Jock Mackinlay, PhD, Robert Kosara, PhD, Michelle Wallace

Data Storytelling
Using visualization to share the
human impact of numbers
 2

The universe is made of stories, not atoms.Muriel Rukeyser (19131980)

Storytelling is a cornerstone of the human experience. The universe may be full of atoms,
but its through stories that we truly construct our world. From Greek mythology to the Bible
to television series like Cosmos, stories have been shaping our experience on Earth for as
long as weve lived on it.

A key purpose of storytelling is not just understanding the world but changing it. After
all, why would we study the world if we didnt want to know how we canand should
influence it?

Though many elements of stories have remained the same throughout history, we have
developed better tools and mediums for telling them, such as printed books, movies, and
comics. This has changed storytelling stylesand perhaps most importantly, the impact of
those storiesover the millennia.

But can stories be told with data, as well as with images and words? Thats what this
papers about.

Table of Contents

The Tales We Tell.................................................................................................................................................3

Why We Tell Stories...........................................................................................................................................4

Storytelling through the Ages..........................................................................................................................6

Story Points................................................................................................................................................................7

When to Tell a Story...........................................................................................................................................9

Storytelling to the People...............................................................................................................................10

About the Authors.............................................................................................................................................12

 3

The Tales We Tell

A little drama makes a big impact
Which do you prefer: raw, unbiased data, or an intriguing tale? It may seem
unfair, butwhen it comes to understanding the world, at leastyour brain
has an unequivocal preference for stories.

Consider the number 131,824. Thats how many 4.0-plus-magnitude earthquakes

weve detected since 1973. This dashboard shows their distribution around the world.
Stories take advantage
of human cognition. They Merriam-Webster defines an earthquake as a shaking or trembling of the earth
that is volcanic or tectonic in origin. But its hard to think about earthquakes
build connections and
without imagining the ground shaking beneath your own feet. By relating to the
context around facts
concept, you can understand it better.
in order to make them
more memorable. On Friday, March 11, 2011, a 9.0-magnitude earthquake struck off the coast of
Earthquakes: Are they are on rise?
Thoku, Japan, resulting in a catastrophic tsunami. As The New York Times
Thousands of About two each year The megaquakes are Megaquake: The Megaquake: The More and more Most earthquakes But major
reported,
earthquakes are
detected each year
qualifyOn Friday,attention
as "major
earthquakes"
athas
what 2:46
drawnp.m. Tokyo time,
Indian Ocean
in 2004.
quake theJapan
quake
2011.
struck.being
quake of First came
earthquakes are
detected
thethe
areroar
being detected in
Pacific Rim
earthquakes
happening m

and rumble of the temblor, shaking skyscrapers, toppling furniture and buckling
highways.
The Then
2011 quake off waves
the coast as was
of Thoku high as 30 feet
a magnitude rushed
9.0 (Mw) underseaonto shore,
megathrust whisking
earthquake. away
It was the most cars
powerful Earthquake Date Time
1/1/2004 to 2/26/2014
known earthquake ever to have hit Japan, and the 5th most powerful earthquake in the world.
and carrying blazing buildings toward factories, fields and highways.
Magnitude
8 to 9.1

Region of the World

All

See this story come together here.

16
s

earthquakes select

Magnitude 9.0 near the east coast of Honshu, Japan

Magnitude
6.0 8.0

While facts and data form the backbone of this story2:46 p.m., 30 feet highits
the flow of the story that ties them all together. Stories also add embellishments
that make data more memorable. Words like roar and shaking add drama to
the facts, making them easier to relate to.

Data tells you whats happening. Stories tell you why it matters.
 4

Why We Tell Stories

The importance of understanding events from beginning to end
What, exactly, makes a story? While weve been telling them since the beginning
of history, we didnt look at their structure until about 335 BC. Thats when Aristotle
claimed that all stories have a beginning, middle, and end.

While this might seem like common sense, it was a pretty big leap at the time.
If you follow an event from start to finish, you begin to understand why it unfolded
that way.

The flow of a story helps us spot causal connections. Getting to the root cause of
an issue can be tricky, but our brains are hardwired to find connections between
events. If you turn them into a well-structured story, it can be easier to spot an
underlying causeand then to share your understanding of that cause.

