Cover image: Andy Kriebel
Cover design: I FOR IDEAS (i-for-ideas.com)

Published by John Wiley & Sons, Inc., Hoboken, New Jersey.
Published simultaneously in Canada.

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750–8400, fax (978) 646–8600, or on the Web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748–6011, fax (201) 748–6008, or online at www.wiley.com/go/permissions.

Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.

For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762–2974, outside the United States at (317) 572–3993, or fax (317) 572–4002.

Wiley publishes in a variety of print and electronic formats and by print-on-demand. Some material included with standard print versions of this book may not be included in e-books or in print-on-demand. If this book refers to media such as a CD or DVD that is not included in the version you purchased, you may download this material at http://booksupport.wiley.com. For more information about Wiley products, visit www.wiley.com.

A catalogue record for this book is available from the Library of Congress

ISBN 978-1-119-51077-2 (Paper)
ISBN 978-1-119-51072-7 (ePDF)
ISBN 978-1-119-51079-6 (ePub)

CONTENTS

Foreword

Acknowledgments

About the Authors

Part I

Introduction
1. What Is Makeover Monday?
2. How Did Makeover Monday Start?
3. The Community Project
4. Pillars of Makeover Monday
5. How to Use this book

Part II

Chapter 1 Habits of a Good Data Analyst
1. Approaching Unfamiliar Data
2. Analysis versus Visualization
3. Take Your Time
4. Build Context Through Additional Research
5. Find Insights
6. Communicate Clearly
7. Ask Questions
8. Summary
Chapter 2 Data Quality and Accuracy
1. Working with Incomplete Data
2. Overcounting Data
3. Sense-Checking Data
4. Is the Data Aggregable?
5. Substantiating Claims with Data
6. Summary
7. Note
Chapter 3 Know and Understand the Data
1. Using Appropriate Aggregations
2. Explaining Metrics
3. Identifying and Correcting Mistakes
4. Time Series Analysis
5. Summary
6. Notes
Chapter 4 Keep It Simple
1. What Is Simplicity?
2. Simplicity in Design
3. Simplicity in Analysis
4. Simplicity in Storytelling
5. Summary
6. Note
Chapter 5 Attention to Detail
1. Typos
2. Punctuation
3. Formatting
4. Crediting Images and Data Sources
5. Summary
6. Note
Chapter 6 Designing for the Audience
1. Creating an Effective Design
2. Designing for Mobile
3. Using Visual Cues for Additional Information
4. Using Icons and Shapes
5. Storytelling
6. Reviewing Your Work to Improve Its Quality
7. Summary
8. Note
Chapter 7 Trying New Things
1. Developing a Sharing Culture
2. Summary
Chapter 8 Iterate to Improve
1. Why Iterate?
2. Examples of Effective Iteration
3. Giving and Receiving Feedback
4. Summary
5. Notes
Chapter 9 Effective Use of Color
1. The Significance of Color in Data Visualization
2. Using Color to Evoke Emotions
3. Using Color to Create Associations
4. Using Color to Highlight
5. Best Practices for Using Color
6. Using Background Colors
7. Using Text as a Color Legend
8. Summary
9. Notes
Chapter 10 Choosing the Right Chart Type
1. Area Charts
2. Stacked Bar Charts
3. Diverging Bar Charts
4. Filled Maps
5. Donut and Pie Charts
6. Packed Bubble Charts
7. Treemaps
8. Slopegraphs
9. Connected Scatterplots
10. Circular Histograms
11. Radial Bar Charts
12. Resources
13. Summary
14. Note
Chapter 11 Effective Use of Text
1. Effective Titles and Subtitles
2. What Is Your Key Message?
3. Instructions and Explanations
4. Summary
Chapter 12 Using Context to Inform
1. The Importance of Context
2. Using Simple Metrics
3. Methods for Communicating Context
4. Summary
5. Note

Part III

The Community
1. Long-Term Contributors
2. Educators
3. Employers
4. Organizations
5. Nonprofits
6. Social Impact
7. Makeover Monday Live Events
8. Makeover Monday Enterprise Edition

Source Lines

Index

End User License Agreement

List of Illustrations

Introduction

Figure I.1: The alphabetical sorting of the pie chart makes ranking the players difficult.
Figure I.2: A bar chart is much better for representing the ranking of the players.
Figure I.3: Collection of the images created by Andy & Andy for Makeover Monday 2016.
Figure I.4: Summary of Makeover Monday 2016 with the 100% club.
Figure I.5: Collection of the images created by Andy & Eva for Makeover Monday 2017.
Figure I.6: 500 people attended Makeover Monday Live at Tableau Conference 2017.
Figure I.7: Example of Sarah Bartlett’s early work.
Figure I.8: Example of Sarah Bartlett’s early work.
Figure I.9: Example of Sarah Bartlett’s more recent work.
Figure I.10: Example of Sarah Bartlett’s more recent work.
Figure I.11: A sample of the visualizations Eva created for Makeover Monday 2017.

