Visual Analytics with Tableau

Introduction

Visual Analytics with Tableau is intended to be a step‐by‐step introduction to the world of visual analytics. My hope is that you will find the provided examples to be useful for learning how to analyze your own data in Tableau. And how to effectively communicate the new insights gained from these analyses, just like the many Tableau users with whom I have had the privilege of working in recent years.

The book should be of interest to the following audiences:

Business professionals who make data‐informed decisions on a day‐to‐day basis.

Analysts and developers who create visualizations and dashboards for their organizations.

Data scientists who want to quickly understand the data in front of them, before perhaps running more sophisticated models on it.

Generally, anybody with access to data and with a desire to understand it.

To follow the contents of this book and to get started with Tableau, you do not need a background in mathematics or any programming experience. The book is suitable for beginners and for those who are looking for a practical introduction to the fields of data analysis and visualization.

That doesn't mean we will restrict ourselves to the basic functionality of Tableau. The first three chapters go through the essentials step by step. But this book goes deeper than that.

Building on this foundation, we will then look at more sophisticated use cases aimed at more experienced practitioners. We will cover a number of Tableau features that should be interesting even for more‐advanced users.

In a few places, you will encounter short calculations and programming scripts. These are kept on the simple side, so that anybody should be able to follow their logic. Of course, more‐sophisticated scenarios can be implemented in Tableau—either with Tableau's own calculated fields, which use a modern programming syntax, or via the integration of R, Python, or MATLAB. However, this is not the focus of the book. Instead of going deep into statistical programming, we will focus on the visual analytics functionality of Tableau.

STRUCTURE

This book has 10 chapters that generally build on each other. The progression of chapters is intended to support you on a continuous learning curve. Chapter 1 starts with an overview that should also help novices get a good first impression of Tableau's capabilities.

Subsequent chapters go deeper into various aspects of the visual analytics process. Chapter 2 details how to find and connect to different data sources. Chapter 3 is in some regards the heart of the book: it provides hands‐on instructions for how to build assorted types of data visualizations in Tableau. I focus on the chart types that my customers and I have found most helpful in commonly encountered use cases.

Chapter 4 introduces calculated fields that allow you to add custom computations to the data. As mentioned, I have tried to make this topic accessible to as wide an audience as possible. Chapter 5 builds on Chapter 4 and covers the more‐advanced, but very useful, Table Calculations and Level of Detail Calculations.

Chapter 6 looks at a very popular Tableau feature: maps. I have seen many Eureka! moments, when new Tableau users figured out how they can easily add their data to different types of maps. We will also talk about how to best bring in additional data without making your maps too cluttered.

Chapter 7 looks at how different statistical methods can be used to augment your visual analytics procedures to provide you with additional insights. We will cover forecasts, clusters, and trend lines, and we will also look at the integration of R, Python, and MATLAB for more advanced statistical modelling.

Chapter 8 shows how individual charts can be combined to create interactive dashboards that allow your colleagues to explore the data on their own terms. Related to that, in Chapter 9, we will look at the different options within the Tableau ecosystem for sharing your work with others: Tableau Server, Tableau Online, and Tableau Public.

Chapter 10 revisits the challenge of data preparation and data cleaning, but goes beyond what is covered in Chapter 2, by focusing on the new application Tableau Prep.

CONVENTIONS

To help you get the most from the text and keep track of what's happening, we've used a couple of conventions throughout the book.

NOTE Note boxes such as this one will provide insights into advanced Tableau functionality and working with different data structures.

TIP Tip boxes are intended to provide additional tips that should make your work with Tableau easier.

COMPANION WEBSITE

All sample data, updates, amendments, and recommended reading materials will be posted to the following website: http://www.visual‐analytics.org/with‐tableau.

Chapter 1

Introduction and Getting Started with Tableau

Tableau was created to empower people to analyze their data regardless of the level of their technical know‐how. At the core of Tableau is VizQL, an innovative visual query language that translates mouse inputs such as drag‐and‐drop into database queries. This allows the user to quickly find insights in their data and to share the results with others.

Crucially, it is not necessary to know what you are looking for or how you want to present your findings. Instead, with Tableau, you can immerse yourself in data. Through visual analysis, you will be able to unearth patterns and relationships in your data that you might not have known existed. In this regard, Tableau is different from other tools, which often require you to know beforehand in what form you want to display your data.

The purpose of this chapter is to introduce you to the different products that make up the Tableau application suite, the Tableau user interface, and to how Tableau processes your data. We will also introduce the sample dataset that is used throughout this book and provide a first glimpse of the possibilities that Tableau gives you for creating data visualizations.

By the end of this chapter you will be able to:

Install Tableau on your computer.

Identify data that is suitable for analysis.

Create your first data visualization in Tableau.

THE ADVANTAGES OF A MODERN ANALYTICS PLATFORM

The first thing you typically do in Tableau is to connect to a dataset. The data can come from simple files, databases, data cubes, data warehouses, Hadoop clusters, or even different cloud services such as Google Analytics. Next, you interact with the Tableau interface to query your data visually and to display the results in various types of charts and maps. Then, you can collate the individual charts in a dashboard in order to put them into the right context.

Finally, depending on the product used, there are different options for communicating the results with others, from sending individual workbooks, to embedding interactive dashboards, to sharing them on social media. Tableau helps you with both the analysis as well as the communication of results, by providing capabilities such as the creation and sharing of explanatory diagrams, data stories, and interactive dashboards (see Figure 1.1).

