Three Design Principles for More Effective Dashboards
November 3, 2020
If you work in data analytics, chances are, you create dashboards on a regular basis. Do you find it easy to make them clear and effective? Does your audience understand them? Are they excited to load the dashboards and look at the numbers?
In my experience, three principles can make or break a dashboard:
Let’s look at each of these concepts in more detail using the dashboard below as an example.
- No Clutter. Remove anything that doesn’t help the reader understand the data. Most often that will include gridlines, axis names and ticks. I love the examples of this from Darkhorse Analytics, #RemoveToImprove. Here’s one.
- White space. Let your dashboard breathe! In the example above, the squares are placed 20-30px away from each other. That helps identify where a graph ends and another one starts.
- Contrast. Is there enough contrast in your dashboard? Can you distinguish the different parts from each other even if you walk a bit further away from the screen?
- Typefaces. Use 1-2 typefaces max and keep their style—size, weight and colour—consistent. In the above example, I used the Inconsolata font for the dashboard title and the caption, and Lato for everything else. Chart titles and BANs are black, whereas the rest is a shade of gray. See where I’m going with this?
- Highlighters. Anything that draws the reader’s attention to a specific part of the dashboard is a highlighter. In this example, the most obvious one is colour: I used teal to separate 2020 from 2019. The key here is to only use that colour for the same reason throughout the entire dashboard.
- Tooltips. You may or may not use tooltips depending on your software. If you do, make sure you format them neatly!
- Structure. Like a story, your dashboard should adhere to some sort of hierarchy. The most common way is to start with summary figures at the top and then zoom into the data as you scroll down. That may not be the best way to represent data each time, so feel free to create your structure, as long as you have one.
- Annotations. The dashboard and the charts should all have titles, and even subtitles when possible. If you can add annotations that explain the ebb and flow of the data, even better! (cf. the line chart above). Remember, annotations help people who aren’t familiar with the data understand what’s going on without asking you.
I could write much more on dashboard design, but I do feel that these three principles are the most impactful in a fast-moving business environment. Everything else is often just the cherry on top.