Chapter 1

Figure 1.1 Upsets in the NCAA basketball tournament championship game.
Figure 1.2 Median age of mothers in the United States when giving birth.
Figure 1.3 Scatterplot of YouTube views vs. channel subscribers.
Figure 1.4 Analysis of life expectancy for different income levels.
Figure 1.5 Visualization of Chicago taxi trips by Pooja Gandhi.
Figure 1.6 The growth of Apixaban prescriptions in the United Kingdom.
Figure 1.7 Example of combining highlighting and context.
Figure 1.8 Combine multiple charts to allow aggregate and detailed analysis.
Figure 1.9 Comparison of White House salaries by Sean Hughes.
Figure 1.10 Plot of solar eclipses by Marc Reid.
Figure 1.11 Design can be used to educate your audience.
Figure 1.12 Turkey vulture migration patterns by Lindsay Betzendahl.
Figure 1.13 Simplify the data to communicate the analysis more effectively.
Figure 1.14 Comparison of ethnicities across major sports leagues in North America.
Figure 1.15 Allowing the user to answer their own question draw them into the analysis.
Figure 1.16 Including additional data is an effective method for explaining the why.

Chapter 2

Figure 2.1 iPhone units sold by year.
Figure 2.2 Year over year change of iPhone units sold.
Figure 2.3 iPhone units sold by quarter.
Figure 2.4 Comparing quarterly iPhone units sold by year.
Figure 2.5 Andy’s American Express spending by Shawn Levin.
Figure 2.6 Andy’s American Express spending by Ravi Mistry.
Figure 2.7 LinkedIn ranking of professional skills by Matt Chambers.
Figure 2.8 Gaps and annotations help make it clear that data is missing.
Figure 2.9 Comparison of beef and chicken consumption for all years.
Figure 2.10 Comparison of beef and chicken consumption with the data before 1965 removed.
Figure 2.11 Count of the total medals won at the SEA Games.
Figure 2.12 Count of medals won by event at the SEA Games.
Figure 2.13 Trump’s Tweets by Mike Cisneros.
Figure 2.14 Dot plot of household income by state.
Figure 2.15 Obesity rates by age group in Pennsylvania compared to all other states.
Figure 2.16 Average obesity rate by state by year.
Figure 2.17 Properly aggregated obesity rate by state by year.
Figure 2.18 Using the correct formula to calculate a profit ratio.
Figure 2.19 An incorrect overall profit ratio resulting from averaging the individual profit ratios.
Figure 2.20 Table of tourism spending in districts in New Zealand.
Figure 2.21 An average of an index is inaccurate at higher levels of aggregation.
Figure 2.22 Analysis of potato prices in Europe by Luke Stoughton.

Chapter 3

Figure 3.1 Pre-aggregated data.
Figure 3.2 Unweighted average of response rates.
Figure 3.3 Weighted average of response rates.
Figure 3.4 Internet access per capita by country.
Figure 3.5 Alcohol consumption rate in Britain compared to 2005.
Figure 3.6 Percent change in alcohol consumption rate in Britain compared to 2005.
Figure 3.7 Rank of regions by sales by product category.
Figure 3.8 Box plot of the rank of regions by sales by product category.
Figure 3.9 Dot plot of sales by region by product category.
Figure 3.10 Box plot of sales by region by product category.
Figure 3.11 Sales summarized to the region and quarter level.
Figure 3.12 Table of average RBIs per player.
Figure 3.13 Table of median RBIs across all players.
Figure 3.14 Bar chart of a count of products sold.
Figure 3.15 Bar chart of unique products sold.
Figure 3.16 Total BMW and Mercedes-Benz prices over time.
Figure 3.17 Median BMW and Mercedes-Benz prices over time.
Figure 3.18 Cumulative visitors to National Memorials.
Figure 3.19 Cumulative visitors to National Memorials since they opened.
Figure 3.20 Relative temperature change by country.
Figure 3.21 Absolute temperature change by country.
Figure 3.22 Duplicated count of medals won in the SEA Games.
Figure 3.23 Correct count of medals won in the SEA Games.
Figure 3.24 Time series plot of CO₂ concentration.
Figure 3.25 Time series plot of CO₂ concentration with a linear trend line.
Figure 3.26 Seasonality plot of CO₂ concentration.
Figure 3.27 Box plot of CO₂ concentration.
Figure 3.28 Dot plot of CO₂ concentration with lines for the median, 25th percentile, and 75th percentile.
Figure 3.29 Monthly CO₂ concentration.
Figure 3.30 Six-month moving average of CO₂ concentration.
Figure 3.31 Change in home prices compared to the first time period.
Figure 3.32 Change in home prices compared to a selected time period.
Figure 3.33 Monthly CO₂ concentration.
Figure 3.34 Cycle plot of CO₂ concentration.
Figure 3.35 Heat map of median CO₂ levels.
Figure 3.36 Heat map of median CO₂ levels in Colorado.
Figure 3.37 Heat map of median CO₂ levels in Florida.

