Data Science 5

Why telling a Story with Data? 

Facts get recorded, stories get remembered

 Memorable  Cause & Consequence More brain parts enabled

Relatable Emotional coupling The same brain parts engaged for listener and storyteller

 Leading to Action Research shows that storytelling can engage brain parts that 

lead to action

Lesson 1 : understand the context 

• Exploratory analysis is what you do to identify interesting things to learn from your data; 

• Explanatory analysis is what you do to communicate these interesting findings to someone else 

• There are a few questions you should be able to answer before you spend much time creating content for explanatory purposes:

– Who is your audience?

– What do you need them to know or do?

– How can you use data to help make your point?

• When consulting for context, the following questions can help to build a robust understanding of the situation:

– What background information is relevant/essential?

– Who is your audience? What do you know about them?

– What biases might your audience have that may make them resistant to your message?

– What data is available that would help strengthen your case? Is your audience familiar with this data, or is it new?

– What would a successful outcome look like?

– If you only had a limited amount of time or words to tell your audience what they need to know, what would you say?

Lesson 2 : choose an effective visual 

• Common types of data visualizations for business communication: 

• When highlighting a single number or two, simple text can work well 

• Tables: for mixed audiences or showing multiple units of measure 

• A well-designed graph will usually get the information across more quickly than a table 

• Line charts are typically best for continuous data; bar charts for categorical data • Bar charts must have a zero baseline 

• Be cautious when using pie and other area charts 

• Let the relationship you want to show guide the type of chart you choose 

• When unsure, solicit feedback from a friend or colleague

Lesson 2 : choose an effective visual 

Which would you use? 

• SCENARIO: You are visualizing data from your employee survey and want to show how employees respond over time to the retention item “I plan to be working here in one year.” Here are four potential ways to graph the same data. 

• Pick a partner and discuss each of the following graphs. What can you easily see/compare with each? What is difficult? Which would you choose and why?

Lesson 3 : identify & eliminate clutter Gestalt Principles of Visual Perception:

• Every single element on the page creates cognitive burden on our audience 

• Leverage the Gestalt Principles to identify and eliminate that which has no informative value 

• Employ alignment and use white space strategically for clean horizontal and vertical lines 

• Avoid diagonal elements and text 

• Use contrast strategically: don’t let your message get lost in the clutter!

Lesson 3 : identify & eliminate clutter

 let’s declutter 

• See the graph below. What visual elements could you eliminate? What other changes might you make to reduce cognitive burden? Spend a moment considering this and make some notes

Lesson 4 : focus audience’s attention 

Preattentive attributes: 

• Employ the power of preattentive attributes and use selectively to direct your audience’s attention and provide a visual hierarchy of information 

• Preattentive attributes can be leveraged in text as well as data visualization to show your audience where to look and make it easier to interact with your visual 

• Determine whether you’re using preattentive attributes effectively by conducting the “where are your eyes drawn?” test 

• Don’t use color to make something colorful—when used sparingly, color is one of your most powerful tools for drawing attention where you want it 

• When choosing color, be mindful of desired tone, company/team branding, and colorblindness

Lesson 5 : tell a story 

• Stories resonate, stick in our memory, and can be retold 

• Text is your friend: use it to title, label, explain, and tell a story 

• Utilize takeaway titles on your slides: use this precious real estate wisely!

• A repeatable, pithy phrase can help reinforce your message and make it memorable 

• Consider the narrative arc—plot, rising action, climax, falling action, resolution—use this to help think through the storyline for your data stories and craft an order, flow, and narrative that will resonate with your audience 

• Conflict and tension are key components of story; consider how you can leverage these 

• Employ tactics like the power of repetition, horizontal and vertical logic, and reverse storyboarding to ensure your story comes across clearly 

• There can be immense value in soliciting a fresh perspective—both in the context of a specific data visualization as well as your broader communication