Book: Information Dashboard Design

  • Read: April 2013
  • Rating: 8.5/10

Information Dashboard Design by Stephen Few is a great resource if you’re designing, building, or thinking about dashboards- specifically, if you’re designing a dashboard with business intelligence information. I’m in the process of building a dashboard (continuously) and dashboard design before and after reading this book is like night and day. It’s incredible how many things I got wrong before reading the book (color usage, location of data, types of charts and graphs that I used).

I’d highly recommend this book to anyone doing anything with dashboards.

My notes

Categorizing -> Role:

  • Strategic - snapshots, high-level, not-interactive
  • Analytical - greater content, comparisons, history, evaluation
  • Operational - dynamic, immediate, specific

Categorize by whose dashboard it is, type of data, data domain, types of measure, span of data, update frequency, interactivity, mechanism of display, portal functionality

13 Mistakes of Dashboard Design

  1. Exceeding the boundaries of a single screen (multi-screen/scrolling)
  2. Supplying inadequate context for the data (not including comparisons)
  3. Displaying excessive detail and precision
  4. Choosing a deficient measure. Communicate meaning - think about the message from viewer’s needs
  5. Choosing inappropriate display media (never use pie charts)
  6. Introducing meaningless variety
  7. Using poorly designed display media
  8. Encoding Quantitative Data inaccurately
  9. Arranging the data poorly
  10. Highlight important data effectively or not at all
  11. Cluttering the display with useless decoration
  12. Misusing/overusing color
  13. Designing an unattractive visual display

Power of Visual Perception

Limits of short term memory

Visually encoding data for rapid perception:

  • preattentive attributes and categories:
    • color (hue/intensity)
    • position
    • motion (flicker/movement)
    • form (orientation, length, width, size, shape, added marks, enclosures)
  • Encoding quantitative data vs. categorical data

Gesalt Principles of Visual Perception:

  • Proximity
  • Similarity
  • Enclosure
  • Closure
  • Continuity
  • Connection

Eloquence through simplicity

  • Summarize = sums/averages
  • Exceptions = something outside realm of normality

Only address one objective/set of objectives

Data-ink ratio => maximize! Reduce non-data pixels; then enhance data pixels

Effective Dashboard Display Media

Text, graphic, combination, graphs, images, icons, drawing objects, text, organizers

  • Bullet Graphs are the bomb
  • Bar graph => nominal/ordinal scale
  • Line graphs => interval scale, used to emphasize trends, cycles, change
  • Scatter plots
  • Treemaps
  • Icons => alert, up/down, on/off. Ideally only use one, maybe 2 tops.

Designing Dashboards for Usability

Organize the information to suport its meaning and use

Maintain consistency for quick and accurate interpretation

Make the viewing experience aesthetically pleasing: colors, high resolution, font

Design for use as a launch pad; design for interaction

Don’t present users with options or alternatives; you own the design