Chapter 4

Figure 4.1 Horizontal layout that fills the space nicely.
Figure 4.2 Vertical layout that flows down the page.
Figure 4.3 White space helps provide focus on the important data.
Figure 4.4 A single chart can be enough to reveal the analysis.
Figure 4.5 Geographic treemap of high dosage inhaled steroids.
Figure 4.6 A well-chosen title and simple visualizations can tell an effective story.
Figure 4.7 Example of using color to highlight and removing unnecessary details.
Figure 4.8 Simplifying the data can help simplify the story.
Figure 4.9 Example of a clean, simple design with minimal colors.

Chapter 5

Figure 5.1 Visual history of Le Tour de France.
Figure 5.2 Example of minimalist line charts.
Figure 5.3 Removing labels from the bars created an uncluttered look.
Figure 5.4 Using color and labels sparingly on a bar chart to indicate the important data.
Figure 5.5 Example of unsynchronized axes using the same unit (lbs).
Figure 5.6 When using the same unit on two axes, the axes should be synchronized.
Figure 5.7 Small multiple-tile map.
Figure 5.8 Example of too many different background colors.
Figure 5.9 Example of simple background colors.
Figure 5.10 Example of effective use of color, fonts, and formatting.
Figure 5.11 Perfect alignment helps guide the audience.
Figure 5.12 Labeling the ends of a line adds context.
Figure 5.13 Labeling specific data points provides focus.
Figure 5.14 Example of using lines to connect sections.
Figure 5.15 Example of color choices that work well together, making them easy to read.
Figure 5.16 When using icons, either create them yourself or give credit as dictated by the creator.

Chapter 6

Figure 6.1 Performance dashboard for the Washington D.C. Metro.
Figure 6.2 Sketching grid by Caroline Beavon.
Figure 6.3 Dashboard of stolen bikes in London.
Figure 6.4 Line chart of sales over time.
Figure 6.5 Big numbers are great for KPIs. Include context if possible.
Figure 6.6 Selection filters take up less space than radio button filters, making them more accessible on mobile devices.
Figure 6.7 German auto production mobile dashboard.
Figure 6.8 Colors that are too similar to one another to work on mobile devices.
Figure 6.9 High-contrast colors work well on mobile devices.
Figure 6.10 Use highlighting to make the most important data obvious.
Figure 6.11 Mobile design using highlighting and contrast.
Figure 6.12 Mobile dashboard of tuition in the US.
Figure 6.13 Information icons are useful for providing extra information to the audience.
Figure 6.14 Icons can help relate the data to the topic.
Figure 6.15 Example of proper image attribution.
Figure 6.16 Too many icons create a cluttered mess.
Figure 6.17 Choosing a subset of the data can make a story more focused.
Figure 6.18 Long-form visualization can be useful for guiding the audience through the analysis.
Figure 6.19 Life expectancy at birth by Mike Cisneros.

Chapter 7

Figure 7.1 Ethnicity distribution in Major League Baseball, by Slavik Glazyrin.
Figure 7.2 Circular chart of gender salary disparity in Australia, by Adam Crahen.
Figure 7.3 Circular time plot, by Jamie Smyth.
Figure 7.4 Trump’s face revealed through a series of tweets, by Curtis Harris.
Figure 7.5 Gamma pattern of solar eclipses, by Athan Mavrantonis.
Figure 7.6 How to find the best place to see an eclipse, by Mike Cisneros.
Figure 7.7 Waffle chart of G-7 employment, by Pooja Gandhi.
Figure 7.8 Waffle chart of 100 people, by Charlie Hutcheson.
Figure 7.9 Small multiple tile map of U.S. household income.
Figure 7.10 Tile map of Africa, visualizing internet usage over time between 1990 and 2015.
Figure 7.11 Line chart tile map of France’s population.
Figure 7.12 Tile map of the deprivation index of Scottish local authorities.
Figure 7.13 Gold prices vs. oil prices using line to divide up the view.
Figure 7.14 Visual history of African American players in Major League Baseball.