For instance, teacher turnover has been a problem in Austin, Texas for a decade.
Austins turnover rate has been higher than the national average for all but one of
the past ten years, and by 2013 it had reached 22.3 percent.

U.S. avg.

In August of that year, a former teacher from J.E. Pearce Middle School in East
Austinan area with one of the highest poverty rates in the citypublished his
long-untold story on Salon.com.

In 2002, the year I started teaching at Pearce, he explains, many of the faculty
had been hand-selected to revive the struggling school. When he entered
Pearce, he was determined to do whatever it took to help these kids overcome
classism and racism and escape poverty.
 5

But within two years, he had quit. Why? Despite his qualifications and efforts,
in the end he came to believe they were in vain. His experiences led him to
conclude that povertyand not teacher qualitywas the root cause that really
needed to be addressed.

As human beings, were constantly receiving information about the world around
us. The amount of sunlight hitting your eye, the level of ambient noise in your
current locationthese are all data points that tell you something about the
current state of the world.

Since you receive each piece of data at a specific time, you automatically fill in
the missing information between each moment. Its something our brains have
evolved to do exceptionally well: We stitch data points together to turn them into
a cohesive, step-by-step story and decide how to act. Sometimes, we do this too
well. We identify patterns and causes that arent really there.

School name School type Avg. Turnover rate

This chart shows the 60.0

10.0 22.0

Teachers

teacher-turnover rate at Mendez Middle school

40.0 39
60
80
20.0

Pearce Middle School 0.0

over the past decade.

60.0

40.0
Pearce Middle school

20.0

0.0

60.0

40.0
Reagan High school

20.0

0.0
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

If the leaders of Austins school system looked at this data in a meeting, each
of them might walk away with a different story about whats causing this rate to
increase. Each would write that story based on his or her individual experiences.
With that story in mind, they would each walk away from the meeting with
separate ideas about how to act next.

But what if the data told a complete story? What if it was told in a clear sequence,
step-by-step from beginning to end? And what if everyone left that meeting
believing in the same storya story founded in data? Could the students in
Austins school system get to their own happy ending, and faster?

John Savage wanted to influence change by becoming a teacher. But now, he

does that by telling stories. As a journalist, he shares information in a relevant,
sequential way so that readers can see important connections between facts. But
what if he used data visualizations to help tell those stories?

Data tells you whats happening, and stories tell you why. When you put them
together, great change is possible.
 6

Storytelling through the Ages

Evolving tools for influential tales
The Austin school districts story hasnt been resolved yet, but one thing is clear:
Theres nothing like a compelling plot to make people care. And since the days of
Aristotle, weve delved deeper into the structure of storiesand learned its not just
about the beginning, middle, and end. A good story also has an arc.

Logical sequence A clear flow turns a collection of facts into a compelling narrative. It glues them
or narrative
together into a structure that makes sense. And the rising action of a story arc
Question Conclusion or
or problem resolution makes it more engaging and memorable.

Thats why weve been telling stories for so long. They are excellent tools for
passing knowledge from one person to another. And theyre exceptionally good
at getting a point across.
A story arc consists of rising
s

action and conflict, presented
logically and fluently to lead to
a conclusion.
In 1854, Dr. John Snow was an epidemiologist in London, a city in the clutches of
a cholera outbreak. Dr. Snow had a hunch about how cholera was being spread and
how it could be stopped, but he needed a compelling way to tell this story to the
people who were making the decisions. So he came up with a new way to tell a data
story: he plotted each death on a map of the city.

When he showed his map to the citys leaders, they immediately understood what the
problem was: an infected water pump in an area with no sewage system. Since most
of the deaths occurred near that pump, they decided to replace its handle. Germ theory
was developed less than a decade later.

The tools we use to share visual stories are only getting better with time. Just a few
centuries have brought us from hand-drawn visualizationslike John Snows mapto
movies and computer graphics. Humans continue to build innovative ways to tell powerful
stories. Tableaus Story Points feature is the next evolution of storytelling with data.
 7

Story Points
Telling and discovering stories with the same tool
Story Points provide a framework for arranging data visualizations sequentially,
so that you can tell stories with a beginning, middle and endeven with large
and complex data sets. With Story Points, you can tell stories with data in the
same tool you use to analyze data, and your presentation tool stays connected
to your data.

