HOW TO WIN WITH

YOUR DATA
VISUALIZATIONS
THE 5 PART GUIDE FOR JUNIOR
ANALYSTS TO CREATE EFFECTIVE
DATA VISUALIZATIONS AND
ENGAGING DATA STORIES

ELIZABETH CLARKE
© Copyright Elizabeth Clarke 2021 - All rights reserved.
The content contained within this book may not be
reproduced, duplicated, or transmitted without direct written
permission from the author or the publisher. Under no
circumstances will any blame or legal responsibility be held
against the publisher, or author, for any damages, reparation,
or monetary loss due to the information contained within this
book. Either directly or indirectly. You are responsible for
your own choices, actions, and results.
Legal Notice:
This book is copyright protected. This book is only for
personal use. You cannot amend, distribute, sell, use, quote
or paraphrase any part, or the content within this book,
without the consent of the author or publisher.
Disclaimer Notice:
Please note the information contained within this document
is for educational and entertainment purposes only. All effort
has been executed to present accurate, up-to-date, and
reliable, complete information. No warranties of any kind are
declared or implied. Readers acknowledge that the author is
not engaging in the rendering of legal, financial, medical or
professional advice. The content within this book has been
derived from various sources. Please consult a licensed
professional before attempting any techniques outlined in
this book.
By reading this document, the reader agrees that under no
circumstances is the author responsible for any losses, direct
or indirect, which are incurred as a result of the use of the
information contained within this document, including, but
not limited to, — errors, omissions, or inaccuracies.
 TABLE OF C ONTENTS

Introduction

1. The Foundation of Data Storytelling - Developing

the Narrative
2. Captivating Your Audience
3. Refining Your Visuals - Choosing the Right Chart
4. Refining Your Visuals - Developing a Winning
Design
5. Crafting a Winning Data Story

Conclusion
Resources
 FREE DATA VISUALIZATION
CHECKLIST!

Make sure you review every visual you create

with this Data Visualization Checklist to
ensure your charts are set up for success.
You also get a bonus chart selection table.
Scan the QR Code or visit
ElizabethSClarke.com to download yours
free!
 INTRODUCTION

“Data are just summaries of thousands of stories –

tell a few of those stories to help make the data
meaningful.”

— CHIP AND DAN HEATH, NEW

YORK TIMES BESTSELLING
AUTHORS.

Data volume is expected t0 reach 79

zettabytes (ZB) by the end of 2021. By the
end of 2025, that figure is expected to more
than double to 181 ZB. To be clear, 1 ZB is 1
billion terabytes! As impressive as those
numbers are, none of those bytes would
amount to much if they were not translated
into an easily digestible format.
Enter left stage… More specifically, data
storytelling.
For decades, the traditional way of presenting
data was to pull out the old faithful pie or
graph chart and spit out facts like a robot
reciting the numbers on a spreadsheet. In
such scenarios, most often, the listeners
needed clothes pins to keep their eyes open.
There was often the request to have the same
information sent over through email or
another method for “review” - a cleverly
disguised way of gaining the information that
was not digested during the boring
presentation.
Ensure that is not the reality that you face
when presenting data by tossing tradition out
the window. Basic charts and numbers just do
not cut it anymore. When most people have
the attention span of a squirrel hopped up on
caffeine, there needs to be a more
compelling, more enticing way of delivering
this data. And this is what data storytelling is
all about – remixing the quantitative and
qualitative nature of data and bringing the
point across in a way that makes other people
want to listen and learn more.
Over the last few years, the data industry has
exploded in growth because this secret is out
of the bag. Politicians use data storytelling to
persuade voters to place the tick on their side
of the ballot with touching commercials
highlighting their contributions to the
community. Marketing departments use it so
that their brands touch your heart, and thus,
you fork over some of your hard-earned cash
before you make a conscious decision to do
so. Data storytelling drives innovation and
product development and adds to the zeros on
the bottom line of major corporations like
Coca-Cola, Mercedes Benz, and Amazon.
The examples of effective data storytelling in
action are almost endless because it has been
realized that there needs to be a change in the
way we present data to conferences,
seminars, potential and existing clients,
customers, business executives, and more.
Through this change, others can happen.
Without this change, most of these many
bytes of information would go unnoticed.
This change does not have to just happen on a
big business level.
This change can start with you.
As a marketing manager in charge of brand
strategy, I have met many people in my line
of work. I have met people new to the
industry who want to leave a mark when
making presentations. I have met people who
have made presentations in the past but failed
to drive the message home to their bosses,
board members, and other listeners. I have
met small business owners who are stumped
about how to win over new clients and
customers when telling their brand stories. I
have met people who simply want to have the
know-how of crafting a captivating and
engaging presentation in their back pocket so
that they can pull it out whenever they need
it.
All these people had one thing in common
even though they came from a vast array of
industries – They were struggling to present
data in an effective, straightforward way that
gets to the heart of the message they were
trying to convey. These people had figured
out that pulling out the old presentation board
and pointing at charts and numbers does not
make an impact and were seeking a better
way.
If you can relate to how these people feel, this
book was created specifically to help you get
ahead of the curve and discover how to
develop storytelling expertise that will give
your data presentations the panache they need
to hit hard and hit home.
Data storytelling and visualization are quite
hefty words for the tongue to lift, but they are
not as complicated as they sound. They do
not always have to involve learning to code
or sweating over complex charts, as many
assume. They are simply methods of clearing
the clutter that zettabytes of data can produce
to reveal a clean, concise goal around which a
story can be created to drive change. This
book helps you clear the clutter and create
that path with ease.
You, too, can impress your boss and other
executives with your data presentation skills.
You can also win over new clients and
customers and convince the existing ones to
recommend you to their peers. You can
convert data to dollar signs for your company
and be on the fast track to your next
promotion. The things that you can do by
supporting data with a story are only limited
by your imagination. You have the power to
drive change and be part of this evolution of
how data is presented and consumed.
You can do all of this and more by learning
the five parts of crafting an effective and
engaging data story. This book was broken
down into five parts to make it easy for you
to understand the strategies and knowledge
outlined in the pages to come. You’ll also
have a guideline to follow when creating
visuals and presentations. Let me give you a
brief breakdown so that you get the gist of the
exciting things to come:
TH E FOUNDATI ON
You need to be able to quiet the noise of all
the bytes of data and focus on what is truly
important – the goal of your presentation. The
goal is what allows for the proper
development of the narrative of your data
story. Your narrative defines the sequences of
events and how you will layer your data for
the best appeal. The narrative structures the
whole story and holds everything together.
Call it your data story glue, if you will.
Without this glue, everything will inevitably
fall apart. Without a solid narrative, not even
the best visuals will salvage the wreck that
the data presentation is bound to be.
This part of the book focuses on helping you
drill down on your data presentation goal and
how to sequence your data story around it for
an easy, natural flow that resonates with your
audience and solves the right solution.

C APTI VATI NG YOUR AUDI ENC E

A good data story is effective not just because
of what you say but how you say it. To
deliver your message most effectively, you
need to understand your audience and speak
their language. Captivating your audience
with an engaging story is what plants the seed
to significant change and growth.

C H OOSI NG TH E R I GH T C H AR T
Now, do not get me wrong. I am not
knocking the tremendous contribution that a
good chart or well-dressed set of numbers can
have to a data presentation. The problem is
that too many people focus solely on this and
forget the narrative to support these data
presentation tools.
The correct chart presented at the right time
can take a simple analysis and give it a visual
form that people can use to develop a mental
image of what you are presenting. This will
help your presentation stick in their minds. It
will make you memorable.
Data visualization is another crucial
component of the data storytelling process
that you must nail on the head. You will
surely learn all you need to know about
presenting the visual aspect of your data
presentation in this part of the book.

A WI NNI NG DESI GN
Sight is the most used of all the human
senses. Use that knowledge to your benefit.
Do not just drop bland, tasteless charts and
graphs on your audience’s laps and call it a
day. That will only leave a bad taste in their
mouths and make you look less than the
competent business person you are.
The face of any data presentation is the
design. Having clean, concise, clutter-free
designs is crucial to keeping your audience
informed and engaged. As they say –
presentation is everything. When it comes to
data storytelling, you need to take the saying
literally and figuratively.
C R AFTI NG A WI NNI NG DATA
STOR Y
The parts discussed above are essential, but
none of them will give your data presentation
the edge it needs alone. You have to bring
them all together to amaze, astound, educate,
and convince your audience. This last part of
the book shows you how to do just that,
painlessly and cohesively.
I run into large amounts of data every day in
my career in marketing and social analytics.
If you are anything like me, you find yourself
fascinated with the components that make up
company growth statistics, consumer and
social analytics, sales figures, expenditure
reports, and all the other numbers that show a
company’s performance. While these figures
excite the senses of the data science nerd in
me, I also realized that they could look like a
foreign language to someone else. I have a
few failed data presentation stories of my
own that I can tell.
When I was new to my career, I struggled to
translate these figures to other people. Still,
experience and knowledge collection have
taught me the best ways to transform any
form of data into a language that anyone can
understand. Translation through storytelling
is a method that cannot be beaten.
Data storytelling is a vital part of any
company’s growth and management. Big
businesses need it and so do small ones.
Executives that have proved their mantle
need it no matter how much they advance in
their careers, and so too does the intern just
starting in a field. My passion is to help as
many people and businesses across the board
take raw data and translate those bytes into
stories that encourage change that allows
meeting goals and targets. Call me quirky,
but I always love when people reach out to
me with stories about how the words I have
written have helped inspire the change they
need to take their professional lives to the
next level.
I hope to hear a similar story from you. A
data story of how much you have achieved
using the words in this book, perhaps.
I suggest you get out a highlighter and mark
anything you find valuable and worthy of
remembering. One of my favourite tricks for
revisiting valuable information with ease.
The data shows that anyone who reads past
this page to the first chapter dramatically
increases their chances of becoming a data
storytelling superstar. So, what are you
waiting for? Turn the page!
 1

THE FOUNDATION OF DATA

STORYTELLING -
DEVELOPING THE
NARRATIVE

“The goal is to turn data into information and

information into insight.”

— CARLY FIORINA, FORMER CEO,

HEWLETT PACKARD

W ith the many, many bytes of information

available for relating to other people,
how do you decide which ones deserve
precedence and should be added to your data
story? That is a fundamental question when
approaching data storytelling. If this question
has come to your mind, you have set yourself
up with the right mindset to present data in
the most digestible way to your audience.
The answer of which bytes of information
you will relate to your audience depends on
your final goal in the presentation. Too many
analysts get stuck on the visual aspects of the
presentation and leave the information that
needs to be relayed as an afterthought. But it
is truly the other way around. The visuals do
not matter if your audience cannot follow a
defined path to digesting the ultimate goal of
the presentation.
Knowing what your goal is allows you to
develop the narrative to your data story. The
narrative is your plot and the goal of your
data story is the climax of that plot. Think of
any good book that you have read or any epic
movie that you have watched. There is a
defined path of events (the plot) that leads to
the climax (goal) of the story. In the
beginning, there is the hook that captures
your attention. There is a midpoint that tells
you that there is no going back for the
protagonist of that story. Finally, the climax
allows you to feel fulfilled in having read or
viewed that story. If any of the critical events
to the plot are missing, you feel dissatisfied
with that movie or book even if you have no
idea why. There is a natural progression to
storytelling that must be adhered to no matter
the type of story, and we are aware of this,
even if only on a subconscious level.
Therefore, the same thing will happen if you
do not apply a narrative to your data story.
Your audience will feel dissatisfied and
unlikely to get the gist of what you portray
with that data story.
However, before you can get to developing
the narrative of your data story, you must
take the time to grasp the goal of this
presentation firmly. Once you have the goal
in mind, you can work backward so that you
develop your narrative. This chapter is
dedicated to outlining precisely the narrative
of data storytelling and how you can create a
compelling narrative that truly allows an epic
climax.
WH AT I S TH E NAR R ATI VE OF DATA
STOR YTELLI NG?
The amount of data collected daily in the
world is truly astounding. From large
corporations to small businesses, every aspect
of a business needs to be recorded. This
collection develops a view of what is
happening in the industry. There is a shortfall
to this many bytes of data accumulating, and
that shortfall is that they do not tell the story
of why these events are happening. This is
where data storytelling comes. Data
storytelling assigns meaning and bridges the
gap of the what and why of the many bytes of
information that businesses collect. The
narrative of data storytelling is the bricks,
mortar, wood, nails, and all other materials
used to put this bridge together. If those
building components are not put together
correctly, that bridge falls apart - indeed, a
crisis that needs to be avoided.
As a result, before presenting any data,
enough time must be spent developing the
narrative of your data story. The first part of
the development is understanding what the
narrative is. That understanding is the focus
of this section.
A data storytelling narrative is about
developing a language that allows for
augmenting data in the most effective way to
deliver to an audience so that the people in
the audience are not left confused and trying
to piece together these bytes of information.
The narrative of the storytelling is the vehicle
that conveys insights on the data that has
been collected to the audience.
Clear language does not just refer to using
plain language that is not filled with jargon
particular to that industry. It also refers to
developing strong plotlines that allow the
data story to have pivotal moments. Those
moments will enable the audience to have
“aha” moments. Such moments are
characterized by the audience having a
sudden realization and wanting to know
more. Such moments have the audience at the
edge of their seats - just like your favorite
story or movie.
For example, one of the most important parts
of the narrative of a good data story is the
development of the hook. The hook comes at
the beginning of your data story. It is what
captures the attention of your audience. The
hook can be a question or stating a problem
that this audience shares as a commonality.
As a result, an effective narrative takes that
audience through a journey that builds
momentum from stating the hook and finally
arrives at the goal of the data presentation,
which is to provide a solution to the problem.
All data story narratives are not the same. If
they were, then that would be quite an
adventure through boring land as your
audience has already seen this particular data
story in a different setting. That is the
exciting thing about data story narratives.
Even though there is a defined path of events
that you can take to develop one, no two
narratives are entirely the same. So, you can
wow your audience time and time again, even
if you are using the same figures of
information.
Still, there are a few familiar data story
narratives that every good data storytelling
must know. They are:
Trends
Presenting trends focuses on how figures rise
and fall over a given amount of time and how
these patterns of number behavior affect the
audience. For example, a business analyst
may present data to the executives that show
sales figures rise towards the end and
beginning of the year but dip around the
middle of the year. Using this data, the
narrative can be focused on why these figures
are so and what can be done to increase sales
when there is traditionally a decrease.
Comparisons
Showcasing comparisons builds on trends
and shows how data changes over time in
relation to specific periods. Using the same
example above where a business analyst
presents to the company executives, the
analyst can compare the sales figures over the
last five years. Perhaps there was a year
where sales were abnormally high. The
comparison will allow that team to develop a
way to make that abnormality a regular
occurrence by analyzing the events
surrounding that particular year.
I will continue using this particular business
analyst example to provide context to the
narrative types below.
Rank Order
Communicating a hierarchy of factors to
make a large amount of detail easier for the
audience to digest. For example, the business
analyst may present a table highlighting the
best performing and worst performing
campaigns. With this data, the executives can
reallocate the budget and focus on their top
converting revenue streams. Also, the
marketing team can focus on promotions that
showcase their best-selling softwares.
Statistical Relationships
Statistical relationships allow the audience to
know the relationship between different data
types to predict how various factors affect
each other. The business analyst may display
the information that shows how the time of
year impacts whether or not specific
campaigns are converting well. For example,
there may be a correlation between the winter
season and their website optimization
software sales as people prepare for the
Christmas rush. As a result, executives can
ensure that the proper software is pushed
during Quarter 4 to increase sales.

“For context, A Quarter is a three-month

period on a company's fiscal year that acts as
a basis for periodic financial reports and the
paying of dividends. A quarter refers to one-
fourth of a year and is referred to as Q1
(January, February, March) Q2 (April, May
June) Q3 (July, August, September), and Q4
(October, November, December).”

C OUNTERINTUITIVE D ATA
This narrative shows surprising data that
contradicts what the norm is. The
counterintuitive nature of data invites further
exploration, especially in instances where it
has an adverse or alternatively, favorable
impact on the business. For example, the
analyst may show the executives a surprising
increase in website optimization software
outside of the peak season. This unexpected
data needs further exploration as to why this
happened to determine if more sales can be
derived from the revealed answer.

UNDER STANDI NG TH E NEED FOR

A NAR R ATI VE
A good narrative data story gives you
direction and purpose. However, there are a
few benefits that you will experience when
you develop your narrative before focusing
on visuals:
A data story narrative allows for only
relevant data to be presented
As I have mentioned several times in this
book already, there are tons and tons of bytes
of information that can be presented to your
audience. Every day, there are tons of it
created in the runnings of every business.
Even when no employees show up to work,
there is still data created about that business.
However, you cannot throw all of this data at
your audience, expecting them to understand
it even if you do. This will only serve to
make your audience feel like a leaf in a
thunderstorm - with no sense of direction of
which byte of data to grasp first.
An effective narrative lets you know what
data you need to pick from the plethora to
present to your audience. Your audience does
not need to know every single thing that
happens in the business. They only need to
know what is relevant to addressing the
particular problem that data presentation
needs to address.
Therefore, you need to keep in mind that you
want to clearly communicate so that they feel
relieved that a solution is being developed to
address their specific problem by the end of
that presentation.
Knowing the final goal of your presentation
and how you are going to get to it aids in
developing a narrative that picks up points of
data that are only relevant to that specific
presentation. The narrative development
allows you to see a way through the fog of all
those bytes of information. It also allows your
audience to have that clearer vision.
An effective data story narrative depends
on citing credible data
From the tons of available data, you need to
remember that not all data sources are
relevant or credible when presenting your
data story. Understanding this allows you to
specify what types of data you will present
and where you cite this data.
Your audience needs to trust that the
information you are sharing with them is
credible. It needs to ensure people look at you
as a credible source of information, and when
they relay that information forward, it is
accurate.
All the data that you need while creating your
presentation might not be available
immediately. Because the narrative of your
data story helps guide the purpose of your
presentation, you can then reason what
information should be included in your
presentation and approach the suitable
sources for that missing data.
Allows for the development of a clear path
of events that get your audience acting as
desired
While a data presentation filled with that
particular language might work for internal
personnel, this will be a recipe for disaster if
you talk to people outside the industry or
company. A strong narrative allows you to
break down data into Bite-sized language that
everyone can understand. A person should
not have to be familiar with the bytes of
information to digest your presentation easily.
This is because a narrative helps define the
events that must be touched on to reach your
goal. Those events can then be broken down
into plain language for a straightforward
presentation. Like all good stories, the
narrative must have a beginning, a middle,
and an end.
Allows for the development of the best
visualizations to best match the goal of the
data story
Graphs, charts, photos, text, maps, tables. All
of these charts and more can be used to create
an effective and engaging presentation.
However, just because they can be used
doesn’t mean you should use them.
Using visuals that are not effective for your
data story can confuse your audience far
more than enlightening them. Therefore, it is
imperative that you choose the proper
visualization to enhance your details. The
narrative allows you to select which
visualizations your audience will most
certainly understand easily. These visuals
need to follow the following criteria:

Appropriate for the data presented. For

example, if you are presenting to your
manager who is familiar with the
information, it would work against the
data story to use a data presentation
 style that is more appropriate for an
executive who is entirely new to the
data.
They are labeled appropriately.
Visually legible. Your audience should
not have to squint to note what your
visuals are presenting. All written
information should be clear to read.
This means that you need to pay close
attention to items such as your font
choice and font size. If it is presented in
an audio format, it should be audible
and pleasing to the ear.
Easy to understand. The information
included in your data visuals should not
omit pertinent information, should not
be manipulative, and should not be
cherry-picked data.