Chapter 8

Figure 8.1 A waterfall approach to data visualization projects.
Figure 8.2 An agile approach to data visualization projects.
Figure 8.3 A very “busy” visualization comparing the price of crude oil and gold.
Figure 8.4 A much simpler way of showing the relationship between oil and gold prices.
Figure 8.5 A green-to-red color palette is used to differentiate ozone levels.
Figure 8.6 A blue-to-red color palette makes this visualization colorblindness friendly.
Figure 8.7 An eye-catching but difficult to read vertical timeline.
Figure 8.8 The horizontal timeline creates a more intuitive design that is easier to comprehend.
Figure 8.9 Small multiple line charts without zero lines make it difficult to compare different states.
Figure 8.10 Small multiple area charts with a clear zero line make comparisons easier and show changes more clearly.
Figure 8.11 The single color of the dot plot reduces the visibility of differences between states.
Figure 8.12 Applying a diverging color palette highlights differences and trends more clearly.
Figure 8.13 White lines with changing thickness for non-UK countries reduce the impact of the pink line representing the UK.
Figure 8.14 Reducing the thickness of all non-UK country lines helps focus on the key message about the UK GDP.
Figure 8.15 The yellow background is uncomfortably bright.
Figure 8.16 The light gray background and light and dark blue lines result in a more professional appearance.
Figure 8.17 Assigning a different color to each team creates a busy visualization with very similar-looking lines throughout.
Figure 8.18 Highlighting a single team in black while keeping all other teams in gray ensures focus without losing context.
Figure 8.19 A deep turquoise background makes the dark purple and blue circles difficult to see.
Figure 8.20 The white background lets the data stand out more easily and makes labels easier to read.

Chapter 9

Figure 9.1 Using a basic line chart to show iPhone units sold over time.
Figure 9.2 A simple color palette shows domestic versus international tourism trends of New Zealand.
Figure 9.4 Key insights about the use of robots to replace workers are accentuated with red color on dark background.
Figure 9.3 Red color highlights the impact of recession on German car exports between 2008 and 2017.
Figure 9.5 Minimalist design and deliberate color choices create associations with Apple as a brand.
Figure 9.6 The design elements of this visualization lead to fast brand recognition with YouTube.
Figure 9.7 Using colors and shapes associated with bees creates a visualization that forms strong associations with the topic.
Figure 9.8 Using a palette that reflects nature’s colors makes the topic visually more accessible.
Figure 9.9 Selecting colors, shapes, and images that are generally recognized as baseball-related supports fast recognition by the audience.
Figure 9.10 Consistent colors linked to specific groups of countries guide the audience through the analysis across multiple charts.
Figure 9.11 Specific data points are highlighted in red against all other data points.
Figure 9.12 The meaning of colors is unclear and confusing because a number of colors are applied to different cities.
Figure 9.13 Reducing colors allows a single city to be highlighted effectively.
Figure 9.14 The use of two contrasting colors for specific cities allows an easy comparison, while gray lines indicate all other cities for context.
Figure 9.15 A black background was used for the topic of solar eclipses.
Figure 9.16 The same solar eclipse visualization is presented on a white background.
Figure 9.17 The beige background was inspired by the Financial Times Style Guide.
Figure 9.18 Notice how the visualization changes when the background color is removed. Removing the beige background leaves no association with the original design inspiration.
Figure 9.19 Colors applied to the words “poultry” and “red meat” in the subtitle work as a color legend.
Figure 9.20 The orange line represents a specific drug, which is made clear by applying the same color to the drug’s name in the title.

