A Round of Introductions
● Pranav Dar
● Senior Content Strategist and
Program Lead
● ~10 years of experience in various
Visualization and Analytics roles
Can you Count the Number of Circles?
Methodology for this Training
Business Sense + Tableau = Data Visualization Expert
Getting Started with Tableau
● The Power of Data Visualization
● Understanding the different Tableau products
Understanding the Length and Breadth of Tableau
● Navigating the Tableau interface
● Connecting Tableau to different data sources
Setting up the Problem Statement
● Understanding the Problem Statement
● Importing Data into Tableau
● Getting Familiar with the Variables
Getting Hands-On with Tableau
● Create your first visualization in Tableau
● Labels, formatting and colors in Tableau
● Filters and Excel-like Pivot Structures
● Learn how to export your Tableau worksheet
Different Chart Types in Tableau
● Learn how to design various charts in Tableau
● Line charts, map visualizations, bubble charts, scatter plots, etc.
● Dual Axis charts in Tableau
Calculated Fields and Parameters
● Feature Engineering in Tableau using Calculated Fields
● Parameter Control in Tableau to Enhance User Experience
Joining and Blending Data in Tableau
● Joins and their Different Types
● Data blending (advanced topic)
● Data joining v Data blending
Building Powerful Dashboards in Tableau
● Understand Dashboards and their importance
● Putting together the different Dashboard components
● Creating an effective business industry-level Dashboard
The Art of Storyboarding in Tableau
● Understand the importance of storytelling for business
● Storyboarding in Tableau and how it’s different from dashboarding
● Build an awesome Storyboard in Tableau
What we will cover in this training
● Getting Started with Tableau
● Understanding the Length and Breadth of Tableau
● Setting up the Problem Statement
● Getting Hands-on with Tableau
● Different Chart Types in Tableau
● Calculated Fields and Parameters
● Joining and Blending
● Building Powerful Dashboards in Tableau
● The Art of Storyboarding in Tableau
● Final Assessment
The Power of Visualization
How Many Squares are in a Rubik’s Cube?
6 Sides, 54 Squares!
The Two Hemispheres of our Brain
The Two Hemispheres of our Brain
● Reading
● Writing
● Speaking
● Reasoning
The Two Hemispheres of our Brain
● Reading ● Visual Perception
● Writing ● Patterns
● Speaking ● Music
● Reasoning ● Expressions
The 3 Vs of Communication
The 3 Vs of Communication
1. Verbal
2. Vocal
3. Visual
Visualizing the Future!
https://www.washingtonpost.com/graphics/national/eclipse
Visualizations have been Around for Centuries
Data is Generated Everywhere & Everyday
We are generating data at an unprecedented pace.
What is Data Visualization and Why We Should use it
Data is Generated Everywhere & Everyday
We are generating data at an unprecedented pace.
Visualization in a Data Science Scenario
So, What is Data Visualization?
Data Visualization is the ability to turn data
into visible and tangible insights that people
can intuitively understand.
Why Should We Use Data Visualization?
To Explore
Why Should We Use Data Visualization?
To Explore
To Analyze
Why Should We Use Data Visualization?
To Explore
To Analyze
To Present
4 Key Elements of an Effective Data Visualization
The 4 Elements of an Effective Visualization
Data
Business
Intuition
Design
Story
Element #1: Data
Data
Element #2: Design
Data
Design
Element #3: Story
Data
Design
Story
Element #4: Business Intuition
Data
Design
Story
Business
Intuition
The 4 Elements of an Effective Visualization
Data
Business
Intuition
Design
Story
Getting Familiar with the Tableau Ecosystem
Understanding the Problem Statement
Welcome to the Superstore
Examples of Other Superstores
Amazon Go Walmart
Understanding the Problem Statement
• A Superstore in the United
States has recently set up
its data science team
• The store has rich data
about its products, regions,
and much more
• The leadership team wants
to know which products,
regions, categories,
segments and so on are
doing well
Understanding the Problem Statement
Insights Time!
