January 1, 2022

Dashboard Design Essentials – KPI Templates

By Spencer Baucke

I am super excited to introduce another addition to the Dashboard Design Essentials series with this post on business dashboard KPI designs featuring work in Power BI. We will be covering how to design BANs, trend lines, tables, and change metrics to create a KPI template that will bring any business dashboard to life.

BANs

The term BAN generally refers to the abbreviation for Big Ass Number. Although I do not know who coined the term, it is a generally accepted term across the BI and analytics industry. A BAN is a big number that is meant to attract attention from end users. Normally BANs are bold, big, and colored in a way that makes users notice them. We want users to notice BANs because often they are the most important thing we are trying to tell them. Because of this, they are most often placed at the top of dashboard (since we know from previous lessons that the important stuff should go up top).

In the dashboard below, there are a couple of BANs that you might notice at the top of the dashboard. Although this is a very exaggerated version of a BAN, it is a great example of their purpose, to get people’s attention and portray a main theme. In this case, I want you to know immediately that we’re talking about wins per team in this basketball rivalry. Now let’s see how we can make similar BANs in Power BI.

Power BI has a lot of great ways to make BANs without having to do much work, so that is nice. Using a data set on NCAA sports financials, I have created three different BANs.

The BAN on your top left is a great example of what most of your metrics are going to look like. I used the Card chart in Power BI to create this. Simple, easy to make, and gets your point across.

Next to that is the KPI chart in Power BI. The subject line being at the top left is a nice feature, but I’d also like the KPI itself to be under there and out of the way so I can make the most of the trend line. This is my least favorite of the three.

That leaves us with our third BAN made from the Multi-row card chart in Power BI. I made the font of the metric a little bigger and used a bold font to really make the number pop. The more subtle gray text for the metric name, the highlight color showing the conference name, and the vertical gray line to the left provides a nice border for the left alignment. Although we won’t use this exact design when we build our final KPI template, it will definitely look most like this one.

Trend Lines

Trend lines offer end users context to your high level KPIs by showing them how that KPI has evolved over time. Trends are very important in understanding your KPI at the present time, which is why in almost every KPI dashboard you will see trend lines.

When determining what type of trend line you should use, you will need to determine what time period is most appropriate for your KPI. This will affect what type of trend lines you will want to use. The more data points that you want to show in your trend, the more likely you’ll want to use a line chart. Since using bar charts take up more physical space on your dashboard, having a lot of data points in your bar chart will take up a lot of dashboard space and will draw a lot of attention instead of providing the subtle context that it’s meant for. See some of the different trend line designs below.

Each of these trend lines has their advantages. The top left chart easily allows you to see month over month change as it’s more easy to associate the value in comparison to 0. For that same reason, it’s easy to see when a value in that trend is negative. The line chart next to it can be more aesthetically pleasing, especially when you’re dealing with a larger amount of data points (like a day level trend for the past 6 months). The downside is that in the 5th year of this trend, the value is negative and it’s very hard to see that using the line chart

The trend at the bottom left is a good combination of the previous too. Although it still takes up a little more dashboard real estate than the line chart, you still get that line chart visual effect and you can also more easily see when a value is negative. The chart on the bottom right is meant to show an example of a trend line that you should not use. There are a lot of things wrong with this chart, but for the sake of KPI trends it’s best that you keep charts to one axis, preferably the value that you’re showing in your BAN.

Change Metrics

Just like trend lines give more context for your KPIs, change metrics are a great way to give end users more understanding and context. Most times you will want to use change percent’s as they are normalized, but there can definitely be exceptions. If your KPI is already a normalized metric on say a scale from 1-100 then an absolute change is acceptable. You will probably want to provide the absolute change someone where in your dashboard, but we will get to how to do that in our final section.

Some of the most common change metrics include Year over Year change (YoY), Quarter over Quarter change (QoQ), and Month over Month change (MoM). These indicators give viewers a snap shot into how the KPI in question has performed over a given time period. This can be helpful in compartmentalizing metric performance. Here is one way to do this in Power BI.

				
					Profits Change Shape = 
IF ([Profits YoY%] > 0, UNICHAR ( 128994 ), UNICHAR ( 128308 ) )
				
			

Using a matrix, I created a measure that returns a unicode shape depending on whether the measure output is positive or negative. In the example above, I used green and red circles. This indicator can be helpful in identifying which metrics have positive/negative change attributes without actually having to look at every change metric and keep mental notes of their changes.

I prefer to use arrows for positive/negative changes because the shape also indicates the change direction, not just the color. There is a way to do this with the red down arrows unicode shape (128315) being the negative symbol and a gray down arrow being the positive symbol then changing the color of that up arrow in the Field Formatting section of the format painter tab in the Visualization pane. You can also use custom svg shapes to achieve the same outcome.

KPI Design Templates

Now that we have covered three essential elements of a KPI dashboard, let’s combine all three to create a KPI template that you can use at your workplace!

First, create a BAN using the multi-row card chart. Make sure to make your number big and bold to attract the eyes. Second, create a trend line of that metric using any of the methods we mentioned above. In my example, I created a trend line with the custom Sparkline from OKViz. Lastly, create your time period change and shape calculations necessary for the time period indicators. Now, let’s combine them on a layout that will make our end-users happy.

As you can see above, this KPI template is sleek, yet spaced out enough to read, and offers viewers all the information they could want without taking up too much room. Our BAN dominates the view, which is good, because we want them to notice that first. The sparkline on the right has enough room to breathe which is important so that users can make sense of the metric changes. And the time period indicators provide a pop of color that immediately indicates good things are happening with the green color scheme.

Keep in mind, this is just one layout that I like to use in my dashboards whether I’m developing in Tableau or Power BI. It works well for me and my clients, and I hope you get some use out of it as well. If you don’t use this layout, perhaps it can provide you some inspiration for your own favorite KPI template!

I really hope this post was helpful in explaining some KPI template ideas and will enable you to apply some of these concepts in your workplace. Need more help making your dashboards a work of art? Our knowledgeable team of experts is here to help!

Data Coach is our premium analytics training program with one-on-one coaching from renowned experts.

Accelerate and automate your data projects with the phData Toolkit