Much like those pesky multiple-choice problems from school (or technical certification exams), you have to choose the best answer, regardless of the number of answers that might be technically correct. The best data visualization is one that can be interpreted quickly, is intuitive and supports actionable analysis. It provides measurable value to the user. To create a visual that meets these standards, follow these cardinal rules:
- Keep it simple but not stupid.
Don’t overload a visual, report or dashboard with more information than is needed, but make it smart. This requires a delicate balance of conveying enough information to yield value without overdoing it. With the vast options and customizations available for visualizations, it’s only natural to want to try them all.
Although the ability to make pie charts on top of a map on top of another chart with dozens of colors might seem cool, it could overcrowd your visual, making it difficult for users to extract what they actually want from it. To make information easier to digest, either cut back on what's displayed in a visual or break data out into multiple visuals. A visualization should have enough context to allow it to stand alone, but it shouldn't force the user to sift through extraneous detail to find the information they need.
- Emphasize important data.
The most important data should immediately draw the user’s eye. People read left to right, top to bottom, and are naturally attracted to the largest and brightest item on a page — use this to your advantage. If it will be in a report or dashboard with multiple visualizations, make it larger and place it near the top left corner to make it stand out.
Tableau has a useful graphic that shows where the most important parts of a dashboard are. Essentially, the top left corner and dead center are where our eyes gravitate first. The bottom right corner is typically the most ignored space on a dashboard. Users want to be able to gather important information quickly, and making vital data stand out with size, color and thoughtful placement supports this.
- Focus on your audience.
Work closely with intended users and understand what actions they want to be able to take based on the data they consume. This will point you, as the developer, in the right direction to create a meaningful and valuable visualization. If the visual doesn’t portray what the user needs, it doesn’t matter how nice it looks, how many interactive features it has or how hard you worked on it — it doesn’t work.
Whether you're creating a single visualization or an entire report or dashboard, all of these rules apply. They may seem basic, but developers often miss the mark by violating one or more of these guidelines. Although there's much more to be said on the topic, keeping these principles in mind will start you on the path to creating great data visualizations.
Learn more about the common mistakes developers make and ways to avoid them.