Chapter 10

Figure 10.1 A stacked area chart makes the total and the lowermost attribute (male) easy to understand but can hinder our understanding of trends for other attributes (female).
Figure 10.2 Individual area charts show trends for each National Park clearly and their proximity allows for comparisons between parks.
Figure 10.3 Individual area charts color coded by category let the audience compare differences quickly.
Figure 10.4 A line chart highlights only the actual data points without the shading of the area below the line and helps to compare different categories.
Figure 10.5 A percent of total area chart gives the top and bottom area a horizontal reference line (0% and 100%), which helps identify trends.
Figure 10.6 A stacked bar chart scaled to 100% shows the contributions of two categories to the total very clearly.
Figure 10.7 Comparing the contribution of the white bars is difficult because their starting point depends on the differing length of the green bars.
Figure 10.8 Using gray reference bars sized to 100% gives your data the appearance to “fill up” the reference bars.
Figure 10.9 Vertical bars between two lines of a line chart highlight the differences between the two categories.
Figure 10.10 In a line chart, trends over time become easier to see than in a bar chart and comparisons between lines can be made with ease.
Figure 10.11 A dot plot moves the focus to the rank of each member within each category.
Figure 10.12 A diverging bar chart is useful when visualizing two opposing concepts, such as two political parties.
Figure 10.13 Labeling each bar with its numerical value makes the data more precise.
Figure 10.14 Highlighting in a diverging bar chart helps bring focus to the larger segment.
Figure 10.15 Shading the area between two dots in a diverging dot plot highlights the differences between the sides effectively.
Figure 10.16 Grouping bars that should be compared together makes analysis of differences between two or more categories easier.
Figure 10.17 A violin plot is an effective way to visualize overall distribution of data and the detailed data points.
Figure 10.18 Vertical lines break these bars into 10% segments and help with comparisons across categories.
Figure 10.19 Filled maps make it difficult to see and compare small areas to large areas, because larger regions or states typically dominate the map.
Figure 10.20 A single color in different levels of intensity is applied for a single metric in this filled map to show impact on different regions.
Figure 10.21 A diverging color palette can be used when a clear midpoint exists for the measurements in the data, such as the classification of good and bad drivers in this map.
Figure 10.22 Bubble plots are useful for showing geographic distribution of the data. Bubbles are sized by a single, common metric.
Figure 10.23 Small multiple-tile maps present patterns over time (left to right) and throughout the seasons (top to bottom).
Figure 10.24 A tile map waffle chart shows each state through an equally sized square, filled according to a single metric.
Figure 10.25 The tile map bar chart also gives each state an equally sized space and effectively highlights income distribution through colors and bar length.
Figure 10.26 Multiple hexagon maps show changes over time, allocating an equally sized hexagon shape for each state.
Figure 10.27 This cartogram shows each state clearly while sizing them according to the number of electoral votes.
Figure 10.28 Pie charts with only two segments can show each category contribution clearly.
Figure 10.29 Adding large numbers in a donut chart with two segments gives exact values to each section.
Figure 10.30 A stacked bar chart sized to 100% works better than a pie or donut chart for data with more than two or three attributes.
Figure 10.31 A stacked bar chart with a color gradient based on a single metric works well to show results across age groups.
Figure 10.32 The difference between two lines can be accentuated by shading the area between the lines.
Figure 10.33 A heatmap focuses more on the patterns in the data and increases the data-to-ink ratio.
Figure 10.34 Stacked bars sized to 100% and colored in two distinct colors allow for easy comparisons and show trends.
Figure 10.35 Bars sized by a specific metric clearly show which segments make the largest contributions to a total and allow for quick comparisons between segments.
Figure 10.36 Comparing the size of bubbles or circles is much harder than comparing bars or rectangles.
Figure 10.37 Small multiple-bubble charts are difficult to compare because the circles change their location to fill the available space.
Figure 10.38 Designing a visualization for a specific topic can be helped by deliberately placing circles and bubbles to create greater shapes.
Figure 10.39 A scatterplot is an effective way to visualize the relationship between two metrics.
Figure 10.40 A histogram can be used to highlight how your data is distributed.
Figure 10.41 Treemaps with a clear focus on a single section make the key message more accessible for the audience.
Figure 10.42 Applying a diverging color palette to a treemap that is structured geographically highlights problem areas effectively.
Figure 10.43 A treemap bar chart shows the ranking of each top-level segment, while providing a ranking of the subsegments within each bar.
Figure 10.44 Bar charts that show categories over time make it easy to compare different attributes, especially when color highlighting is used.
Figure 10.45 Color highlighting to show the largest bar helps when sorting by size does not apply.
Figure 10.46 Stacked bars sized to 100% with labels for both segments provide clear and specific answers.
Figure 10.47 A waffle chart focuses more on each individual segment.
Figure 10.48 Small multiple slope charts become more informative with labels for the attributes selected by the user.
Figure 10.49 Labeling only selected lines in a slope chart highlights their relevance for the message you aim to communicate.
Figure 10.50 Two contrasting colors work effectively to highlight increases and decreases in a slope chart.
Figure 10.51 Arrows show the progress of customer satisfaction between two points in time, with decreases additionally highlighted in red and increases in green.
Figure 10.52 A dot plot focuses on the position rather than the slope of the difference.
Figure 10.53 Connecting individual dots to lines in the direction of change between two points in time shows each age group’s development over time.
Figure 10.54 A scatterplot with a line connecting the dots of each attribute over the years shows most states progressing to more households being in higher income brackets.
Figure 10.55 Clearly labeling the start and end point of a line in a connected scatterplot accentuates the development over time.
Figure 10.56 While this connected scatterplot shows dramatic changes over the years, the start and end point of the data are fairly close together as highlighted by their labels.
Figure 10.57 A simple line chart focuses your audience’s attention on the trends over time.
Figure 10.58 Adding vertical lines or bars between two lines accentuates the magnitude of the difference between the two lines.
Figure 10.59 A diverging area chart with a common baseline can be effective for showing the different trends of two attributes over time.
Figure 10.60 A diverging line chart highlights the differing or similar patterns of two attributes across the same variable.
Figure 10.61 This dot plot highlights criteria that are more important for women compared to men and also indicates the overall importance.
Figure 10.62 Changes over time are shown in a clockwise order in this circular histogram, which displays all the data at one time.
Figure 10.63 This circular histogram highlights outliers in the data with lines that are far longer than most other lines.
Figure 10.64 With bar length representing the magnitude of black participation, and labels for key points in time, this circular histogram clearly indicates changes over the years.
Figure 10.65 Line charts can be used to effectively show patterns in the data across multiple regions at once.
Figure 10.66 Giving the audience multiple visualizations of the same data provides a comprehensive picture of trends, changes, and key findings.
Figure 10.67 Shading the area between two data points highlights the differences between two attributes.
Figure 10.68 Seasonal trends and changes over time are easy to spot in a heatmap coloring values by a single variable.
Figure 10.69 Radial bar charts can be easy to understand but can distort the viewer’s perception of the data.
Figure 10.70 While the outer bar dominates this view, the chart provides an overall impression of each bar’s contribution to the whole.
Figure 10.71 Straight rectangular bars are more effective because they do not require the audience to interpret the curvature of the bars as in a radial bar chart.
Figure 10.72 A dot plot with shading that indicates the minimum and maximum value shows the spread of the data as well as the value of each attribute.
Figure 10.73 Connecting the data points in a dot plot with a line shows the spread of the data. Adding nonselected data as small gray circles provides context.
Figure 10.74 A unit chart sized to 100% is easy to understand and can be enhanced with large numbers showing the actual value for each age group.

