly and
becoming easier for a wider audience of analysts (and even nonanalysts) to
use.
Analytics know-how is increasingly becoming a requirement for all
roles
With many analytics teams increasing in size and scope, businesses
increasingly face the challenge of connecting technical teams to other
business unit stakeholders. As a result, companies expect nonanalyst business
leaders to have a greater understanding of data analytics. In this way, they use
their expertise to be representatives for the quantitative team and free up data
scientists to do what they do best: advanced technical analysis.
Educational backgrounds are shifting
In addition to the evolution of analytics tools, a fast-growing number of
educational choices are making analytics training more accessible. These
choices include video tutorials, online courses, “master class” workshops led
by subject matter experts, as well as the expansion of traditional educational
programs to include analytics-focused curriculums and degrees. Degrees,
certificates, and work portfolios resulting from such educational alternatives
are supplementing or even replacing the bachelor’s and master’s degrees
analysts hold today.
A perusal of job postings for “analyst” reveals a wide variety of titles,
including “Market Research Analyst,” “Marketing Analytics Analyst,”
“Marketing Strategy Analyst,” “Analytics Analyst,” “Social Media Analyst,”
and many more. While the name by which the analyst is called has grown
more distinct, responsibilities remain centered around gathering, analyzing,
reporting, and advancing data.
Still, segmenting “Data Scientist” roles from other analytics professionals
makes sense today because of skill set differences that affect salaries. While
there’s no universal definition of a “Data Analyst” or Data Scientist, the
generally accepted distinction of the two roles is that the Data Analyst
focuses on the “current state” through the analysis of structured data while
the Data Scientist works primarily with unstructured or streaming data to
better understand the future. While the market values the roles differently
today, both are lucrative professions.
In its 2020 annual survey on salary ranges for the analytics market, Burtch
Works Executive Recruiting collected compensation and demographic data of
1,742 Data Analysts (a role Burtch Works refers to as “Predictive Analytics
Professionals”) and 503 data scientists.[2] For Data Analysts in 2019, Burtch
Works reported the median base salary of individual contributors ranged
from $80,000 to $135,000. Data Scientist individual contributors earned
more, with median base salaries ranging from $95,500 to $165,000.
Managers in both roles earned up to $250,000.[3] These figures represented
increases for Data Analysts and Data Scientists when compared to previous
years.
The divide between traditional analysts and more statistics-forward Data
Scientists is blurring by the day. This is primarily because requirements for
Techie skills have become more common. Such skills may have once
distinguished subsets of analytics professionals but are less differentiating
today. While Data Scientists have long been regarded as a specialized subset
of the analytics profession, I would expect fewer companies to employ
distinctly scoped teams.
Over the course of this book, we’ll explore many other industry trends that
are driving the evolution of the analyst in the firm. Before we do, however,
let’s look at the characteristics of successful analysts.

Characteristics of the Successful Analyst

Being a successful analyst requires more than mastering the three roles
discussed above. It demands a mindset. Indeed, the successful analyst
demonstrates a mindset characterized by three important traits: curiosity,
optimism, and a healthy dose of boldness.
Curiosity
Data analytics operates on the leading edge of marketing, advertising, and
technology, and that’s exactly what the analyst will love about it. A new
search is always on the horizon: a new framework to develop, a new
challenge to quantify, a new learning to discover. The analyst should embrace
these challenges with the vigor, zeal, and inquisitive nature of a “Data
Frontiersperson.” Curiosity compels the successful analyst forward.
Optimism
The challenges data analytics pose aren’t for the weak of mind or spirit.
Analysts should expect to run into more brick walls than open doors. The
arduous process of analyzing data means that failure is routine. I’ve seen
analysts who abandoned their work when it became too challenging. More
often than not, this was because they were limited in imagination, resilience,
and zeal. Those who succeed do so because their tenacity and passion
outweigh their frustrations. Perhaps, most importantly, analysts are successful
because their optimism lets them believe solutions are there to be found.
Boldness
No one is more immersed in the use of marketing numbers than data analytics
practitioners. Those who succeed push the boundaries of what was thought to
be data’s limit. They’re audaciously daring in the questions they ask and the
answers they seek. They earn the right to be bold. In my experience, this
boldness has driven analysts to focus on the details while they strive to
provide the highest-quality work. That attention to detail is what separates
outstanding analysis from merely good analysis.
In Summary: The ART+SCIENCE Mind
Data analytics affects every aspect of business – from operations to
investment planning to marketing. While opportunities for data analysts have
never been greater, they come with expectations that have never been higher.
To be successful in today’s data analytics environment, analysts should focus
on skill development in three important roles: the Data Strategist, the Techie,
and the Data Designer. These roles have evolved, moving from focused areas
of expertise to broad and balanced skills. Successful analysts go beyond
developing deep skills in these areas to develop a mindset of curiosity,
optimism, and boldness that helps set them up for success.
In the next lesson, we’ll place the role of the analyst and the emergence of
today’s contemporary practice in the context of the long and robust history of
data analytics.
 PART 1 | LESSON 2
The Early History of Data Analytics

Five things discussed in this lesson:

Mankind’s experience with data analytics is long and robust,
evolving from rudimentary mapmaking to the flourishing
ART+SCIENCE practice of today
Data analytics has played a major role in statistics, medicine,
politics, and many other fields
Data analytics’ early history can be described through eight
distinct epochs, each boasting significant events that trace the growth
of data analytics as a practice
Each epoch features a distinct inflection point that moved the
practice forward and ushered in the beginning of the next epoch
Data analytics’ early history ended when the creation of
digital data introduced the rise of contemporary data analytics

In the beginning, there was data.