Let’s quickly check how much of yesterday’s session you can remember. ☺
• How do we separate the First Name and the Last Name from a column?
• Measures are blue and Dimensions are Green. True or False?
• Data Extract is a subset of the data that doesn’t need to be refreshed regularly. True or False?
• Which is the category with the highest sales?
• Which segment has the lowest sales?
Different Chart Types in Tableau
● Learn how to design various charts in Tableau
● Line charts, map visualizations, bubble charts, scatter plots, etc.
● Dual Axis charts in Tableau
Calculated Fields and Parameters
● Feature Engineering in Tableau using Calculated Fields
● Parameter Control in Tableau to Enhance User Experience
The Different Chart Types
*Image courtesy EDraw
The Different Chart Types
“There is magic in graphs. The profile of a curve reveals in a flash
a whole situation — the life history of an epidemic, a panic, or an
era of prosperity. The curve informs the mind, awakens the
imagination, convinces.” - Henry D. Hubbard
“Graphical excellence is that which gives to the viewer the
greatest number of ideas in the shortest time with the least ink
in the smallest space.” - Edward Tufte
The ‘Show Me’ Button in Tableau
Line Scatter
Bar Charts Heat Maps
Charts Plots
Stacked Dual Axis
Histograms Geospatial
Charts Charts
Bubble Area
Box Plots Pie Charts
Charts Charts
Advanced Charts in Tableau (Waterfall)
Advanced Charts in Tableau (Gantt)
Time Series Charts
What is a Time Series?
Time Series can be Represented as..
Line charts!
Why do we Need Line Charts in Business?
To Monitor
Changes over Time
Why do we Need Line Charts in Business?
To Monitor
Changes over Time
To Analyze Trends
Why do we Need Line Charts in Business?
To Monitor
Changes over Time
To Analyze Trends
To Compare
Patterns
*Image courtesy The Hindu
Why do we Need Line Charts in Business?
To Monitor
Changes over Time
To Analyze Trends
To Compare
Patterns
Map Visualizations
The Rise of Location Data
The Rise of Location Data
What is Geospatial Analysis?
Country/Region City State Postal Code Region Sales Quantity Discount Profit
United States Henderson Kentucky 42420 South 261.96 2 0 41.9136
United States Henderson Kentucky 42420 South 731.94 3 0 219.582
United States Los Angeles California 90036 West 14.62 2 0 6.8714
957.577
United States Fort Lauderdale Florida 33311 South 5 0.45 -383.031
5
United States Fort Lauderdale Florida 33311 South 22.368 2 0.2 2.5164
United States Los Angeles California 90032 West 48.86 7 0 14.1694
United States Los Angeles California 90032 West 7.28 4 0 1.9656
United States Los Angeles California 90032 West 907.152 6 0.2 90.7152
United States Los Angeles California 90032 West 18.504 3 0.2 5.7825
United States Seattle Washington 98103 West 407.976 3 0.2 132.5922
What is Geospatial Analysis?
Use Cases of Geospatial Analysis
Transportation &
Sales & Marketing Situational Intelligence
Logistics
Strategic Growth
Risk Analysis Satellite Imagery
Optimization
Calculated Fields
Calculated Fields in Tableau
What are Calculated Fields in Tableau?
Calculated fields allow us to create new columns (or
data) from our already existing data. Think of it like the
power to do feature engineering in Tableau!
What can you do with Calculated Fields?
Convert data
Segment data Aggregate data
type
Filter results Calculate ratios
And much, much more!
Types of Calculations
Basic LOD Table
Calculations Expressions Calculations
Transform values at: Transform values at: Transform values at:
● Data source ● Data source ● Visualization
● Visualization ● Visualization
Data source = Row-level calculation
Visualization = Aggregate-level calculation
Example of a Calculated Field
Memory Jogging Time
Let’s quickly check how much of yesterday’s session you can remember. ☺
• What was the formula we used to count the number of orders?
• How can we reverse the colors in any Tableau visualization?
• Can you recall the steps to correct locations if Tableau cannot automatically detect it?
• Can you spot what's wrong with this formula?