Chapter 11

Figure 11.1 With a title in form of a question, this visualization guides the audience to look for and find the answer in the data.
Figure 11.2 Questions in titles encourage your audience to take a position and see if the data supports their view.
Figure 11.3 Making a clear statement in your title sets the tone for your analysis and highlights your insights from the start.
Figure 11.4 Titles are commonly used to describe the data set and the visualization.
Figure 11.5 Many topics in Makeover Monday leave room for creative titles that make your audience smile.
Figure 11.6 Not all analysis has to be serious and a humorous title supported by imagery brings out the lightheartedness of this visualization.
Figure 11.7 The subtitle is an element that can be used to effectively state your key message, reflecting the insights you derived from your analysis.
Figure 11.8 The subtitle states the key message, supports the short title, and summarizes the analysis visualized in multiple area charts.
Figure 11.9 Being very specific in your title or subtitle helps your audience to understand your findings easily.
Figure 11.10 A single large number in a visualization draws the audience’s attention and provides the focus for the key message.
Figure 11.11 Large numbers provide an effective summary of each medal count in this visualization.
Figure 11.12 Dynamic BANs based on user selections show the impact of the parameter changes and highlight the most important metrics.
Figure 11.13 Including a call to action for your audience to donate, sign a petition, access further information, and so on can make your visualization very impactful.
Figure 11.14 Adding simple instructions to your visualization encourages your audience to interact with the visualization and explore the data further.
Figure 11.15 Clear and concise descriptions of dashboard actions show your audience that more information can be accessed by through interactivity on the visualization.
Figure 11.16 Coloring specific parts of the title to correspond with the visualization gives your audience an immediate understanding of the significance of the color chosen.
Figure 11.17 Applying two different colors to the title text relates artists clearly to their respective bars in the bar chart.
Figure 11.18 Annotating your charts helps guide your audience through different parts of your analysis and the key data points in your visualization.
Figure 11.19 Annotations typically focus on the most important, interesting, or surprising data points and provide contextual information for your audience.
Figure 11.20 Stating the actual values of different data points adds precision and shows the contribution of each attribute.
Figure 11.21 Percentage-of-total labels give the audience a better understanding of the significance of Snapchat compared to the other social networks.
Figure 11.22 Text is used throughout this visualization to explain the data and findings and make the analysis more accessible for the audience.
Figure 11.23 Tooltips that appear as a user hovers over the tile map show additional information for each state.
Figure 11.24 The tooltip lists the total number of births per month and the median age of the mother. This information is provided in complete English sentences.