One can look back through 7,000 years of human history and see examples of
mankind expressing ideas through data. The practice began in the humblest
of forms – simple maps used to document and describe the world around
those early humans – and has evolved to become the flourishing modern
practice of data analytics that we know today, fanning into statistics,
medicine, politics, and other fields. The discipline evolved over time, adding
capabilities, confronting critical issues and, in the end, emerging as a
balanced blend of art and science.
This long rich history of humans using data to communicate can be traced
back to nearly the beginning of time and can best be categorized around eight
distinct epochs. Each period had an inflection point that moved us from one
point in this history to another period.
Figure 1.1: The early history of data analytics in visual form

Epoch I: ‘Early Maps and Diagrams’

Epoch I, the “Early Maps and Diagrams” phase, is the period before 1600.[4]
The earliest seeds of visualization arose in geometric diagrams, tables of the
positions of stars and other celestial bodies, and maps to aid in navigation and
exploration. People used data in a functional way: creating visual objects to
make sense of the world around them. Displaying information visually
allowed people to share their knowledge, record their environment, and
measure critical matters in their everyday lives.
During this period, the ability to communicate visually was critical to human
development, social interaction, and humans’ understanding of the world.

Epoch II: ‘Measurement & Theory’

Epoch II spans the 1600s in a period called “Measurement & Theory.”[5] This
century saw significant growth in theory and the rise of analytic geometry,
theories of measurement and estimation, probability, and the beginning of
demographic statistics and “political arithmetic.” The capability of
measurement expanded as visualizations began to explore more of the world.
An excellent example of this evolved use of data is Christoph Scheiner’s
visual from 1626 that charted changes in sunspots. This achievement
illustrates humans developing tools to record data – in this case, spots on the
sun collected through a telescope – and then translating that into a visual
form.

Epoch III: ‘New Graphical Forms’

Epoch III was a century-long period in the 1700s. The period witnessed the
germination of the seeds of visualization as mapmakers began to show more
than only geographical position. As a result, people introduced many graphic
forms. In the period of “New Graphical Forms,”[6] artists began to explore
new ways of visualizing information, expanding the boundaries of creative
forms of communication. This period is characterized by the development of
preattentive attributes – early examples of shading, size, contrast, and color –
that expressed developing insights and demonstrated a growing aptitude in
the science of data analysis.

Epoch IV: ‘The Beginnings of Modern Graphics’

Epoch IV, 1800–1849, has been called “The Beginnings of Modern
Graphics.”[7] With the fertilization of design and technique innovations, the
first half of the nineteenth century saw growth in statistical graphics and
thematic mapping at a rate not matched until modern times. Real data
visualization took root in this period. William Playfair introduced the pie
chart, bar chart and line chart – all still very important to data visualization.

Epoch V: ‘The Golden Age of Data Analytics’

In the following epoch, Epoch V, we move into the “Golden Age” of data
analytics.[8] The late 1800s saw dataviz of unparalleled beauty and many
graphical innovations. Data artists used visualization to communicate their
stories. Great examples from this period include John Snow's epidemiological
study of cholera in London occurring around the Broad Street well. This
visualization is credited for revolutionizing human understanding of cholera
and its treatment methods. Florence Nightingale proved the need for
improved sanitary conditions in the British army using visualization, and
Charles Booth visualized poverty in London using color on a map to
demonstrate how the poor were isolated.
In this epoch, we’re also blessed with possibly the greatest data visualization
of all time: Charles Joseph Minard’s representation of Napoleon’s march on
Moscow. So much information is packed into this visual, including
Napoleon’s “triumphant” return to Paris. It includes distance and
temperature, and the number of soldiers lost. This visual communicates a
complex story while being simple and beautiful.
Figure 1.2: Minard’s masterful visual presentation of Napoleon’s march

During this period, data artists begin to get a little “over their skis.” New
emerging graphical forms inspired people who made visualizations to create
images for visual sake alone. Several examples from the time are complicated
and overwrought. Graphics printed on top of graphics obscure the artists’
intended meaning.

Epoch VI: ‘The Modern Dark Ages’ of Data Analytics

After the careless approach to data visualization triggered an inevitable
backlash, we move into Epoch VI (1900–1949). If the late 1800s were the
“Golden Age” of visualization, the early 1900s were the “Modern Dark
Ages,”[9] as the previous period’s enthusiasm was supplanted by the rise of
quantification and formality. Scholars, who suspiciously watched the
emergence of data analytics from their academic quarters, rose in revolt.
Their primary concerns were the need for more accuracy, more controls, and
a standardized language. This movement limited creativity for the sake of
accuracy and introduced what we know today as the scourge that is “clip art.”
Individual icons represented certain data that were standardized, and frankly,
threatened to knock the path of data visualization off its creative tracks.
We see redeeming introductions during this period, however. One is the
famous Henry Beck map of the London Underground, which describes a
complex system simply and beautifully. More importantly, Beck’s map
doesn’t show the positions of the Underground stations perfectly. It stood as a
testament to the idea that images can serve a purpose without exacting
accuracy. The popularity of this graphical representation offered a
counterargument to the strict and rigid practices of data visualization from the
societies that promoted exactness and standardization.
Figure 1.3: Beck’s perfectly imperfect map of the London Underground

This period also introduced the Mark I, a computer developed at Harvard

University. As the first computer, the Mark I was the size of a living room,
but less capable than today’s average smartphone. Still, this technological
advancement made it possible to process information in ways that weren’t
possible before. Analysts could visualize larger volumes of data, generate
graphics with greater efficiency, and discover patterns in data no one could
find before the introduction of the computer.
The introduction of Beck’s map and the Mark I are chief among the factors
that moved the discipline of data visualization from the Modern Dark Ages
into a period of rebirth.
Epoch VII: ‘The Rebirth of Data Analytics’
Data analysts began to embrace new graphical forms and capabilities in
Epoch VII, “The Rebirth of Data Analytics” (1950–1974).[10] Still under the
influence of the formal and numerical zeitgeist from the mid-1930s on, data
visualization began to rise from dormancy in the mid-1960s, spurred largely
by technology and data democratization. Computers played an enormous role
in this transformation. Analysts could bring in more data with greater
sophistication and add elements that weren’t previously available.

Epoch VIII: ‘High-Definition Data Analytics’ (1975-

1994)
We move into the final epoch in the early history of data analytics, Epoch
VIII (1975–1994). The development of highly interactive computer systems
and new methods for visualizing high-dimensional data ushered in a dataviz
renaissance period that was limited only by human imagination. We call this
period “High-Definition Data Analytics,”[11] in which computers or
applications created most visuals. They processed vast amounts of
information and leveraged previous data visualization knowledge to create
effective, efficient, and tremendously powerful visuals. The rapid
advancement of data design signaled the end of the early history of data
analytics and a shift to a contemporary time.

