Call UK: +44 1483 549027 / USA: +1 408 733 5818
Dashboard Design Services
Why Dashboards?
When asked in a recent Forrester Research survey what their priorities were for achieving future success and growth, 90 percent of senior executives pointed to improving the customer experience. To meet that goal of customer-centricity, customer feedback must be collected and then used to drive performance improvement. The challenge is that companies in many cases are overflowing with data in different formats residing on different databases having different platforms. Operational and financial data, as well as customer attitudinal data, all play a role in describing the intricate interrelationship of company performance, customers’ perceptions of their experiences, and financial health of the company. Furthermore, in this age of consolidation, acquisitions have been on the rise, further exacerbating the situation of disparate databases in use throughout the enterprise.
Well-designed dashboards that pull in key information from those disparate data sources allow users to quickly analyze and monitor how their group is doing, with alerts highlighting those metrics that are falling outside a pre-defined range. Complex information is communicated quickly, resulting in a rapid understanding of business performance and thereby driving smarter and faster decisions. With time a valuable commodity in short supply in most companies, few other than dedicated analysts are able to commit to delving into mountains of data on an ongoing basis. By providing a top-level view of the Key Performance Indicators (KPIs) and then the necessary detail through drill-down to supporting metrics, the impact on business can be immediate.
CustomerImpact Five-Step Process
CustomerImpact designs and delivers dashboards that display the critical metrics in a visually engaging manner, connecting business users with their information and driving action. Our five-step process is as follows:
1) Discovery Phase.
▪ Project objectives are confirmed, timelines established, resources committed, and budgets established.
▪ A key task during this phase is to gather requirements and expectations from stakeholders with regard to their KPIs. They should be asked what business questions they need answered, what other questions does each question raise, and what action might this lead to should results be problematic
▪ If corporate and departmental KPIs do not exist, we gather input through interviews with key personnel and analysis of existing data from previous market research and customer satisfaction studies.
▪ Since customer feedback is vital, we work with clients who do not have a robust Voice of the Customer program in place to design, implement, and manage such a program for them.
2) Design Phase. Discussions and agreement about:
▪ Supporting metrics for each of the KPIs, e.g., providing sales in units and dollars at the Account Executive level. This data would be accessed from the Sales Growth chart on the Sales department dashboard. Mapping of data to objectives underlies this step.
▪ Data sources that house these metrics
▪ Functionality of the dashboards and drill-down screens, with navigation paths
▪ How the dashboard parts are arranged
▪ How to present the data most effectively, e.g., which formats for different kinds of data, positioning of the web parts, use of color, ways to highlight the most important data.
3) Build Phase. Technical implementation, including:
▪ Integration of data from various data sources, including the more complex customized enterprise applications, such as CRM
▪ Defining and programming the algorithms that will output the data needed, such as averages, indexes, top-box percentages, and correlations
▪ Creation of the user interface
▪ Creation of the charts, graphs, and drill-down charts
▪ Defining trigger data points and threshold levels that will trigger alerts
▪ Implementation of security rules that control which users see which information
▪ Testing for accuracy of data
▪ Planning for integration with the corporate network environment
4) Deploy to Production
▪ Implementation of security requirements
▪ Integration with corporate network environment
5) Manage.
▪ CustomerImpact monitors database generation and presentation on an ongoing basis to ensure that the specs are met.
▪ As conditions and expectations change, we enhance the dashboard, working with clients to assure quality and timely delivery.
Best Practices in Dashboard Design
Best practices in creating dashboards fall into two categories: selection of metrics and visual communication of data.
Selection of Metrics
Each user should be targeted so that the KPIs selected are appropriate and meaningful. They must be the right tool for the right job. KPIs are, in fact, metrics that are tied to targets. Therefore, they need to be shown in context, so that it is clear whether they represent acceptable or unacceptable levels of performance. Visual comparisons can highlight how the group is doing relative to corporate strategic goals, competitive benchmarks, and prior period performance.
Information that conveys the condition of the KPI and helps users understand why that is the case also needs to be identified. These data will constitute the detail found in the drill-down screens. For example, in examining the health of the Support function, the KPI might be responsiveness and the drill-down would show this information per agent. The hierarchies of KPIs and supporting data should be mapped, with data sources shown.
A maximum number of metrics should be shown. Too many and the scene is cluttered with no conclusions readily apparent. Therefore, much thought must be given to which ones to highlight. CustomerImpact works with clients to make those important decisions.
Visual Communication of Data
There is an evolving body of knowledge that applies the core principles of visual perception to the design of dashboards and the placement of web parts in that dashboard. Some guidelines are:
-
Colour should be used when needed for the purposes of highlighting and indicating items that are grouped. Vivid and subtle colors have different uses and the hues and intensity of the same color should be considered, as well as the context of where color is placed (the same color can look different if placed in differently colored backgrounds).
-
Methods of highlighting certain data (excluding the use of color) include outlining the object, positioning it on the dashboard so that it jumps out at the user, orienting it differently from the other data points, making it larger, or shaping it differently.
-
Using the appropriate format, such as pie charts, bar graphs, line charts, and gauges is critical. Pie charts, for example, usually are used to convey parts-to-whole information, but they are less than ideal if there are too many parts (difficult to grasp the relative size of each part) and especially if the parts are not in a logical order.
-
Setting the scale on the axis appropriately ensures that the relationship between the data points is interpreted correctly.
-
Clarity has to be favored above decoration, even though the latter can be very attractive. Extra grid lines that are not needed to convey what the values of the data points are, pictures in the background, even 3-D bars, serve only to distract.