Figure 1.1 An interactive sales dashboard. We will build this in Chapter 8.

MY PERSONAL TABLEAU STORY

I first came across Tableau in 2009, when I was writing my thesis at CERN, the European Organization for Nuclear Research in Geneva. I was exploring the landscape of available tools for the visualization and communication of data because I was not happy with the clunky, inflexible solutions that were commonly used back then.

Like most of my colleagues at CERN, I spent a lot of time aggregating data in Python, a popular universal programming language, only to then visualize it in another tool, the command‐line tool GnuPlot. It was a struggle to keep all the scripts well maintained, and even small changes required a lot of time and effort.

When new data came in, the scripts had to be re‐run. The resulting visualizations were, of course, static and didn't offer any interactivity to the end user. And the software packages I used had a lot of dependencies that had to be resolved every time a new version became available.

When I eventually learned about Tableau, I was amazed by the ease of use of the graphical interface and the possibility of being able to interact with my data directly. Every time I dropped another measure or dimension onto the canvas, I got new insights from my data. What used to take me hours could now be done in minutes, and it was fun, to boot! The interactivity of the resulting dashboards and the ability to have them automatically refresh when the underlying data changed sealed the deal for me. I was a fan. I still feel as passionately about Tableau today as I did back then, and I hope to be able to impart some of that enthusiasm to the readers of this book.

THE TABLEAU APPLICATION SUITE

Some readers may have bought this book because they already have one or more Tableau products installed on their machine and would like to jump right in and learn how to use them. But for those who are not so familiar with the different Tableau products, here is a quick overview:

Tableau Desktop Tableau Desktop is an application for Windows and Mac, appreciated by both analysts and business users. In Tableau Desktop, you can connect to flat files (such as Excel and CSV files) and save your workbooks to your local hard drive. To tap into an organization's IT infrastructure, you can also use Tableau Desktop to connect to a host of different database solutions, and you can share your workbooks via Tableau Server or the cloud‐based Tableau Online.

Tableau Prep Tableau Prep is the latest addition to the Tableau product suite and is designed to help you prepare your data before you analyze it in Tableau Desktop. The visual interface allows you to quickly merge differently formatted datasets, clean the data, and unify the level of aggregation. Tableau Prep fits seamlessly into your analysis workflow.

Tableau Server Tableau Server is a platform for data analysis and is used by small family‐run businesses and large Fortune 500 companies alike. It is intended for the organization‐wide provision of data visualizations and dashboards that can be viewed in a browser and are frequently embedded into the organization's intranet.

Tableau Online Tableau Online is a Tableau‐hosted solution for storing and deploying dashboards. It provides similar functionality to Tableau Server but is a cloud‐based service. No purchase and maintenance of server hardware is necessary here.

Tableau Public Tableau Public is a hosting service for the publication of data visualizations to the web. It is used by newsrooms and bloggers but also by companies, research institutes, governmental bodies, and non‐governmental organizations that aim to get their data stories into the public eye. The interactive visualizations can be viewed in the browser directly on the Tableau Public platform, or they can be embedded into blogs and websites.

Tableau Reader Tableau Reader is a free desktop application that allows you to open and interact with Tableau workbook files that have been created in Tableau Desktop. However, it is not possible to make any changes to the visualizations in Tableau Reader.

NOTE   The figures throughout this book show Tableau Desktop version 2019.1, unless stated otherwise. The web‐edit screen of Tableau Server and Tableau Online contains a number of features that you might recognize from Tableau Desktop. But the functionality of the browser‐based products is still limited when it comes to creating new visualizations and dashboards. Therefore, I advise you to install Tableau Desktop on your machine, especially if you are still new to Tableau. The following section will provide more information about the system requirements and the installation process of Tableau Desktop.

INSTALLING TABLEAU DESKTOP

Installing Tableau Desktop is a simple process and takes only a few minutes. Therefore, this will be a very brief section.

System Requirements for Tableau Desktop

Before installing Tableau Desktop, be sure your machine meets the necessary requirements for this application. Tableau Desktop is available for Windows and Mac.

These are the official minimum requirements for a Windows installation:

Microsoft Windows 7 or later (64‐bit)

Microsoft Server 2008 R2 or later

Intel Pentium 4 or AMD Opteron processor or later

2 GB RAM

At least 1.5 GB of free hard disk space

These are the official minimum requirements for a Mac installation:

iMac/MacBook 2009 or later

OS X 10.10 or later

At least 1.5 GB of free hard disk space

Should you wish to work with large datasets, I recommend the following additional specifications:

Latest service pack or update for your operating system

Intel Core i3/i5/i7/i9 or AMD FX processor or later

At least 8 GB RAM

Solid‐state drive (SSD) with at least 20 GB of free space

Full‐HD resolution (1920 × 1080 pixels) or higher with 32‐bit color depth

Downloading and Installing Tableau Desktop

If you don't already have Tableau Desktop installed on your machine, use this link to download the latest trial version: https://www.tableau.com/products/desktop.

Make sure you are logged in to your machine as administrator and that you have the rights to install software on the machine. Run the installer as you normally would, given your operating system:

On a Windows Machine Open the setup (EXE) file, and accept any safety prompts from your