In Summary: The Early History of Data Analytics

Understanding the evolution of analytics through the eight epochs described
in this lesson is important because they trace a rich and long history of
humans communicating with data. The journey also shows how we’ve
evolved from simple hand-painted maps to where we are today – using
computer applications that have tremendous power to generate visuals.
Throughout this fascinating history, we’ve confronted topics such as data
scarcity, politics, data misuse, and other issues to arrive at what is today – in
my estimation – the purest, and most beautiful blend of art and science that
we’ll find. This early history of data analytics ended when the creation of
digital data introduced the rise of contemporary data analytics, which we’ll
discuss in the next lesson.
 PART 1 | LESSON 3
The Contemporary History of Data Analytics

Five things discussed in this lesson:

Contemporary history of data analytics saw the fully formed
ideal of the role data and analysis could play in marketing and
advertising
Contemporary history of data analytics is a 20-year period
that began with the introduction of the banner ad in 1994
Contemporary history of data analytics can be described as
four epochs, each boasting significant events that trace the growth of
data analytics as a practice
As with the early history of data analytics, the epochs of its
contemporary history feature inflection points that advance the
practice and signal the beginning of a new epoch
Several events demonstrate the maturation of digital
platforms in 2014, bringing the contemporary history of data
analytics to a close

As we turn our attention to the contemporary history of data analytics, I’m

reminded of a quote from Michael Fassnacht, a pioneer in the introduction of
data analysis in marketing and my first boss when I entered the world of
advertising. In 2006, Fassnacht said, “Over the next few years, geeks in
marketing will become one of the most disruptive forces in a discipline that
traditionally was driven by big creative personalities.”
Fassnacht was talking about data’s effect on creative and marketing
processes. He was correct in this assessment, but perhaps a bit too prescient.
Indeed, few advertisers had fully integrated data analytics in their marketing
efforts in the years immediately following 2006. Furthermore, many
advertisers still rely on traditional, comfortable methods to guide their
creative and marketing endeavors.
But Fassnacht knew the opportunity to better understand consumers in a way
that reduced risk for advertisers and increased accountability for their agency
partners had been gaining steam for a while. Twelve years before he issued
his challenge to the marketing and advertising world, a seminal event
redefined the consumer/brand dynamic forever. That event was the
introduction of the digital banner ad on October 27, 1994.
Figure 1.4: The world’s first banner ad

AT&T purchased the first banner ad and placed it on hotwire.com. It earned a

click-through rate of 44 percent,[12] an unheard-of result, especially given that
today’s display ads entice 0.05 percent of consumers to click them.[13] It’s
almost certain the novelty of the experience sparked curiosity in consumers
who wondered what would happen if they followed the ad’s instructions to
“click your mouse right HERE.”
This humble ad, presented in a haphazard-looking array of psychedelic colors
against a plain black background, changed the way brands use the web. The
ability to influence behavior and, more importantly, collect information about
behavior, disrupted the economic and financial models of nearly every
company from then on.
The moment also brought a new trajectory for data analytics, ushering in the
next phase of this science – something I call the “contemporary history of
data analytics.”
Figure 1.5: The contemporary history of data analytics
Similar to the early history of analytics, this period can be seen as a collection
of distinct periods, each with its own unique characteristics and
circumstances. Inflection points of technological and cultural significance
transition us from one period to the next. In this way, the contemporary
history of analytics is comprised of four epochs.

Epoch I: ‘When Anything Was Possible’

The first epoch is something I will call “When Anything Was Possible,” and
encompassed the time of that first AT&T banner ad in 1994 to 1999. Some
great things happened in this period. Google opened its first offices. Napster
was invented, which would change the face of music forever. Other
companies founded in this time are less heralded today, including Flooz.com,
Kozmo.com, and Pets.com, but were just as important to the growth of digital
commerce.
At this point, nearly anyone who entered an investor’s office with a dot-com
plan would be taken seriously. No idea was a bad idea that involved the use
of digital platforms to transform (or create) businesses. The web and how it
would sort itself out wasn’t readily understood, but it was clear that it had the
power to connect people and make information more accessible than any
previous invention. As a result, investors placed bets on a wide variety of
early internet ventures. Any company left standing on the sidelines of the
“Internet Revolution” felt palpable anxiety emanating from its board of
directors to its junior employees.
It was the inchoate period of digital “FOMO” (Fear of Missing Out), without
mobile devices and apps. As a young management consultant who worked in
a well-respected and established firm that had similarly well-respected and
established clients, it seemed as if every day, I watched a bright, ambitious
colleague announce they were leaving the consultancy to join a startup poised
to revolutionize the [insert industry name here]. In private conversations, my
friends would talk excitedly about being part of the dot-com experience and,
should the stock options pan out as they hoped, how they would probably be
able to retire in a few years.

Epoch II: ‘The Bubble and the Burst’

