As discussed in part one of this two part blog series – Dashboards for Business Intelligence: 18 Design Don’ts (part one) – personalized dashboards are critical delivery mechanisms for Business Intelligence (BI), allowing decision-makers involved in different business functions throughout an organization, to stay abreast of relevant, KPIs at a glance.
Dashboard don’ts: The back nine
Last week, we presented the first half of our checklist of fundamental flaws and oversights to avoid when constructing a BI dashboard. Today, we offer up our final thoughts on common, but evadable, stumbling blocks.
When planning and compiling a dashboard, don’t:
- Try to deliver BI dashboards to everyone straight away: Begin by delivering dashboards to a small and specific user group, building on successes, and learning from mistakes.
- Design dashboards in isolation of intended user groups: Business users are the ultimate dashboard ‘customers’. Think of a dashboard as a product. Why would people want, and want to use, a product that does not offer them the capacity to improve the efficiency and effectiveness with which they can complete their routine workplace duties? Let business needs drive the technology and its development – not the other way round.
- Organize the information without considering its intended use: Not only is the information displayed via a dashboard integral, but so is the way it’s structured. Arrange information logically according to priority level and intended use. The three most common types of dashboard structure involve organizing information according to:
- Flow: Arrange information so that it can be read from left to right, and/or up and down in a logical progression.
- Relationships: Position related reports together such as revenue and expenditure.
- Grouping: Group reports according to common characteristics – you might position all your geospatial reports at the top of your sales dashboard.
- Present and update information without considering the frequency of use: The frequency with which a particular user group accesses their dashboard; be it daily or weekly; will affect the type of reports displayed and the level of detail required.
- Provide no or little context to the information displayed: Numbers in isolation are incapable of empowering a user to monitor business processes and take action when required. To provide true insight and enable a process or task to be effectively monitored, a data set must be compared to other figures, such as quantitative targets or historical information.
- Assume one sizes fits all: Measure/display KPIs relevant to the specific department and job function of the individual receiving the dashboard, taking into consideration such factors as:
- The amount of time a user has to view and react to information presented on their dashboard on a daily basis
- Data expertise and skill
- Only create dashboards based on departmental factors: Decide on the overriding purpose of each specific dashboard to help guide the type of dashboard developed and delivered. For example, is your dashboard:
- Broad or specific
- Strategic or operational
- Historical or real-time
- Uniform or customizable
- Neglect to provide interactivity: Dashboards by their very nature provide a summary overview of key metrics and business drivers. Users must be able to interact with their dashboards to drill down or through summary reports to obtain deeper understanding, filter reports for specific data, and share important information. If users have the capacity to interact with data to attain more detailed information, it will be simultaneously easier to produce uncluttered, clean, clear and easily consumable dashboards.
- Neglect to present actionable information: A dashboard is a tool to facilitate action. If users do not act as a result of reporting and analytics, then it is not possible to derive Return on Investment (ROI) from BI projects. Consider how the dashboard helps users to understand information in a way that facilitates and empowers them to act.