• Votes for this article no votes for this yet
  • Dashboard Insight Newsletter Sign Up

How to Design Effective Dashboard Displays

by Wayne EckersonThursday, February 17, 2011

The charts contain the same data but are arranged to communicate different messages. The first chart is ideal for regional managers who want to see sales in their region and compare their performance to other regions. The second chart is ideal for a product manager who wants to see product sales across regions to understand which products sell best where. Designers enable these quick comparisons by placing data in the appropriate axes (including the legend) based on what users need to see.

stacked bar charts
EXHIBIT 12.7 Comparing Comparisons

Given the wrong chart, managers would have to work twice as hard to glean the same information even though each chart contains identical data.

Side by Side. To communicate comparisons clearly, it is best to place numbers or items side by side rather than far apart. This enables the viewer to consume the information in a single visual gulp rather than having to look back and forth across the page to make the same comparison. Charts that require less eye movement are easier to consume.

Scalar Proportions. It’s also important to ensure that charts portray relationships between variables correctly. This means designers must pay attention to the quantitative scales that they use. For instance, one outlier can wreck the visual relationships in a bar chart, such as when sales for every product range between $10,000 and $25,000 but one exceeds $500,000. In this case, the chart’s scale would be so big that it would obscure the relationship between all products except the outlier.

In another case, if variables have similarly high values, designers may be tempted to create a bar chart whose scale begins at an arbitrary high number so the relationships among bars is more obvious. Although is an admirable gesture, it can create a visual mismatch in which the proportional length of the bars is different from the actual numeric relationship between the items. For this reason, bar charts always should start at zero. (This does not apply to line charts, however.)

Use Preattentive Processing

Preattentive processing is visual perception that occurs below the level of consciousness. It detects specific visual attributes at rapid speed. Dashboard designers can leverage these attributes to highlight critical data values and relationships that viewers should notice when glancing at a dashboard or chart.

Preattentive Processing
EXHIBIT 12.8 Attributes of Preattentive Processing

Exhibit 12.8 shows nine visual attributes that dashboard designers can exploit to make things jump off the dashboard page and immediately get the attention of viewers: shape, size, saturation, enclosure, markings, color, line width, orientation, and position. You can see quickly how each attribute makes a single element stand out from the rest.

For example, color is a common technique for highlighting poor performance in a KPI. Rather than creating a stoplight image, dashboard designers merely need to place a red dot next to a subpar KPI. Or they can use saturation to highlight a single bar among many in a bar chart that they want users to focus on. (See Exhibit 12.9 .)

Predefine Drill Paths and Interactivity

Drill Paths. A chart should enable viewers to drill down effortlessly to see detailed data. It is best to predefi ne drill paths that users need and bake them into the system. Ideally, users only see drill paths that are pertinent to their role and aren’t overwhelmed with too many options.

However, some users eventually will feel constrained by predefined navigation paths and request more latitude to explore data. When this happens, administrators should activate a right-click feature that exposes additional navigation options for these individuals only.

Functions. In the same way, some individuals may desire additional functionality to manipulate the output of charts and tables. Administrators should be able to turn on chart - specific toolbars that expose new functions to these users, such as sort, calculate, annotate, export, and switch between chart types, among other things.

horizontal bar graph
EXHIBIT 12.9 Use of Preattentive Processing

Be careful not to expose too much functionality too quickly to a general audience. Users can be so distracted by icons and overwhelmed by options that they stop using the tool. Think how popular Google has become by exposing a single function on its home page — keyword search — even though it has close to 100 applications to offer. Less is more.

Drill Actions. Software vendors have yet to devise a standard way to perform drill-downs, and many techniques employed today are not intuitive. Some dashboard products require users to right click to view a dialogue box — an awkward movement for most casual users. Others require viewers to click on one or more drop-down boxes to specify the parameters and then click a "go" button. Although power users like having multiple drill paths and parameters, casual users do not.

The ideal way for users to drill down is by left clicking on the actual metric name, alert, bar, or other attribute that catches their attention. They click once to view a new table or chart populated with data. They can click again to drill down even farther. Once they become familiar with the navigation metaphor and desire more interactivity, administrators can activate right-click options.

Tweet article    Stumble article    Digg article    Buzz article    Delicious bookmark      Dashboard Insight RSS Feed
 
Previous Page  Next Page
1 2 3 4 5 6 7 8
Other articles by this author

Discussion:

No comments have been posted yet.

Site Map | Contribute | Privacy Policy | Contact Us | Dashboard Insight © 2018