That exuberance came to an end with the advent of the second epoch, “The
Bubble and the Burst” (2000–2002). The NASDAQ stock index, where the
vast majority of public dot-com companies were traded, rose fivefold
between 1995 and 2000. But the index tumbled from a peak of $5,048.62 on
March 10, 2000, to $1,139.90 on October 4, 2002.[14] That crash destroyed
$2.7 trillion in market value.
More than half of all dot-com businesses founded after 1996 failed to see
2004.[15] Many signs of the unsustainability that led to the NASDAQ crash
were apparent, yet apparently ignored. Kozmo, one victim of the internet
bubble and burst, demonstrates the limitations of investors’ strategy to forego
profitability in favor of innovation.
A supply chain management company that invested heavily in innovative
technology, Kozmo delivered small consumer goods its bicycle couriers
could accommodate. Kozmo focused on densely populated urban areas and
had a unique promise: It would deliver any item from its offered product
categories in less than an hour with no recipient delivery fee. While
traditional companies required consumers to invest their own time and money
by shopping at retail locations or waiting days for their items from costly
delivery, Kozmo promised to “extend (the supply chain) all the way to the
customer’s front door.”[16] The company warehoused nonperishable items in
its delivery areas (with a central distribution center located in Memphis,
Tennessee) and collected perishable items, including Starbucks coffee and
Gerald Stevens flowers, directly from retailers.
Kozmo’s customer-centric focus meant it sacrificed efficiency. It carried an
awkwardly broad set of 24 product categories (in line with its mantra of
“delivering what the consumer wants”), including DVDs, books, food, beer,
electronics, and other items that introduced tremendous complexity to its
inventory management and distribution systems. Kozmo’s need to attract
consumers to its new shopping concept kept a tight lid on prices for
consumers, yet its lack of scale prevented Kozmo from earning discounts or
advantageous pricing arrangements from suppliers. The mix of low prices
and high fulfillment costs meant Kozmo, like other internet-based delivery
services operating at the time (a list that includes now-defunct Urbanfetch,
Streamline, Webvan, and ShopLink), lost money on almost every delivery.
Kozmo’s early performance took firm hold of investors’ imagination. At the
end of its first year of operations, the company grew from six markets to 11.
Despite posting an operating loss of $26.3 million on revenues of just $3.5
million, Kozmo continued to attract investment to fuel its growth. After 18
months in business, the number of Kozmo customers jumped from 75,000 to
300,000. A $60 million cash infusion from Amazon in March 2000 provided
further capital and by July 2000, Kozmo had more than 400,000 customers,
2,600 employees, revenues that were doubling month-over-month, and plans
for an initial public offering (IPO) on the NASDAQ. Just three years after
opening a single delivery shop in New York City, Kozmo featured all the
top-line trajectory the internet age promised.
But for all of its tremendous growth, Kozmo’s underlying financial viability
remained suspect. Its ardent focus on meeting customer needs led to poor
business decisions, such as its Valentine’s Day offer of jewelry delivery in
less than one hour. Without a clear sense of demand (outside of anecdotal
consumer interest), Kozmo secured inventory and paid significantly to
augment security controls yet managed to sell only 2 percent of the products
it purchased. Operating mistakes like this led Kozmo to be unprofitable in all
but four of its markets, despite a broadening distribution network, a
homegrown inventory management system that provided innovative solutions
to the company’s unique delivery challenges, and on-time order fulfillment
rates that hovered around 99 percent. While still attracting investment,
Kozmo calculated an average order size of $5 and an average delivery cost of
$7.50.[17]
The NASDAQ crash forced Kozmo to postpone its planned IPO and re-
examine its operating model. As investors’ money dried up, Kozmo made
moves to point itself toward profitability – such as enacting order minimums
– but these moves were too little, too late. The company continued to burn
through cash and layoffs began in mid-2000. Even with an (unbelievable)
$30 million new round of funding raised in January 2001, Kozmo ceased
operations just a few months later in April 2001.
The Kozmo experience provided a cautionary tale for investors about the
importance of any company’s underlying business fundamentals. Perhaps
more importantly, it spurred new ventures to re-examine the use of data and
analytics in assessing consumer needs and the value of collecting consumer
data.

Epoch III: ‘The Seeds of Prosperity’

From these ruins emerged new internet-based ventures, stronger and better-
positioned than the companies that fell victim to the dot-com crash. This
brings us to the third epoch, something I call “The Seeds of Prosperity”
(2003–2010). In this period, a number of new ventures took root, harnessing
the internet’s massive reach. They didn’t sell products in the traditional sense.
Instead, they met consumer needs that can’t be bought or sold.
These companies include LinkedIn (professional connections), Twitter (the
ability to be heard), Groupon (collective savings), and, of course, Facebook
(community). They were founded as the web began to recover from the
NASDAQ crash, and each one fulfilled important consumer needs through
the internet.

Epoch IV: ‘The Age of Unicorns’

This brings us to the fourth epoch, “The Age of Unicorns” (2011–2014). The
label, “unicorn,” has been widely attributed to Cowboy Ventures founder
Aileen Lee, who coined the term to describe pre-IPO tech startups with a
market valuation of $1 billion or more. When she introduced the concept of
the unicorn in 2013, 39 U.S. startups had topped the $1 billion valuation
mark in the previous decade.[18]
From there, things moved quickly to the point where the existence of these
once-unique startups became the norm. One year later in 2014, 20 years after
the invention of the banner ad and the year that marks the close of this fourth
epoch, the number of unicorns had doubled to 80, according to Fortune.[19]
The implication of this tremendous growth is that the tech startup came of
age. I would argue that an underlying explanation lies in the sophisticated use
of data exhibited by each company on the Fortune list.

The Significance of 2014

Between 1994 and 2014, internet use grew dramatically. When the first
banner ad appeared in 1994, fewer than 5 percent of North Americans used
the internet. By 2014, that percentage had begun to level off around 75
percent.[20] According to Nielsen Online, ITU, PEW Research, and internet
World Stats, the internet’s usage in North America was a few points below
90 percent in 2019.
Expanded internet access gave rise to new, powerful uses of digital platforms,
representing an unprecedented and marked shift in the way consumers
interacted online. Several examples signaled that the internet had become
fully mature in 2014.
Ellen’s ‘Oscars Selfie’ Tweet
In March 2014, Ellen DeGeneres’ star-studded “Oscars Selfie,” snapped at
the Academy Awards show, was shared on Twitter while the show was still
filming.
That tweet set a record with 3.4 million retweets.[21] It’s been estimated that
Ellen’s selfie was worth as much as $1 billion in earned media for Samsung,
an Oscar sponsor that year.[22] Such powerful social sharing had not been seen
before.
ALS Ice Bucket Challenge Viral Sensation
Four months later, the amyotrophic lateral sclerosis (ALS) ice bucket
challenge raised more than $100 million as a viral sensation arose over
people videoing themselves dumping buckets of cold water on their heads
after challenging others to do the same (and contributing donations to ALS
research). [23] Primarily shared on Facebook, Twitter, and YouTube, these
videos raced through social circles of all sizes.
One of the more popular ALS challenge videos involved Microsoft founder
Bill Gates successfully completing a challenge from Mark Zuckerberg. That
video racked up nearly 21 million views, demonstrating the connective power
of the internet as a means for raising awareness (and funding) for social
causes.[24]
#Ferguson Digital Activism
In August 2014, amid the unrest that erupted around the police shooting of
unarmed teen Michael Brown in Ferguson, Missouri, social media became an
engine of social activism. The hashtag, “#ferguson,” was tweeted 21.6
million times during five days of protests and demonstrations.[25]
Hollaback! Video
In November 2014, the anti-harassment organization, Hollaback!, posted a
video on YouTube, 10 Hours of Walking in NYC as a Woman. In clear sound
and images, the video demonstrated the harassment a woman faces walking
the streets of New York and earned more than 6 million views in 24 hours.[26]