In addition to these factors, you need to

ensure that your visuals have appropriate
color and imagery.

STEPS FOR DEVELOPI NG TH E

NAR R ATI VE OF YOUR DATA STOR Y
All of your favorite novels and books
incorporate a few key elements in the
narrative that get you hooked from the first
scene or page. These elements include:

The plot. This address is how you build

momentum from the hook to the final
course of events that deliver on the goal
of the storytelling. When related to data
storytelling, this describes how you can
effectively and efficiently solve the
problem that your audience is
experiencing.
The setting. In a traditional novel or
movie, the setting describes where and
when the events take place. Related to
a data story, this describes placing your
audience in a position where they
understand the data and insight that you
are delivering. Strategically placing
your audience in a helpful position in
your data story includes providing
historical data, benchmarks in the
industry being discussed, and processes
 that are currently being undertaken.
The characters. In a novel or movie,
this describes your protagonist,
antagonist, and supporting characters.
The characters in a story are critical in
forming a stance on how the story will
be relayed to listeners, readers, and
viewers. Related to data storytelling,
this describes the tone and priorities
that will be delivered to be most
effective with your audience.
The climax. This is the end of the story
and describes the culmination of the
challenges faced throughout that story.
The climax describes the solution
presented for the problem stated at the
story’s beginning.

No matter the type of story you are telling, all

of these elements need to be incorporated,
and this is no different for the business world.
Your data story must include these elements
for its narrative to be effective.
Luckily, you do not have to be a great
novelist or screenwriter to incorporate the
same elements into an effective data story.
There is a formula for this, and it follows the
following steps:
Step 1 - Identify the Goal of your Data
Story
To set up the goal for your data story and
subsequently, set up the foundation for the
narrative, there are a few questions that you
need to ask yourself before you do anything
else. These questions include:

What is the problem that will be

presented in your data story?
What is the possible solution or
solutions for solving that problem?
What would you like to achieve by the
end of your data story?
What call-to-action can you
strategically place in your presentation
that will likely cause your audience to
react in the way you would like?
What takeaways would you like your
audience to live with by the end of your
data story?
Answering these questions thoroughly will
allow you to set up the plot for your data
story. You will know the beginning, middle,
and end of that story so that you can extract
data in alignment with that vision.
The important thing to remember when
answering these questions is to let the data
guide the narrative. Sometimes we believe
that data is driving us in a particular
direction, but picking out only irrelevant
information will misguide us. To make sure
you have the correct data and it is guiding
you in the proper direction involves doing a
few activities. Such activities include:

Making comparisons of different sets

of data. This will allow you to form
correlations to see how these datasets
relate to each other.
Look for trends. Trends allow you to
note how different aspects of the
business are developing, changing, or
remaining stagnant.
Noting anomalies. Anomalies are sets
of data that do not align with what you
 expect or are outside the norm. Noting
anomalies will prompt you to look at
why this is happening and whether this
is in favor of your business or
detrimental to its activities.
Noting counterintuitive data.
Counterintuitive data is surprising or
not what you would expect out of
evaluating specific trends or making
comparisons. Like anomalies, such
information needs to be analyzed to
determine whether or not they are
favorable to the runnings of the
business or detrimental to it.

It is essential that you take this time initially

to do this analysis so that you realize the true
goal of telling your data story rather than
what you believe it to be. Analyzing data is
quite an adventure as it can bring up
unexpected twists and turns.
Step 2 - Align Your Data Story Goal with
Your Audience
This step aligns with creating the setting, like
in a fictional novel or a movie. It sets up the
who and when of your data story. The “Who”
is your audience. To tell a data story relevant
to your audience, you need to take the time to
learn about them. You need to understand
factors such as their:

Demographic
Age
Knowledge of the subject matter that
will be prominent in your data
presentation
Careers and educational background

Knowing these details and more allows you

to develop your data story in a way that is
most relatable to your listener. This will
allow you to see if you can add more specific
jargon to your data story or if you need to
simplify the language for clearer takeaways
while you are presenting. More importantly,
knowing your audience allows you to
understand why they care about the problem
and the most specific solutions.
Knowing your audience allows you to
develop the when of your data story. The
“when” describes how far back the
information needs to be provided and what
current analysis and future predictions are
specific and relevant to that group. These are
the specific types of information that will be
included in your data story to have the most
impact on this group.
Step 3 - Develop the Structure of Your
Data Story
With your audience and goal in mind, you
can start developing the structure of your
narrative. This structure will include the
following elements:

1. The context of your data story. This

speaks to why the story is relevant and
worth telling to your audience.
Knowing the context of your story
allows you to develop a hook that
engages your audience and makes them
invested in it from the get-go.
2. The key players that are related to the
context of that data story. Like in a
 book or movie, some characters are
significant to the advancement of that
story. These players could be
executives, customers, clients, and
more.
3. The problem that needs solving. The
problem stated in your story is the
whole reason this presentation was
necessary in the first place. There is a
conflict that your audience is invested
in, and you need to state this clearly so
that this group knows why they are
being presented with this data.
4. The solution to the problem. Of course,
if there is a problem, you cannot leave
it unsolved. So, the follow-up to letting
your audience know what the problem
is to provide them with possible
solutions. These solutions need to be
broken down into clear action steps that
can be taken to solve this problem. You
also need to provide your audience
with this solution in a relatable way
that allows them to know what value
they are gaining by following the
action solutions you have provided
 instead of other solutions.

It is best to tell your data story linearly.

Following the structure from context to key
players, and then problems and solutions. Just
like you would not tell a story in a book or
movie by starting with the middle or the end,
you also need to deliver your data story in a
way that makes sense to your audience.
Step 4 - Consider Your Data Visuals
This is the climax of your data story. When
you get to this point in your data story, you
should be able to tick off your ultimate goal
in having created that story in the first place.
Your data visuals bring all the previous
elements together in a small flow that
engages your audience and allows you to
retain their attention so that you can deliver
the climax of your data story, which is the
solution to the problem.
Data storytelling is not just about telling your
audience. You need to show your audience,
and that is the role of creating visualizations.
Visualizations enhance the story you are
telling and help simplify the information so
that the most important parts are highlighted
clearly and immediately.
There are a plethora of visual options. Some
of your choices include:

Bar charts
Pie charts
Road maps
Country maps
Area maps
Flowcharts
Tables
Pictographs
Scatterplots
Venn diagrams
Hierarchy diagrams

Remember that the visuals that you choose

need to be relevant and engaging to your
particular audience. The visuals need to align
with the interests of that audience. These
visuals need to capture their attention
immediately and hold onto it for the duration
of your presentation.
Infographics vs. Data Visualization
On the topic of data visualizations, it is
pertinent that we take the time to address a
point where many people feel confused. The
fact is whether or not data visualizations and
infographics are the same. The answer to that
is they are not one and the same even though
they are similar types of visual content.
Data visualizations are translations of
datasets through individual charts that make
that data easy to understand in a visual
format. Whereas looking at massive
spreadsheets with figures upon figures can be
confusing, using data visuals like a bar chart
or a map allows you to digest that
information in a much easier, faster way.
The purpose of infographics is to allow
persons to note a large amount of information
through the combination of text, icons, data
visualizations, and illustrations to make an
informed decision. As you can note from this
definition, infographics include the use of
data visualizations. Whereas data
visualizations are typically brief visual
content that communicates a certain point,
infographics allow more information to be
condensed in an organized flow. It is typical
to note the use of infographics in areas such
as the homepages of websites, landing pages
for marketing campaigns, brochures, and
social media.
Both of these types of visual content can have
quite an impact when presenting data.
However, which one you use depends on the
purpose of the presentation. Infographics are
typically used in marketing campaigns. Data
visualizations are more prominently used for
data storytelling.

TH E 5 KEY C ONC EPTS TH AT YOUR

DATA NAR R ATI VE MUST FOLLOW
TO BE EFFEC TI VE
The preceding section is about creating the
bones of your data story. You have the
general structure. But it is still not ready to be
presented to an audience so that it is most
effective. There are a few key concepts that
you must apply to the narrative of your story
so that you communicate your vision clearly
to your audience. Without the application of
these key concepts, you still run the risk of
confusing your audience even with a good
structure and flow of content.
This section is dedicated to showing you how
you can pretty up your narrative to be
appealing and engaging to your audience.
There are five concepts to doing so, and they
are:
Concept 1 - An Effective Data Story Is a
Natural Extension of the Problem to Be
Solved
There needs to be a natural, linear
progression of your data story from beginning
to end. From the jump, the problem needs to
be stated in your hook. You need to provide
at least one solution to this problem in the
climax of your data story.
Remember that your audience will feel like
something is off if you do not give them the
structure that naturally flows from point A
(the problem) to point B (the solution), even
if they are unsure where that feeling
originates. Like a plant blooming from seed,
you need to show that your data story comes
from that central point, which is the problem
to be solved.
Every word you use and every visual you
present needs to naturally lead back to this
point.
Concept 2 - Understand Your Audience to
Avoid Visual Complexity
Your job as a data storyteller is to simplify
the many bytes of data available about a
business and present them in an engaging
way that helps the audience solve a problem.
Fundamentally, you get to know your
audience as part of developing your narrative.
This allows you to create a data story
presented in a language that resonates with
that group. Your problem might be complex,
but you need to break down that complexity
into language that the audience relates to.
Visuals that are too complex often showcase
a few characteristics that include:

Not understood by the majority of the

audience and so, the audience shows
visible signs of confusion
Overly detailed
Inappropriately labeled
Often highly colorful
Looks sales pitchy
Has a promise that cannot be delivered
on

Overall, if a visual is too complex, it sends a

confusing message to the audience. This
leaves a lot of room for misinterpretation.
This does not align naturally with the
problem that must be solved from that data
story. This also leaves your audience with a
few assumptions about you, none of which
are positive. Such assumptions can include:

You are biased towards a specific

solution or opinion even if this solution
or opinion is not supported by the data
You are arrogant
 You want to obscure important details
about the presentation
You are creating a data story that only
supports a belief rather than one meant
to inform that audience.

The ironic thing is often when we are

confronted by visual complexity, the
underlying problem is not nearly as complex.
This is why it will be easy for the audience to
assume any negative things about the data
storyteller.
Concept 3 - Trust Your Audience to Avoid
Visual Simplification
Simplifying your data presentation to
language that your audience understands does
not mean that you must patronize your
audience. You never want your audience to
feel that you are insulting them with
oversimplified language.
Just like with the irony stated above, where
low problem complexity can lead the
inexperienced data storyteller to use high
visual complexity, we have the opposite issue
here with high problem complexity. Many
experienced storytellers can resort to using
low visual complexity to highlight more
complex problems. Again, there is an
unnatural extension of the data visuals to
support the problem that needs solving.
This point and the point above highlight why
it is so important that you become very
familiar with your audience and the things
they favor, such as the language they speak.
If you are speaking to engineers and the
problem is related to engineering, you only
use technical terms related to that group.
"Dumbing down" your language is likely
going to come off as insulting to such a group
if your topic is engineering-related.
Keep the language on par with what data is
being presented and who you are presenting it
to. They need to focus on your solution from
the data, not what specific vocabulary means.
Concept 4 - There Are Tradeoffs Involved
in the Use of Proper Data Visualization
The effective and efficient use of data visuals
falls between the problem complexity and the
visual complexity needed to support that
problem complexity. Many data storytellers
spend what can seem like an eternity
pondering how complex they should make
data visuals for presentations with the thought
of pleasing every single member of the
audience.
The fact is that just as with other things in
life, you cannot please everyone, especially
when the group is on the larger side. There
will always be personal preferences that turn
at least one person off. For example, one
member of your audience may despise the
shade of green you use even if they found
value in the data story you presented.
As a result, you need to aim for presenting
visuals that please the majority of your
audience. There will be trade-offs when
developing data visuals that support your data
story because of this.
Concept 5 - Respect Your Audience to Win
at Data Storytelling
The majority of your success when making a
data presentation relies on the receptiveness
of your audience. Your audience can make or
break your presentation. Because of the
weight of their opinion, your audience
deserves respect. They deserve the time and
effort that you put in before the presentation
to learn about their particulars. You can have
the best data visualizations and a superb
narrative. Still, this data presentation is
doomed to failure if you patronize, talk down
to, or do not correctly match the problem
complexity to your audience’s understanding.
Far too many data storytellers approach data
presentations as if they are teachers talking
down to a group of kindergarten students.
You are not superior to your audience
because of your knowledge about the data
being presented. You need to ensure that you
have it in your mind that you are on a level
playing field with your audience. Just as your
audience can learn from you, you too can
learn from your audience.
 2

CAPTIVATING YOUR
AUDIENCE

“Presentations aren’t about the presenter; they’re

about the audience and what the audience needs.”

— SIMON RAYBOULD

B ydatanow,storyyoupresentation’s
would have realized that your
success depends
on the depth of knowledge that you have
about your audience. You need to have a
profile of these people before you take even
the first step in putting your information
together and certainly before you start
designing that presentation. Doing otherwise
is like hunting in the dark and hoping to hit
the target - a nearly impossible task.
Ensure that you hit the mark as close to the
bullseye by putting in that preliminary
groundwork about the people you will be
presenting to. Ensure that you captivate these
people to have the highest chance of eliciting
the change you want from making that
presentation.
The question that has many junior analysts
abandoning this vital task is the how of it.
How exactly do you go about finding what
you need to know about your audience so that
you cater your data story to touch their
emotions and better get your point across?
Luckily, this is not a matter of guesswork.
You can implement proven methods to gather
the information you need to know about this
group. This chapter focuses on the “how” of
getting this information and using it to
maximum effect.

I DENTI FYI NG YOUR AUDI ENC E

The process of learning your audience starts
with asking lots of questions. You need to
have that inquisitive nature ingrained in you
to learn about this all-important group of
people. Only by asking questions can you
determine what is essential to these people
and, therefore, how you can touch them most
deeply with the message you want to convey.
You have to remember that the message is
not as important as how you deliver it. And
the how of it is dependent on who you deliver
it to.
Think about it in your everyday life. There
are some people that you can tell things
straight up and bluntly, and they will not be
offended. In fact, they are more respectful of
that straightforwardness. On the other hand,
there are people in your life who sometimes
need you to sugarcoat the same message so
that it is received in as efficient of a manner
as possible. No one approach is better than
the other. It is simply a matter of how best to
build a bridge of effective communication.
How we deliver messages is dependent on the
receiver - the audience - in everyday life and
data storytelling.
There are many questions that you can ask
and, therefore, develop answers about your
audience. However, just as you can get
bogged down by the many bytes of data you
want to convey to your audience, you can
also get overwhelmed by the sheer number of
questions you can ask about them. That does
not have to be your reality. The solution here
is to keep things as simple and as
straightforward as possible. Only ask
questions that allow you to determine the
most important factors about your audience
and why your message would be important to
that group.
This section focuses on some of the main
questions that typically run across the board
when learning about your audience. The
answer to these questions gives you a solid
foundation that gets you to the heart of the
members of that group as quickly and as
efficiently as possible.
Although all this background work can be
rather tedious ill leave you with a quote to
shift your perspective:

“Give me six hours to chop down a tree and i will

spend the first four sharpening the axe”

— ABRAHAM LINCOLN

We don’t just want to create some visuals, we

want to win with them! A fancy chart won’t
make up for lack of preparation. Now, let’s
figure out who our audience is to have a
better chance at winning them over.
Who Is Your Audience?
This is the first question you need to ask
yourself because it directly impacts how you
convey information to the group and what
format needs to be relayed. The answer to
this gives your audience a face. It makes you
feel like you are talking to real people while
compiling your data and preparing the visuals
to support your message. This will
undoubtedly help you build a better
presentation compared to having an obscure
image in your mind of what this group of
people looks like.
With the power of knowing what your
audience “looks” like, you are better able to
shape a message that will be received by that
audience.
There are five main categories of audiences
that you will encounter as a data storyteller.
They are:
The Novice
This is the type of audience’s first exposure
to the subject matter. In such a case, you want
to simplify the information so that it is easy
to understand. Therefore, jargon and
technical terms should be avoided. Stick to
layman’s terms and common language. On
the flip side, you do not want to insult the
audience's intelligence by oversimplifying the
information.
The Generalist
This type of audience is more aware of the
topic of discussion but still lacks
understanding in certain areas. As the data
storyteller, it is your job to fill in these gaps
and provide this audience with knowledge on
the major themes connected to the data being
conveyed.
The Manager
Such an audience has an in-depth, actionable
understanding of intricacies and
interrelationships about the data due to their
experience and access to details. As a data
storyteller, you do not have as much to do
concerning explanations, but that still does
not take away the importance of conveying
your message to this group in an efficient and
effective manner.
The Executive
This audience understands the importance
and probable outcomes of certain situations
but still needs the details. It is your job as the
data storytelling to build the bridge between
the data and those possible outcomes.
The Expert
This audience has the most in-depth
knowledge about the data and does not rely as
heavily on your data presentation to be
informed. This group is more interested in
gaining more information compiled in a
cohesive way to generate faster conclusions
about the data.

I DENTIFYING your audience's background and

the depth of their knowledge of the data
allows you to develop a strategy for how
specific you need to be when explaining
terms and designing your visuals. For
example, a novice audience to the marketing
strategy might need a rundown on the
marketing budget allocation before they can
understand why advertising performance is
dropping. At the same time, a group of
marketing managers or media buyers do not
require such an explanation. Knowing what
your audience already knows determines how
your data story will play out for the most
effective communication. Keep in mind,
being a novice doesn't mean it's their first day
on the job. A higher-level executive might be
unfamiliar with the allocation of the new
advertising budget and needs a simple
explanation to understand your point entirely.
Knowledge of your audience and
understanding of the data also allows you to
know what the listener is trying to gain from
your data story.