Chapter 12

Figure 12.1 A visualization without sufficient contextual information can leave your audience confused as to what your message is.
Figure 12.2 Metrics that are easy to understand and labeled clearly help you communicate your message to your audience
Figure 12.3 Using multiple BANs in different sizes and font colors across your visualization makes them less effective as they compete for attention.
Figure 12.4 Adding well-positioned BANs in your visualization shows your audience what to focus on and aids their understanding.
Figure 12.5 Using more than a dozen colors challenges your audience to move their eyes constantly between chart and color legend to make sense of the data.
Figure 12.6 Applying minimal color to highlight the data and including a color legend creates clarity for your audience.
Figure 12.7 The reference line becomes almost invisible among the multitude of colored lines and is therefore ineffective at providing contextual information.
Figure 12.8 The black reference line is darker and thicker than the gray lines and can easily be compared to the red line.
Figure 12.9 Tooltips that are not designed with the user in mind are a wasted opportunity to provide additional information for your audience.
Figure 12.10 The tooltip contains the most important metrics with simple descriptions and clearly states the month and year of the data point.
Figure 12.11 The subtitle is used to state the metrics and fields used in the charts, instead of stating some of the insights from the analysis.
Figure 12.12 This subtitle clearly states the key findings, what the chart shows, and what the results mean for the environment.
Figure 12.13 These indicators do not specify the magnitude of profits or losses and how the different regions compare to each other.
Figure 12.14 Indicators are more effective in combination with percentage figures and actual consumption statistics.
Figure 12.15 Comparing quarterly results on a single continuous timeline is difficult because quarterly differences are not as easy to see.
Figure 12.16 Dividing the chart into quarters and showing performance over the years makes the comparison of quarters easier.
Figure 12.17 Economic performance is often viewed in quarters to account for the impact of summer months and holiday seasons.
Figure 12.18 Without normalizing the data, very populous nations like China and India will appear as outliers for resource consumption at a total level.
Figure 12.19 At a per capita level, comparing countries is much more appropriate and the results for China now appear very differently.
Figure 12.20 Highlighting detailed metrics at a per capita level for an individual country creates a country profile that can be compared to a benchmark.
Figure 12.21 Supplementing your data set with related data can enrich your findings and make them more relevant for your audience.

Foreword

The world of data is changing rapidly. We can talk about machine learning, artificial intelligence, and automation as much as we want, but what will always be needed is the ability to communicate insights. Only a creative being, a human, fluent in the language of data, can do this effectively. Only a human can combine the skills of data wrangler, communicator, designer, and artist to create visualizations that convey emotion, create an impact, and influence opinions.

Learning to be that kind of person is more vital today than ever.

There are hundreds of books about the principles of data literacy, packed full of examples. From those, you will learn the theory and principles of data literacy. What this book shows is how to build a framework for you to practice. To be an expert in any field requires practice, and this book, along with all the weekly Makeover Monday projects, gives you the framework you need to become one.

In this book you will find hundreds of examples of makeovers, made by people like you: everyday data warriors, trying to make sense of the world through data. You will learn how to develop your own skills and become as fluent as some of the most advanced authors featured in the book.

The Makeover Monday project began its growth as an experiment between Andy and me in 2016. After a year, I stepped back, and Eva Murray stepped in. Together, the two of them expanded the project greatly, and this book is the culmination of their efforts. Now everybody—even those not partaking every week—gets to benefit. Watching the project grow over the years has been truly inspiring; Eva, Andy, and I had no idea Makeover Monday could become so popular. To see the impact on individual people as they develop a portfolio, get involved in a community, and even get new jobs has been a great honor.

I can think of no better guides to help you on your journey than Andy and Eva, two people who are experts in the field, superb teachers, and true community builders.

I hope this book inspires you as it has the thousands of people who have been involved in Makeover Monday so far.

Andy Cotgreave
Brill, England
June 2018

Acknowledgments

From Andy and Eva

First and foremost, we must thank the Makeover Monday Community. Without all of you, the project and this book would not exist. Your constant participation and enthusiasm keeps us going. We enjoy seeing you evolve in your development and watching your careers flourish.