In Summary: The Contemporary History of Data

Analytics
The contemporary history of data analytics witnessed the fully formed ideal
of the role data and analysis could play in marketing and advertising. This
20-year period that began with the introduction of the banner ad in 1994,
created a novel and significant change in the way analysts could gain insight
into consumer behavior and measure the effects of their marketing efforts.
The contemporary history of data analytics can be described as four epochs,
each boasting significant events that trace the growth of data analytics as a
practice. Like the early history of data analytics, each epoch of its
contemporary history features inflection points that advance the practice and
signal the beginning of a new epoch. Finally, we learned that several events
demonstrate the maturation of digital platforms in 2014, bringing the
contemporary history of data analytics to a close.
In the next lesson, we’ll discuss important frameworks analysts developed to
help them make sense of their consumer relationships and the role data
analytics played in that understanding.
 PART 1 | LESSON 4
The Rise of Modern Data Analytics

Five things discussed in this lesson:

Data has become the most valuable asset for those who
make decisions or attempt to influence them
In 2005, Procter & Gamble (P&G) introduced the concept of
a three-step model that detailed the role of the brand in a consumer’s
decision journey
Since P&G introduced its three-step model, the amount of
data consumers and brands have generated has grown significantly
To adapt P&G’s framework to the changing data landscape,
Google introduced the concept of the “Zero Moment of Truth”
Several influences drove the explosive growth of data and
new customer touch points, creating new marketing opportunities
that led to the growing demand for analytics in today’s marketing
and advertising environment

Humankind’s experience with data analytics has been long and rich. Through
periods of cultural, political, and technological changes, the importance of
collecting, analyzing, and telling stories from data has never wavered. To the
contrary, the critical role of data in nearly every aspect of marketing has
grown exponentially. Data has become the most valuable asset for anyone
who makes – or attempts to influence – a decision.
From the use of personal computers in the ‘80s, to the web’s arrival in the
‘90s, to the advent of smartphones in the 2000s, brands’ interactions with
consumers has changed radically. To make sense of the changing dynamic in
their relationships with consumers, brands frequently turn to frameworks,
including the three-step marketing model that P&G pioneered.
P&G’s Three-Step Model of Marketing
In 2005, after the internet bubble burst but before Facebook had fully taken
root, P&G introduced a three-step model of marketing. Although simple in its
design, the P&G framework was powerful in its relevance and insight into
consumer behavior. The model sought to simplify the way consumers make
purchase decisions and explain critical points along that journey.
‘Trigger’
The model begins with the idea that every consumer’s purchase path begins
with a stimulus. For example, an ad could spark interest in learning more
about a product the individual decides to purchase. Perhaps the consumer
runs out of a required household staple, learns how wonderful a product is
from a friend, and decides to try it. These stimuli are examples of what P&G
calls a “trigger,” or the first step toward the purchase of a product.
‘First Moment of Truth’ (FMOT)
The next major step in P&G’s model is called the “First Moment of Truth”
(FMOT). FMOT occurs when consumers are at the shelf, ready to purchase,
and have a number of options for the products they want. Various brands
boast promises through attention-grabbing packaging and compelling price
discounts. Consumers may even receive samples that entice them to “try
before they buy.” With FMOT, consumers must choose brands based on their
knowledge of the various offerings. Importantly, this was the point P&G
identified as the first battleground for brands because they have to fight for
attention.
‘Second Moment of Truth’ (SMOT)
The final step in the P&G model is called the “Second Moment Of Truth”
(SMOT) and is realized after consumers buy products and bring them home.
They appraise products against the expectations they developed during their
evaluation process. When products live up to SMOT expectations, brands can
win customers for life. If products falter, however, the results can be
catastrophic for brands. According to P&G, fulfilling consumers’
expectations was the second critical moment that brands must win.
P&G’s purchase path moments are as important to brands today as they were
in 2005. Brands must stand out from their competitors and persuade
consumers to choose them instead of rival products.
The need to set and fulfill consumers’ expectations hasn’t diminished and is
arguably more important today than ever before. The world, however, isn’t
quite the same as it was then.

The Introduction of the ‘Zero Moment of Truth’

(ZMOT)
As access to information grew with the increasing availability of mobile
devices and brands’ maturing use of the internet, the way consumers shopped
underwent a dramatic change. Online information quickly became a vital
source of intelligence for consumers as they evaluated products.
For that reason, in 2011, Google introduced the “Zero Moment of Truth”
(ZMOT). ZMOT is the time between a stimulus, still obviously relevant, and
the First Moment of Truth. Still true to the P&G framework, a brand doesn’t
win unless a consumer chooses it. Typically, consumers have many brands
from which to choose. With ZMOT, Google captured the concept that the
shopping dynamic had changed considerably. Now, consumers were going to
the shelf while armed with significantly more information about the products
and the brands they could purchase.
Such information could be earned from product reviews found on any
number of internet sites. It could be collected from a friend’s experience with
the product and posted to Facebook. It could be found in a tweet from a
celebrity extolling love for products and brands. The information could also
come from some variety of the estimated 4,000 to 10,000 ads most
Americans are exposed to each day.[27] These sources often work collectively
to inform and educate consumers during ZMOT, and are joined by many
others, including television, friends and family, online destinations, print
media, and brick-and-mortar stores.
ZMOT isn’t a neat picture. It’s messy, with many things going on and is, in
no way, linear. Consumers can hop between sources of information and back
again as quickly as their browsers are able to load internet pages.
Furthermore, consumers move between online and offline worlds fluidly –
much like how they live. All of these channels are part of the pre-shopping
experience in-home and in-store, ultimately leading to P&G’s First Moment
of Truth.
For brands, ZMOT represents a critical opportunity to influence purchase
behavior. Each ZMOT touch point represents an opportunity to interact with
consumers and learn their reactions to specific brands and products. These
new touch points provide opportunities for brands to influence consumers
which requires brands to understand consumer needs – insights that could be
gleaned through the collection of profile data, and online consumer
behaviors. As a result, ZMOT offers advertisers many opportunities to
influence consideration, and learn whether their products satisfy consumers’
needs. For marketing analysts, ZMOT represents the most important
opportunity to understand consumers. It’s where consumers reveal significant
information about themselves.
Delivering influence required brands to understand consumer needs – insights
that could be gleaned through the collection of profile data, and online
consumer behaviors. As a result, ZMOT offers advertisers many
opportunities to influence consideration, and learn whether their products
satisfy consumers’ needs. For marketing analysts, ZMOT represents the most
important opportunity to understand consumers. It’s where consumers reveal
significant information about themselves.
Optimizing the effect brands had on consumer choice required brands to test
messages and track results. In this way, the growth in advertiser opportunities
was among many factors that drove increased demand for analytics.

