I was flipping through Stephen Few's book, Information Dashboard Design, and was very interested in the section on the thirteen common mistakes in dashboard design. This blog will be the first part of a short summary of the key messages for each mistake.
“Exceeding the boundaries of a single screen”
To ensure data is easily understood, viewers should not have to scroll or switch between multiple screens. All related context should be presented together for better comprehension. A dashboard is considered "bad" if viewers must rely on their memory, since human short-term memory can only hold a few chunks of data simultaneously.
“Supplying inadequate context for the data”
Your dashboard should provide a visual display of the primary information and context. Your key metrics should have the right supporting context. For example, include comparisons or trends. The amount of context is also crucial. We don't want to overwhelm viewers with details, as they might miss the basic message.
“Displaying excessive detail or precision”
Dashboards should provide a high-level overview with just enough information for viewers to grasp it quickly. If viewers see too many details, such as numbers that are overly precise, they will try to filter out what's important. This slows down their information processing.
“Choosing a deficient measure”
For a measure to be meaningful, you should know what is being measured and what unit represents it. Deficient measures are those that do not support the intended message or context. You should carefully consider the message that most directly supports the viewer's needs, then choose an appropriate measure to convey that message.
“Choosing inappropriate display media”
It's very important to carefully consider which information or message to convey—for example, showing trends, comparisons, or parts of a whole—and then choose an appropriate chart to display it.
“Introducing meaningless variety”
There is often a tendency to use different types of displays just to prevent viewers from getting bored by repetition. However, you should always choose the display that works best for viewers to understand the context. Ensure consistency in your choice of displays, which allows viewers to interpret data in the same perceptual way, saving them time and energy.
That’s it for part one! Let’s meet again in part two!