WH AT I S TH E PER SONALI TY TYPE

OF YOUR AUDI ENC E?
Answering the above question lets you know
who your data visualization is tended for and
what this audience already knows about the
data you will present. Next, you need to know
how best to address this audience, and this is
where knowing the various personalities of
your audience applies.
Of course, there will be a wide array of
individual personality types in your audience.
However, when people come together in a
group, they tend to adopt one dominant
personality type. This is typically a product of
the setting and company culture. For
example, some businesses have a more laid-
back and amicable approach to doing
business. On the other hand, some companies
prefer a more straightforward and analytical
approach between team members. Your job
as the data storyteller is to find out exactly
how this group is used to communicating
with each other and develop that style of
communication when developing your data
story.
Before we outline the primary audience
personality type that you will encounter, I
want to take a moment to point out that your
personality type should not be the one
highlighted in your data story. Many junior
analysts tend to fall back on simply
developing a data story around one that
resonates with their personality type. But it
always needs to be at the forefront of the data
storyteller’s mind that this presentation is not
for them. Instead, it is catered to the audience
and what the audience resonates with first
and foremost.
There are four main audience personality
types. Whether you present to novices,
generalists, managers, executives, or experts,
you need to consider this personality type. Do
not simply assume that an audience will
adopt a personality style dependent on how
familiar they are with the data.
I will state each personality type below and
outlined how best to cater your data story to
communicate most effectively with such a
group.
Analytical
Analytical people are thoughtful and
purposeful individuals driven to maintain
order and strive for organization. They are
often likened to perfectionists. Keep this
image in your mind when you think of the
individuals that make up an analytical
audience. This audience is looking for you to
organize your presentation and provide data
in an ordered fashion that insinuates
collective thinking.
This is the personality type you might most
closely relate to or resonate with. An
analytical audience is used to processing data
to make informed decisions. They are the
ones surely turning over the piece of your
data stories in their heads so that they get a
clear understanding of the situation. Often,
even if this group is made up of novices, such
individuals would have done research
beforehand so that they come into this data
story delivery with an underlying knowledge
that allows them to better understand the data
that will be presented. This can make your
job as a data storyteller easier in many ways,
as there may be less explanation necessary to
powerfully deliver your main points.
However, this can also pose problems as the
audience can make assumptions beforehand,
and if the data does not support these
assumptions, you will have your work cut out
to convince them otherwise.
Also, while another audience may want you
to skim over a few details and simply present
what you feel is most important, an analytical
audience might wish you to provide even
more information outside what is represented
in your data visualization.
Your job with such an audience is to prepare
for any of such circumstances and still deliver
an exceptional and informative data story.
Competitive
On an individual level, one can spot a
competitive person through the high amount
of confidence that this person exudes. Such a
person is highly action-oriented and not as
highly detail-oriented compared to the
analytical personality type. Competitive
personality types naturally gravitate toward
leadership positions, but they also naturally
respect strong authority figures. Another
name for competitive personality types is
drivers. They certainly like to be in the
driver's seat.
An audience described as competitive will
not want to be bogged down with details, so
instead, focus on the big picture as the data
storyteller. Based on this picture, they are
very decisive, and you can count on them to
get things done.
However, you have to know how to approach
such a group to get this action-oriented
attitude on your side. You need to be very
direct and focused. In addition, you need to
place yourself in a position of authority. Your
position is being truthful and reliable in your
delivery of that data.
Amicable
People described as amicable personalities
are patient, hard to offend, and do not like
offending others. Do not let this description
fool you into thinking that this personality
type is one of the pushovers that can be
convinced easily. They can be quite stubborn
as well. Therefore, they will dig their heels in
if they feel that your data story is lacking in
some way.
Therefore, it is important that you show
yourself as trustworthy in the development of
your data visualizations. Also, this audience
feels an emotional connection with you and
the information. You can build that bridge by
making this audience type feel involved in
your presentation. Do this by asking them
questions frequently and listening to their
contributions.
Expressive
Expressive people can be described as social
butterflies. There is never a dull moment in
the company of an expressive personality
Because they love to laugh and talk. They
love to feel included as well as make others
feel included.
Therefore, as an audience, expressive
personality types want to feel as if they are
included in the discussion about the data that
you are presenting. Such an audience also
leans toward trusting their feelings over data
even when presented as hard facts. You will
have your work cut out for you when
convincing them of the value that will be
derived from placing merit on your data
visualizations.
One such way of convincing them of the
value of the data is to focus on the more
creative aspect of data storytelling, which is...
well, storytelling. Captivate this group by
sharing stories and human experiences that
might not be translated easily in data
visualizations.

I T IS possible that you will get an audience

that has overlapping personality types and not
one strong dominant personality type. You
might notice this, mainly when working with
audience members from different teams, such
as marketing, management, human resources,
and other departments in one company. So,
what do you do then? The solution here is to
again research your audience so that you are
aware that the members are coming into this
with several strong personality types that you
need to cater to. You should not show
preference to any one personality over the
other as you will make some of your audience
members feel pushed to the side. They will
feel like their opinion is undervalued, and
therefore, this will not have a reaction that is
conducive to your call to action. Instead,
what you need to do is develop your data
presentation in a way that maximizes your
chances of being understood by your
audience. This may mean incorporating
several techniques for getting through to
different personalities in one data story. This
can seem like a tall order, but it is likely to up
your audience’s chances of acting as you
would prefer.

WH AT AR E YOUR AUDI ENC E'S

LI TER AC Y AND NUMER I C AL
LI TER AC Y LEVELS?
While this question can immediately find
basic literacy and number literacy aptitude,
that is not the case. Instead, it is about
judging how well your audience relates to the
data that you will represent. For example, if
you are telling a data story to IT professionals
about an IT-related subject, then this
audience's literacy and numeracy literacy
concerning IT-related subjects will be one
where you do not have as much explaining to
do. On the other hand, you may be presenting
IT data to an audience made up primarily of
HR representatives, and even if this group of
people has excellent overall literacy and
numeracy literacy aptitude, this group may
find it challenging to understand the
particular language, both literature, and
numeral, relating to that subject matter.
Similarly, if you are telling a data story about
HR metrics, IT professionals may not have as
high of a literacy and numeracy level to
understand that particular language. In either
case, you need to break the data down into
manageable bite-sized pieces that are easy for
the audience to digest.
Many data storytellers are wary of
approaching an audience with a lower literacy
and numeracy literacy level in that particular
subject area, but this is instead an opportunity
in disguise. See, the thing is, the actual
mantle of a data storyteller is not made by
how aesthetically pleasing the data visuals are
or how well they put all the data together.
Instead, this mantle is proven by how well
this person can get the audience to follow
their train of thought. Suppose you can get an
audience with no prior knowledge of a
subject area to understand that subject area.
In that case, you deserve the title of a true
professional data storyteller. Do not worry if
you do not quite live up to that title yet, as
this is a learned skill rather than one that
some people are born with.
Some of the techniques that you can use to
better improve the understanding of an
audience that might have a lower literacy or
number literacy levels include:
Avoid Using Technical Jargon as Much as
Possible
All industries and niches have a unique
language with acronyms and technical
terminologies that will confuse outsiders who
are not as familiar with the happenings of that
setting. This particular language is called
jargon. The last thing you want to do is leave
your audience confused when they exit your
presentation. Therefore, the best practice is to
avoid using unique languages unless your
audience is made up of people familiar with
that industry or niche.
When it is necessary to use technical jargon
with a novice audience, ensure that you take
the time to explain what the terms mean and
how they relate to the context of the
presentation.
Be Humble and Use Humor
One of the best ways to break down jargon is
to use humor. A good joke to break the ice is
a great way to make everyone feel
comfortable, yourself included. Humor is also
an excellent tool for ensuring that your
audience sees you as a person just like them
and not someone talking down to them
because of your superior knowledge of the
data.
Be sure to convey through language and body
language that you are willing to explain
things that are unclear to your audience.
When presenting a data story, the aim is not
to impress your audience with how smart or
informed you are. Instead, it is to inform your
audience to make a sound decision about how
to act thereafter. That means using language
that the audience understands.
Pay Attention to Your Audience's Cues
There may be times when your presentation
does not immediately resonate with your
audience. This does not automatically mean a
failed data story. Instead, you can learn to
read the non-verbal cues that your audience
expresses to adjust your language for
effective communication to take place.
Simply use a conversational tone to explain
whatever information that you just imparted.
Reading the room is a skill that you must
develop, as data storytelling is fluid and
requires you to change and adjust on the spot
to reach your audience no matter the
circumstances.
Use the Power of Storytelling to Invoke an
Emotional Response
Human beings love stories. It is why we read
fiction novels and watch movies. It is why we
cannot help but be engrossed by the
dilemmas going on in other people’s lives. It
will also help many audiences, especially
those composed of amicable and expressive
audience members, feel more connected with
you and, by extension, your data story.
Relating your data to stories can give the
audience a better mental visualization of the
information you are trying to get across.
Allow Your Data Visualizations to Help
Explain Technical Information
Of course, your verbal communication needs
to be top-notch to reach your audience
effectively. Any written paraphernalia you
hand to your audience must also be relayed as
professional, easy-to-understand content.
However, you need to keep in mind that the
human brain more easily deciphers visual
content. Therefore, you need to take full
advantage of the potential that your data
visualizations afford you. This extends across
the board and allows you to break down
technical language into concise
communication.
Focus on the Information That Your
Audience Relates to
All the parts of the data involved in your
presentation might be fascinating to you.
However, you have to remember that your
audience is here for a particular reason, and
things that you find fascinating might fail to
capture the attention of your audience when
that information ranges out of that scope.
Therefore, it is best that you highlight the
things that your audience finds informative
and relates to. For example, you might be
presenting to an audience of marketing
specialists who want to understand more
about media buying. Such a presentation
should focus on the process rather than its
history or old tactics media buyers previously
used.

H OW TO C APTI VATE YOUR

AUDI ENC E
No matter how well of a package you have
put together to develop your data story, if
your audience does not get on board with
your vision by finding that data helpful or
fascinating, your call-to-action will be left
unanswered. Luckily, there are a few
techniques that you can use to up your
chances of making that vital connection to
keep your audience informed and decisive
about the following steps to take.
The rest of this part includes nine key
strategies for engaging your audience so that
your data story is a fruitful one.
Focus on Connection Rather Than Making
an Impression
Of course, you want to be seen as
knowledgeable and authoritative as you
deliver your data story to your audience. You
want to make a great first and lasting
impression. You want to wow these people.
You want to be memorable. There is nothing
wrong with such wants. However, they
should not be your first priority.
Focusing on making a good impression with
your audience makes your data story about
you, when first and foremost, it should be
about your audience and what they need to
gain out of the presentation. Your audience
needs to be changed in some way that is
valuable to them by the time you say the last
word of your data story. That change may be
that they are now informed in a way that they
were not before. It may be that they now
understand the process of making a more
informed decision in the future. It may be that
they now understand data they did not
previously. Any of these positive changes are
aided by your hand, and so, your audience
will develop a positive connection with you
and your data story.
That should be your focus - building a
positive connection rather than making a
good impression. The best thing is that by
prioritizing that connection, you up your
chances of making that good impression.
Have a Strategic Plan
With all the bytes of data that you need to
wade through to develop clear key points to
deliver to your audience, sharing all that you
know can seem more straightforward. That
urge will be especially strong if your focus is
to impress your audience with your
knowledge. Your audience will not be
impressed by data dumping. Instead, they are
more inclined to feel overwhelmed and
confused.
Avoid this by developing a strategy for
turning all that data into key insights that
develop the message you need to give to your
audience. Think of the one thing that you
want your audience to take away from your
data story and develop a story and visuals to
support that. Be as clear and as concise as
possible as you do this. Everything that goes
into your presentation needs to add to this
core message. If it does not, remove it from
your data story.
Bring Life to Your Data Story With Your
Excitement
The first person who needs to feel enthused
about your data story is you. If you are bored
while thinking about it or while preparing the
story, then the chances are that boredom will
extend to your audience. You need to get
fired up about your presentation to transfer
that energy to your audience.
This excitement should not be faked,
however. Your audience will feel that energy
right away, and the effect will be the same.
They will not be excited about your
presentation any more than you are. Instead,
add life to your story by infusing some of
your personality through the use of your
natural body language, facial expressions,
changes in intonation and pace in your voice,
and eye contact. Just remember to not go
overboard with this. The personality type of
your audience will dictate just how much of
your personality you add to your data story.
For example, a dry joke here and there will
suffice with an analytical audience, while
full-blown jokes may get to the heart of an
expressive audience.
No matter the audience type, though, if you
are rigidly delivering your presentation in a
monotone voice and standing in the exact
same position the entire time, you will lack
the upbeat energy necessary for transferring
excitement to your audience.
Use Stories to Make Your Data Story
Unique
One sure way to add excitement to your data
stories is to use your storytelling skills. The
use of stories makes numbers and figures
relatable. There is a time and place for hitting
people with hard facts and figures. It is even
appropriate at times during your data story.
However, continually hitting your audience
between the eyes with figures and hard facts
will leave them lost. Soften the blow with
stories. Not only do stories soften the blow,
but they also make the data relatable,
memorable, and more digestible to the
audience.
Stories can come from anywhere that is
appropriate to your data. They can be your
personal experiences. They can be real-life
examples that marry well with the data. Even
jokes can be used to tell stories. Just
remember to make these stories relevant to
the data. An effective approach is to craft a
story related to the data and convey it
throughout the presentation to give context.
People remember stories a lot better than they
do numbers.
Use the Sandwich Approach to Highlight
Key Insights
There will be a few key insights that support
the main point of your data story. These
insights are then supported by data points.
Using such a structure allows you to know
what is relevant to present to your audience.
A strategy called the Sandwich Approach
promotes the stating of insight followed by
delivering relevant data to support the insight.
The insight is then repeated to reinforce that
information. Think of the insight as the pieces
of bread on the top and bottom of the
sandwich and the data as the filling to
complete the sandwich. The intended effect is
to show the audience how the data is relevant
to the keep points that they will be taking
away from the data story.
Avoid Vague Generalizations
You need to be specific and concise when
delivering figures and hard facts about your
data. As mentioned earlier, there is a time and
place to soften the impact of numbers and
facts, but when you do indeed deliver them,
do not be wishy-washy about the act. Deliver
them so that impact is immediate and
deliberate.
You might be worried about overwhelming
your audience when you make these
deliveries, but the way to avoid this is to
ensure that you are not delivering fluff. Any
information that you make part of your data
story needs to earn its place there. It needs to
contribute to the insights and ultimately to the
core message of the presentation. Trim the
data down to the minimum and deliver it
comprehensively.
Spreadsheets… Don’t Use Them in Your
Data Story
Spreadsheets are beautiful things that help us
compile, calculate, and track data. They are a
necessary tool for making sense of the many
many bytes of data created every day. As
useful as spreadsheets are to you as a data
analyst, they are not something you should
deliver to your audience. Think of it as being
the director of a movie. You do not show
your audience the script or any of the other
raw products used to make that movie.
Rather, you present the final product, which
will educate or entertain. The same analogy
needs to apply to data storytelling.
Your job as the data storyteller is to use that
information to design a final product that
allows the audience to easily comprehend the
relevant data which has already been
compiled, calculated, and tracked.
Use Trends and Patterns
Imagine having thousands of small balls of
all different colors sent at you simultaneously
and being told to catch only blue ones in less
than 1 minute. An overwhelming and
impossible task, correct? This is essentially
the situation that some data storytellers
present their audience with.
Being presented with many small pieces to a
whole and then being asked to put the pieces
together is a formidable task, especially when
they are millions of pieces. Your audience is
coming to you to avoid this overwhelm, and
the best way to facilitate that path of easy
understanding is by compiling relevant pieces
of data into trends and patterns. In essence,
you will be putting all the blue balls together,
all the green balls together, and so on.
Trends and patterns allow the audience to
grasp how the data is developing over time
and what predictions can be made for the
future from the present and past data. This
facilitates more informed decision-making.
Leave Your Audience With Practical
Advice and Takeaways
As I said earlier, the mark of a good data
storyteller is leaving the audience with a
positive change by the end of a presentation.
Do not leave this up to chance, though. Make
it easy for your audience to know what you
want them to take away from the data story
so that they can start acting immediately to
make good on your call to action. Give them
a detailed outline of how they can use the
information they gather from you to pave a
better way forward.
H OW TO KNOW I F YOU H AVE GOT
I T R I GH T WI TH YOUR AUDI ENC E
Unfortunately, there is no way of knowing
whether or not you have hit the nail on the
head with your audience until you are in the
thick of things while presenting. This is why
it is so important that you do the preliminary
work necessary for gathering information
about this audience and how best to approach
them with your data.
However, while presenting, there are signs
that you can look for to see if you have gotten
it right with your audience. One of the main
ways is to note if your audience is actively
engaged in your data story. Do they look
interested, or are they instead fighting yawns
and reaching for their phones? Are they
asking questions when the floor is open, or do
they remain silent during that time? Are the
questions asked on-target, or do they veer
wildly off course from the topic? Do they
look eager to know more, or is their body
language closed off with arms folded and
confused looks?
Do not despair if you do not quite get it right
with your audience. There is no such thing as
a failed data story. What there is are data
storytellers who fail to learn lessons when
things do not go quite their way. Analyze
how every presentation goes. Gather data
about it. As much as I would love to tell you
all there is to know about getting it right with
your audience, a lot of it boils down to
experience. You have to act and note the
results from each experience. The more
knowledge you gather under your belt, the
more you will be able to fine-tune your skills
to connect with your audience the next time
and the time after that.
 3

REFINING YOUR VISUALS -

CHOOSING THE RIGHT
CHART

“Data visualization is the language of decision

making. Good charts effectively convey
information. Great charts enable, inform, and
improve decision making.”