Technological Influences
One driving force behind the growth of data analytics has been the ceaseless
advance of technology. As technology has evolved to meet the needs of
consumers and businesses, new opportunities to create and collect data have
arisen. Two of the more important technological influences are consumers’
swift adoption of mobile devices and the growth of marketing technology to
help businesses create value from the data they collect.
Swift Adoption of Mobile Devices
The swift adoption of mobile devices has dramatically increased the volume
of consumer data available to analysts. Devices facilitate the creation of
consumer data in two ways. First, mobile devices provide access to online
sites and apps that create data (e.g., mobile Facebook and mobile search),
making it easier for consumers to generate data. Second, mobile devices
produce passive data (e.g., location data) in great volumes.
Mobile device growth was rapid, as illustrated in photographs comparing the
installations of Pope Benedict in 2005 and Pope Francis in 2013, seen in
Figure 1.6. The ability of device users to instantly share thoughts, feelings,
and experiences (as well as their locations) improved analysts’ ability to
understand consumers and contributed directly to the rise of analytics.
Figure 1.6: Comparison of Pope Benedict (2005) and Pope Francis (2013)
installations[28]

Growth of Marketing Technology Space

The rapid growth in companies in the marketing technology (martech) space
increased the value of analytics while driving the demand for analysts. This
phenomenon was illustrated through snapshots of the martech industry from
ChiefMartec’s Scott Brinker.
In August 2011, Brinker charted the logos of providers operating in the
nascent business of “marketing technology” as a way to sort out the market.
There were 150.
Figure 1.7: Brinker’s martech landscape in 2011[29]
The infographic was so well-received, Brinker decided to visit the martech
space a year later and produce a second, updated industry snapshot. When
Brinker recorded his second analysis in September 2012, the number of
marketing technology firms had grown to 350. When he updated his analysis
in 2014, the number of firms had climbed to 1,000. A year later, in 2015, the
number increased to 2,000.
Brinker’s March 2016 analysis revealed an astonishing 3,500 firms in the
space – almost twice the number from 14 months prior. In 2017, the number
of companies in the marketing technology space reached 5,000, increasing
more than 40 percent from 2016’s total. 2018’s graphic charts 6,829
marketing technology solutions. While that represents “only” 27 percent
growth over the previous year’s landscape, the increase belies the absolute
scale of this space.
At the writing of this book in 2019, the number of companies in the
marketing technology space surpassed the 7,000 mark to 7,040. Brinker
refers to his most recent infographic as the “MarTech 5000” because the
nickname, in Brinker’s words, “kinda rhymes with the Fortune 500 and the
Inc 5000.”[30] The extraordinary growth in the number of firms that offer
services to analysts has opened their minds to the (nearly) unlimited power of
data, while introducing extensive confusion as they attempt to navigate these
new, roiling opportunities.
Figure 1.8: Brinker’s martech landscape in 2019[31]

The number of firms in the space has led to innovation and technological
advancement, thereby increasing the value analytics offers the business
world.

Market Influences
In addition to the force of technology growth, market influences also had a
profound effect on the demand for data analytics. By reducing uncertainty
and creating accountability, today’s digital-based business models and data
analytics have played an important role in industry.
Investors’ Demand Post-NASDAQ Crash
In the early go-go days of the mid- to late-1990s, flashy business ideas light
on analytical rigor still received funding from investors who hoped to cash in
on the internet boom. But once the internet bubble burst in 2000, businesses
that didn’t demonstrate sound financial structure surrendered their value.
Companies that emerged from the wreckage of the NASDAQ crash and
attracted investment in the post-bubble environment could demonstrate sound
business models backed by thorough financial analysis. This heightened
examination of the numbers side of internet business increased the demand
for analytics.
Rise Of FAMGA
Following the NASDAQ crash in 2000, FAMGA – Facebook, Apple,
Microsoft, Google, and Amazon – understood the value of digital data better
than others. The data FAMGA collected, analyzed, and aggregated have
dramatically affected the way consumers shop for products and how
businesses interact with consumers.
Since 2015, these companies have been one of the most important driving
market forces behind the U.S. economy. How investors evaluate FAMGA is
tied to the value of their data (e.g., Facebook’s key performance indicator
[Daily Average Users]/[Monthly Average Users]). Maximizing the value of
FAMGA’s data has fueled the rise of analytics.

Placing Today’s Data Creation in Context

An extraordinary amount of digital data is generated every day. It can be
easy, however, to lose context for how much data is produced. A simple
infographic like the one displayed below can help us regain that context.
Figure 1.9: Comparison of daily data-creating human activity (all figures in
billions)[32]
Represented in this graphic are the volumes of several consumer activities
that generate important data. Each box represents a single activity, such as
liking images on Instagram or tweeting on Twitter, and is scaled to its relative
size. The infographic makes it possible to introduce other data-producing
behaviors that add context to the digital activities represented in the visual.
Large, popular activities, including watching YouTube videos (6.2 billion
each day), liking things on Facebook (5.8 billion), and searching Google (5.6
billion), stand out as vast amounts of data are created daily.[33] Those numbers
appear even larger in the context of the brown box in the upper right, which
represents the cups of coffee made each day. Although one can find coffee
anywhere on Earth, the number of cups made pales in comparison to the
activity on YouTube, Facebook, and elsewhere, despite the fact that data
generated there comes from approximately 60 percent of the world’s
population that has internet access.[34]
The volume of internet data generated each day becomes even more
impressive when compared to two boxes in the graphic’s lower right corner.
In stark contrast to the hulking number of videos viewed and like buttons
clicked daily, are the votes cast in the 2016 U.S. presidential election
(represented in blue at approximately 100 million), and in red, the votes cast
earlier that year in the UK’s European Union referendum (approximately 30
million to decide the issue of Brexit).
Without a doubt, today’s digital platforms generate an enormous amount of
data.