Calculated Fields and Parameters
● Feature Engineering in Tableau using Calculated Fields
● Parameter Control in Tableau to Enhance User Experience
Joining and Blending Data in Tableau
● Joins and their Different Types
● Data blending (advanced topic)
● Best Practices
Building Powerful Dashboards in Tableau
● Understand Dashboards and their importance
● Putting together the different Dashboard components
● Creating dynamic interlinked Dashboards
Parameters in Tableau
Parameters in Tableau
What are Parameters in Tableau?
Parameters are dynamic values that can
replace constant values in calculations, filters,
and reference lines.
4 Steps to Create Parameters
1) Create a new parameter
2) Show parameter
3) Create a calculated field using the parameter
4) Apply this calculation on your Tableau sheet
A Sales/Profit Parameter
Data Joining in Tableau
The Concept of Joins
What is a Join?
Joining is a method for combining the related data among common fields in
different tables. The result of combining data using a join is a virtual table
that is extended horizontally by adding columns of data.
Types of Joins in Tableau
Inner Join
Table 1 Table 2
Name State Quantity Name Title Tenure (in years)
Neha California 2 Scott Data Scientist 2
Adam Florida 4 Jasna Sales Head 4
Scott Illinois 2 John Marketing Intern 1
Jill Colorado 1 Neha Product Lead 3
Rohit California 3
Inner Join
Table 1 Table 2
Name State Quantity Name Title Tenure (in years)
Neha California 2 Scott Data Scientist 2
Adam Florida 4 Jasna Sales Head 4
Scott Illinois 2 John Marketing Intern 1
Jill Colorado 1 Neha Product Lead 3
Rohit California 3
Inner Join
Table 1 Table 2
Name State Quantity Name Title Tenure (in years)
Neha California 2 Scott Data Scientist 2
Adam Florida 4 Jasna Sales Head 4
Scott Illinois 2 John Marketing Intern 1
Jill Colorado 1 Neha Product Lead 3
Rohit California 3
Inner Join
Name State Quantity Name (Table 2) Title Tenure (in years)
Neha California 2 Neha Product Lead 3
Scott Illinois 2 Scott Data Scientist 2
Left Join
Table 1 Table 2
Name State Quantity Name Title Tenure (in years)
Neha California 2 Scott Data Scientist 2
Adam Florida 4 Jasna Sales Head 4
Scott Illinois 2 John Marketing Intern 1
Jill Colorado 1 Neha Product Lead 3
Rohit California 3
Left Join
Table 1 Table 2
Name State Quantity Name Title Tenure (in years)
Neha California 2 Scott Data Scientist 2
Adam Florida 4 Jasna Sales Head 4
Scott Illinois 2 John Marketing Intern 1
Jill Colorado 1 Neha Product Lead 3
Rohit California 3
Left Join
Name State Quantity Name (Table 2) Title Tenure (in years)
Neha California 2 Neha Product Lead 3
Adam Florida 4
Scott Illinois 2 Scott Product Lead 3
Jill Colorado 1
Rohit California 3
Right Join
Table 1 Table 2
Name State Quantity Name Title Tenure (in years)
Neha California 2 Scott Data Scientist 2
Adam Florida 4 Jasna Sales Head 4
Scott Illinois 2 John Marketing Intern 1
Jill Colorado 1 Neha Product Lead 3
Rohit California 3
Right Join
Table 1 Table 2
Name State Quantity Name Title Tenure (in years)
Neha California 2 Scott Data Scientist 2
Adam Florida 4 Jasna Sales Head 4
Scott Illinois 2 John Marketing Intern 1
Jill Colorado 1 Neha Product Lead 3
Rohit California 3
Right Join
Name State Quantity Name (Table 2) Title Tenure (in years)
Scott Illinois 2 Scott Data Scientist 2
Jasna Sales Head 4
John Marketing Intern 1
Neha California 2 Neha Product Lead 3
Full Outer Join
Table 1 Table 2
Name State Quantity Name Title Tenure (in years)
Neha California 2 Scott Data Scientist 2
Adam Florida 4 Jasna Sales Head 4
Scott Illinois 2 John Marketing Intern 1
Jill Colorado 1 Neha Product Lead 3
Rohit California 3
Full Outer Join
Name State Quantity Name (Table 2) Title Tenure (in years)
Neha California 2 Neha Product Lead 3
Adam Florida 4
Scott Illinois 2 Scott Data Scientist 2
Jill Colorado 1
Rohit California 3
Jasna Sales Head 4
John Marketing Intern 1
Combining Data from Multiple Sources
What is Data Blending?