— DANTE VITAGLIANO

o, you have done it. You have taken the

S time to develop your narrative by better
understanding your goal and the solutions
you could provide to the underlying problem
that prompted the need for that data
presentation.
However, no matter how grand your narrative
is, there needs to be something that attracts
your audience to the data that you have to
present. Think of it like this, many people say
that we should not judge a book by its cover.
But guess what? Most of us do, as we rightly
should because first impressions are often
correct.
The charts that you choose represent what
your audience’s first impression will be.
Those visuals will draw your audience in for
a further examination of what this represents
or what will turn them away. Your data
visuals need to correlate with the type of data
that you are presenting to that audience. It is
what will make that data understandable and,
therefore, engaging to your audience.
Because data visuals are such a massive part
of making the most impact on your audience,
it only fits that those visuals have the
spotlight in this chapter. Therefore, the
coming pages will focus on the importance of
balanced visuals that genuinely represent the
narrative of your data story and why it is vital
to choose the correct chart, and how to do so.

TH E I MPOR TANC E OF VI SUAL

R EPR ESENTATI ON
The sense of sight. It is one that we often take
for granted, even though it is what we most
heavily rely on—approximately 90% of the
information that our brains process daily is
provided via sight. Most of how we interpret
and interact with the world around us results
from what we see.
The process that happens from the time we
see something to the brain processing what
this visual means is called visual perception.
It describes the process of the brain analyzing
and then interpreting the information it gains
from our sight. This process happens so
quickly - in an average of only 13
milliseconds - that it is easy for us to discredit
the importance that it plays in our daily lives.
Of course, visual perception is a vital process
that has allowed the continuation of the
human species. Still, it is also essential that
you understand how this process can enhance
your data story. The human brain processes
visual information far faster than textual
images - 60,000 times faster, in fact. In this
day and age where we are constantly being
bombarded with images on billboards and
screens, our brains are seeking out these
images even if we are not aware of them most
of the time. At least half the time, your brain
is actively processing visual information.
I am not just providing these statistics to
enhance your mental muscles. I am providing
these small tidbits of data to show you that
the visual aspect of your data story can be
processed far faster than any bit of text that
you think to provide to your audience. The
human brain is better acclimated to seek out
and process visuals, so the many bytes of data
can be conveyed to your audience better in
this way.
By better conveying information to your
audience, you increase the chances of moving
them to act on the solution you have provided
via your data story. Here is another
astonishing statistic for you - a visitor to a
website usually takes less than 10 seconds to
decide whether or not they will continue to
peruse that website. Why? Because of what
they see on the web pages that they land on.
Apply that same sentiment to your audience
viewing the visuals that you present with
your data story. Within seconds, your
audience will have a visual perception of the
charts and other data visualizations you have
provided and decide whether or not they are
interested in what you have to say.
Often pictures speak louder than words.
When giving your presentation, you need a
powerful visual representation that supports
the strong narrative you should have
developed beforehand. If you take the time
and use the resources available to you to
create appropriate and visually impactful
charts, here out of benefits that you can
expect to receive:
Allows for Compacting More Information
in Less Space
Here is a comparison for you. It takes an
average of almost 2 minutes to read one page
of a book. Your brain can visually perceive
the same information in that text format in
mere milliseconds if presented as a chart.
Good data visuals allow you to feed your
audience more information in less time to feel
that they gain more value per second through
that presentation.
Higher Engagement Rates
As time passes, with so many means of
distractions like social media, the average
human adult’s attention span has decreased.
In the year 2000, the average adult attention
span was 12 seconds. By 2015, that figure
had decreased to 8.25 seconds. This means
that human adults have a shorter attention
span than goldfish, which have an attention
span of 9 seconds. Once you realize that the
average human being picks up their phone
more than 1500 times a week and spends
more than 3 hours doing that task, you
understand how human attention is such a
fleeting thing.
Creating good data visuals is the one aspect
of a data story that you can use to ensure that
you not only capture your human audience's
attention but also hold onto it for more than 8
seconds. However, you do not only want to
capture and hold your audience's attention.
You want to also make them feel invested in
that data story. This will increase the chances
of the audience engaging you for more
context to the information being shared.
Good data visuals have the power to gain you
that objective.
A Higher Rate of the Audience Performing
the Call-To-Action Delivered at the End of
Your Data Presentation
Let's be honest here. Data storytellers may
enjoy creating data stories, but they are not
creating these stories just for the heck of it.
There is a point to it all, and that point is to
get your audience to ask questions that align
with the solution you are providing to a
particular problem.
With the increased engagement rate that good
visuals provide, there is a higher probability
that the audience will act in a way that aligns
with the call-to-action that you will provide
in the climax to your data story. The fact is
that the more visually stimulated we are by
something, the more emotionally attached we
will become to that thing. From that comes
the higher likelihood we are to act on these
emotions. Therefore, translating your data
into appropriate visuals allows your audience
to respond quicker to the action steps that you
provide in your data story.
Allows for a More Everlasting Effect
It will do you no good if your story is
forgotten the minute your audience steps
outside the room. You need to ensure that
your presentation was impactful enough to
stick in their memory. Your data story needs
to have a lasting effect that encourages your
audience to follow through with the call-to-
action and perhaps convince others to
participate in fulfilling their call-to-action.
Attractive and informative data visualizations
give you the power to stick in your audience's
memory and thus, increase engagement.
C H OOSI NG TH E R I GH T C H AR T
So we have established the value of having
attractive, engaging charts in your data story.
However, the question remains - how do you
choose the correct chart to align with your
narrative and allow the natural progression
from the problem to be solved? The anxiety
that this problem can induce is only
compounded when this data story requires
multiple charts. How do you keep your
message from being lost in the noise of using
the wrong charts? How do you use data
visuals to enhance the content you are
presenting rather than take away from its
value?
Luckily, you can avoid the anxiety of these
questions by following the advice provided in
this section.
Choosing the correct chart starts with
examining the narrative you have developed
and then asking yourself what type of data is
being represented. You will most likely be
dealing with data falling in one of these four
categories:

Comparison
Composition
Relationship
Distribution

Each of these types of data is best showcased

by certain types of charts. Therefore, we will
break down what each data type means and
the charts typically best suited to make that
representation.
Comparison
This data type shows how one set of data
compares to at least one other group of data.
With this type of data, there may be multiple
variables from different sets of data or
various categories within one data set. For
example, if the data you are presenting
focuses on salary comparisons, diverse
datasets may show salaries within different
science communities. On the other hand, a
college may show salary variables in the
various departments.
Data comparison is often used in data stories
because it is simple in concept and
application yet allows powerful results. Data
comparison provides for:

Tracking how data changes over time

Showing the differences and
similarities between different sets of
data
Showing the differences between past
and current data
Showing the results before and after
solutions and applications have been
applied

When making comparisons about particular

items in relation to different sets of data,
some of the best charts to use include:

Column charts
Bar charts
Tables

When comparison data shows the movement

over time, some of the best charts to use
include:

Line charts
Column charts
Circular area charts

Composition
This type of data allows noting how part of a
data set can compare to the whole data set.
Data stories that show composition can be
static or show change over time. Also,
composition data may be expressed in
absolute numbers or in relative forms such as
percentages to show the variations of parts of
the whole. Just like comparison data,
composition data is a widely used type of
data.
Examples of visualizations that can be used
to show static composition include:

Pie charts
Waterfall charts
Column charts
In the case of composition data that changes
over time, commonly used charts include

Column charts
Area charts

Relationship
This type of data shows the connection
between at least two variables in a given set
of data. An example of a relationship with
only two variables may be children's height
relative to their age. Another data set
exploring relationships with multiple
variables includes website conversions from
specific demographics such as age, gender,
etc.
If someone just dumped a bunch of numbers
on you, there is no way of finding the
correlation between these numbers until they
are grouped to show relationships. Good
visualizations help determine these
relationships.
Scatter plots and bubble charts are typically
used to show relationships in data
presentations. Scatter plots are more
commonly used when there are only two
variables, while bubble charts are more
commonly used when there are more than
two variables.
Distribution
Useful in developing trends, this type of data
shows how variables in a set of data or
multiple steps of data are distributed over
time. With trends, probabilities can be
developed to offer predictions of possible
outcomes based on historical information. For
example, data from a swim club may show
that swimmers in different height categories
swim at different paces. This data can be used
to make future predictions as to who might be
the top competitors based on different swim
categories. Column charts, line histograms,
scatter plots, and 3D area charts commonly
show distribution data.

W HILE THESE FOUR types of data are

considered the pillars of data visualization,
there are more techniques that you can use to
determine what is the right visuals
appropriate for your particular data story.
This, of course, involves asking yourself a
few questions. One of the commonalities that
you might have noticed mentioned in the
types of data outlined above is the number of
variables in datasets. Therefore, one of the
first questions you need to ask yourself is
how many datasets are represented and how
many variables are outlined in each data set.
Once that has been determined, you also need
to question how many data points will be
displayed for each variable chosen to be
outlined in that data story, as well as you
need to determine whether or not these data
points will be plotted over a period of time or
another variable that shows progression or
grouping.
Once you have adequately established the
data type and answered the questions above,
you can determine the best visualizations for
that data story. But as a rule of thumb, here
are a few rules that you can help guide the
way for developing your data visualizations:

1. Pie charts should be used to show

simple composition data types and, in
fewer cases, relationship data types.
They should never be used to show
distribution or comparison data types.
They allow for viewing parts of a
whole of a simple set of data.
2. Use line charts to show the
development of trends.
3. Bar charts are a good data visualization
for showing comparison data types and
reflecting change over time.
4. When a large data set needs to be
visualized, scatter plots are helpful.

TYPES OF C H AR TS AND WH EN AND

WH Y TO USE TH EM
There are tons and tons of different types of
charts that can enhance the narrative of data
stories. There is no way that we can delve
into all of them, but we can look at those that
are more widely used. You must understand
the basics of charts and how and when you
can use them before you dive into the use of
more complex visuals.
While there was nothing wrong with leaning
on more complex charts when it is warranted
(remember that balance between problem
complexity and visual complexity), there is
no disputing the fact that clean, simply-put-
together charts can most often get your point
across more efficiently and effectively when
supported by a good narrative compared to
more complex visuals. Again, this is
subjective to the data story that you are
presenting. Still, you need to know what the
foundational charts are, and how, why, and
when to use them before you go onto the use
of complex graphs.
Some of the foundational charts that every
good data storytelling need to know how to
use include:

BAR GR APH S
This type of chart has a lot of aliases. It also
goes by the name of a column chart or bar
graph. It is so named because it allows data
visualizations where numeric values are
featured in the form of bars. The levels of
these bars are plotted on one axis while the
values are plotted on the other axis. Each
category of data is highlighted on one axis,
and the length of that bar corresponds to the
value on the other axis. Bar charts can make
use of either vertical or horizontal bars. The
categories are placed on the horizontal axis
when vertical bars are used, and the opposite
is true when horizontal bars are used. Vertical
bars are the norm, but horizontal bars are
good practice when working with long
category labels. Whether you use vertical
bars or horizontal bars, the thing you need to
ensure is that you accurately label each axis.
FIGURE 1 Bar graph
 FIGURE 1.1 Horizontal bar graph

Another common type of bar chart is called

the stacked bar chart. The name comes from
the fact that individual bars are divided into
sub-bars stacked on top of each other to show
the correlation between different categories.
For example, a marketing team may use a bar
chart to compare their marketing budget from
2018-2020. The team will further break down
each year in a stacked bar chart to show the
budget allocation. The height of the bar will
establish the total budget, and the bar will be
divided into different sections showcasing the
portion of the budget for that year.

FIGURE 1.2 Stacked bar graph

Bar charts allow for the easy comparison of

data variables, so bar charts are commonly
used when comparing data types featured in a
data story. However, bar charts are not just
limited to show comparison data. They are
also used to indicate the distribution of data
points. So many data stories that showcase
groups of highest or lowest, or most common
to least common variables showcase bar
charts.
To have the most impact when using bar
charts in your data presentation, there are a
few rules that you should stick to. These rules
include:
Use Zero (0) As Your Baseline Value
Ensure that both the axes on your bar chart
start with a zero value baseline. There are two
good reasons for engaging in this practice if
you use bar charts in your data presentation.
First, it allows your audience to compare the
length of the bars in your chart easily.
Second, the practice maintains the integrity
and truthfulness of your presentation. A
diversion from this practice can easily
misrepresent the data, even if that is not your
intent.
To be clear, this is a rule that is generally
followed for most types of charts.
Order Categories According to Bar Levels
Bar charts are commonly used to show
comparisons and trends, and the standard
conventionally is to place bar categories from
longest to shortest. This allows the audience
to interpret comparisons easily and to realize
trends.
This rule is not hard set. If categories are
inherently ordered in a specific way to serve a
particular purpose, then that takes precedence
over the longest to shortest ordering
convention.
Use Rectangular Shapes
You might be tempted to get fancy when you
are doing your bar charts but resist this
temptation. Ensure that the shapes of your bar
fit a rectangular form with straight edges.
You might see rounded bar shapes used in
some bar charts, but these types of bar charts
can easily be misinterpreted as the audience
will find it difficult to tell where the bar
indicates the value on the axis.
Also, avoid using 3D bars, even if you might
see these being more commonly used. Again,
they make it difficult for the audience to
interpret the bar’s actual value and add
unnecessary visual noise.
Use Color in Moderation
You might be tempted to go all out with the
use of color when developing bar charts, but
again, this is a temptation that you must
resist. Color must be used with purpose when
being integrated into your data visualization.
One of the best uses of color is highlighting
datasets with commonality or showing
different sets of data.
Do Not Replace Your Bars in Your Charts
With Other Items
Some data storytellers want to differentiate
themselves from others by replacing the bars
in bar charts with other images, but this is not
the way to go if you’re going to retain the
truthfulness and integrity of your data
presentation. Bars charts are so commonly
used because they allow for the
straightforward interpretation of data.
Deviating from the use takes away that ease
and makes it easy for your audience to
misinterpret the message you are trying to get
across.
If you feel that images best represent your
data visualization, then a bar chart might not
be the best data visualization option. In such
a case, the use of a pictogram chart may be
the better option for you.

PI E C H AR TS
Pie charts are so commonly used in data
visualization that some people might say that
they are overused. We will delve into when it
is appropriate to use pie charts and times
when it is inappropriate. Also, we will look at
how to determine what is proper and
improper related to pie chart usage. However,
before we get to that, let's look at pie charts
and their value to data visualization.
Circular in nature, pie charts are data
visualization tools that use slice sizes to
depict parts of a whole or highlight the
relationship between multiple datasets.
Investors share percentage, for example.
The primary use of the pie charts is to
compare the groups contained within one set
of data. Suppose we reuse our investor share
example from above. In that case, we can
either develop one pie chart showcasing what
percentage each investor owns. Or, we can
create three separate pie charts to show each
investors shares in relation to the whole. In
this case, creating three different pie charts
doesn’t make all that much sense, and those
two types of data cannot be compiled into the
same pie chart as these data points would
confuse your audience. Compilation is best
done using bar charts.
FIGURE 1.3 Non effective chart - Not the
best option for this set of values as it is
doesn’t visually represent the data
effectively or draw an easy conclusion.
 FIGURE 1.4 The more effective choice-
A better option for this set of values as
you can immediately distinguish who
owns what % of shares in relation to each
other.

Just as there are rules for creating the best bar

charts possible when doing data
presentations, that also applies to the
development of pie charts. Some of these
rules include:
Use Annotations
It is often difficult to determine the exact
proportion of each pie slice by sight alone.
Do not burden your audience with the task of
trying to make these determinations but using
annotations. These annotations can take up
the form of fractions, percentages, or whole
numbers.
Use a Limited Number of Pie Slices
Can you imagine if you used a pie chart to
present the budget allocation for an entire
company? You would need a magnifying
glass and lots of time on your hands to go
over all that data. That is certainly not a task
you want to give to your audience. Therefore,
limiting the number of pie slices that make up
your pie chart is essential. The use of five
categories or less is the standard practice with
pie charts. Creating a pie chart with more
slices makes the visual look cluttered and
hard to decipher. Coming back to our earlier
example, if the values have a noticeable
difference then it might work in your favor to
use a pie chart. As you can easily distinguish
who owns how much of the whole at a
glance.
 FIGURE 1.5 When to use a pie chart - If
you have minimal values and can easily
distinguish the parts of the whole, a pie
chart may be an effective option.

Personally, if I need more than three or four

categories, I'll switch to a bar chart. Pie charts
are best used for a simple part of the whole
analysis, nothing more. Even very few slices
can aid confusion if the figures are similar to
each other.
Order Your Pie Slice for Easy Reference
Like with bar charts, the order you choose to
deliver your data can enhance the experience
for your audience or detract from it. The
standard practice is to order the slices from
biggest to smallest. Also, just like with bar
charts, that regular convention takes a
backseat to inherent orders that allow for a
better viewing experience.
Use Flat Shapes to Represent Pie Charts
The use of 3D shapes is becoming more and
more popular in data visualization with the
popularity of 3d modeling, but they leave too
much room for misinterpretation. Avoid
using anything apart from flat shapes that
best show the proportion of data represented
by a pie slice.

LI NE GR APH S
Also called a line chart or a line plot, a line
graph is just what it sounds like. It is a type of
data visualization that shows continuous
progression using lines from left to right to
show changes in value. This constant
progress is shown on the chart’s horizontal
axis, while the vertical axis shows the value
metrics that highlight that change. For
example, a marketing agency might use a line
chart to show how the website traffic from
their top advertising campaigns for one of
their clients has been distributed over the first
3 quarters going into Q4. The line will show
the progression from January to September
on the horizontal axis, while the vertical axis
will show the metric value, which is the
number of website visitors and how it differs
throughout the first 3 quarter’s. With this
information they can see how their campaign
performance is and note some possible
projections going into Q4.
FIGURE 1.6 Comparison - An effective use of a line
graph comparing the performance between campaign
A and campaign B over the first three quarters.

Because of this structure, line charts are great

for showing trends and distribution. To make
the most out of the use of line charts, here are
a few rules that you can stick to when
creating them:
Choose Appropriate Measurement
Intervals for Both the Horizontal and
Vertical Axis
Also called a bin size, the proper interval
between measurements plotted on both axes
of a line chart is important for a quick and
accurate interpretation of data. There is no
strict science on how to choose an
appropriate interval. Instead, this relies on
your knowledge of the data and how best to
translate it to the audience.
For example, if we go back to showing the
website traffic for May, a daily interval on
the horizontal axis is likely appropriate. On
the other hand, this would be inappropriate
for visitors over a year. Instead, a monthly
interval would be more appropriate because it
will be less tedious for your audience to read
- 12 intervals rather than 365.
Limit the Use of Lines
More than one line can be used to show the
progression on a line chart. For example, a
marketing agency might have individual lines
to show website visits, add to carts, or
purchases over a year to track the conversions
rates.
However, while multiple lines are great at
highlighting certain pieces of data, too many
lines can confuse your audience and lead to
misinterpretation. As a rule of thumb, limit
the number of lines used to 5 or less.
Limit the Use of Dual Axises
There are times when you will come across
line charts with dual horizontal axes. For
example, they may be used to show negative
and positive values with the line progression.
A company can use this to show an audience
the periods when it makes a profit compared
to when a loss was made over each month of
one year.
While a dual-axis can enhance the
understanding of your audience, this is
typical in a point of confusion. Therefore,
where it is possible to communicate the data
without a dual-axis, do so even if it means
using another chart type.