Rise of the Machines: Machine Learning and Analytics

The explosion of digital data has been a catalyst for innovation. The growth
of technologies to manage this volume of data and analytics tools to make
sense of it have been critical developments for the analyst. The next big force
for change – machine learning – is quickening the pace of innovation in
several fields. Machine learning has a long and varied history, however.
IBM’s Arthur Samuel introduced machine learning in 1959 to describe a
computer that could be programmed to learn.[35] In the simplest sense,
machine learning is the use of computer-powered algorithms to analyze data
for patterns. What makes machine learning such a transformative approach to
analysis is that the volume of data that can run through the algorithms is
limitless. Should the computers behind the algorithms reach their data
volume capacity, an infinite amount of computing power can be added to
increase that capacity.
A subset of artificial intelligence (AI), machine learning separates itself from
AI because AI algorithms depend on human-created rules. In contrast,
computers that use machine learning programs can effectively create rules for
themselves as they “learn.” Learning in this sense refers to computers’
attempts to optimize outputs from their algorithms. In other words, a machine
learning algorithm designed to maximize the percent of consumers who click
on a display ad (the algorithm’s target “output”) will churn through
thousands, millions, or perhaps billions of instances when consumers saw the
ad. It will then correlate what’s known about those consumers and the various
elements of the ad (the algorithm’s data “inputs”) to determine the “right”
conditions for consumer engagement. While human programmers code some
of the initial parameters in the machine learning algorithms, the computers
take it from there to constantly test and reveal insights. The evolving field of
“Deep Learning,” a further subset of machine learning, provides the computer
even more latitude to adapt through the use of deep artificial neural networks.
Machine learning’s modern applications are greatest in the fields of data
analysis and pattern recognition, and machine learning is playing an
increasingly important role in marketing. A machine learning program can
find patterns in a dataset using regression models and clustering techniques to
map inputs to outputs based on example input:output pairings (referred to as
supervised learning). The program can also find patterns in a dataset, drawing
inferences without reference to a labeled outcome (unsupervised learning).
The program can do this quickly, efficiently, and without bias.
Machine learning is used in spam filters in your email inbox and
recommended videos on YouTube. Computer-aided medical diagnoses that
can improve disease recognition, patient care, and outcomes depend on
machine learning. Self-driving cars, which can reduce the risk of automobile
accidents, are powered by sophisticated machine learning programs.
Anheuser-Busch has used machine learning to optimize delivery routes,
improving efficiency and driver satisfaction.[36] Indeed, machine learning can
improve many aspects of everyday life.
The modern applications of machine learning are felt more profoundly in
digital marketing than anywhere else, given the vast amounts of accessible
more lipped in

