Book Summary: Storytelling With Data by Cole Nussbaumer Knaflic

Better communicate insights from data by knowing how to select and use the right visualization tools to tell a compelling story that reveals a clear and consistent message in your data.

📖 The Book in 3 Sentences

1. Understand your audience and take clear steps on how you are communicating to them by seleing the right type of visualization and data relationships you would like to show.
2. Learn how to focus your audience's attention by identifying and eliminating clutter, using contrast strategically, aligning elements, use of pre-attentive attributes like color, size and position to guide analysis of your data.
3. Craft a story with a clear beginning (plot), middle (twists), and end (call to action).

🖼 Impressions

Storytelling With Data is a guide on how to effectively visualize data and effectively tell stories from them.  It. helped me learn the right way to select charts, know how to leverage the strengths of each and techniques on how to tell the story to reveal insights on your data.

👥 Who should read it?

Storytelling With Data is definitely for data analysts who build visualizations, reports and dashboards who are looking to improve their technique.  Given that this is a theory book with concepts, I do recommend to retain the concepts is to get hands on with practice which the second book Storytelling with Data : Let's Practice helps with!

💡 How the book changed me

• I am more thoughtful about my audience, messaging and outcome I am working torwards when visualizing data
• I now have a better idea on picking and choosing the right chart to select my data
• I know effective techniques to improve the look and feel of my data visualizations to quickly communicate my message
• I can communicate with data effectively

✍️ My Top 3 Quotes

It’s easy to spot a hawk in a sky full of pigeons, but as the variety of birds increases, that hawk becomes harder and harder to locate. - Colin Ware

Well-designed data visualization - like a well-designed object - is easy to interpret and understand.  When people have trouble understanding something, such as interpreting a graph, they tend to blame themselves. In most cases, however, this lack of understanding is not the user's fault; rather, it points to fault in the design.  Good design takes planning and thought.  Above all else, good design takes into account the needs of the user.  This is another reminder to keep your user - your audience - top-of-mind when desining your communications with data.

When we construct stories, we should do so with a beginning (plot), middle (twists), and end (call to action).  Conflict and tension are key to grabbing and maintaining your audience's attention.  Another central component to story is the narrative, which we should consider in terms of both order (chronological or lead with ending) and manner (spoken, written, or a combination of the two).  We can utilize the power of repetition to help our stories stick with our audience.  Tactics such as horizontal and vertical logic, reverse storyboarding, and seeking a fresh perspective can be employed to help ensure that our stories come across clearly in our communications.

📒 Summary + Notes

Chapter 1. Understand the Context

Build a clear understanding of who you are communicating to, what you need them to know or do, how you will communicate to them, and what data you have to backup your case.  Employ concepts like the 3-minute story, the Big Idea, and storyboarding to articulate your story and plan the desired context and flow.

Book: Data Points by Nathan Yau

Book: Resonate by Nancy Duarte

The more specific you can be about who your audience is, the better position you will be in for successful communication.  Avoid general audiences, such as internal and external stakeholders or anyone who might be interested.

The method you use to communicate with your audience has implications on several factors, including the amount of control you have over how the audience takes in the information and the level of detail that needs to be explicit.  With a live presentation, you are in full control.

3 minute story concept is where you only have 3 minutes to tell your story, get to your points and be succinct like an elevator pitch

Big Idea concept is to distill your idea to a single sentence. It must:

1. Articulate your point of view
2. It must convey what’s at stake
3. it must be a complete sentence

Storyboard with pen and paper and break it down into the following sections:

• Issue:
• Demonstrate Issue:
• Ideas for overcoming issue:
• Recommendation:

Chapter 2. Choose an appropriate visual display

When highlighting a number or two, simple text is best.  Line charts are usually best for continuous data.  Bar charts work great for categorical data and must have a zero baseline.  Let the relationship you want to show guide the type of chart you choose.  Avoid pies, donuts, 3D, and secondary y-axes due to difficulty of visual interpretation.

Book: Show Me The Numbers by Stephen Few

This chapter is quite important as it provides guidance on how to use certain data visualizations effectively.

Simple Text: When you have just a couple of numbers to share, simple text can be a great way to communicate.  When you have more data that you want to show, generally a table or graph is the way to go.

20%

of children had a

traditional stay-at-home mom

in 2012, compared to 41% in 1970

Tables: When you use a table in a live presentation, consider whether you are losing too much by doing so.  Consider including the full table in the appendix and a link or reference to it, if necessary.

Design wise, make sure tables fade into the background letting data be the focus.  Use light borders and white space to set elements apart.

Heatmap: A heatmap is a way to visualize data in tabular format, where in place of the numbers, you use colored cells that convey the relative magnitude of the numbers.  To reduce mental processing, you can use color saturation to provide visual cues.

Graphs, Points

Scatterplot: Scatterplots are useful for showing the relationship between two things, and they are more frequently used in scientific fields.  They can be used to show the relationship between miles driven and cost per mile, for example.

Lines, Linegraph: When you're graphing time on the horizontal x-axis of a line graph, the data must be in consistent intervals.  Be consistent in the time points you plot.