AR EA C H AR TS
A slightly more complicated chart is the area
chat. What makes the area chart special is
that it combines a bar chart and a line chart to
show the progression of a variable compared
to another set of data. This progression is
usually demonstrated over intervals of time.
The difference between a line chart and an
area chart is the shading notable between the
lines and the horizontal axis.
Area charts are typically used to show
comparisons between multiple variables or
how one set of data is divided into different
proportions. Because of this, there are two
main types of area charts.
The first one is called an overlapping area
chart. This type of area chart shows the
comparison between variables and different
sets of data. This type of chart offers the
standard line, but each point plotted on the
vertical axis indicates the value for every
variable in the different datasets. Each plot
point has shading between the line and the
horizontal axis. Of course, this shading can
add a little panache to an area chart, but it
shows the greatest value in each variable and
differentiates each variable from the others.
As a result, such a chart will typically be
distinct as it has figures that look like
mountain peaks. An example of an
overlapping area chart could be monitoring
website traffic during a product launch
throughout the day from different sources.

FIGURE 1.7 Comparing landing page

traffic - Comparing website traffic from
separate sources during a product launch.
Can easily distinguish the better
performing funnel to focus advertising on
 that specific source in the future.

While overlapping area charts are great for

showing how different datasets correlate and
are differentiated from each other, you should
limit the number of data groups placed in one
such chart. As a rule of thumb, limit the
variables to three or less.
The second type of area chart is a stacked
area chart, and it is used to show how
individual categories of one set of data
progress. Such a chart helps track a total
value and break that one set of data down into
separate categories. Such charts make use of
multiple lines, and just like a stacked bar
chart makes use of different colors to show
subcategories, so does a stacked area chart.
 FIGURE 1.8 Stacked area chart -
comparing trends over time while easily
understanding each value amount.

An example of a stacked area chart would be

an airport noting the number of persons
entering the country and where each visitor is
arriving from for the year.

SC ATTER PLOTS
Mainly used to show the relationship between
at least two variables, scatter plots use dots to
represent values based on these variables and
how they correlate in relation to one another.
These points allow for reporting the
relationship between two variables and show
patterns in the distribution of that data.
The relationship between the plotted points
can show various positive, linear, and strong
patterns. Such a pattern shows the
distribution of these dots in a line that has an
upward trend. On the other hand, the
distribution of dots may indicate no clear
relationship between the two variables as
these dots are plotted all over the chart. Of
course, other relationships can be shown
between these two extremes, such as one that
is non-linear but still strong. To make the
general trends that are developed by the plot
points in this type of chart easy to spot, it is
helpful to draw lines based on the distribution
of those points. This line is known as a trend
line.
Scatter plots are not just limited to the use of
only two variables. A third variable can be
added, but more variables are not
recommended as this will lead to creating a
cluttered chart that is hard to understand. The
plotting of too many variables is called
overplotting, and it is so-called because
having too many variables and dots makes it
challenging to understand the relationship
between them.
As great as scatter plots are, there is one
possible problem that you may run into when
using them. While you can note the
relationship between two variables, you
cannot determine what causes this
relationship based on the points plotted on a
scatter plot. The counter to this limitation
with this type of chart is that it invites further
investigation.
Here are some examples of possible scatter
plot outcomes:
FIGURE 1.9 No relationship known.
FIGURE 1.10 Moderate, negative, linear
relationship.
FIGURE 1.11 Strong, non linear
relationship.
 FIGURE 1.12 Strong, positive, linear
relationship.

USI NG TABLES
At this point, I am sure that you might be
wondering one thing - so, what about tables?
Don't worry. I have not forgotten about them.
Tables are often the source of the data you
will develop into visualizations throughout
your career developing data stories. You
might even find that they make good data
visualizations when showcasing comparison,
composition, and relationship data types.
Especially when there are few variables or
data points to be outlined to your audience.
Other times when it would be appropriate to
use tables in your data presentation include:

When the data cannot be easily

represented in a visual format.
When you need to showcase precise
values to your audience or bring the
audience’s attention to unique datasets.
When the data that needs to be
communicated does not involve trends
but is instead of a quantitative
informative nature
When the data involved has multiple
units of measurement
When making comparisons
When you need to showcase individual
values to your audience.
When highlighting causes of the
patterns shown in the data presented.
When specific parameters need to be
shown to highlight particular datasets.
Apart from that, charts are the best data
visualizations to impress upon your audience
the message contained within the data or
when you want to highlight the relationship
between datasets or groups within a dataset.
Of course, you are not limited to just the use
of tables or just the use of charts. As long as
the narrative of your story remains concise,
use whatever visual properties you think will
enhance your message.
How to Make Charts
Now that you have a better understanding of
choosing the right charts to enhance your data
story, you might run into how to create these
charts. Although an in-depth guide will have
to be saved for another book, I will point you
in the right direction if you don't already have
your tool of choice. Here are my top three
options:

Tableau

Tableau is a data visualization tool that can

visualize data and get a clear opinion based
on the data analysis. It is the top dog for
visualization tools and can turn data into a
helpful visualization very efficiently. It is
proficient in handling large and changing
datasets due to its integration with some
advanced database solutions, including
Amazon AWS, My SQL, Hadoop, Teradata,
and SAP. It also has a very high level of
security.

Qlikview

Qlikview is one of Tableau's top competitors.

It is highly customizable and has a wide
range of features. It can be a bit more of a
learning curve to get a feel for it. It also
offers powerful business analytics and
intelligence reporting capabilities. Qlikview
can be used alongside Qlik Sense which
handles data exploration and discovery.

FusionCharts

FusionCharts is a javascript-based charting

tool. It can prudence about 90 different chart
types, integrate with many platforms, and has
a lot of flexibility. What makes FusionCharts
so attractive is that you can choose from
many live templates and simply plug in your
data sources.

H ONORABLE MENTION : Excel

Although Excel is a spreadsheet, it is a very
flexible tool and has a number of data
visualization options that are easy to utilize if
a quick chart is needed.

I T DOES NOT MATTER what tool you are using.

What matters is that you follow the specific
principles that allow your charts to be most
effective when combined with the narrative
of your data story. Therefore, I encourage
you to explore and experiment with tools for
chart creation to find the one that works best
for you.
Now that we have gotten the basics of the
visuals down, let’s jump right into the next
chapter, where we will define how you can
refine your visuals for an even harder impact
when presenting to your audience.
 4

REFINING YOUR VISUALS -

DEVELOPING A WINNING
DESIGN

“You can achieve simplicity in the design of

effective charts, graphs, and tables by remembering
three fundamental principles: restrain, reduce,
emphasize.”

— GARR REYNOLDS

aving the right visualizations that

H enhance your narrative is a step in the
right direction in hooking your audience
before you even say a word when you make a
data presentation. However, if these data
visualizations aren’t designed to appeal to the
aesthetic this audience resonates with, all that
effort beforehand would have been for
nothing. Let’s ensure that you are not wasting
your time or your energy.
Always, and, I cannot stress this enough,
always keep in mind that the data you are
presenting is only as valuable as it is
understandable to your audience. Deciding on
the type of charts helpful in conveying your
message is indeed an important one. Still, the
work needs to be followed up, making it
attractive and easy to digest mentally. With
just one glance, you want to capture the
attention of this audience. Remember how
short the human attention span is currently.
You need to make sure that this audience
wants to learn more. Your aesthetic plays a
vital role in developing that want. Therefore,
while your chart might best represent the
data, the audience can be pulled away from
the message if the fonts interrupt the reading
experiences, if the colors are too loud, or if
the headings do not correspond with the
delivered data.
With so many options when it comes to
design, you can get caught up making the
visual data pretty. But you must remember
that the primary function of these visuals is to
inform. The look must serve as a complement
to that purpose. Even a visually unappealing
chart that informs is better than a chart whose
message gets lost in the quest to make the
chart visually attractive.
Still, you want to hit your audience with a
double whammy with charts that are both
attractive and informative. You have taken
care of the informative part with the
development of the narrative and choosing
the right time of charts. Next comes making
decisions that make that data appealing to the
eyes of the audience.
This part of this book focuses on how you
can create clean and simple data
visualizations that allow you to capture your
audience's attention immediately. Only after
you have hooked their attention can you
deliver the key points precisely and with a
resounding impact that increases the
likelihood that your audience will be moved
to act on your call-to-action. The winning
design is free of clutter, has the proper use of
color, is clean and precise, and is founded on
design psychology. We will focus on each of
these critical points and more now. Let's jump
right in.

TH E PSYC H OLOGY OF DESI GN

We have talked about visual perception and
how it plays a part in how your audience
perceives the data visualizations you deliver
to them. Let’s take a deeper look into how the
brain relates what you see and how this plays
a part in the success of your data
presentation.
There is a science as to why we find certain
things visually appealing in comparison to
others. That science extends to why your
audience will be attracted to certain data
visualizations instead of others, even when
these charts display the same information.
Therefore, the first thing that needs to be
addressed when developing a winning
strategy for creating visuals that your
audience wants to look at is not color
schemes or font types. Instead, you need to
delve into how humans mentally process
images. Only then can you design and
manipulate these mental pathways in your
favor.
These mental pathways are known as artistic
psychology. More precisely, it goes by the
name of pre-attentive attributes. This is the
process by which information catches a
person’s attention based on the visual images
delivered to the brain so that the data
delivered can be processed.
Think of times that your eyes have been
trained in a particular direction, but you do
not remember a single time from that period
of looking. This is because nothing during
that time captures your attention visually.
Therefore, the brain did not allocate resources
to translate what that visual input might
mean.
On the other hand, there are times when so
much catches your attention visually that you
do not know where to look next. That is pre-
attentive attributes at work. That is the effect
you want to stimulate when your audience
looks at your data visuals.
With the knowledge of pre-attentive
attributes in your back pocket, you can design
visuals that catch your audiences’ attention.
There are four qualities about an image that
affect whether or not that image catches
attention. These qualities are:

Form
Color
Spatial position
Movement

Let’s take a look at what each of these

elements means.
Movement refers to the use of flicker and
motion elements related to images. A classic
example of this is used in traffic lights.
Banner ads also make frequent use of this
element. These can, of course, be used in data
visualizations; however, the use must be
carefully weighed as it can be more
distracting than aiding in capturing your
audiences’ attention. The audience may
become captivated by the flickering effect
rather than the information being relayed.
Spatial positioning speaks to the perception
gained about an object’s position relative to
another object or one’s self. It also refers to
how a person might perceive that this object
is turned relative to themselves. More
specifically, does this person believe that the
object is behind, in front of, left or right of,
above, or below their person? 2D and 3D
usage are examples of spatial positioning at
work in data visualizations. The use of spatial
positioning apart from 2D positioning can
add depth to your data visualizations. Still,
the use is not typically recommended
because, just like flicker and motions usage,
they can detract from the message rather than
enhance it.
We will discuss color more in-depth later in
this chapter. Let’s take a deeper look at the
element of form now.
Form refers to the structure of the elements
that make up your data presentation.
Therefore, this takes into account things like:

Length
Width
Orientation
Size
Curvature
Shape
Added marks
Enclosure
Hue
Grouping
Positioning
Intensity
FIGURE 2 Pre-attentive attributes

...and more in relation to your chart. For

example, the form would refer to the bars’
height, width, degree of curvature, and hue in
a bar chart. The wonderful thing about form
is that it can help attract attention to certain
parts of your data presentation while
detracting attention from other parts.
On the other hand, the improper use of form
can make an element distracting. For
example, if the tops of the bars in your bar
charts are curved instead of straight, the
audience may spend more time determining
the actual value being represented instead of
the connections you are trying to convey.
You want to make it easy for your audience
to extract information from your chart and
move on swiftly. This will significantly
increase your engagement and the incidence
of the audience members being moved to the
call-to-action.
The form also allows for creating uniformity
in your charts. This makes for a
straightforward interpretation of data as well
as differentiation of differing elements. For
example, each unit of a pie chart represents a
certain percentage or fraction of the whole
and allows for easier interpretation of the
proportions of data being presented.
Likewise, the height of each bar on a bar
chart allows for that data to be mentally
digested.
All of the qualities that contribute to the pre-
attentive mental process do not happen on a
conscious level. They are not things that you
have to think that you have to process for
them to happen. They just do, and it only
takes fractions of milliseconds for human
attention to be captured by such elements.
Let’s have a look at how we can use these to
our advantage. As you can see in this
horizontal bar chart there is valuable
information, but someone seeing this chart
for the first time won’t necessarily understand
the significance of it.
 FIGURE 2.1 Not very memorable -
Charted values with no known
significance.

Rearranging the bars and highlighting

specific information in green will better get
your point across.
 FIGURE 2.2 Use attributes to your
advantage - Highlighting best performing
sources in green to better understand the
data.

The same can be done if you want to

showcase the areas that didn’t perform as
well as expected.
 FIGURE 2.3 Using color to draw
attention to specific values.

Adequately incorporating these elements into

your data visualizations gets your audience's
subconscious mind on board with the
message you are trying to convey… once you
do it the right way. Once performed well,
though, incorporating these elements into the
design of your data story allows for seamless
interpretation of even the most complex data.
The 8 Laws of Artistic Experience
Just as there is a science to the data
visualization experience, there is also an
artistic approach to enhancing that audience
experience. These two facets of the visual
experience are interconnected. There are
reasons why people generally tend to
gravitate toward particular aesthetics instead
of others. Luckily, you do not have to be an
artist to develop your data visuals based on
that balance. You simply have to be aware of
the laws that govern how the brain interprets
visual data for the most effective processing.
These eight laws are:
The Peak Shift Principle
This law of artistic experience focuses on the
psychological reaction when human beings
focus on an exaggerated feature out of a
collection of items. You, as a data storyteller
using expertly designed visuals, can use this
law to focus your audience's attention on the
most important data being relayed.
For example, a marketing manager displaying
a bar chart to their marketing team can
highlight a particular month’s sales in red
while muting the other aspects of that bar
chart in grey tones. The exaggerated feature,
which is the red bar, will more than likely
capture the audience’s attention while
reducing an emotional response to the
irrelevant information highlighted in grey.
This is a simple example, but the idea can
easily be expanded upon to other visual
elements.

Isolation
This law focuses on removing objects that are
not necessary to impact information so that
the focus remains one of the relevant
elements. As a data storyteller, this means
that you keep your visuals as simple as
possible. All information that is irrelevant to
the presentation should be kept off your
visuals. For example, if the marketing
manager wants the marketing team to focus
on 2021 sales, there is no reason why they
should also display the sales for 2020 in their
visualizations.
Grouping
This artistic law focuses on the similar
grouping of objects. This allows the audience
to note the commonalities that these features
share and the minor differences.
The marketing manager may choose to group
the sales of the best-performing months so
that the marketing team can see the
similarities between them. This can then lead
to further information on what causes sales to
be high during these particular times and
what caused the disparity between these
months.
Contrast
Contrast allows the audience to better note
the communicated information due to
different colors placed next to each other.
Data tends to become misconstrued by the
brain when it all looks the same, even when
there are differences. Therefore, using this
artistic law allows your audience to better
note your information as there are clear
disparities between each set.
The marketing manager may choose to
display 12 months of sales information on a
bar chart by utilizing colors such as red and
green so the marketing team can get a more
truthful perception of each month’s
advertising performance. Green can be used
for the profitable campaigns and red for the
not so profitable campaigns.
Perceptual Problem Solving
This artistic experience is about getting the
mind to actively work to reach a conclusion
about the visual information that is being
displayed. Such an artistic experience relies
on the artist creating a display that allows the
audience to follow a path of logic to reach
that conclusion. In other words, it relies on
that “aha” moment after a moment of
thought.
For example, the marketing manager may
display two sets of information side by side.
These datasets are the sales information and
changes made to the main website landing
page at certain times of the year. They aim to
get their audience to follow a path of logic
that shows that the placement of their best
sellers on the website landing page converts
better than showcasing the cheapest products.
The Generic Viewpoint
This artistic rule ensures that an audience
should not have to take a unique perspective
to gain information from watching a display.
This means not allowing the visual display to
remain up to interpretation. You do not want
your audience to be confused. Make the
visual display as straightforward as possible
so that the audience's innate understanding of
that data is supported no matter how this
person looks at it.
Visual Metaphors
In literature, a metaphor is defined as one
thing that is a representation of something
else entirely. From an artistic standpoint, this
is about relating seemingly unrelated ideas.
Using Visual metaphors gets your audience’s
mental juices flowing and allows them to be
more engaged with your visualizations. The
marketing manager's display of seemingly
unrelated information such as the sales data
and placement of landing page items can be
seen as a visual metaphor.

Symmetry
This artistic experience relies on making
objects proportionately pleasing to the eye.
Think of it like watching a drawn skyline.
Suppose there is a building that seems
crooked in relation to the others. In that case,
this makes for a visually displeasing
experience to most people, even if they are
unaware of what aspect is causing this
displeasure.
The same thing will happen if your
visualizations have aspects that seem like
they do not originate from a central axis.

C OMMON MI STAKES MADE WH EN

C R EATI NG DATA VI SUALI Z ATI ONS
Often to get a good picture of how you
should perform a task correctly, you need to
know how not to perform that task. There are
a few data visualizations mistakes that are
common in the data visualization field. These
mistakes can be off-putting to the audience
and leave them with a bad taste in their
mouth about that presentation. It might even
leave them questioning the ethics of that
analyst.
However, typically the data analyst does not
make those mistakes out of malicious or
destructive intent. They are simply honest
mistakes born out of ignorance. You may find
that even you have been guilty of making
these same mistakes when utilizing data
visuals. If you are indeed guilty, note that this
is not the end of the world, especially now
that you are on a path to knowing better.
To set you on the right path, we will go over
some of the top mistakes made when
designing data visuals, how you can spot
them, and how you can correct them.
Misleading Color Contrast
Color can add that special panache that your
data visuals need to convey accurate and
adequate amounts of information to your
audience when used correctly. The improper
use of that color can lead to your audience
being confused and deceived about what you
are trying to convey with the information that
you are presenting.