Heres an example of how Story Points works:

More than 131,000 earthquakes have been detected worldwide since 1973. Weve
also seen many recent high-impact quakes, like the one that struck near Thoku,
Japan in 2011. But is the number of quakes around the world actually increasing?

At the top of this visual, a series of annotations (shown above) guide viewers
through their exploration. By clicking through these annotationslike pages in a
booka series of interactive visualizations appear.

This story begins at the global level, and then focuses in on some of the larger
quakes weve recently seen. After surveying areas such as Japan, it turns to
overall trends of earthquake detection.

Immediately, a trend pops out of the visual. More earthquakes are being detected
than ever before, but most of the increase is actually coming from lower-magnitude
quakes. It seems that earthquakes are not on the rise; rather, weve improved our
methods for detecting them.

Of course, this does nothing to lessen the human impact of larger quakes. But
when youre looking at a complex data setwhether in a research lab, a board
room, or somewhere elseyou need to be able to decide how youll act upon it.

Earthquakes: Are they are on rise?

Earthquakes: Are they are on rise? Thousands of About two each year The megaquakes are Megaquake: The Megaquake: The More and more
Earthquakes: Are they
Most earthquakes
are on rise?
But major
earthquakes are qualify as "major what has drawn Indian Ocean quake Japan quake of earthquakes are are being detected in earthquakes are not
Thousands of About two each year The megaquakes are Megaquake: The Megaquake: detected
The eachMore
yearand moreearthquakes"Most earthquakesattention But major in 2004. 2011. being detectedof
Thousands the Pacific
About two Rim
each year happening
The more are
megaquakes oft.. Megaquake: The Megaquake: The More and more Most earthquakes But major
earthquakes are qualify as "major what has drawn Indian Ocean quake Japan quake of earthquakes are are being detected in earthquakes are not earthquakes are qualify as "major what has drawn Indian Ocean quake Japan quake of earthquakes are are being detected in earthquakes are not
detected each year earthquakes" attention in 2004. 2011. being detected the Pacific Rim happening more oft.. detected each year earthquakes" attention in 2004. 2011. being detected the Pacific Rim happening more oft..

The 2011 quake off the coast of Thoku was a magnitude 9.0 (Mw) undersea megathrust earthquake. It was the most powerful Earthquake Date Time
1/1/2004 to 2/26/2014
knownmagnitude
A big reason people are more aware of earthquakes are the recent "mega-quakes". These quakes are those whose earthquake
is ever to have
Earthquake Datehit Japan, and the 5th most powerful earthquake in the world.
Time The 2011 quake off the coast of Thoku was a magnitude 9.0 (Mw) undersea megathrust earthquake. It was the most powerful Earthquake Date Time
1/1/2004 to 2/26/2014 1/1/2004 to 2/26/2014
greater than 8.0 and they often cause significant damage and lose of live. known earthquake ever to have hit Japan, and the 5th most powerful earthquake in the world.
Magnitude
Magnitude 8 to 9.1 Magnitude
8 to 9.1 8 to 9.1

Region of the World

Region of the World All Region of the World
All All

16 16 16
earthquakes selected earthquakes selected
earthquakes selected

Magnitude 9.0 near the east coast of Honshu, Japan Magnitude 9.0 near the east coast of Honshu, Japan

Magnitude
Magnitude
Magnitude 6.0 8.0
6.0 8.0
6.0 8.0
 8

This story was built using visualizations connected directly to the data source.
As earthquakes continue to be detected around the world, the visualizations in the
story will update to include the new data.

Stories help you understand whats going on beneath the surface. From start
to finish, they walk you through a series of events so that you can see whats
causing them to happen.

The data tells you whats happening, but the story guides you to an understanding
of why. And once you understand that cause, you can decide how to act.

Earthquakes: Are they are on rise?

Thousands of About two each year The megaquakes are Megaquake: The Megaquake: The More and more Most earthquakes But major
earthquakes are qualify as "major what has drawn Indian Ocean quake Japan quake of earthquakes are are being detected in earthquakes are not
detected each year earthquakes" attention in 2004. 2011. being detected the Pacific Rim happening more oft..

