4 tips for better BI dashboards
With business intelligence (BI) tools, businesses have access to more data than ever before. Managers and executives have hundreds, if not thousands, of different figures and statistics to sift through to help them make decisions and drive change. The challenge, then, isn’t a lack of data—it’s that it’s impossible to separate the signal from the noise.
Dashboards help provide clarity
This is where dashboards come in. Dashboards allow users to analyze data quickly and easily, customize what sort of information they want to see, and make it far easier to share data with others. Instead of picking through some thick reports or diving in the deep layers of a transactional database to find the data you need, dashboards show you exactly what you need, exactly when you need it.
Business users and teams of all kinds can benefit from creating and analyzing data dashboards. Using modern BI tools, these teams can create dashboards on their own, without the need for IT or technical staff. With very little training required and keeping key considerations in mind, business users can follow best practices to construct their own dashboards.
Considerations with dashboards
In reality, dashboards are an imperfect tool for sipping from the data firehose. Dashboards rely heavily on visual design principles to get their message across, but the average person making a dashboard has no formal training in visual design. In the BI space, this is especially true, as the employees tasked with creating BI dashboards tend to have data analysis and computer science backgrounds.
This leads to a lot of visually unfocused, cluttered, unhelpful dashboards. A dashboard is intended to pluck out insights from the mountain of data you collect. If a dashboard is filled with useless information and doesn’t present what it needs to, it causes the exact same problem the dashboard was meant to solve—there’s no way to pick out the useful information from the rest.
The average business can’t keep a team of UX designers on staff just for dashboards, though. Like it or not, every employee needs the ability to create useful, visually interesting dashboards.
It can be difficult to learn the basics of UX design from scratch. In this article, we outline four simple tips to make your dashboards stand out.
1. Know the purpose of the dashboard.
The first step in creating visually effective dashboards is knowing why you’re making the dashboard. This may seem like a simple concept, but it’s slightly more complicated than just reading the project brief and starting work. The key here isn’t just to know what information needs to be presented, but why users need that information.
Some dashboards are intended to get across critical, time-sensitive information quickly and simply. The information on a dashboard like that should be presented differently than a dashboard that’s intended for slower, more cerebral decision-making. Both these dashboards should look different than a dashboard intended to drill deep into analysis of one facet of a business operation.
Essentially, a dashboard’s function should drive its form, and not the other way around. Every dashboard is intended to provide the answer to some business question or the other. When making a dashboard, know what that question is, and let that drive the dashboard’s layout.
2. Know who your audience is
When working on a dashboard, be conscious of who the end-user of the dashboard will be. Who are they, and why will they be looking at this dashboard? How much information will they need to know? An executive may want a more complete picture of a given topic than a middle manager or entry-level employee needs. There may be metrics that matter a lot to entry-level employees but are meaningless to executives.
This point is closely linked to the first point. The purpose of a dashboard is impossible to nail down without an understanding of who the dashboard is for. Dashboards provide answers to business questions, but the kinds of answers people want to change based on their situation. Two people can look at the same dashboard and draw different conclusions from it based on their own perception.
Also, consider the data literacy of the dashboard’s end user. Some BI tools offer hundreds of data visualizations, each more complicated than the last. For some dashboard creators, the temptation to use these bespoke visualizations is irresistible. If your audience doesn’t know how to read them, however, then it’s best to stick with simple charts, graphs, and tables.
3. Know what data visualizations to use
BI tools offer almost endless customizability in how dashboard builders can visually represent data. However, not all of these visualizations are a good choice for any data source. Some designers may use a fancy graph that needlessly complicates their information, while others may ignore useful tools they’re not familiar with.
Data visualizations have strengths and weaknesses. Not all charts and graphs are good at conveying the same kinds of information. Before choosing a data visualization, figure out what that visualization is intended to display.
Consider using bar charts or scatter plots to show the relationship between variables. These visualizations are great for showing how two data sources interact. For showing the distribution of a data set, a scatter plot is useful, but consider histograms and other column charts as well. BI tools also usually have map features that allow dashboard creators to show how a data set is distributed geographically. Line graphs are the best choice for showing how a variable is changing in real-time and comparing performance, and pie charts or treemaps are excellent for displaying composition.
4. Understand graphic design best practices
At its core, graphic design is about finding the best way to present information in a visual way. A dashboard, at its core, is a visual way of presenting lots of information. To understand how to build an effective dashboard, you have to understand graphic design, at least a little bit.
Graphic design is all about making parsing information as natural as possible. Visualizations should be intuitive; users should be able to instinctively understand how a dashboard is organized so that they can read it without trying too hard.
We have a whole article on how to use graphic design for good here, but we’ll go over the basics in this article as well.
First, make sure that important metrics are emphasized and draw the eye naturally. The metrics that a user wants to look at first should be the ones that capture the eye first. Less important metrics can be deprioritized and moved to the sides or edges.
Next, put related metrics next to each other. There should be a clear flow to the data on a dashboard so that users know where to look for information on a given topic. Nothing is more frustrating as a dashboard user than having to hunt for a statistic because the dashboard is badly organized.
Lastly, build your dashboard with a consistent strategy. The whole dashboard should be designed to flow together and have natural relations, not just individual metrics. Many BI tools have templates you can use to organize an entire dashboard in a consistent way.
The path to dashboard success
Without clear goals and some basic graphic design skills, dashboard builders will never be able to create dashboards that can effortlessly, intuitively communicate all the important information that users need to do their jobs effectively.
There are a few keys to making useful, effective BI dashboards. First, designers need to know the purpose of the dashboard and the goals that they want to achieve with it. Second, they need to take stock of their audience and consider what information they’ll want the dashboard to have. Third, have a good understanding of data visualizations and how they can be used to properly communicate data insights. Lastly, understand some basic graphic design principles to help focus and structure the board design.