In some cases if your line shows a summary statistic you can also add a sense of range by showing minimum, average and maximum wait times at passport control for an airport over a 13 month period.

Slopegraph: Slopegraphs are useful for highlighting the single category that decreased over time in the preceding example.  They are not always easy to create, and they may not work with all data.

Bars, Vertical bar charts, Stacked vertical bar charts

Bar charts: are easy for our eyes to read.  Our eyes compare the end points of the bars, so it is easy to see which category is the biggest, which is the smallest, and also the incremental difference between categories.

The rule is that bar charts must have a zero baseline.  This rule does not apply to line graphs.  With line graphs, since the focus is on the relative position in space rather than the length from the baseline or axis, you can get away with a nonzero baseline.

There is no hard-and-fast rule when it comes to the width of bars, but in general they should be wider than the white space between the bars.  The bars should not be so wide that your audience wants to compare areas instead of lengths.

The vertical bar chart is the plain vanilla bar chart.  It can be used to display single series of data, two series, or multiple series.  Be aware that as you add more series of data, it becomes more difficult to focus on one at a time and pull out insight.

Stacked vertical bar chart: Use cases for stacked vertical bar charts are limited.  They are meant to allow you to compare totals across categories and see the subcomponent pieces within a given category.

The stacked vertical bar chart can be structured as absolute numbers, where you plot the numbers directly, or with each column summing to 100 percent.

Waterfall chart: The waterfall chart can be used to pull apart the pieces of a stacked bar chart to focus on one at a time, or to show a starting point, increases and decreases, and the resulting ending point.

Horizontal bar chart: The horizontal bar chart is the easiest chart to read.  It is especially useful if your category names are long, as the text is written from left to right, making your graph legible for your audience.

When designing any graph showing categorical data, be thoughtful about how your categories are ordered.  If there is a natural ordering to your categories, it may make sense to leverage that.  Otherwise, think about what ordering of your data will make the most sense.

Stacked horizontal bar chart: Stacked horizontal bar charts can be used to show the totals across different categories, and they can be structured to show either absolute values or sum to 100%.  They can be used to visualize portions of a whole on a scale from negative to positive.

Area graphs: Area graphs are difficult to read, so I avoid them unless I need to show numbers of vastly different magnitudes.  The second dimension you get using a square for this allows you to do this in a more compact way than possible with a single dimension.

Avoid use Pie charts and 3D: Pie charts are difficult for people to read.  They are best replaced with a horizontal bar chart, which is easy to understand and interpret.

The only exception to the rule of never using 3D is when you are plotting a third dimension.  Do not use 3D to plot a single dimension.  It introduces unnecessary chart elements and graphs strange things when plotting values in 3D.

Avoid secondary y-axis: When you're faced with a secondary y-axis problem, consider which alternative shown in Figure 2.  27 will better meet your needs.  Alternative 1, where each data point is labeled explicitly, puts more attention on the specific numbers.

Chapter 3. Eliminate the clutter.

Identify elements that don't add informative value and remove them from your visuals.  Leverage the Gestalt principles to understand how people see and identify candidates for elimination. Use contrast strategically.  Employ alignment of elements and maintain white space to help make the interpretation of your visuals a comfortable experience for your audience.

The most important aspect of your visual communications is the cognitive load on the part of your audience.  Try to minimize this as much as possible.

The Gestalt School of Psychology set out in the early 1900s to understand how individuals perceive order in the world around them.  They came away with the six principles of visual perception that define how people interact with and create order out of visual stimuli.

Decutter: step by step

1. Remove chart border
2. Remove gridlines
3. Remove data markers
4. Clean up axis labels
5. Label data directly
6. Leverage Consistent Color

Chapter 4. Focus attention where you want it

Employ the power of pre-attentive attributes like color, size and position to signal what's important. Use the strategic attributes to draw attention to where you want your audience to look and guide your audience through your visual. Elevate the effectiveness of pre-attentive attributes in your visual by applying the "were are your eyes drawn?" test.

Your iconic memory is super fast. It is tuned to a set of pre-attentive attributes, which are critical tools in your visual design tool belt. Information stays in your iconic memory for a fraction of a second before it gets forwarded on to your short-term memory.

The best way to demonstrate the power of preattentive attributes is to show it.  Figure 4.  2 shows a block of numbers.  Count the number of 3s that appear in the sequence.  The correct answer is six.

756395068473

658663037576

860372658602

846589107830

versus

756395068473

658663037576

860372658602

846589107830

Graphs, without other visual cues, can become very much like the count the 3s exercise or the block of text we’ve considered previously.  With no clues about what’s important or should be paid attention to, it’s the count the 3s exercise all over again.

When you’re doing exploratory analysis, you should mostly avoid using preattentive attributes.  When it comes to explanatory analysis, however, you should have a specific story you are communicating to your audience.  Use preattentive attributes to help make that story visually clear.

Size: The importance of size is important to remember when designing your visual communications.  If you're showing multiple items that are of roughly equal importance, size them similarly.

Color is one of the most powerful tools you have for drawing your audience's attention.  Use it sparingly, use it consistently, and be mindful of the tone color conveys.