FIGURE 2.4 Unnecessary color - The use

of color in this graph has no significance
and may have your audience drawing
conclusions that aren’t warranted.

Color is not just used for purely aesthetic

reasons, even though it can be pleasing to the
eye. It is a persuasive element that shows
degrees of contrast so that your audience
notes disparities and differences in the
information presented. If highlighting
differences in values isn’t necessary, avoid
adding unnecessary colors.

FIUGRE 2.5 Using color effectively -

Keeping colors uniform and ordering
graph appropriately will warrant a clear
understanding of the data.

Highly contrasting color elements can cause

the audience to believe that there is a greater
degree of disparity than there really is.
Therefore, it is best to use similar colors such
as different shades of the same color for
highlighting information with similarities and
use higher levels of contrast like two
completely different colors to show
differences in the data presented.
Improper Use of 3D Graphics
Many data analysts are trying to adopt 3D
data modeling but simply are not using it
correctly. You should only use a 3D chart
when a third dimension must be highlighted
in your data story. While these graphics add
visual interest, they do not always benefit the
presentation. While data 3D graphics can be
engaging, they can lead to confusion about
the scale of differences and similarities
between different datasets. They can also
possibly obstruct how the audience perceives
this data.
There is a simple way to avoid this potential
mishap occurring in your data presentation.
Simply do not use 3D graphics unless there is
a very good reason for doing so, which is the
addition of dimensions that merely cannot be
expressed effectively in 2D. The use of
traditional 2D graphics eliminates these
problems. The age-old saying of don’t fix it if
it ain't broke applies here.
Too Much Data
With so many bytes of data being produced
about a business every single day, it can be
challenging to determine what needs to be
included in your data presentations and
visualizations and what does not need to be
included. Being so close to your data, you
may feel that every single byte of information
is crucial to share to tell the entire story.
However, too much information will simply
overwhelm your audience. This overload of
information makes the members of your
audience zone out away from what you are
trying to convey. This leads to less
engagement and your call to action not being
acted on.
 FIGURE 2.6 Too much data means
clutter - Cluttering your graph with data
and unnecessary elements will lead to
confusion. Every element including the
data must add significance to the graph so
the viewer can easily understand what it is
conveying.

The solution is to keep things as simple as

possible. This means condensing the data into
a limited number of visualizations or using
multiple visualizations at different points in
your presentation to present the audience with
easier-to-digest information in smaller
quantities. Lets have a look at another
example:

FIGURE 2.7 Overwhelming data - having

too much data with no specific meaning
will be confusing to most audiences.

When dealing with a data set similar to this

one, try simplifying the data. Showcase the
top five sets of values and mark the rest as
“other”. If your audience is needing more
information about the “other” values you can
create a few extra charts showcasing each
specific point.
 FIGURE 2.8 Simplifying the data -
Simplify the data by showcasing the top
five values in order. Expand upon any
specific points or values if needed.

Omitting Baselines and Truncating Scale

To make data more digestible to the audience,
some analysts choose to manipulate scales on
charts. An example of this in action is to omit
the baseline or start somewhere above the
zero mark on the chart’s Y-axis. This is
typically done to make the differences in data
more noticeable. Another example of this in
action is replacing or shortening the X-axis
value in datasets to be more comparable to
lower values in that dataset.

FIGURE 2.9 Starting graph above zero -

Misleading your audience with an
exaggerated scale may have viewers
drawing conclusions that are inaccurate.

While these practices can indeed make it

easier for some audiences to digest the data,
they can also make it confusing and
misleading to most. Essentially, these
practices exaggerate or minimize the
differences in datasets, which is unethical
when presenting data to an audience. The
solution here? Simply do not incorporate
these practices in your data visualizations.
Make sure to keep your visuals accurate and
clear.

FIGURE 2.10 Proper scale - Compared to

figure 2.7, you can see the values are
closer to one another and do not warrant
any serious action that figure 2.7 might
have hinted to.

Choosing the Wrong Visualization Method

Choosing the right charts to support your
presentation is not a game of eeny, meeny,
miny, moe. There is an art and a science to
this because data visuals are not one size fits
all. For example, using bar charts can make
differences and similarities between different
datasets more apparent. In contrast, when
doing a simple parts of a whole analysis a pie
chart might be well suited as long as you can
easily distinguish the values.

FIGURE 2.11 Not the best choice for this

particular set of data. Your graph should
enhance the meaning of the data, not make
it more confusing.
 Figure 2.12 A better graph for this
particular set of data.

FIGURE 2.13 Using the proper chart

allows for better interpretation of the data
at first glance.

The type of data visual you choose can get

your audience on board with what you are
trying to present or make them more
confused than when they first started viewing
your data presentation.
Avoiding that problem by taking the time to
understand what your visuals must convey
and then choosing the right charts to correlate
with that ultimate vision.
Confusing Correlations
Using visuals to show correlations between
different datasets is a helpful way of giving
your audience a broader understanding of the
data being presented. One of the best ways to
show correlations is to overlay them in the
same chart. However, having too many
different datasets highlighted in one chart can
lead to confusion. Instead of showing
connections and inciting an “aha” moment, it
can lead to the opposite effect, which is the
“huh?” Moment. Too many data analysts try
to use correlations to show the cause of what
drives that data to be what it is. This will
always fail because correlation is not
synonymous with cause.
Instead of using what can be an unethical
practice, it is better to use multiple
visualizations to show how different datasets
can be connected rather than overlaying them
in a single chart. These multiple
visualizations can still allow for that “aha”
moment when the audience connects the
relationships between different datasets.
Biased Text Descriptions
The inclusion of text is also part of designing
your data visualizations. Just like everything
else in the design process, the text needs to be
considered carefully. All texts such as titles,
captions, and labels need to support and
provide an unbiased view of the data
displayed in your charts. These texts are part
of what persuades your audience to perform
your call to action.
However, the problem lies in the fact that
some data analysts use this text to manipulate
how the audience perceives the data. This
practice creates a bias towards a certain
opinion or view. This is an unethical practice.
Avoid doing it.
Cherry-picking Data
Cherry-picking data is when an analyst may
only visualize specific data points to better
support their narrative while leaving out
crucial, contradictory evidence. For example,
if sales are steady throughout Quarters 1-3,
then dives Q4, they may only present quarters
1-3, claiming they are on a steady growth
trajectory. In reality, the marketing methods
working for them might need some adjusting
due to the downturn.

DEC LUTTER I NG YOUR DATA

VI SUALI Z ATI ONS
Think about it. If you walk into a room with
furniture, trash, all sorts of miss-matched
pieces everywhere, you will be confused
about what you should look at and how you
should act in that room. This causes mental
confusion. Most people are not able to be
productive in a space that is cluttered like
this.
The same analogy applies to your data
visualizations. Data visualizations that are
cluttered cause mental confusion, so your
audience will not know where to look first or
next. All that mental noise will cause most
people to zone out and become less engaged
with your data presentation. Avoided that
mental confusion by applying the apt saying,
less is more.
Two different charts presenting the same data
can have vastly different reactions because of
the amount of information placed and how it
is placed. Obviously, you want to be on the
side of the fence where you present a good
chart that is aesthetically pleasing and gives
context to your audience. Tips for obtaining
such an uncluttered chart include:
Start Your Bar Chart at Zero (0)
Think of the X and Y axis of your charts as
the skeleton. Things will be skewed if you are
missing bones from that skeleton. Zero is part
of the makeup of the skeleton, and removing
it will skew the look of that chart. Starting
your chart above zero makes the bars of your
chart misleading, and this skewed view will
confuse your audience, and you would be
representing the data inaccurately.
Ensure you have a “proper” decluttered chart
by starting the origin as zero. Follow that up
with ensuring that your axes are evenly
scaled, placing uniformed spaces between the
bars, and placing the bars in chronological
order or in order of size.
Remove the Chart Border
Chart borders do not add information value to
your visualizations. Often it is visually more
appealing to watch white space rather than
the clutter that these elements add.
Do Not Use 3-Dimensional (3D) Effects
As mentioned before, 3D charts can obstruct
information and be misleading if they do not
serve a very specific function. We will
discuss this more later in this book. In
addition, they can make charts harder to read
and add clutter that is unnecessary to the
presentation. 2D graphics are typically more
suitable for effective data storytelling.
Steer Away from Dark Gridlines
Your data visuals need to be simple enough
that the audience can note from a single
glance what is being presented along each
axis of your chart. Therefore, the need for
gridlines should be zero or minimal.
The best practice is to eliminate the use of
gridlines as they add unnecessary noise. In
cases where they are necessary, use soft, grey
gridlines instead of harsh, black gridlines.
Avoid Overuse of Bright, Bold Colors
Colors should be used to show similarities
and differences and to provide context to
your data story. Just as you should not use
high color contrast to avoid misleading the
audience, you need to also limit the amount
of color you use. Bold, bright colors pull
attention in several directions at once and
make it hard to concentrate on specific
elements of data visuals. Let’s look at a chart
that has some of these distracting elements.
 FIGURE 2.13 Cluttered chart - Bar graph
with many distracting elements that don’t
add to the data story.

Always use a single color to represent the

same type of data and avoid using color
patterns like stripes, and polka dots as those
can be too distracting. You also need to be
mindful of the number of colors that you use
in a single data visualization. Do not use
more than six colors. If more values are
required, try breaking it up into two charts or
leaving out information that isn’t highly
relevant. Lets look at the corrected version:
 Figure 2.14 What it should look like - Bar
graph with proper scale and removal of
distracting elements such as gridlines,
chart border, and unnecessary color.

The best practice for using color is to use a

single color and vary the shades to show
similarities and differences or use a spectrum
between two similar colors to show a range of
data.

U SE the tips above to create charts that quiet

the noise and steer your audience to the
message you convey. Loud, cluttered charts
tend to push audiences away rather than draw
them in, but your data visuals will be a lot
more impactful if you apply the above
principles to create clean concise charts.
Here is a quick tip if you are having trouble
designing data visuals that stun - visit
ElizabethSClarke.com and download my free
data visualization checklist. This tool can be a
game changer for new data analysts as you
can adequately assess your charts to make
sure all the elements are correctly executed
and set up for success. Its absolutely free and
a lot better than memorizing this whole
section!
TH E C OLOR EXPER I ENC E
A compelling data story is memorable and
easy to digest. The same can be said about
effective data visualization. Color is one
element that makes it easier to achieve that
memorability and easy digestion of
information.
However, while color can add depth and
dimension to your data visualization, it can
also distract your audience from the
information you were trying to convey.
Unfortunately, most data storytellers have a
distracted audience because of the improper
use of color when designing data visuals.
That poor association is often the result of not
understanding color theory and how to use
color palettes when creating aesthetically
pleasing yet informative visuals. You cannot
just throw colors together and hope for the
best. You have to be strategic about how,
why, and when you use them.
We have touched on several aspects of color
theory above. It is time to dive into how you
can choose a color palette that compliments
the intentions of your data visuals rather than
distracts from it. While you can, of course,
stick to one color in your charts, using a
limited range of colors can add something
unique to the same information. This range of
colors is your color palette.
This range is not thrown together randomly,
though, if you want that color to be effective.
There are three color palette types that you
can fall back on to make your charts pop in a
good way, even if you do not have a design
bone in your body. These types are:
Qualitative Palettes
With such a color palette, the colors used are
distinctive. For example, if four colors are
used in a line chart, they may be green,
purple, orange, and yellow. Qualitative
palettes are typically used when the variables
are categorical and clearly different. For
example, a line chart may be designed to
show the unemployment rate in different
countries over five years. Each line would
represent a different country, which is a
different category. This is distinct, and so
each line would be colored differently.

FIGURE 2.15 Qualitative color palette.

When using a qualitative color palette to

design your chart, you may be tempted to go
buck wild with the color input, but this is a
temptation that you need to resist. Ideally,
you should limit the number of colors used to
6. On the far end, you may go up to 10 colors
but no further. The use of too many colors
brings up the audience’s inability to
distinguish between the colors used as the
chart becomes cluttered. This leaves far too
much room for misinterpretation. If you find
that your categories exceed ten total, the
alternative is to bundle similar categories
together or bundle categories with smaller
values together and label this as “other.”
Here are a few rules to stick by when using a
qualitative color palette to design your charts:

Ensure that the colors are used to

complement each other so that the chart
remains visually appealing.
Ensure that different categories are
distinct by ensuring different colors are
used. This can be done by adjusting
color saturation and lightness.
Ensure that the color differences are
not too significant, as this can lead the
audience into thinking that some
categories carry more importance than
others.
Do not use the same colors more than
once, unless the categories have a
relationship of some kind and a similar
hue can be used.

Sequential Palettes
This type of color palette makes use of color
by adding variations with different saturation.
For example, there may be six variations that
need to be presented on a chart. The color
used on this chart to show these data
variations is pink. Each of these variations
will have its value represented by a different
shade of pink.

FIGURE 2.16 Sequential color palette.

Sequential color palettes are typically used

when the variable values are numeric or
ordered. For example, a chart showing wage
changes in different companies over the four
quarters of a year may be designed with a
sequential palette. On the other hand, a
diagram showing the percentage variables of
a whole may also use a sequential palette.
Here are a few tips for gaining the most value
when using a sequential color palette when
designing your charts:

Use lighter colors to depict lower

values and darker shades to depict
higher values when your chart is
plotted on a white or light-colored
background. Plot the variations from
lighter to darker.
Use darker colors to depict lower
values and lighter colors to depict
higher values when your chart is
plotted on a dark background. Plot the
variations from darker to lighter.
More than one color may be used as
well. This is done by also playing with
the saturation of the colors used. For
example, a chart may show the
increasing temperature in a region by
designing a chart that moves from a
color hue like shades of blue to a
warmer hue like shades of yellow and,
finally, a hotter hue like shades red.
The visual designer can use a discrete
or continuous color gradient when
 plotting the values with this color
palette. A discrete palette is one where
there is a clear distinction between the
color saturations. Discrete color
palettes may even make use of text-like
number values to highlight that
distinction. On the other hand, with
continuous palettes, the saturation
appears to merge into each other. The
use of either type of palette is
dependent on the goal of the visual
designer. A discrete palette helps the
audience easily digest the data. So, this
is great for times when there is a
greater range of data. Discrete palettes
are also great for use when the variable
ranges are vastly unequal in size.
Continuous palettes are better when the
varying ranges are relatively equal in
size and when the range of data is
shorter.

Diverging Palettes
When plotting variables with a central value
like zero, divergent palettes are typically the
go-to color palette. Two different sequential
palettes are combined to show the movement
of values. Values on either side of that central
value are assigned a different color gradient.
One of the most common situations where
divergent color palettes are used is when
negative and positive values are highlighted
in one chart. The same rules for designing
when using a sequential palette apply when
creating charts with diverging palettes.

FIGURE 2.17 Diverging color palette.

The data color picker by Learn UI Design is

an excellent tool to help you pick sequential
and diverging palettes. Check it out here:
https://learnui.design/tools/data-color-
picker.html
BEST PR AC TI C ES WH EN USI NG
C OLOR
In addition to keeping in mind the type of
palette you use, here are a few tips for
ensuring that your charts remain visually
appealing to your audience.
Use Colors to Create Associations
Human beings have particular associations
with color. For example, we see red and think
stop or danger. Likewise, green is typically
associated with nature and safety. These color
associations invoke certain emotions within
us. You can use these associations when
designing your data visual to your advantage.
Keep such associations in mind when you
pick color palettes. For example, you may be
presenting a data story to people from a
certain university. When designing your
visualizations, using their university flag
colors can immediately invoke camaraderie
in the audience and make them more
receptive to your message.
Use a Single Color to Show Continuous
Data
In situations where it is not permitted that
your audience knows exact figures but rather
that they recognize a trend, the use of
continuous sequential and diverging palettes
may be best. This allows the audience to
grasp the trend quickly, whether increasing,
decreasing, or unchanging so that you can
link other information to that movement.
Use Contrasting Colors to Show
Comparison and Contrast
Help your audience easily distinguish
between datasets that are different with the
use of different colors. For example, a social
media analysis can easily distinguish the
conversion rate of using organic traffic versus
ads on Facebook with the help of the metrics
being colored green and blue. On the other
hand, if both metrics were colored the same,
it would be harder to determine this at a
single glance.
Use Color to Highlight Important
Information
If you want your audience to focus on a
particular piece of information, use a brighter
color or a higher saturation of the single color
used for that data set. This makes the
information stand out from the rest. You can
also choose to only color that set of
information and leave the rest, less pertinent
information, colored grey.
Do Not Pick Colors that Easily Merge Into
Each Other
Once your audience starts to squint at your
chart, you have failed to make the
information easy to interpret with a glance.
Avoid this situation by making the colors
used when designing your chart easy to
distinguish from each other. For example, if
you are designing a line chart, using different
shades of one color will incite that squinting
effect. On the other hand, using a qualitative
palette will allow for easier interpretation of
the separate data trends with one look.
Keep Your Color Count at a Minimum
The colors of the rainbow are at your disposal
when you design your charts. That does not
mean that you should pick every color to
infuse into your chart design. Make the colors
relevant and also keep the color count down
to avoid clutter. The maximum number of
colors used should be kept at six. Any more
and you risk your audience becoming
overwhelmed and unable to interpret the
information easily. If necessary, try and
separate the data into two visuals.
Account for Accessibility
Color vision deficiencies are more prominent
than most people realize. Some people can
distinguish between separate colors. For
example, orange and purple may look the
same to some people. This is called color
blindness. It is more common than you would
think. Approximately once in twelve men are
color blind. As a data storyteller, you need to
account that some of your audience may
suffer from such difficulties, and you need to
cater to them as well. Be sure to research
whether anyone in your audience faces visual
challenges so that you can factor that into the
development of your color palette.
Sometimes, this information is not easy to
source, so consider simply just designing
every chart with this in mind.