Magnitude
16K 4 to 9.1

15K

14K

13K
Since 1973 there has been a steady increase in
the number of earthquakes being recorded with
12K
a particularly strong increase in 2003 to 2005.
11K

10K
Number of earthquakes

9K

8K

7K

6K

5K

4K

3K

2K

1K

0K
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013

Earthquakes: Are they are on rise?

Thousands of About two each year The megaquakes are Megaquake: The Megaquake: The More and more Most earthquakes But major
earthquakes are qualify as "major what has drawn Indian Ocean quake Japan quake of earthquakes are are being detected in earthquakes are not
detected each year earthquakes" attention in 2004. 2011. being detected the Pacific Rim happening more oft..

Magnitude
5 to 9.1
2000

Region of the World

1800 APAC
EMEA
North & South Am..
1600

1400
Number of earthquakes

1200

1000

800

600

400

200

0
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
 9

When to Tell a Story

Dashboards versus data stories
Data stories may be groundbreaking, but theyre nothing new. John Snows cholera
map makes that clear. So why cant we just use dashboards to tell data stories?
M alaria af f ect s over half t he world populat ion, t he map below shows t he cases
and deat hs due t o M alaria f or each count ry. Concent rat ed around t he equat er
Sub-Saharan Af rica has t he largest problem.
Highlight af f ect ed count ries by
number of cases or deat hs
We can.
Cases

Just as Dr. Snow adapted a versatile mediumpaper, pen, and inkto share
a data story, so too can dashboards be shaped into compelling visual stories.
Total Worldwide Cases : 207,420,000
However, theyre already pretty darn good at what theyre usually designed for.

Lets take a look.

This visualization, by Matt Francis
s

of the Wellcome Trust Sanger When it comes to monitoring your datakeeping its daily pulsean interactive
Institute, shows cases of malaria
around the world. Once you see dashboard is a great tool. And if you notice a change or problem in your data, you
whats going onthat malaria can use a dashboard to investigate the cause. But when you need to communicate
is mainly striking Sub-Saharan
Africachances are you want to your findings to otherswhen you need to convince them to take actionyou
know why. And then you want to
need to show them why.
tell that story.

By putting data in a sequence, you can help viewers understand the context
and make useful connections between each point. Francis uses a series of
visualizations to guide readers through a thought process in a focused, user-
friendly way.

...t hat means over 1700 children, t he equivalent of 32 school buses, will die f rom
In 2013 an est imat ed 627,000 people, 90% of t hem children under 5 died f rom What could you spend $10 on?
malaria
M alaria, t hat s almost t he ent ire populat ion everyonsingle
of Washingt DC... day, t hat s one f or every minut e of every day
City
Item
Det roit Ticket t o t he M ovies
Bost on
Seat t le 3 gallons of gas
Denver Long Last ing Insect icidal Net
Washingt on
Deat hs f rom M alaria One mont h subscript ion t o rdio
Port land Pharrel Williams New Album
Oklahoma Cit y
Las Vegas 5 Dunkin Donut s Cof f ees
Kansas Cit y
10 KFC Hot Wings
At lant a
$1.00 $2.00 $3.00 $4.00 $5.00 $6.00 $7.00 $8.00 $9.00 $10.00 $11.00
0K 50K 100K 150K 200K 250K 300K 350K 400K 450K 500K 550K 600K 650K 700K
Population

Dashboards tell you whats happening, but stories explain why.

 10

Storytelling to the People

How to be a game-changer with data
What if youre not the one making decisions with data? In 1854, John Snow
believed that water was spreading cholera in London, but needed to convince
the citys leaders too.

Likewise, the former teacher John Savagewho recently wrote about Texas
struggling schoolshas a major hunch about whats causing their problems.
Could he do what Dr. Snow did in 1854: follow his instincts, dive into the data,
and tell a life-changing story with it?

Savages story impacts teachers all over the east side of Austin, Texas. And if
Austin's Teacher
you look at regionalTurnover Problem
data, its clear that Austins east side struggles more than
its (wealthier)
Austin's teacher
turnover rate has
west.
Teacher turnover is a
problem across
The problem is
particularly actue at
Salon.com published
a piece called "I
"At Reagan, 19 of
teachers were giving
A struggle between
teachers and a
In 2014 the S
wrote,
been above the nati.. Austin, although the.. some schools. taught at the worst s.. lessons in subjects f.. principal drove high .. "Multimillion-d

Use the slider to see rate..