When we use too many colors together, beyond entering rainbow-land, we lose their preattentive value.  When we consider what these colors represent, it's not necessarily where we want our audience to look.

Design for color blindness

vischeck.com - see what colorblind people see

colororacle.org - full screen color filter

checkmycolors.com is a tool for checking foreground and background colors and determining if they provide sufficient contrast for color-sight deficiency

informationisbeautiful.net/visualizations/colors-in-cultures

Chapter 5. Think like a designer

Offer your audience visual affordances as cues for how to interact with your communication: highlight the important stuff, eliminate distractions and create a visual hierarchy of information.  Make your designs accessible by not over-complicating and leveraging text to label and explain.  Increase your audience's tolerance of design issues by making your visuals aesthetically pleasing.  Work to gain audience acceptance of your visual designs.

The form and function of your data visualizations should be considered when designing them.  The audience should be able to do what you want them to do with the data, and the visualization should allow for this with ease.

When it comes to the perfection of data visualization, the decision of what to cut or de-emphasize is just as important as what to include or highlight.  Consider broadly what information is critical and what is not.  Eliminate unnecessary, extraneous, or irrelevant items or information.

Don't overcomplicate!

The more complicated a visual is, the more time your audience believes it will take to understand it, and the less likely they are to spend time understanding it.  Avoid this by making your visuals easy to read and understand.

Make it legible: use a consistent, easy-to-read font (consider both typeface and size)

Keep it clean: make your data visualization approachable by leveraging visual differences

Use straightforward language: choose simple language over complex, choose fewer words over more words, define any specialized language with which your audience may not be familiar and spell out acronyms (at minimum, the first time you see them or in a footnote).

marriage data: The All category was removed altogether.  The data being plotted is Number of newly married adults per 1,000, and it is strange to discuss the number of adults using decimal places.

The title bar at the top of your PowerPoint slide is a precious real estate that should be used wisely.  It is the first thing your audience encounters on the page or screen, and it should be used for an action title.

When it comes to communicating with data, is it really necessary to make it pretty.  The answer is a resounding Yes.  People perceive more aesthetic designs as easier to use than less aesthetic designs, and this has been proven in countless studies.

Chapter 6. Tell a story

Craft a story with clear beginning (plot), middle (twists), and end (call to action). Leverage conflict and tension to grab and maintain your audience's attention.  Consider the order and manner of your narrative.  Utilize the power of repetition to help your stories stick.  Employ tactics like vertical and horizontal logic, reverse storyboarding, and seeking a fresh perspective to ensure that your story comes across clearly in your communication.

Story is a powerful tool that can be used to communicate with audiences.  It is a time-tested structure that humans have been using to communicate with stories for centuries.  We can learn from the art of storytelling to better tell our own stories with data.

Story is what ties together information, giving your presentation or communication a framework for your audience to follow.  You can use the idea of beginning, middle, and end to set up the stories that you want to communicate with data.

The bulk of your communication develops what could be, with the goal of convincing your audience of the need for action.  You retain their attention through this part of the story by addressing how they can solve the problem.

The final aspect of a good story is its ending, which should be clear about what you want your audience to do.  You can do this by tying it back to the beginning, or recapping the problem and the need for action.

When you storyboard at the onset of building a communication, you craft the outline of the story you intend to tell.  Reverse storyboarding does the opposite.  You take the final communication, flip through it, and write down the main point from each page.  The resulting list should look like the storyboard or outline for the story you want to tell.

Chapter 10 Where to go from here

Inspiration for good examples:

• Eager Eyes (eagereyes.org) - thoughtful content on data visualization and visual storytelling.
• FiveThirtyEight's Data Lab (fivethirtyeight.com/datalab) - minimalist graphing style on a large range of news and current event topics
• Flowing Data (flowingdata.com, Nathan Yau)
• The Functional Art (thefunctionalart.com, Alberto Cairo) - An introduction to information graphics and visualization, with great concise posts highlighting advice and examples.
• The Guardian Data Blog (theguardian.com/data) - News-related data, often with accompanying article and visualizations, by British news outlet.
• HelpMeViz (helpmeviz.com, Jon Schwabish) - "Helping people with everyday visualizations" this site allows you to submit a visual to receive feedback from various readers or scan the archives for examples and corresponding conversations.
• Junk Charts (junkcharts.typepad.com, Kaiser Fung) - self-proclaimed "web's first data critic," focuses on what makes graphics work and how to make them better.
• Make a Powerful Point (makeapowerfulpoint.com, Gavin McMahon) - Fun, easy-to-digest content on creating and giving presentations and presenting data.
• Perceptual Edge (perceptualedge.com, Stephen Few) - No-nonsense content on data visualization for sensemaking and communication.
• Visualising Data (visualisingdata.com, Andy Kirk) - Charts the development of the data visualization field, with great monthly "Best visualizations of the web" resource list.
• VizWiz (vizwiz.blogspot.com, Andy Kreibel) - Data visualization best practices, methods for improving existing work , and tips and tricks for using Tableau Software.
• storytelling with data (storytellingwithdata.com) - the blog for this book!