Notes I made while reading Storytelling with data: a data visualization guide for business professionals from Cole Nussbaumer Knaflic.

Chapter 1: The importance of context

  • Exploratory: What you do to understand the data
  • Explanatory: Selected knowledge with purpose

  • Use explanatory for presenting

  • Who: Your audience
    • Define specific
    • Your relationship with the audience
    • Expectations
  • What: Action
    • Propose, ask for action
    • Choose right medium (presentation, email)
    • Avoid mix and compromise (slideument)
    • Choose the right tone
  • How
    • What data is there that is useful
    • Ignore nonsupporting data
  • Reduce to the max
    • 3 Minute story
    • Elevator pitch

Avoid PowerPoint early, use analog tool like storyboarding (PostIt notes)

Chapter 2: Choosing an effective visual

Simple text / Big number

One number that stays with the audience Simple text

Tables

  • Audience reads values from left to right
  • Design should fade in the background, avoid fancy Excel formatting

Table formatting

Heatmap

  • Helps the reader to see the different values
  • Make sure to add a legend to clarify the colors

Heatmaps

Graphs

  • Interact with visual system => processed faster

Scatterplot

  • Shows datapoints on two axes
  • Shows relationships of two things
  • Discover trends and similarities
  • Take time to explain to audience

Scatterplot

Line Graph

  • Continuous data, often timeseries
  • Label series directly

Line graph

Slopegraph

  • Two series of data
  • Want to show the differences
  • Example: Survey before/after on several topics
  • Works best if lines do not overlap

Slopegraph

Bar chart

  • Very flexible usage (vertical, horizontal, stacked, etc.)
  • Easy to understand
  • Good comparison between values

Bar charts

Area Graph

  • Show relationship
  • Use Rectangular shapes instead of round

Area charts

Graphs to avoid

  • Everything with angles and round shapes (Pie charts, donut charts etc.). Reason: Humans are not good at estimating round areas and distances.

  • Graphs that are too complicated (Area charts, Bubble charts etc.). Too much information in one chart. This leads to a lot of effort of you explaining the graph instead of explaining your idea.

Image sources: Knaflic, Cole. Storytelling With Data: A Data Visualization Guide for Business Professionals, Wiley, © 2015.