2013

Reach program schools

NO
YES

Teachers
6
50
100
161

Avg. Turnover rate

10.0 22.0

Savage also alludes to recovery efforts in Austins schools. Did they work?
About six years after he left Pearce, an article in the Austin American-Statesman
reported on a new study that claimed teacher quality was higher in the west than
in the eastand that at Reagan High School in Northeast Austin, 19 percent of
teachers were giving lessons in subjects for which they were not certified. It went
on to describe an incentive program aimed at improving teacher performance.
 11

But Savage believes its poverty that needs to be addressed, not teacher
performance. And the turnover rate at Reagan only went up after that article
was published.

School name School type Avg. Turnover rate

10.0 22.0
60.0

Teachers
40.0 39
Mendez Middle school 60
80
20.0

0.0

60.0

40.0
Pearce Middle school

20.0

0.0

60.0

40.0
Reagan High school

20.0

0.0
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

About three years later, the Statesman reported that millions of dollars spent
on teacher incentives had done little to fix the problem. State data would show
teacher turnover to be higher than ever before.

By telling this whole storyfrom the emotional impact its had on real people to
the data that points to the true causea single person could change the game for
teachers and students across Austin.

From improving public education to preventing disease and better understanding

natural disasters, data stories have unlimited potential. What would happen if
there was an explosion of authorship around the world? Imagine the impact of
billions of people telling their stories with numbers.

Stories have always been for everyone. Since ancient times, weve been using
stories to conserve and pass on information. With game-changing inventions like
the printing press, widespread access to information has become a reality. Were
also seeing an explosion of authorship around the world today.

With a data story, hundreds of millions of rows can be distilled into a single
narrative. By following that narrative, its much easier to understand whats going
on beneath the surface. The data can tell you whats happening, and a story can
elucidate the cause.

Perhaps most importantly, though, stories motivate action. And in todays world of
information overload, its crucial to be able to make informed, actionable decisions
with all of your data.
 12

About the Authors

Jock Mackinlay

Jock Mackinlay is Tableaus Vice President of Visual Analysis. At Stanford

University, he pioneered the automatic design of graphical presentations of
relational information. He joined Xerox PARC in 1986, where he collaborated
with the User Interface Research group to develop many novel applications
of computer graphics for information access, coining the term information
visualization. Much of the fruits of this research can be seen in his book,
Readings in Information Visualization: Using Vision to Think. Jock has a PhD
in computer science from Stanford University.

Robert Kosara

Robert Kosara is a researcher in Tableaus Visual Analysis group. Before

he joined Tableau in 2012, he was a professor of computer science at UNC
Charlotte. Robert has created visualization techniques like parallel sets and
performed research into the perceptual and cognitive basics of visualization.
Recently, his research has focused on how to communicate data using tools
from visualization, and how storytelling can be adapted to incorporate data,
interaction, and visualization.

Michelle Wallace

Michelle Wallace is a product marketer at Tableau Software, where she tells

stories about how thought-leaders around the world are making an impact
with data. She holds a BA in English from Western Washington University with
a minor in astronomy. Prior to Tableau, Michelle was a magazine writer who
covered local interests and histories around the United States.
 13

About Tableau
Tableau Software helps people see and understand data. Tableau helps anyone quickly analyze,
visualize and share information. More than 19,000 customer accounts get rapid results with Tableau
in the office and on-the-go. And tens of thousands of people use Tableau Public to share data in
their blogs and websites. See how Tableau can help you by downloading the free trial at
www.tableausoftware.com/trial.

Additional Resources
Download Free Trial

Related Whitepapers
Why Business Analytics in the Cloud?

5 Best Practices for Creating Effective Campaign Dashboards

See All Whitepapers

Explore Other Resources

Product Demo

Training & Tutorials

Community & Support

Customer Stories

Solutions

Tableau and Tableau Software are trademarks of Tableau Software, Inc. All other company and product
names may be trademarks of the respective companies with which they are associated.