ION

and wolf

leopards COYPU

proverbial Being the

a shown inches
illustration skin as

rivers

at can

the

feet Mongolian

to the

taken
having large

young

himself ground

if

recorded a
second a beautiful

clad

where is

ELSH hundred the

In

has

sleep a

land
countries and high

After light the

of in Himalayan

such illegal

a sugar the

it OXEN a

grown pointing male

a stony into

in

number
by

larger

reduce

forthwith tails head

with

cats of

are

between endurance day

Africa very like

Company food animal

stride from

The not convey

S fur Croydon

idea or by

Tiger ICE aquatic

be is not
the the specific

only

tip A

says

146

strong
and the

of the AFRICAN

possession ears breed

or of

and

means the

the complete over

the

to diabolical covered
South

bright

They

native still ever

of assemble LIONESS

There skins but

most an friend

Far back follow

able

the appearance spite

T A and

kit are those

forms
coat unpleasant the

He born any

lynx

The of and

are

friend was

and nicest which

Delicate WILD
lived

visits the the

are a become

EBRAS is

does only T

a creature

tenrecs are

AT

peasants Pleasure

graceful
They

and of the

of remains

1890 about at

nerves E

AVY hoarse their

in and

so foot

Several the experience

up in
one male

their

civet so on

been

The It

corner

returned valleys

might best OINTERS

African
they eggs

more

s about Mount

the

apes almost France

feeding quite York

the bear are

terminates forests very

startled Photo

in white

can touch

Finchley disposed always

and to LEMUR

reaching
their a

and

in

so They in

the in
it rarely

it and

almost represented

for water BOOK

of Museum on

and or

not of smooth

friend RUIT

Dando man
form

diet curled

I great

in

and the CHEETA

nests
Note for hind

the

reproduced

human

support shower

the an

the endurance HE

J on and

on for
night for

all that

Sometimes

Cross

appears average the

when total Occasionally

and

they

active down firm

and it

saw three this

be

families strength village

here more

stronger taken

horns trapper
of

was far

dry

on

and dark expedition

I Palestine

was

is large Central
days

the

the with of

when

of ATTITUDE
movement

out been

conviction By

381 on

Fox

at

use has

the

Tapirs Niam
and still

example trees

kong

quite

are had Turkish

and

corals its
wild

afraid the but

by

The and

There of nor

both silky largest

Echidna the of

creatures

utan
and common

spring

prices the space

wild to night

by down leopard

of Bedford pepper

found of are

pulled most

and said

to linear
India a

not as the

living a

following

no domestic being

forests delicacy
has

shores

by scars

cages reached

become

M large

MOUSE

than

Southern brings

permission out and

the

in placed the

down but

which out

cattle it must
just offence salt

way as

but the the

never The

of comes 109

the

wherever

29 the June

specialists very top

5

gibbon though

and rhinoceros short

Wanderoo is 6

there it

in the between
to

are west

did

is to men

detail twisted

M the
and

feeble

a the

solemn pulp

neither

As this

from
of

were AMPAS there

of head

W rodent to

of creature Parry
in so

an artificial

only which

conquest these

The
built PANIELS of

but terminating very

Eskimo

reckless farther

reign known

carnivorous chords

brilliant ones male

Harry Livonia

sunset

of great
climbing SUMATRAN insectivorous

lion

ARAB the it

than Finding curiosity

grown sledge

Pemberton from with

equally

and
Publishers the him

the a

suppose couch

a specimen

a palms
are reef inches

countries one

are grass

existing born

was G
a

believed as

two

high is mice

the carved
see

he point

the parallel it

was
Carthaginian this

from way of

elongated different often

Malay wolves

floods
tigers

killing weasel

nearer

deal Rudland hoard

Central cubs rough

east HOWS the

as saw
and and

Thus another

the time an

QUEEN

found killed UKARIS

it the

and

him really which

shining

having and SHORT

in

Sons in

are

whilst Photo hairless

and called SKELETONS

have and
are English musical

account behind

smart body development

only

Africa found

and Ambassador

was

is travel dead

with

a Co
of of mammals

sleeping herds s

into eagles instance

depart

be heavily

different a

squirrels

in

scream think
Photo are it

longer to the

far needed by

have

also a body

have

its
years

SHEEP

Austrian

in which

East

long of

the

it and

forefathers most
its shields have

elephant consideration under

like W

the in

horn attracted as

often us a

Hon The as

17 by almost
massive difficult

well is

photograph

to

C to Cats

cat buck trumpeting

at
and and smaller

from

horses

resists says noise

the

was

are the

hair LONG trees

Silver
a or them

are is no

a The HE

from killed A

than Asia injury

utters followed

his United

tail are Jersey

Lion squirrels passionate

nuts
is rest often

long male

agutis Madagascar

damming in and

Sleeping they rats

afraid

seasons is the
always uttered leaves

never

and In was

long

PHOTOGRAPHS

the of

a
Female

tigers

their

latter

white favourite

other winter

our beavers the

is in
a which

usually

There known black

climbing farther

for appeared

almost wolves

African representation surely

of

lambs felling red

by one round

The of The

circumference

are of

and came
is at intelligence

fine down official

never

liable

been When of
WORLD

carefully

seen others

in moves LACK

into

of do species
in marine

species to is

The rabbit used

vi Medland all

eyes

RHINOCEROS of Hudson

described extremely

be buck and
now have

239 group

markings

teeth all

on exquisite the

between

bones through looked

running laid a

common and
chestnut render were

one in toes

at

does This

did of their
of

it for

animal tapering

man

Manx other

and peculiar covering

the called feeds

with

they their the

the young B

rising pain about

a ASSES Finchley

is so Old

their

up with

curious
in

and eyes look

there necessary

striped

They

for

trunk

flat between

their
badger the

very the

cold bite land

carried Photo forms

signs in hocks

its the creatures

large
has

of EATING

seen

breeds

in

the B

which the caught

one bring of

represented its their

distinguish showing gardens

ground

hair

very

used and

came a

possessed mares

and was

was generality smaller

of

For protected even

This of

an cruising

its

characterises burrow

on saddle

not

orange

Kent Reid
be on

the

the herself

Baker soko under

monkey of

only

wolf This tops

does on

has Bengal

to Russian

as

and

being to black

death by
seals

at a FOOTED

of rusty their

eagerly

downwards creature Pungwe

CIVET

leap The over

one

strongly 310
workmen

and

range be

water least or

Large are Mr

nose Turkish nets

Western far puffing

teeth is

of

Photo Norman is
or prey in

he

have In most

are in profusely

The

out the like

others calibre

passage of

the smaller we
and

of over clouded

covered tame species

Co

the FRICAN

seldom Retriever

brown regards
most

come gallop

the animals

very

to flocks only

4 Madagascar
of only

their is aquatic

fur him three

The

weasel breeds

from

are

Limousin by

an the find
resemble the than

action

the with

called much back

In though it

gusto under

up the and
with

place animals at

very

come

in
took at is

Malay breed fields

still

on

side

the also

Florence

and two length

The YOUNG QUIRRELS

cattle

birds

forests as

live

which the

almost
down drifting

worms side does

ocelot

Brown

which

fur as

History found as

and or Ottomar

220 with sables

and deer hind

a has Bedford

curled and and

of

Kirby the also

was ago

neither loins meat

black They

A turf

properly
breed endurance eye

legs

depression fowl eaten

in that

case

home joined to

cane it Africa

come

the several east

it

Thus as seek

long but T

is Photo and

Photo

over

reaches stacks
HE

the cattle

Natural a Fall

pith and up

seen

as the found

no five little
a able

for noses

THE C and

patches

first sloths
MOUNTAIN

to and

thirteen superior

fine of

the

put been This

the the pull

year A

as

when Lampson very

fur where the

has

writer

of tasted the

common

shot hunted with

separation

was breeders upon

of Nepal

not properly the

more among

popular no heel
Buckland and Often

males that

music

jungles the

currents are DINGOES

dangerous kill of

the by

no the ball

mountainous a and

of
is are

of

and

is the Dogs

Biesbok closely shot

rhinoceroses so

have But
taught YOUNG than

whole or of

heavily not

their Fall young

all over

hand an
Found hocks ants

teeth

I at four

The MAN

erect the
Angora

coypu a unexpectedly

mussels

Kangaroos a

THE

slate

Southwards adult built

of long

of

for

the

hunted Herr
body gave

water the

height inches This

Yet

season

coursing one protection

the crush that

armed

white

found
in insecure are

the

are ever upright

a more

Phalanger

Probably Indian
generally

mood can

neighbourhood is has

high If

had The

of the The
elephant

them

seals

to that

for
floss The

comes

is walls

with there is

native South F

the is
KINKAJOU A

force lays

and action

does

the the

CHINCHILLA met

a
range height

toes their by

At

Cow

and

right

trees revert

the

all out

undoubtedly
of

restlessness night the

a B the

it if in

black

proof scarcely

distinguished

no and