At a first glance, raw data doesn’t offer anyone but data experts many actionable conclusions. The high volume and complexity of consolidated data sources in most organizations create a visual overload, hiding essential truths about your business and your customers in plain sight. Yet with the right data visualizations, hidden patterns and trends can spring forth from the screen and inform decisions. The question is: how do you ensure the data visualizations within your dashboard will tell a story that guides your next steps? If your organization adheres to the following data visualization best practices, you’ll be equipped to uncover hidden patterns that enhance your awareness and refine your decision making.
1. Follow Design Best Practices
The value of your data lies in the clarity of your visuals. Good design fosters the right conclusions, allowing the cream to rise to the top of its own accord. And bad design? It baffles your users and confuses your findings – even if the patterns and trends from your reports would otherwise be straightforward.
The importance of data visualizations is only really apparent when you take time to review the worst of the worst. This Tumblr provides plenty of mindboggling examples of data visualizations gone wrong. Moreover, it helps to communicate the importance of following design best practices:
Appreciate White Space
The readability of your data visualization is a key consideration. Not every inch of a visualization needs to be crammed with information (we aren’t designing for Where’s Waldo). The image on the left illustrates just how important it is to give your graphical visualization some breathing room.
One way is to break up the information. Your dashboard should have no more than 15 objects or tiles at a time. Any more and the content can make it difficult to focus on any single takeaway. If the visuals are intertwined, following the rule of thirds (dividing your dashboard vertically or horizontally into thirds) can prevent clutter and draw the eye to key observations.
Adopt A Color Palette
The colors within your dashboard visuals matter. By choosing colors that reflect your brand or fall within a clear color palette, you allow the visuals to stand out on their own.
When you pick colors that are not complementary, you fail to draw the eye to any insight, encouraging the key findings to be overlooked at a first glance. The image on the right is a perfect example of how your takeaways can all too easily fade into the background.
Respect The Message
Visualizations should never prioritize style over substance. You want pleasant graphics that do not distract from the actual message. Clarity is always key.
In the image on the left, the vertical axis representing height makes it seem as if women from Latvia tower over those from India. Most users can still extract value but it calls the dashboard visualization’s credibility into question.
Provide Clear Answers
Complexity is another area to avoid. Your business users shouldn’t need to perform mental gymnastics to figure out what you’re trying to communicate. Any extra steps they need to take can muddle the actual findings of a report.
In the image on the right, there are a variety of factors to calculate the number of avocado toasts it takes to afford a deposit on a house (bad idea already) that are not clear from a first glance.
Each piece of avocado toast on the graphic represents 100 toasts, making it next to impossible to gauge the amount represented by the incomplete pieces for Mexico City or Johannesburg. Plus, people would need to calculate the cost based on the average price in each city, adding an additional step to verify the data. Regardless, nothing is clear and the “insight” doesn’t merit the additional work.
At the end of the day, you don’t have to have an exceptional eye for visual design but the person implementing visualizations into your dashboard does. This can be an external resource (if you have one) or a partner experienced in following data visualization best practices.
2. Cater to Your Target Users
Great data visualizations are tailored to their target end users. The data sources, the KPIs, and the visualizations themselves all need to align with their goals and challenges. Before implementing your data visualization, ask these questions to determine the perfect parameters:
- What is the background and experience of your end users? Your data visualizations need to rise to the level of your users. Executives need visualizations that offer strategic observations about revenue streams, operational efficiencies, and market trends. Employees on the front-lines need answers to allow them to evaluate performance, KPIs, and other tactical needs.
- What experience do they have with reporting dashboards or data visualizations? Frame of reference matters. If your users have primarily used Excel, you need to strike a balance between enhancing their reporting and creating a sense of familiarity. That might mean visualizing KPIs that they’ve previously created in Excel or building an interface that mirrors the experience of any previous tools.
- What are their most common usages? There’s finite space for data visualizations within your dashboards. Every graphic or chart you provide should apply to frequent use cases that provide the greatest value for your team. Otherwise, there’s a risk that your investment will fail to earn the fullest ROI.
- Which pain points do they struggle with most? Visualizations are meant to solve problems. As you are determining what to illuminate with your dashboard visualizations, you need to reflect on the greatest needs. Are there challenges pinpointing customer motivation and behaviors? Have revenues stagnated? Are old processes inefficient? The squeakiest wheels should be oiled first.
Answering all of these questions creates a foundation that your partner can implement as they create meaningful dashboards. Data visualizations that cater to these audiences are better at satisfying their needs.
For example, if you are creating a dashboard for a restaurant, your visualizations should cater to the pressing needs and concerns of the owner or manager. Expenses by supply category, total sales vs monthly expenses, special orders by count on receipt line items, and other KPIs can supply owners with quick and essential insight that can enhance their business strategy.
Regardless of industry, the data visualizations within your dashboards should balance immediate needs with actionable insight.
3. Show Overlooked Relationships between KPIs
Reviewing one benchmark alone gives very narrow insights. Good data visualization can help organizations to connect the dots between KPIs. In fact, there are plenty of instances where the connection between one KPI and another is not apparent until that data is visualized. That’s when it helps to have an experienced partner guiding your work.
Let’s use our home health staffing firm partner as an example. We helped them to implement a Snowflake data warehouse and one of the key lessons they wanted to learn was about the response they should take during COVID-19. They were eager to review cases across the United States, but that would only provide limited insights.
We suggested they visualize the data against a few other parameters. A timeline functionality could help to create an interactive experience that showed the growth of outbreaks over a period of time. In other scenarios, organizations could do a side-by-side comparison of KPIs like (CO2 emissions, automotive traffic, or other conditions to measure the impact or trends related to the virus).
What’s equally important is that you do not overlook outliers. There are organizations that will get in the habit of burying statistical anomalies, not realizing when those outliers become the norm. Working with the right partner can give a fresh perspective, preventing essential findings from falling outside of your awareness.
Rachel Stewart is a Data Management and Analytics Consultant at Aptitive. In her role, Rachel enables companies to discover insights and access the information held by their data.