We’d like to thank Tableau Software, and specifically the Tableau Public team, for their support in helping the project grow and for providing a free sharing platform. We’d also like to thank data.world for providing a platform for sharing the data and creating an excellent discussion forum. While Tableau has been the most prominent tool used for Makeover Monday, we’re very excited to see more and more visualizations from Power BI, Yellowfin, Google Data Studio, and R. The more tools we are all exposed to, the better we will become as data analysts. Thank you also to BrightTALK, who provided us with a webinar channel that gives us a direct connection to our community for sharing feedback, discussing visualizations, and presenting about various data-related topics.

We’d like to thank Bill Falloon and Purvi Patel from Wiley for taking a chance on us first-time authors and for helping us navigate the book-writing process. Without your continuous, positive support, we would have never gotten this across the finish line.

We also want to thank Andy Cotgreave for his support and advice throughout the book-writing process and for introducing us to Wiley. Andy’s contributions in 2016 helped launch Makeover Monday into what it is today.

We’d like to thank Pablo Gomez and his team at iForIdeas for creating the Makeover Monday branding. The cover you created looks amazing! Thank you for all of your support, attention to detail, and recommendations throughout the process. This book looks great because of you!

In addition to those listed above, we would like to thank everyone who helped make this book possible: Adam Crahen, Ann Jackson, Carl Allchin, Carsten Weidmann, Charlie Hutcheson, Cole Knaflic, Deborah Thomas, Donna Coles, Emma Whyte, Jeff Shaffer, Luke Stoughton, Maureen Stone, Pablo Sáenz d Tejada, Pooja Ghandi, Ryan Sleeper, Sean Jackson, Tim Ngwena.

The cover design of this book is based on a visualization Andy created for a Makeover Monday challenge of the artwork from the Tate Collection.

Special thanks to the team at iForIdeas who turned Andy’s visualization into a stunning cover image as well as a visual identity for Makeover Monday.

From Andy

Thank you, Eva, for being an incredible partner on this journey. You’ve pushed me to grow in ways I never knew existed and have filled me with encouragement and positivity. You’ve brought so much to this project, pushed me to write this book, and you do an amazing job of keeping us both organized and keeping me in check.

I’d like to thank The Information Lab, The Data School, and Tom Brown for supporting my Makeover Monday hobby and giving me the time to help others grow. Your support has been incredibly influential in helping Makeover Monday flourish.

I’d like to thank Jonathan MacDonald and Craig Bloodworth from The Information Lab for supporting the Makeover Monday website infrastructure. Calling Jonathan while he’s on holiday to fix a website I crashed tells you all you need to know about him as a person. Thank you, Craig, for using TIL resources to give Makeover Monday a reliable home on the internet.

I’d like to thank Tim Ngwena for his support in helping produce high-resolution images. Thank you for your wizardry and for being so patient with me.

Finally, and most importantly, I’d like to thank my wife, Beth, and our four amazing children, Michael, Oscar, Elizabeth, and Henry, for their support throughout the writing of this book. Thank you for giving me the time and space to get this done when I needed it most. We’ve sacrificed many nights and weekends together to get this done. It’s been many years now that I’ve been addicted to data visualization and you have never wavered in your support.

From Eva

Thank you, Andy, for taking a gamble and asking me to join you in running the Makeover Monday project. While I underestimated the impact it would have on my life and my career, I love being part of this and having a platform for helping others grow their skills and develop their careers. Thank you for encouraging me to tackle the things that scare me and for saying yes to almost all of my ideas, even if they result in more work for us initially. Makeover Monday has been transformational, not just for many of our community members, but for me as well.

I’d like to thank Exasol and Aaron Auld for supporting my involvement with the data visualization and analytics community and for encouraging me to do what I love doing. Thank you for also giving me the opportunity to host and travel for live Makeover Monday events, which have been a great way to connect with the participants of our project.

I’d like to thank the local community and user group leaders who invited us to meet their people through Makeover Monday live events. What you do in your communities and organizations extends the work we do and takes the resources to many more people than we can reach by ourselves. Thank you for being such advocates for Makeover Monday and for helping others to grow, learn, and develop their skills. We appreciate all you do, Fiona Gordon, Josh Tapley, Kai-Ming Cheah, Michael Mixon, Neil Richards, Sarah Bartlett, Simon Beaumont.

Last but not least, thanks to my family and friends for being supportive and understanding and for patiently accepting that my availability between work, book writing, and cycling was much less than usual. Having you cheer me on along the way helped me through this writing process and on my path of becoming an author.