Making the Grade: Four Tips for Better Data Visualizations

Blog: The Tibco Blog

What makes a great data visualization? Data alone just isn’t enough. Depending on context and presentation, viewers may take away exactly the message you’ve crafted or leave with no better connection to the concept. Here’s a four-point plan to make sure what viewers see is exactly what they get.

1. Critical Intentions

According to Computer World, (CW), the first step toward great data visualizations is determining purpose: What do you want to communicate, and who’s viewing the data? The CW article references a complex and highly-detailed chart of Napoleon’s Russian campaign, along with a simpler pie chart showing two simple facts: That 98 percent of his army died and just 2 percent survived. Effectiveness here is a question of audience—history buffs versus the general public—and the same is true of your organization. If you’re presenting to IT professionals about a new tech development, for example, charts and graphs heavy on the numbers are probably your best bet. If you’re communicating the same information to managers—opt for a streamlined, high-level approach.

2. Leverage Text

As noted by MIT News, one can improve the impact of data visualizations by creating the right title. While visual elements are critical, simply repeating the presented data in your title does not grab the viewers’ attention. For example instead of a title like “Number of Security Breaches Over Time”, could be rewritten as “Security Incidents Hit All-Time Low Last Quarter.” By using more concise, descriptive language that sums up the take-away message – our visualization becomes more engaging, directive and helps frame your data in context.

3. Perfect Pictures

Of course the ideal visualizations also leverage powerful imagery to get your message across. Here, MIT researchers discovered that while pictograms—which were often seen as “dumbing down” content—don’t reduce the effectiveness of your data and when used correctly can increase the impact and memorability of your visualization. Bottom line? Simplicity is your goal here: If your pictures are complex enough they require an explanation, you need a re-design.

4. See It Again. And Again. And Again.

Repetition is the final link in the visual chain: Information presented once doesn’t have the same impact as data presented multiple times in different ways. Want to convey a gap or increase? Provide the information itself but also contextual repetition—numbers fully spelled out or occupying graphics of varying sizes to give a sense of scope. Outright repetition also works but is best used when paired with subtle changes to the presentation style each time the same information is shown.

Want better data visualizations? Be clear about your purpose, use great text, don’t be afraid of simple pictures, and finally, repeat, repeat, repeat.

Next Steps

Try Spotfire and start discovering meaningful insights in your own data.
Subscribe to our blog to stay up to date on the latest insights and trends in Fast Data and Big Data analytics.