We explore best practices to creating smart and impactful dashboards!
Target Your Audience
Now this might seem a bit obvious (because, honestly, it is) but you have to start here. Often this step is taken care of because there's a certain individual asking for the dashboard.
Don't Go Overboard
A dashboard is supposed to give an individual an fast way to understand what the data is saying. Now you might be thinking, “But the devil’s in the details!” And you would be right; the devil is in the details. But that doesn’t mean every metric you have should be in the dashboard!
This is the best of both worlds. The overall message of the data is preserved, but the viewer still has the ability to go deeper if they are so inclined. Below, we explore interactivity which can offer another way to preserve details and simplicity, simultaneously.
Use The Right Chart Type
If you haven’t already, check out our blog post on choosing the right chart types. Certain chart types are better at different things and you should be aware of these best practices.
For instance, the human brain can actually tell differences in length better than area. This is good news for the bar chart, but bad news for the pie. Our article lays out how to approach choosing the right chart in detail.
In Slemma, you can separate your charts into sheets. A single subject dashboard (for which we advocated above) often has sub-categories. Using sheets is the best way to allow the viewer to mentally segment the charts.
The overall message will be communicated without losing the secondary information in the subsequent sheets.
Make Them Interactive
When you are presenting a more detailed analysis of the data, there are two ways to go about it.
- The first is to create the filtered charts yourself for each important data segment. You would create a dashboard where each sheet is a different trading block for example or a different consumer segment.
- The second (and more efficient) way is to use controls. With Slemma, you can add interactive filters to dashboards that allow the viewer to filter, without editing rights. For each dimension you can create one of these controls.
The result? Instead of many, many charts that explore subsections of data you can have a small set of charts with the power to explore all of the different subsections.
The latter is almost always preferable, unless you want to bring the viewers’ attention to a specific filter combination.
Pay Attention to Aesthetics
I can already hear the grumbling; and I know, making a dashboard look pretty seems like a gigantic waste of time (especially for internal reports). The fact of the matter is, however, that your viewers are human beings! They are not data vacuums that automatically absorb anything you put in front of them.
By making the dashboards beautiful, you enable the viewer to interact with it longer. This will increase the likelihood they will gain valuable insights which is the intended purpose.
- Color coordinate by subject.
- Take advantage of conditional formatting.
- Use shapes to group related visualizations.
- Choose colors that don’t burn your eyes into oblivion.
Our final tip for creating effective dashboards is to use the text box. Especially when doing KPI dashboards, it’s important to be sure everyone is on the same page. There are many different ways of calculating certain metrics.
Text boxes allow you to elaborate on how you came up with certain measures. Now you might be saying, “C-level guys don’t care about the dirty data crunching I did.” But, honestly, they should and probably do! The way something is calculated can drastically change its meaning which impacts decision making.
What are some of your best practices? We would love to hear them! Write a comment on social media or send us a note at firstname.lastname@example.org. Happy Dashboarding!