TH E PR OPER USE OF TEXT

A problem that many new analysts run into is
that they feel the need to explain every part of
the data visualizations with the inclusion of
excessive amounts of text. The thing that you
do not realize to avoid repeating this mistake
is that the data visualizations need to be
strong enough to speak for themselves with a
minimal amount of text.
When text is included, it needs to be done in
a clean, concise manner that does not include
lengthy paragraphs or unnecessary
descriptions. Compelling text usually only
consists of the labeling of axes if required and
memorable, effective titles.
In keeping with the element of text inclusion,
be sure to use a font that is appropriate for the
data presentation style and visualizations that
you have chosen. If the font style and overall
feel of the data presented do not mesh well,
the audience will be put off from that
experience.
While you should keep the use of text to a
minimum in your data visualization, there is
no denying that text does play an essential
role in conveying the right message to your
audience. Common text elements in data
visualizations include captions, labels, titles,
legends, and labeled icons. Despite their
scarce use, the proper implementation of text
in your data visuals is a must. Here are a few
tips for making the appropriate use of texts in
your charts:
Limit the Use of Uppercase Text
We live in a day and age when uppercase text
can be perceived as a rude expression such as
shouting. You certainly do not want your
audience to feel that you are shouting data at
them. Therefore, the use of uppercase text
needs to be limited and carefully
implemented when used. If you decide to use
uppercase texts, do so in a manner that calls
attention to a particular element.
Keep Chart Captions Short And Clear
Chart captions are used to summarize the data
being portrayed in your charts. You need to
analyze this text with an eagle eye, and any
word that does not add value needs to be cut
off swiftly. This means you need to chop all
adjectives. Articles like the, a, and an should
be removed where they are not required. All
words that have a shorter synonym need to be
replaced. Think short and crisp when you use
captions.
So, what should be included in your chart
captions? The first notable item is the units of
measurement to represent the data in the
chart. This should, of course, correspond with
the data plotted in the chart. For example, you
cannot have inches as the measurement in
your chart and have centimeters in your
caption. They need to be the same. Time
periods should also be included if they are
relevant to your chart. These are typically
added in brackets. As for location, as a safe
bet, center your caption as this generally is
the most visually appealing. As a final note
on this, ensure that you use sentence-style
capitalization when creating captions that end
those sentences with a period.
Shorten Data Labels
Labels help your audience identify the
categories represented in your charts. They
allow those categories to be associated with
the corresponding value. These labels need to
align with the categories represented in your
chart. They need to be as short as possible for
easy reading, especially if your chart factors
in several variables. It is acceptable to use
abbreviations to ensure this short length.
Suppose you find that it is not possible to
shorten the length of your labels past a point
where the labels do not run into each other,
slant them. This diagonal view is still easily
legible to most audiences. Do not rotate them
as this will be hard for the audience to read.
Data Legends Must Match the Data
Plotted
A legend acts as the key that clues your
audience in on what you are conveying with
the elements added to your charts. For
example, the legend may indicate what each
color in your chart represents. The first thing
that you must do is ensure that the legend
indeed corresponds to the data being
represented. The order also needs to
correspond to the order that these elements
appear on the chart. For example, if red is the
first color depicted on your chart, the first
item in the legend needs to indicate what red
represents. Lastly, the legend must be placed
outside of the information plotted on the chart
to ensure the chart remains uncluttered.
Titles Should Be Clear and
Straightforward
Just like captions need to be kept minimally
worded, so too do titles and headings. Do not
confuse a title for a caption. The title of your
chart needs to hold a lot less information
compared to a caption. Ensure that the graph
is titled in the shortest, crispest way. Those
words are meant to draw the audience by
capturing attention. Titles should also be void
of jargon and be easy for any audience to
interpret.

FIGURE 2.18 Ineffective title - title

should effectively state the values seen in
the graph.

Let’s revamp the chart above and make it

more presentable.
 FIGURE 2.19 Proper use of text - Clear
and effective title explaining what the
chart is representing. Abbreviated x-axis
for clean look. Additional values on top of
bars can make it easy to read when
specific values are required.

Ensure that Your Fonts and Other Text

Elements Do Not Clash With Your Color
Palette
You might be tempted to get fancy with it
when using texts since this use is sparse, but
again, this is a temptation that you must
resist. The first thing you need to do is ensure
that your fonts complement the color palette
you use. Also, ensure that the color of the
fonts matches or complements the color
palette.
A few more tips:

Do not use bold or italics unless there

is a clear purpose for doing so.
Do not use bold and italics at the same
time.
Do not use more than two different
fonts on any one chart.
 5

CRAFTING A WINNING DATA

STORY

“Visualizations act as a campfire around which we

gather to tell stories.”

— AL SHALLOWAY

M ovies,
They
novels, and even data stories.
are modern examples of
storytelling. However, storytelling is not
new to humanity. Cave paintings show that
our cavemen ancestors have been doing it a
millennium back. In fact, there are cave
paintings found that date back between
17,000 and 15,000 BCE.
Of course, with verbal communication came
oral story storytelling, and with written
language arose written storytelling. Written
stories have been found dated back as far as
3,400 BCE. Those stories were written on
clay tablets.
Luckily, some of the stories we find most
precious are not written on such fragile
things. With technological advancement, we
can now listen to, read, and tell stories on
radios, televisions, tablets, computers, and
even on the go with our smartphones. What a
way we have come!
To have withstood millennia of human
changes, storytelling must be powerful
indeed, and it has been shown that its powers
are unlimited. Storytelling has the power to
shape how we view the world and ourselves.
This one thing has the ability to determine the
prejudices that we develop (and yes, we all
have biases) and the values and morals that
we hold. It helps us understand and remember
valuable information. It helps us cohabitate
with each other via communication. Of
course, we can deliver communication via
hard facts, but stories give us context and
deliver valuable insight that would have
otherwise been lost.
All of these powers and more are yours to
harness because by using storytelling, you
can translate data into something
understandable, relatable, and actionable.
Through storytelling, data analysts can turn
numbers that mean nothing to an audience,
into a vehicle to drive change.
Just as there are steps and processes involved
in creating a magnificent story through a
novel or a movie, there are also steps and
processes involved in creating a magnificent
data story. Winging it will not do. Just as the
human body is composed of specific parts to
make complete anatomy, your data story
needs to hold certain features for it to be
complete and deliver value to your audience.
Without these components, you will have a
poorly executed data story.
This book was written to ensure that your
data story has complete anatomy. It was
written to give you the steps and processes
necessary for developing a magnificent data
story each and every time. This part of this
book shows you how to bring all the steps
and processes together beautifully.

H OW TO PR ESENT YOUR DATA

It is all good and well to understand the steps
and processes that go into creating a
magnificent data story. However, you also
need to know how to put these steps and
processes together to have a well-oiled data
storytelling machine. Such a machine has
eight main components. These components
are:
The Data Is Clearly Visible to the
Audience
This component can seem obvious, but it is
often the case where audiences need to squint
to make out figures and texts on data visuals.
What can seem clearly visible to you during
the design phase of your data story can be
hard to make out for your audience.
Avoid the embarrassment by getting a second
opinion as to how visible the information on
your charts is before you put it in front of the
eyes of your audience.
The Data Illustrates the Key Points
Remember that your data means nothing to
your audience until you slice it, dice it, and
spice it up with the proper condiments. Then
it becomes something tasty that your
audience wants to digest.
Always uphold the structure of the main point
of your data story, which is supported by key
insights that are then further supported by
points of relevant data. If your presentation
does not follow that hierarchy, you need to go
back to the drawing board. Anything else
leaves far too much up to the interpretation of
your audience. You are the one who needs to
provide a clear path to solutions for your
audience.
The Data Analyst Only Shares 1 Key Point
from Each Chart at a Time
Each chart you develop as part of your data
story has one function: to give context to a
key insight. Some data analysts try to take the
lazy way out and stuff several key insights
into one chart. The only thing such an action
will achieve is to confuse your audience.
The charts in your presentation need to be
strategically aligned with the narrative you
develop. The narrative needs to be paced well
so that your audience does not become
overwhelmed. The charts help maintain that
strategic pace.
The Components of Each Chart Are
Clearly Labelled
Your audience needs to gain as much
information from a glance as possible without
being overwhelmed. As a result, every
component of your chart needs to be clearly
and simply labeled. They also need to be
visually impactful. Avoid that dreaded
squinting effect from the audience.
Also, try to avoid abbreviations when
possible in component labels as their
extension might not be obvious to all
audience members.
Lastly, add the component labels to each
chart. Never assume that the audience will
remember from viewing a previous chart.
The Data Visualizations Guide the
Audience to Pre-planned ‘Aha’ Moments
An effective data story is a guided path to a
conclusion. Think about any good tour that
you have been on. There are stops along the
way for you to marvel and point at before you
reach that big finish. You need to provide the
same experience with your data story. The
conclusion is your big finish, but you also
need to provide stops along the way that wow
your audience. Those stops are your ‘aha’
moments. They are the key insights.
Make them special by verbally pointing them
out as well as adding them to your charts.
While you, of course, want to gently guide
the audience to such moments and make the
realization themselves, assume that it will fly
over the heads of at least some of the
audience members. Ensure that this does not
happen by clearly stating these moments.
The Chart Titles Reinforce the Key Points
Another tool that leads your audience to their
key insights (AKA the ‘aha’ moments) is the
title of your charts. Just like those special
stops along a guided tour spell out what the
visitors will be seeing, you need to provide
that part of the experience for your audience.
Ensure that the title complements the charts
so that your audience is not confused.
Imagine being on a tour and stopping at the
“mango trees” attraction only to see apple
trees. You will doubt everything else to come
on that tour. The same analogy applies to
titling your charts. This may seem obvious to
some, but I have seen one too many titles that
make little to no correlation with the data.
The Data Analyst Presents to the Audience
Rather Than to the Data
Eye contact. It is a nonverbal communication
method that can do a lot of things. It can
intimidate. It can tell of interest without a
word. It can express delight. Among the
many other things it can do is that it can build
a connection. Use that nonverbal
communication device to your advantage
when you present your data story.
Of course, there is nothing wrong with
glancing at your visualizations to make
references. However, far too many data
analysts keep their eyes trained on the charts
that they are presenting rather than their
audience and miss that opportunity to form a
human connection with these people. Not
only that, but they also give up the chance to
observe cues from the audience as to whether
the audience is captivated or bored.
Observing this allows the data analyst to keep
on the current path or make changes on the
spot to develop and maintain the connection.

H OW TO EXPLAI N YOUR C H AR T
As great as your visuals may be with
aesthetic colors, appropriate titles and labels,
and all the like, they need to be followed up
with verbal communication that reinforces
what is seen and adds relevant context.
I get it. Verbal presentations may not be some
of our strong suits. And left to our own
devices, we tend to make a muck of things.
But that will not be the case. You can fall
back on a methodology to make the most out
of explaining your charts, and it is called
Schneiderman’s mantra.
Ben Schneiderman developed this mantra.
Born in 1947, he is an American computer
scientist and a professor at the University of
Maryland. He proposed his mantra as a way
for data analysts to understand how people
visually engage content. An overview of this
visual engagement goes like this:

1. Overview of the information

2. Zoom and filter specific parts of the
information
3. Look for relevant details on demand

This mantra is used highly in interactive

visualizations, but the principles are still
applicable when static charts are used.
His mantra is the fifth step of developing
chart presentations that wow and that are
effective.

1. The five steps are:

2. Think about everything necessary to be
included in your charts.
3. Label everything from the titles and
legends to the X and Y axes.
4. Provide the context of how the data
was processed to develop the charts.
5. Be intentional about the visual
encoding, such as the colors and text
used to develop your charts.
6. Use Schneiderman’s mantra to explain
the charts to the audience.

Steps 1, 2, and 4 are done before you start

making your presentation. We have covered
the bases as to how to do these steps. Step 3
is a matter of courtesy and allows you to
avoid confusion. Simply outline your process
for assimilating your data into the charts
briefly in the introductory part of your
presentation.
Now let’s focus on step five, which outlines
Schneiderman’s mantra. By knowing how
human beings tend to engage visual content,
you can approach explaining your data
visuals in a way that aligns with this natural
flow. Here is a step-by-step guide to exactly
how you can do this.
Provide the Overview First
You feel it when you are confronted with
something visually for the first time. Your
eyes move back and forth, trying to take in as
much as possible in one glance. We often
describe the motion as not knowing where to
look first. The reason for this is because our
brains are trying to decipher exactly what we
are looking at. The brain has enlisted the
function of the eyes to go on this fact-finding
mission.
 FIGURE 3 Overview of graph first -
Overview of graph lets audience have a
general understanding of values before
making any conclusions.

Take advantage of the audience's natural

inclination and aid the eyes’ mission to
decipher what is being visually presented to
them by explaining exactly what they are
looking at in the form of your visualization.
This does not have to be long-winded. You
only need to provide a sentence or two to
describe what the chart is about. In essence,
you are giving a slightly more detailed
version of the title of your chart. And
remember, your title is the showcase of the
key insight highlighted by that chart.
Zoom and Filter
The next thing the brain is inclined to do after
getting an overview of visual information is
narrow in on specific things that are
particularly attention-catching. It filters
through all the information that is being
presented and zooms in on specific details.

FIGURE 3.1 Zoom and filter - Highlight

 key points to better get your message
across. Slowly lead your audience to your
conclusion.

As an effective data analyst, you would have

provided visual cues that are particularly
attention-catching. For example, these may
be the only colored bars in your bar graph or
the highlighted trends in your line graph.
These attention-grabbing details are what you
will use to provide that ‘aha’ moment. Such
details point your audience in a particular
direction to reach the conclusion, which is the
main point of your data story. Think back to
our guided tour analogy, and you know what
I mean. Those details support your insight
(the stop before reaching the end of the tour).
This is where visual cues such as the colors
you use can be of aid. For example, you may
want to point out a specific line on your line
chart to expand on that data. In that case, all
you would have to do is say something like,
“Notice this red line here…” and follow that
up with your explanation.
Provide Details on Demand
Once the brain has drilled down on specific
details that it finds interesting, curiosity is
aroused. Therefore, it will want to know more
about these particular details.

FIGURE 3.2 Details on demand - Explain

key information in your chart related to
specific highlighted points.

Again, this is something that you can

capitalize on while delivering your data story.
After you have made sure that your audience
has been thoroughly captivated by the
highlighted key insights, provide them with
more data about them. Give them the relevant
information about why this insight is relevant
to the ultimate conclusion of your data story.

T HE BEST THING about using Schneiderman’s

mantra is that this three-step approach can be
applied to explain any type of chart for any
kind of situation. It is a solid approach that
helps the audience understand what has been
presented to them and allows data analysts to
have a foolproof method of giving the
audience relevant context to what is being
visually presented to them. Always have this
manta in the forefront of your mind when
explaining your charts - overview first, zoom
and filter, and finally, provide details on
demand.

USI NG TH E R I GH T VOC ABULAR Y

From the beginning
Now that you’ve chosen an effective graph
and designed it to work in your favour, it’s
time to get an idea of what vocabulary you
should use when describing graphs, charts,
and diagrams. Here are some phrases I like to
use to catch my audience’s attention right
from the start.

If you look at this graph, you will

notice…
To illustrate my point, let’s look at
some charts…
Let’s turn to this diagram…
I’d like you to look at...
If you look at this graph, you will
notice…
Let me show you this bar graph...
Let’s have a look at this pie chart…

Describing important elements

When you are describing any form of visual
information, it is important to guide your
audience to the key points you are trying to
make. Some examples would be:

The colored segment is for…

 The vertical axis shows…
The shaded area describes...
The horizontal axis represents…
The curve here illustrates…
The solid line shows…
The green bar indicates….

PER FOR MI NG A GR EAT

EXEC UTI ON
One of the best storytelling tactics is to give
real-life examples, and I will show you what
this looks like by giving you an example of
what an effective data story looks like in this
section. This section is a run-through of what
an effective data story looks like in a real-
world setting.
For such a scenario to work, you need to have
gone through the steps previously outlined in
this book. There is no skipping the creative
process that ultimately leads you to the
presentation of your charts. You would have
needed to develop an engaging narrative,
compile data that is relevant that aligns with
that narrative and design visually appealing
charts to support this relevant data. Without
these supporting aspects, there is no hope of
executing an excellent presentation.
With those aspects in the bag, you are
prepared for the big day. You will prove your
mantle as an engaging storyteller supported
by data visuals. Again, this is not something
that you have to develop a new method for.
There is a methodology that you can fall back
on to tell great data stories.
An engaging data story has four main
sections, and they are:

The Introduction
The rising action
The climax
Conclusion

F OR EVERY PART of the data story section, you

will use Schneiderman’s mantra to break
down each chart used. Always state the main
point of each graph, then using key insights
and relevant data to support this.
Let’s jump right into outlining what telling a
data story looks like in the real world.

TH E I NTR ODUC TI ON
In a book or film, this section is a
representation of the main characters going
about their daily lives before they are
ultimately thrown through the wringer to
come out as a changed person. It is the setup
to show where this character is now so that
the audience has context as to how they will
move through the main parts of the story. It is
what allows the audience reading or watching
to develop a connection with this character.
When the audience cares, the introduction has
performed its function and hooked these
people.
In data storytelling, the introduction has the
same function. It is a setup meant to give
context and hook the audience. This is where
the analyst will show the audience what
problem brought everyone to that setting on
that day, the benefits to solving that issue,
and possible solutions to solving the problem.
The audience needs to care about moving
through the guided tour that starts with the
problem and ends with the solution.
Let’s use an example to illustrate. A
marketing manager is presenting to the CEO
and CFO in regards to ad performance, and
they are introducing the problem by saying:
“Since we introduced our advertising
campaign in Quarter 2 of 2019, traffic to our
website has grown steadily by approximately
15% to 25% each quarter. We anticipated
that this trend would continue to be the case
for quite some time. However, that prediction
has not come to fruition. In the first quarter
of 2021, we observed a staggering 30%
decrease in website traffic, and it got us
wondering why…”
FIGURE 3.3 Ask the right questions - Ask the right
questions so your audience can draw the proper
solution on their own without you needing to persuade
them.

As you can see from this example, the

manager has set the stage by showcasing the
norm and highlighting the problem that led to
this data story’s need. They also alluded to
finding the cause of the problem and the
delivery of a solution to the problem.
They have provided context and delivered the
hook.
TH E R I SI NG AC TI ON
In books and films, this is where the main
character's life changes. A catalyst causes this
change and ensures no turning back from the
path ahead of this person. The introduction
has gotten the audience to care about this
person, but this part will have the audience at
the edge of their seats wanting to know what
happens next.
In data storytelling, again, the analogy
applies. The catalyst that incites change is the
delivery of analyzed data to support the
findings that highlight the problem. Many
junior data analysts make a mistake here:
they throw data point after data point after
point at the audience. The function of this
part of the story is to build anticipation of the
solution to the problem. Continue to give
context to the problem and support this with
only relevant details. Do not be stingy but
build anticipation by leaving the audience
wondering where you are leading them. Keep
that image of the guided tour in your mind.
You do not want to see everything the tour
has to offer upfront. Otherwise, what would
there be for you to look forward to?
Continuing with our example of the
marketing manager’s presentation, they might
continue by saying:
“Of course, all due diligence has been paid
here. We have analyzed our advertising
campaign strategy and reviewed our content.
We have also compared data concerning how
our competitors are performing. Our other
advertisements were performing as usual.
The problem is isolated to this particular
campaign. The advertisements that were
bringing us the most traffic suddenly dropped
in performance and stopped attracting new
website visitors. The question remains… Why
has this happened?”
FIGURE 3.4 Comparisons- Line graphs are great for
trend comparisons when a drastic change has to be
accounted for.

