Presenting data in a meaningful way empowers informed decision-making. Here are three key elements to help present analytics most effectively. A lot of hard work goes into obtaining, processing, analyzing, and displaying data. But data’s true magic lies in presenting insights in a meaningful way to empower informed decision-making. After all, it’s estimated that 65% of people are visual learners and we tend to process images 60,000 times faster than text. After creating and analyzing thousands of reports representing different types of data points over the last 12 years, I’ve found three key elements that help present analytics most effectively.
3 tips for intuitive and visually compelling analytics
Tip #1: Less is more
When designing a dashboard or analytics solution, the simplest solution is often the best. Any additional complexity should be justified. Information overload leads to fatigue and negatively impacts productivity, the very opposite of what we want from dashboards and analytics solutions. To avoid information overload, developers need to deeply understand the business context and how end-users need to interact with the data to gain insights.
Trinnex follows a structured process for making this happen. We call it SEED. It’s the same process we use to help our clients achieve digital-first resiliency, which combines existing data and institutional knowledge with digital solutions to enable smarter infrastructure decision-making.
In the context of delivering intuitive and visually compelling analytics using data that’s already available, the SEED process can be broken down as follows:
- Stakeholder Engagement – This step is all about finding the right people and asking the right questions to identify and build a deep understanding of opportunities for data-enabled improvement. It also helps identify the comfort levels within an organization of using new technologies.
- Exploration – Here’s where we start to get our hands dirty. We obtain sample data very early in the process and start exploring it to come up with creative ways of unlocking insights.
- Evolution through Iteration – This is where the magic happens. We leverage sample data to explore or create some prototype solutions, which are evolved through a highly collaborative, iterative, and fast-moving process with our stakeholders.
- Data Enabled Decision Making – This is the culmination of our process, which leaves end users with clean, simple, and beautifully designed software that provides new, actionable insights to optimize performance and empower smarter decision-making.
Tip #2: Use color wisely
Thoughtful and intentional use of color is key, as colors trigger emotions and convey moods. Color attracts our attention. When it’s overused or mis-used it can be distracting and sometimes lead us to (either consciously or subconsciously) ponder the significance of the colors before gleaning important insights.
Stephen Few, thought leader in effective data visualization, rightly states that “different colors should only be used when they correspond to differences of meaning in the data.” Additionally, muted colors are preferred over bold colors, which can cause visual fatigue. Finally, it’s important to be cognizant of color blindness and remember that there are appropriate substitutes for red/green contrast.
Tip#3: Choose visualizations thoughtfully
Some obvious ground rules exist around choosing the right visualization for your data. For example, a line chart conveys trends over time more effectively than a bar chart. There are also some strong, perhaps well-grounded, opinions about the ineffectiveness of certain chart types.
Take the pie chart for example. Most prominent data visualization thought leaders (think Stephen Few and Edward Tufte) agree that bar charts are more effective than pie charts because our brains are more effective at perceiving differences in length (bars) than differences in angle (pies). The bottom line is that thoughtfully choosing the right visualization for your data is critical. Failure to do so can cause gaps in perception and missed insights.
How data charts can go horribly wrong
Here’s one of my favorite examples of how data charts can go horribly wrong. The example comes from a publicly available data set of sewer overflows recorded in a major metropolitan area over a 12-year period. The images below use this exact same data set represented in two different visualizations: a stacked line chart and a 2-dimensional heatmap.
The data should show an emerging trend happening over the last few years. Overflows are becoming more concentrated in the late spring and early summer months, whereas before, the data shows high volumes of overflows scattered throughout the year. The data also shows that this pattern may be improving in 2021, with fewer overflows in late spring/early summer and throughout the rest of the year, relative to previous years.
Example-1 shows several distracting lines of peaking and dropping data points and fails to present the emerging trend in a clear way.
Example-2 presents the trend more clearly. You can see a more consistent pattern of purple boxes in the spring and summer months of the past few years, indicating increasing sewer overflows during these months relative to previous years. You can also see fewer overflows (more orange boxes) during winter and fall months in recent years relative to previous years. Additionally, you can see the trend may be starting to change in 2021,as the color of the boxes during spring and summer months is shifting from purple to pink, indicating fewer overflows. Example-2 is a more effective visual way to present this type of data.
Take your analytics to the next level
There are thousands of analytics programs to choose from, ranging from hands-off commercial software to industry-specific digital solutions built by experts who’ve worked first-hand with similar data sets and can help point out meaningful insights. Consider what matters most to you given the current resources you have and the priorities you need to address. The Trinnex team can help take your analytics to the next level - reach out today.
Sources:
https://uxplanet.org/designing-compelling-dashboards-10-tips-for-more-powerful-designs-298d94db1b13
