8 Tips for Making Your Dashboards More Effective

First of all, it’s important to make clear that what many people call a “Dashboard” is actually an exploratory analysis panel. These are different concepts and therefore call for different visual approaches.

Generally speaking, a dashboard is the equivalent of the “front page” of a news portal. The front page should contain the most important information from the entire site, at a minimal level of detail, but enough for the user to “scan” it and get a sense of what’s happening at that moment. If something catches their attention, they can dig deeper by clicking through to a page with the full content.

According to Stephen Few,

A dashboard is a visual display of the most important information needed to achieve one or more objectives; consolidated and arranged on a single screen so the information can be monitored at a glance.

Few makes a point of highlighting the great confusion that has existed for some years between dashboards and other data analysis interfaces:

The greatest clarification that is needed today is a distinction between dashboards, which are used for monitoring what’s going on, and displays that combine several charts on a screen for the purpose of analysis.

With that said, below we list some fundamental points for making your dashboards more effective. And if you’d like to learn a bit more about the DAR (Dashboard, Analysis, Report) methodology, feel free to read this post.

Add context to your indicators

Imagine “total sales for the month” as a numeric indicator on the dashboard.

“$2.4 Million”

Is that good or bad? Compared to the previous month, is it better or worse? And compared to the same period last year? Does this value hit the sales target set for the period?

It’s essential to provide context for the data, allowing users to make comparisons and draw conclusions more quickly.

Don’t over-detail

If the goal of dashboards is to convey the most important information for decision-making as quickly as possible, what’s the point of including tables with detailed values?

Following our example, if my company brings in revenue in the millions, why put the cents on the dashboard? Is that relevant enough to be on our “front page”?

Respect the screen limits

While our analysis or report screens may occasionally have vertical scrolling (although it’s not recommended), the dashboard needs to “fit” on end users’ screens, without objects being hidden behind buttons or scrolling.

Balance numeric indicators and charts

Numeric indicators can be used in dashboards, as mentioned above, as long as they’re presented with context. Prefer positioning numeric indicators, in a larger size, at the top of the data area, leaving charts and gauges just below. But be careful, since using too many numeric indicators will slow down the reading of the dashboard (our brain processes reading, even if the numbers aren’t spoken aloud), so favor gauges and charts, using text only for the most important indicators.

Build “back to front”

You can, but it will take much longer, if you start building your application from the dashboard.

Leave this screen for last, because by building the analysis screens first, you’ll already have the indicator calculations ready, along with the visual standard and chart types defined. If you leave the dashboard for the end, your job will basically be to “cut” detailed information that doesn’t make sense for the dashboard, which is much easier than creating charts “from scratch.”

Speedometers, only in case of emergency

Even if users love speedometer gauges, try to avoid using them on the dashboard. Prefer horizontal “gauge” styles: they convey exactly the same information and take up less screen space. Generally, in the space where only one speedometer would fit, you can fit 2 to 3 horizontal gauges.

Highlight only negative anomalies

Imagine you have 15 indicators on your dashboard, where 12 of them are performing as expected (positive) and only 3 are performing poorly (negative). An interesting visual approach would be to use a “traffic light” as a visual indicator reinforcing this performance: green for good and red for bad. In the end, your screen will be full of green and red dots, and your eyes won’t know where to look (which information is most important and needs to be investigated first?)

That’s why you should visually highlight only the indicators that show poor performance, since business users will need to pay attention to those first, not the ones that are doing well.

Show what’s happening right now

Don’t turn your dashboard into a historical analysis panel. Leave that task for the application’s “internal” screens, the “analysis” ones. Prefer displaying information that helps users understand the snapshot of the data at that moment. That’s what matters most. Leave the “investigative” work for the internal screens.