Data blending is a method for combining data from multiple sources.
We can bring in additional information from a secondary data source
and Data blending displays it with data from the primary data source
directly in the same view.
Data Joining v Data Blending
Data Joining Data Blending
● Data comes from the same source ● Data comes from multiple sources
● Joins the two sheets/tables ● Keeps the two data sources separate
● Joins happen at the row-level ● Results are aggregated to the
appropriate level
Data Joining v Data Blending
Data Source 1
Order ID Order Date Ship Date .... State Sub-Category Sales
CA-2018-152156 11/8/2018 11/11/2018 .... Kentucky Bookcases 261.96
CA-2018-152156 11/8/2018 11/11/2018 .... Kentucky Chairs 731.94
CA-2018-138688 6/12/2018 6/16/2018 .... California Labels 14.62
US-2017-108966 10/11/2017 10/18/2017 .... Florida Tables 957.5775
US-2017-108966 10/11/2017 10/18/2017 .... Florida Storage 22.368
Data Source 2
Month State Target
1-19 California 12,000
2-19 Florida 8,000
3-19 Arizona 6,000
4-19 Kentucky 9,000
Data Joining v Data Blending
Data Source 1
Order ID Order Date Ship Date .... State Sub-Category Sales
CA-2018-152156 11/8/2018 11/11/2018 .... Kentucky Bookcases 261.96
CA-2018-152156 11/8/2018 11/11/2018 .... Kentucky Chairs 731.94
CA-2018-138688 6/12/2018 6/16/2018 .... California Labels 14.62
US-2017-108966 10/11/2017 10/18/2017 .... Florida Tables 957.5775
US-2017-108966 10/11/2017 10/18/2017 .... Florida Storage 22.368
Data Source 2
Month State Target
1-2017 California 12,000
2-2017 Florida 8,000
3-2017 Arizona 6,000
4-2017 Kentucky 9,000
Data Joining v Data Blending
Data Source 1
Order ID Order Date Ship Date .... State Sub-Category Sales
CA-2018-152156 11/8/2018 11/11/2018 .... Kentucky Bookcases 261.96
CA-2018-152156 11/8/2018 11/11/2018 .... Kentucky Chairs 731.94
CA-2018-138688 6/12/2018 6/16/2018 .... California Labels 14.62
US-2017-108966 10/11/2017 10/18/2017 .... Florida Tables 957.5775
US-2017-108966 10/11/2017 10/18/2017 .... Florida Storage 22.368
Data Source 2
Month State Target
1-2017 California 12,000
2-2017 Florida 8,000
3-2017 Arizona 6,000
4-2017 Kentucky 9,000
Welcome to the World of Dashboards!
The Power of Dashboards
The Power of Dashboards
The Power of Dashboards
What is a Dashboard?
A dashboard is a visual interface that provides multiple views of
key performance indicators (KPIs) for a business or a process.
Elements of an Impactful Dashboard
Elements of an Impactful Dashboard
Know your Audience
Elements of an Impactful Dashboard
Know your Audience Structured Story
Elements of an Impactful Dashboard
Know your Audience Structured Story Key Takeaway(s)
Use Cases of Dashboards
Interlinking
Reports
Use Cases of Dashboards
Interlinking
Reports
Information
Sharing
*Image courtesy Tableau
Use Cases of Dashboards
Interlinking
Reports
Information
Sharing
Measuring and
Monitoring
Performance
*Image courtesy Tableau
Use Cases of Dashboards
Interlinking
Reports
Information
Sharing
Measuring and
Monitoring
Performance
Forecasting!