As can be seen, by this explanation, the

manager is giving more context to data. At
this point, they are also showcasing charts
that support this context. Notice that their
wording gives more details and leads the
audience’s focus to a particular point, which
is why this problem has arisen and how it can
be solved.

TH E C LI MAX
This step in storytelling represents that
turning point in the story where the changes
that will come to fruition become clear. It is
the highest point of tension in the story. It is
the height of the anticipation built in the
rising action step. Remember that this part
will not serve its purpose unless you set the
foundation with a good introduction and the
anticipation developed in the rising action.
In data storytelling, the climax is your ‘aha’
moment. It finally reveals the things that the
presenter was alluding to in the introduction
and rising action portions. Things should
become clear to the audience then, leaving
them with a sense of fulfillment.
Back to our example, the marketing manager
may say something like this:
“It took some digging, but then we found a
problem. At the time that the traffic decrease
was recorded, we had lowered the
advertising budget to test other markets
beyond our traditional audiences.
Additionally, we have found that due to this
budget drop, we started to lose out on front-
page bids. Due to this, our competitors' ads
were outbidding us for the top spots, gaining
more visibility with our primary audience.”

FIGURE 3.5 Be resourceful - Always check the data

of related information to see if there are any patterns.
In this case, the ad spend was lowered right around
the time of the performance drop.

This speech has revealed the why of the

problem that was stated from the introduction
and given context in the rising action. The
audience should feel relieved to know the
cause of the problem and look forward to the
conclusion which leads the way forward.
C ONC LUSI ON
In a book or movie, this is the part that gives
closure to the audience as it shows how the
main character is settling in with the new
changes. Data storytelling summarizes the
entire presentation in as brief a language as
possible and defines the steps that should be
taken next.
The following steps can be hypothetical, a
solution to how the situation can turn out if
certain actions were taken or avoided, or
recommendations for future action.
The marketing manager can give a conclusion
stating:
“Clearly, the update to the advertising
campaign, which was a lowered budget along
with the targeting of alternative markets, did
not go as planned. There are a few possible
solutions that can be implemented to revamp
the advertising campaign to what it was and
potentially yield better results. We can
replace the budget or potentially raise the ad
spend. This raise can be facilitated by moving
some of the budget from our lesser-
performing advertisements. We also need to
retarget these advertisements to our original
market for the highest conversions. With
these changes, the campaign should perform
as usual.”

FIGURE 3.6 Always project - Creating projections of

what can happen or what should’ve happened is a
great way to ensure your performance is on a steady
growth track.

Like everything else with data storytelling,

you do not have to reinvent the wheel to
explain your charts right. The template has
already been prepared for you. First,
introduce the problem that has warranted the
need for the development of this data story.
Get your audience anticipating a solution by
giving them context and hinting why this
problem has developed with the rising action.
Provide the ‘aha’ moment in the climax of the
data story by plainly explaining the why that
has been anticipated. And finally, leave the
audience fulfilled with a conclusion that
outlines possible solutions to solving this
problem.
 JOIN OUR COMMUNITY!

My goal is to create a thriving community

where we can all share our data visualization
and storytelling techniques and learn from
one another in one of the most crucial
decades for data! Join our facebook group
and be a part of our growing community. Feel
free to ask questions and share what you
know.

www.facebook.com/groups/junioranalystsmastermind/
 FREE GOOD KARMA

“We make a living by what we get. We

make a life by what we give.”
— WINSTON S CHURCHILL

If you enjoyed reading this book and gained

the intended value, Please take 60 seconds to
leave a review! Every review helps this book
reach another striving data analyst or business
professional looking to perfect their craft and
excel in their career.
The feedback is greatly appreciated, and I
always love to see what people get out my
work and how I can improve on future
projects. Thank you so much for being patient
on this small detour. Now continue on with
the book for some final insight.
Yours truly,
Elizabeth Clarke
 CONCLUSION

“The effectiveness of data visualization can be

gauged by its simplicity, relevancy, and its ability
to hold the users hand during their data discover
journey.”

— JAGAT SAIKIA

Every single day, data, so much of it, is being

created. It would all look like nonsensical
babble if not for people like you. Data
analysts and business professionals are the
translators of data, and without them, so
many audiences would not know how to
proceed with this data influx.
However, there is an art and a science to
delivering data to an audience in an
understandable, relatable, and actionable
way, and that is data storytelling. You have
gotten to this part of this book, and I
commend you for that dedication to
mastering data storytelling. You have all the
tools necessary for delivering a powerful data
story to your audience each and every time.
From here, you can continue to expand upon
your knowledge and grow your expertise to
translate more complex and compelling data.
As a quick reference, here is a brief summary
of what it takes to create a data story that
wows and informs:
The Foundation
This is the narrative of your data story. The
entire structure will depend on how
effectively you develop the narrative. Your
goals will be established here. You will also
drill down on the context of your data story
here. The context gives the why of this data
story - why it is necessary that this data story
be told. What is the problem that needs
solving, and how can it be solved in
actionable steps? Developing the narrative
also allows you to develop an understanding
of who the key players are and what
information is already at their disposal.
Figuring out such details will also help you to
develop an introduction that hooks your
audience and keeps them captivated
throughout the presentation.
Captivating Your Audience
The narrative, the chart development, and all
other effort and time you have put into
developing a data story will be for nothing if
you fail to connect with the audience you are
delivering to. These people rely on you to
clearly state the problem that needs to be
addressed and provide possible solutions.
Therefore, how to present data to them should
always be at the forefront of your mind. You
need to become familiar with the needs of
these people, their familiarity with the
subject, the best ways of communicating with
them, and the best methods of influencing
them in taking action after all is said and
done.
Refining the Data Visuals: Chart
Development
With the narrative developed and the
audience figured out, you need to recreate the
details in visual form. This is the function of
your charts. However, not just any chart will
do. You need to create naturally paced charts
and designed visuals to keep your audiences
informed but not overwhelmed.
This phase also means choosing the right
charts to showcase data in the best light. This
choice is dependent on the type of data that
needs to be highlighted during your
presentation. The types of data are
comparison, composition, relationship, and
distribution.
Refining the Data Visuals: The Design
Chart development is not just about having
relevant data presented. Charts must also be
visually appealing to your audience. This is a
delicate balance that you must achieve to
keep your audience engaged and interested in
seeing your data story to the end. The
psychology of design is something that you
must fall back on to choose colors, textures,
texts, and other design details that positively
capture the attention of the audience and help
them understand the key insights being
delivered.
Crafting a Data Story That Wins
You have all it takes to execute a winning
data story once you follow the steps outlined
above. These tools guide you to crafting a
presentation that fulfills its purpose. The last
step is execution. Luckily, Schneiderman’s
mantra allows you to clearly explain your
charts with these steps:

1. Overview the information

2. Zoom and filter specific parts of the
information
3. Look for relevant details on demand

A data story being effectively delivered has 4

main sections, and they are:
 The introduction
The rising action
The climax
Conclusion

Follow that structure, and you will have this

in the bag!
If there is one thing that I want you to take
away after reading this book, it is that anyone
can craft an engaging data story. The steps
and tools have already been clearly laid out
for you. This book was written in plain
language to show you step by step how this
methodology plays out so that you have the
best chances of creating winning data stories
every time. This last bit only you can
perform, putting these steps and tools into
good use. Don’t overthink it. Just follow the
methodology, expand upon it, and become an
awesome data storyteller. What are you
waiting for?
 I NDEX

Introduction

1. The Foundation of Data Storytelling - Developing

the Narrative
What is the Narrative of Data Storytelling?
Understanding the Need for a Narrative
Steps for Developing the Narrative of your Data
Story
The 5 Key Concepts That Your Data Narrative
Must Follow to Be Effective
2. Captivating Your Audience
Identifying Your Audience
What is the Personality Type of Your Audience?
What Are Your Audience's Literacy and Numerical
Literacy Levels?
How to Captivate Your Audience
How to Know if You Have Got It Right With Your
Audience
3. Refining Your Visuals - Choosing the Right Chart
The Importance of Visual Representation
Choosing the Right Chart
Types of Charts and When and Why to use Them
Bar Graphs
Pie Charts
 Line Graphs
Area Charts
Scatter Plots
Using Tables
4. Refining Your Visuals - Developing a Winning
Design
The Psychology of Design
Common Mistakes Made When Creating Data
Visualizations
Decluttering your Data Visualizations
The Color Experience
Best Practices When Using Color
The Proper Use of Text
5. Crafting a Winning Data Story
How to Present Your Data
How to Explain Your Chart
Using the right vocabulary
Performing A Great Execution
The Introduction
The Rising Action
The Climax
Conclusion

Conclusion
Resources
 RESOURCES

Boyd B. (2018). The evolution of stories:

from mimesis to language, from fact to
fiction. Wiley interdisciplinary reviews.
Cognitive science, 9(1), e1444.
https://doi.org/10.1002/wcs.1444
Boyd, R. L., Blackburn, K. G., &
Pennebaker, J. W. (2020). The narrative arc:
Revealing core narrative structures through
text analysis. Science advances, 6(32),
eaba2196.
https://doi.org/10.1126/sciadv.aba2196
Dettori, J. R., & Norvell, D. C. (2018). The
Anatomy of Data. Global spine journal, 8(3),
311–313.
https://doi.org/10.1177/2192568217746998
Elliot A. J. (2015). Color and psychological
functioning: a review of theoretical and
empirical work. Frontiers in psychology, 6,
368.
https://doi.org/10.3389/fpsyg.2015.00368
Hattab, G., Rhyne, T. M., & Heider, D.
(2020). Ten simple rules to colorize
biological data visualization. PLoS
computational biology, 16(10), e1008259.
https://doi.org/10.1371/journal.pcbi.1008259
Lee, J. C., & Livesey, E. J. (2018). Rule-
based generalization and peak shift in the
presence of simple relational rules. PloS
one, 13(9), e0203805.
https://doi.org/10.1371/journal.pone.0203805
Li Q. (2020). Overview of Data
Visualization. Embodying Data: Chinese
Aesthetics, Interactive Visualization and
Gaming Technologies, 17–47.
https://doi.org/10.1007/978-981-15-5069-0_2
Marković S. (2012). Components of aesthetic
experience: aesthetic fascination, aesthetic
appraisal, and aesthetic emotion. i-
Perception, 3(1), 1–17.
https://doi.org/10.1068/i0450aap
Martinez-Conde, S., Alexander, R. G., Blum,
D., Britton, N., Lipska, B. K., Quirk, G. J.,
Swiss, J. I., Willems, R. M., & Macknik, S.
L. (2019). The Storytelling Brain: How
Neuroscience Stories Help Bridge the Gap
between Research and Society. The Journal
of neuroscience: the official journal of the
Society for Neuroscience, 39(42), 8285–8290.
https://doi.org/10.1523/JNEUROSCI.1180-
19.2019
Mastandrea, S., Fagioli, S., & Biasi, V.
(2019). Art and Psychological Well-Being:
Linking the Brain to the Aesthetic
Emotion. Frontiers in psychology, 10, 739.
https://doi.org/10.3389/fpsyg.2019.00739
Midway S. R. (2020). Principles of Effective
Data Visualization. Patterns (New York,
N.Y.), 1(9), 100141.
https://doi.org/10.1016/j.patter.2020.100141
Plante, T. B., & Cushman, M. (2020).
Choosing color palettes for scientific
figures. Research and practice in thrombosis
and hemostasis, 4(2), 176–180.
https://doi.org/10.1002/rth2.12308
Ranganathan, P., & Gogtay, N. J. (2019). An
Introduction to Statistics - Data Types,
Distributions, and Summarizing Data. Indian
journal of critical care medicine: peer-
reviewed, official publication of Indian
Society of Critical Care Medicine, 23(Suppl
2), S169–S170. https://doi.org/10.5005/jp-
journals-10071-23198
Suzuki, W. A., Feliú-Mójer, M. I., Hasson,
U., Yehuda, R., & Zarate, J. M. (2018).
Dialogues: The Science and Power of
Storytelling. The Journal of neuroscience: the
official journal of the Society for
Neuroscience, 38(44), 9468–9470.
https://doi.org/10.1523/JNEUROSCI.1942-
18.2018
Wolfe, J. M., & Utochkin, I. S. (2019). What
is a preattentive feature?. Current opinion in
psychology, 29, 19–26.
https://doi.org/10.1016/j.copsyc.2018.11.005
L.C.T. (2020, July 7). How to Tell a Story
With Data. Lucidchart. Retrieved November
2, 2021, from
https://www.lucidchart.com/blog/how-to-tell-
a-story-with-data
Goldmeier, J. M. (2019, November 12). The
only data visualization guide you’ll ever need
(in 5 principles). LinkedIn. Retrieved
November 2, 2021, from
https://www.linkedin.com/pulse/only-data-
visualization-guide-youll-ever-need-5-jordan-
goldmeier
Goldmeier, J. M. (2019, November 12). The
only data visualization guide you’ll ever need
(in 5 principles). LinkedIn. Retrieved
November 2, 2021, from
https://www.linkedin.com/pulse/only-data-
visualization-guide-youll-ever-need-5-jordan-
goldmeier
Infogram. (n.d.). How to Choose the Right
Chart for Your Data.
https://infogram.com/page/choose-the-right-
chart-data-visualization
Gulbis, J. B. (n.d.). Data Visualization – How
to Pick the Right Chart Type? EazyBI.
Retrieved November 2, 2021, from
https://eazybi.com/blog/data-visualization-
and-chart-types
Yi, M. (2019, August 23). A Complete Guide
to Bar Charts. Chartio. Retrieved November
2, 2021, from
https://chartio.com/learn/charts/bar-chart-
complete-guide/
Yi, M. (2019b, September 24). A Complete
Guide to Stacked Bar Charts. Chartio.
Retrieved November 2, 2021, from
https://chartio.com/learn/charts/stacked-bar-
chart-complete-guide/
Yi, M. (2019b, August 29). A Complete
Guide to Pie Charts. Chartio. Retrieved
November 2, 2021, from
https://chartio.com/learn/charts/pie-chart-
complete-guide/
Marr, B. (2021, July 13). Why You Shouldn’t
Use Pie Charts In Your Dashboards And
Performance Reports. Bernard Marr.
Retrieved November 2, 2021, from
https://bernardmarr.com/why-you-shouldnt-
use-pie-charts-in-your-dashboards-and-
performance-reports/?
contentID=1779#:%7E:text=From%20a%20design%20po
Yi, M. (2019c, September 13). A Complete
Guide to Line Charts. Chartio. Retrieved
November 2, 2021, from
https://chartio.com/learn/charts/line-chart-
complete-guide/
Yi, M. (2019d, September 16). A Complete
Guide to Area Charts. Chartio. Retrieved
November 2, 2021, from
https://chartio.com/learn/charts/area-chart-
complete-
guide/#:%7E:text=An%20area%20chart%20combines%20
R. (2020b, March 8). Presenting data
visualization to engage your audience.
Medium. Retrieved November 2, 2021, from
https://uxdesign.cc/presenting-data-
visualization-to-engage-your-audience-
815eb6a43a62
Yi, M. (2019f, October 16). A Complete
Guide to Scatter Plots. Chartio. Retrieved
November 2, 2021, from
https://chartio.com/learn/charts/what-is-a-
scatter-plot/
Brown, L. (n.d.). Wondershare Fotophire
Online Support Center. Wondershare.
Retrieved November 2, 2021, from
https://photo.wondershare.com/graph-
maker/best-graphing-software.html
Verma, R. (2021, February 26). Data
Visualization: Top 5 Most Important Things
to Know. Loginworks. Retrieved November
2, 2021, from
https://www.loginworks.com/blogs/data-
visualization-top-5-most-important-things/
V. (2019, February 13). Preattentive
Attributes in Visualization - An Example.
Daydreaming Numbers. Retrieved November
2, 2021, from
http://daydreamingnumbers.com/blog/preattentive-
attributes-example/
Preattentive Visual Properties and How to
Use Them in Information Visualization.
(2018, October 2). The Interaction Design
Foundation. Retrieved November 2, 2021,
from https://www.interaction-
design.org/literature/article/preattentive-
visual-properties-and-how-to-use-them-in-
information-visualization
Horne, J. (2020, June 5). Neuroaesthetics and
Informative Art | iDashboards Blog.
IDashboards |. Retrieved November 2, 2021,
from
https://www.idashboards.com/blog/2017/08/23/neuroaesth
and-informative-art/
Baltusevičius, G. (2021, February 2). How to
Do Storytelling with Data Using
Visualizations. Blog | Whatagraph. Retrieved
November 2, 2021, from
https://whatagraph.com/blog/articles/data-
using-visualizations
Bowers, M. (2020, October 20). Numbers
Shouldn’t Lie – An Overview of Common
Data Visualization Mistakes. Toptal Design
Blog. Retrieved November 2, 2021, from
https://www.toptal.com/designers/ux/data-
visualization-mistakes
I. (2018, November 29). Dos and Don’ts:
Data Visualization Tips Before and After.
Medium. Retrieved November 2, 2021, from
https://medium.com/@Infogram/dos-and-
donts-data-visualization-tips-before-and-
after-f1d65a7b6402
M. (2021, June 17). 7 Best Practices for
Using Color in Data Visualizations. Sigma
Computing. Retrieved November 2, 2021,
from
https://www.sigmacomputing.com/blog/7-
best-practices-for-using-color-in-data-
visualizations/#:%7E:text=Why%20color%20use%20in%
Makulec, A. (n.d.). Identifying Your
Audience. Slideshare. Retrieved November 2,
2021, from
https://www.slideshare.net/AmandaMakulec/identifying-
your-audience-40086476
Sleeper, R. (2021, January 2). Vital Question
1: Who is the Audience? Playfair Data.
Retrieved November 2, 2021, from
https://playfairdata.com/vital-question-1-
who-is-the-audience/
S. (2018b, May 25). Presenting complex
data? Engage your audience with these 10
tips. Medium. Retrieved November 2, 2021,
from
https://speakerhubhq.medium.com/presenting-
complex-data-engage-your-audience-with-
these-10-tips-232509301a4d