Posted by Peder Enhorning
February 4, 2015
Dashboards have become strategic tools for delivering decision making information. And ideally, those dashboards are automatically updated to reflect daily, weekly or monthly trends. Furthermore, much of the information should be KPI based. If properly constructed, the KPIs should immediately inform how things are trending and what action is needed.
To properly deliver on these promises, dashboards need to be visually effective. They don’t necessary have to look attractive, but need to be impactful and allow for interpretation without too much effort. In Stephen Few’s 2006 book, Information Dashboard Design, he stated that: “A dashboard is a visual display of the most important information needed to achieve one or more objective; consolidated and arranged on a single screen so the information can be monitored at a glance.”
Following on that premise we offer seven ways to make your dashboards as visually impactful and useful as possible.
Although we will focus this discussion on appearance, we can’t ignore what the dashboard needs to contain. The most important dashboard consideration is the message you are conveying. An attractive looking dashboard is useless unless it delivers valuable information and all required information fits on the dashboard. So dashboards need to be actionable and relevant to the intended audience. Delivering KPIs is critical, so the user can immediately determine how things are trending. And KPIs should be created with the corporate goals and objectives in mind.
KPIs can be defined as one of four types of metrics:
And they should all be shown trending over time. So, if the objective for the organization is to reduce costs by increasing customer self-service, then an effective KPI might be the ratio of online registrations to all registrations, month over month. That will give us an indication at how well the online registration system is working. Reporting systems don’t typically deliver these kinds of numbers, so you may have to do these calculations before plugging them into your dashboard.
Contact a KPI consultant to learn more about our workshops that guide you through the KPI Karta methodology to identify the best KPIs for your business.
Playing, exploring, and experimenting with the charts is what keeps users engaged. Interactive dashboards enable your audiences to perform basic analytical tasks such as filtering the views, drilling down and examining underlying data – all with little training. Viewers need to be able to get the big picture from the dashboard that everyone sees and then be able to drill down into a view that tells them what they need to know to get their jobs done. Allowing your viewers to interact with dashboards lets them find hidden relationships in the data that you may never even have intended or thought of.
The Toronto 2011 Parking Infractions dashboard shown below contains several elements which are all joined together by street name, type of infraction, infraction time of day, and so on. As any part is interacted with, the entire dashboard is updated.
Sparklines are data-intense, design-simple, word-size graphics that provide a quick sense of historical context. When designing sparklines in reports, it is helpful to also highlight the minimum point and the maximum point.
Sparklines can also be used as thumbnails to indicate the necessity of drilling into the dashboard for further detail. That way you can hide the complexity of the numbers and show them only if requested.
It’s easy to get overly ambitious and want to provide highly detailed, real-time dashboards covering each and every business challenge. But it is very bad practice to cram and jam too much into a single view.
Instead, produce dashboards that are tailored for specific audiences. Most systems will enable selectable areas to be part of a dashboard view based on user log-in credentials. If further detail is required, it can be accessed by drilling into that specific portion. This is a great use of sparklines.
Only include information relevant to your audience. That means customizing dashboards for unique people or groups. HR may find sales figures interesting but it’s just distracting if it’s not valuable to their job. It’s better to have less information on the dashboard and encourage the user to drill in to get more detail if required.
A dashboard should bring together common data elements from disparate sources such as customer information from CRM, email marketing and in-house sales order systems. If you are looking at a single customer view it only makes sense that it is all contained within one view!
Research shows that people read a dashboard the way they read a book. That means top left to bottom right. The visualizations which show the key metrics should be at the top and to the left of the dashboard.
Make sure that related visualizations where a comparison can be made are placed next to each other. The dashboard should be arranged to make the end user consume the dashboard in the order the dashboard creator intended it to be consumed.
Less is more. If it doesn’t fit on one page, it’s not a dashboard, it’s a report! Delivering the complete business snapshot on one page, providing that clarity has not been sacrificed, allows for faster data inspection and understanding. Don’t force users to scroll. Cognitive research – studies examining how the human brain processes information – has demonstrated that the ability to visualize all interrelated information sources together empowers people to more easily and clearly grasp the significance and overall meaning of an information set with greater accuracy.
When designing your dashboard, embrace the use of colors, shapes, lines, thicknesses, degrees of shading, and other treatments that leverage visual perception. Things to avoid include overly cute widgets, 3D graphic treatments, and distracting color schemes.
Accountants have always known that visual pop-outs such as color can be used to highlight and draw attention to certain numbers. Negative values are set in red to quickly allow the viewer to find areas of concern. Similarly, the size of a circle can signify relative amount in sales or number of units sold,and the shading of that shape can inform another measure such as profit. Using a different shape is a great way to indicate another item or region. Humans are good at understanding values based on shape, colour, and size and less good at greys, texture and orientation.
Another great reason for using different shapes is that 12% of men and, therefore, 6% of the population is color blind. The most common color blindness is an inability to distinguish reds and greens.
Using different colors in the dashboard can draw attention to unique parts and using similar colors can tie two parts of a dashboard together. Generally, a complimentary color scheme can drastically improve the appeal of your dashboard as a coherent color scheme draws the whole dashboard together and makes it look more professional. A good starting point is to identify colors which are prominent in any image or visualization you have used on your dashboard and then use a color scheme to compliment these colors.
The two examples below both show the same data. However, it’s often difficult to gain insight and find patterns when looking at numbers in a Figure 1. Instead, when those values are combined in colored boxes, important information jumps out at the viewer as shown in Figure 2. In this case, size indicates sales volume and color shows profit. And less space is used on your dashboard.
Pie charts are generally considered poor data visualization for any data set with more than half a dozen elements. A bar graph is a better alternative in this case. Pie charts are problematic as it is very difficult to discern proportional differences with a radially divided circle, except in the case of a small data set that has large value differences within it. Pie charts also pose a problem for labeling as they are either dependent on a color or pattern to describe the different data elements, or the labels need to be arranged around the perimeter of the pie, creating a visual distraction.
In this example, having a different color for each column in the column graph is unnecessary and even distracts the user from the data. Having just one color for all columns actually works better in this case.
People seem to think that because cars and planes contain gauges, we must have them in business intelligence dashboards too. Actually, they don’t provide many added benefits. Gauges show very little information and take up a lot of space. A bullet graph is much more effective in conveying the same information but in a much smaller space. Real estate is precious on a dashboard, and the space saved by using a bullet graph instead could be used for other information.
3D charts are best used in the scientific community for seeing patterns in n-dimensional data. Do not use 3D bar charts just for the sake of doing so or because you think they look cool. 3D charts can easily be misinterpreted and often distort 2D data. The best practice is to keep your charts simple.
Visualization is often viewed as a personal preference. But it’s not. There are firm rules and best practices that will enhance the delivery of information. Follow them, and your dashboards will be useful and impactful.
We offer Tableau Training including a course on Tableau Visualization Best Practices that incorporates visualization best practices for performance, visualization, and dissemination of Tableau dashboards.
Ramp up your team on Tableau fundamentals in just 2 days.
To get details on customized, ON-SITE training, leave your email.