Creating the Blueprint for our Dashboard
Why Tableau?
Blueprint for our Dashboards
Primary Dashboard
State Sales
Blueprint for our Dashboards
Primary Dashboard Secondary Dashboard
Back
Sub-Category Sales
---------------------------------
State Sales
Trend/Forecast
Sales
Blueprint for our Dashboards
Primary Dashboard Secondary Dashboard
Back
Sub-Category Sales
---------------------------------
State Sales
Trend/Forecast
Sales
Back
Detailed Report
Blueprint for our Dashboards
Primary Dashboard Secondary Dashboard
Back
Sub-Category Sales
---------------------------------
State Sales
Trend/Forecast
Sales
Back
Detailed Report
Calculated Fields and Parameters
● Feature Engineering in Tableau using Calculated Fields
● Parameter Control in Tableau to Enhance User Experience
Joining and Blending Data in Tableau
● Joins and their Different Types
● Data blending (advanced topic)
● Best Practices
Building Powerful Dashboards in Tableau
● Understand Dashboards and their importance
● Putting together the different Dashboard components
● Creating dynamic interlinked Dashboards
The Art of Storyboarding in Tableau
● Understand the importance of storytelling for business
● Storyboarding in Tableau and how it’s different from dashboarding
● Build an awesome Storyboard in Tableau
The Art of Storytelling
The Art of Storytelling
The Art of Storytelling
Here’s a Real-world Story
*All images taken from Zendesk.com
The Art of Storytelling
“Stories are how we remember. We tend to
forget lists and bullet points.” - Robert McKee,
Best-Selling Author and Lecturer
Sometimes, our Storytelling Goal is Obvious
Delivering a TED Talk
Sometimes, our Storytelling Goal is Obvious
Addressing your employees about a new strategy
Sometimes, our Storytelling Goal is Obvious
Launching a new product
Wait, What if the Topic is Mundane?
• Delivering a project status to the client
• Talking about budgetary/accounting
issues
• Elaborating on technical details
Understand the Objective of your Presentation
Giving a presentation is about persuasion,
not distributing information.
Engage Through Storytelling – Nancy Duarte
Source: https://www.youtube.com/watch?v=GY3u6QuZXEs
3-Step Storytelling Framework
The Art of Storytelling
Establishing a Framework for
Storytelling
Many People Start with a Vague Theme
We have a new strategy for customer
satisfaction.
The credit risk model didn’t work out
as we thought so we changed the way
we approached the problem.
We are going to increase focus on
product testing.
We plan to revamp the homepage of
our site.
3-Step Storytelling Framework
Villain
Victim
Hero
A Simple Software Example
An all too familiar situation!
Plug in our Framework
Villain The software update killing the battery life of the phone.
Victim People who updated their phone.
A new update that will either roll back the previous iteration or
Hero
improve on the current release.
Understanding the Problem Statement
You have recently joined ABC company as a data
scientist.
ABC is a nascent e-commerce website with a
catalogue of thousands of products across a variety of
functions. All the users are currently served a ‘Most
popular products’ section since there is no
recommendation engine on the portal.
Use the Villain-Victim-Hero framework to prepare a
story which you will then use for your eventual
presentation.
Plug in our Framework
No personalization on the e-commerce portal for customers
Villain owing to a lack of a recommendation engine.
Victim ● All customers visiting the portal - both new and loyal ones.
● Our company’s bottom line!
Build and deploy a recommendation engine to personalize
Hero
recommendations for customers and drive up revenue.
Sketching the Story Blueprint
Group Exercise: Storytelling in Tableau
The Superstore leadership team was thrilled by our profit story - they appreciated the
region-by-region breakdown. They want to dive a level deeper and have asked you to create
another storyboard. This should showcase the sales AND profit by segment and category
(both features should be included).
Use appropriate captions and annotations to bring out your story and highlight the parts the
leadership team should notice.
Hint: You might find the use of parameters useful here (recall the 2 in 1).
How About a Donut?
