Book Review: Information Dashboard Design
A book review on the "Effective Visual Communication of Data"
I get commissions for purchases made through links in this post.
Written by Stephen Few
Copyright January 2006 - First Edition
Publisher: O'Reilly Media
When developers write a program, they always envision a product that a user could just walk up to and immediately start using it. For most developers, a dashboard is exactly that.
A dashboard is considered a combination of a really simple interface and a way to provide an at-a-glance view of critical information necessary to use the application or provide relevant data for decision-making. The different sections of the dashboard can provide drill-down points for digging deeper into the application for additional details and, as always, the user is provided with a way to get back to their "home" page or main dashboard.
O'Reilly and Stephen Few has done an exceptional job with Information Dashboard Design. The cool thing about this book is that it does the opposite of what most books do. Information Dashboard Design first displays a gallery of existing dashboards from various systems (good and bad) and then goes into the depths of what is considered good design for a truly "effective visual communication of data."
Chapter 1 gives the reader a history of dashboards and provides a gallery of dashboards from various systems for reader to understand what a dashboard is and contains from a visual perspective.
Chapter 2 discusses the three types of roles of dashboards: Strategic, Analytical, and Operational. Stephen Few takes each role and explains the different types of typical dashboard data and what it provides to the user. With a total of eight pages, this is definitely the smallest chapter in the book (definitely a quick read).
Chapter 3 reminds me of a very long, quality post that could've been on a web site. The chapter is called Thirteen Common Mistakes in Dashboard Design.
For most developers who are building dashboards from scratch, this is an excellent checklist to have around for your project. I won't go into the thirteen mistakes, but I will tell you that the mistakes do cover a lot of ground and are clear and right to the point. Throughout the chapter, Stephen explains each mistake in detail.
Now you know why I thought this could've been an excellent blog post.
For a dashboard to be appealing, it has to communicate important data through different types of visual characteristics. These characteristics are discussed in Chapter 4 which is called "Tapping into the power of visual perception."
The chapter briefly touches on the limits of short-term memory and how we see with our brains as opposed to our eyes.
The majority of the chapter explains how using visual clues in a dashboard can bring attention to certain components. Clues such as color, position, form, and motion can provide an attention-grabber for the user to see critical data.
Since you now have a firm understanding of what a dashboard contains, Chapter 5 dives into a few strategies of how to create effective dashboard displays with a second paragraph of defining characteristics of a well-designed dashboard.
A couple strategies discussed in this chapter include the goals of your dashboard, reduce the non-data pixels, enhance the data pixels, and provide relevant data to the user.
This chapter took a couple of reads, but I definitely understood what Stephen Few was talking about after the second read. Bottom line: If the data isn't important, remove it. If it is important, make it stand out or enhance it in some form or manner. Enhancing the pixels would mean referring back to Chapter 4 and applying the visual characteristics to the dashboard.
Now that you have all of your data, how do you display it properly in your dashboard? Chapter 6 talks about the different types of media available for a dashboard.
The following types of media could be used when building your dashboard:
- Drawing Objects
Each type of media is discussed thoroughly in this chapter and examples are provided.
Your dashboard is almost complete. Chapter 7 focuses on the usability of dashboards and making sure your dashboard isn't a "cluttered mess."
Stephen Few talks about organizing your data in a fashion that works with the user, not against them. Some of the topics discussed include using your "dashboard as a launch pad" (which I described above), make sure the "information supports its meaning and use," and making sure your design has been tested for usability.
Finally, Chapter 8 reveals how to put all of your data together on one screen so it's relevant to your users. Stephen finalizes this chapter with a bang. He provides four outstanding sample dashboards:
- Sample Sales dashboard
- Sample CIO dashboard
- Sample telesales dashboard
- Sample Marketing Analysis dashboard
These dashboards provide guidelines of how developers and designers should create their own dashboards. Each dashboard has their own objectives and goals for what they want to convey to their users. These objectives and goals are defined beside each dashboard and should be considered a template for future dashboard designs.
Information Dashboard Design is what I expected from O'Reilly when I first heard about the book. The book is a definite buy for my library of software books.
I've read the book twice and since then, I've used it for reference purposes. The material is solid, well-written, provides a good understanding of what a dashboard is, and takes the reader by the hand to fully understand the concept behind every dashboard that exists on the market today.
It's an exceptional book on a subject that is rarely discussed and shows that haphazard dashboard design will not portray quality software. First impressions matter the most and if a dashboard design is in your application specs, you should run to purchase this book and apply said techniques.