Data visualizations are powerful communication tools. Their visual nature makes understanding complex data relationships easier to break down and digest, and their colorful presentation ups the energy on even the most mundane topics.
Are you looking to find and illuminate insights hidden within your data? Visualizations are the way to go.
But like any art or science—and in many ways, data visualizations are both—there are best practices that can help guide you.
Keeping these five simple tips in mind will help steer you toward creating successful, impactful visualizations.
- Keep your visualizations simple
This rule is a perfect place to start when designing anything. It applies doubly to visualizations. Data is already complicated enough without adding distractions like ornamentation or backgrounds. Strip away non-essentials and let the data speak for itself. Avoid gimmicks like 3D charts or illustrations. They add visual noise to the visualization and often interfere with data fidelity and accuracy. Finally, limit yourself to highlighting one or two insights within a single visualization; trying to pack too much meaning into a single chart or graph can overwhelm your data narrative. A great way to incorporate additional insight is by housing secondary and tertiary info in interactive elements. For instance, a tooltip can be exposed on rollover or on tap of visual elements, like a bar or line.
Speaking of data narratives…
- Know your data narrative
Every visualization tells a story. With visualizations, strive for nonfiction rather than fiction: your narrative should be born out of the data, not the other way around. Learning to interpret data and translate it into the ‘proper’ visualization is key. To do this, you should know the types of visualizations that are available to you and how each is utilized. Determining factors can be the size of your data set, and whether you are organizing, contrasting, searching for and highlighting relationships, searching for and highlighting trends, or breaking down the data into its simplest dimensions. Each of those goals suggests its own visualization.
- Label, label, label
Without labels tying data to what it represents, visualizations remain abstract and open to interpretation. We can’t overstress the importance of:
- Always label your axes
- Always incorporate a legend explaining how colors and patterns map to your data
- Ensure your units of measure are clearly presented and make sense given the scale of the data
While simplicity is key, this is one dimension of visualization where it can’t hurt to overreach. If in doubt, label it.
- Color where it counts
Don’t overwhelm. We prefer to reserve color for representing data. Keep titles, labels, and gridlines a muted neutral color. This allows the data to stand out on a page. Be consistent across visualizations: colors are a language, and they take on very specific meanings within a visualization system. Visualization color theory is a whole blog post in and of itself. There is a massive amount to discuss here: accessibility considerations, simultaneous contrasts, gradient progressions, red/yellow/green as sentiment and branding.
- When in doubt, bar chart
This is less a tip and more of a fortune cookie. The humble bar chart is easy to read and keeps data relationships simple. And while the pie chart (and its modern offspring, the donut chart) are popular and look snazzy, they are difficult to ‘read.’ If you find yourself in doubt about what visualization to apply, try sketching out a bar chart first—or a grouped bar chart—and see how it works for you.
While these five tips can lead to better results when designing data visualizations, there’s a lot of detail we can’t go into here (or else we’d end up with a blog post). If you want to be certain you are getting the best possible visualizations into your experience, reach out to us at Main Digital. We’ll be happy to consult with you and your team and put our decades of design and process experience to work for you.
You don’t have to go it alone. Bring Main Digital’s Experience Team along with you.
Contributed By: John Clarkson