To better focus your dashboard project, here are some key questions you should ask yourself before building a dashboard.

1. What purpose does this Dashboard serve?
   
   1. Do I need this to provide analytical reports to mid-level managers?
   2. Do I need to monitor daily/hourly transactions of a particular commodity?
      (Note: This is where your vision needs to be clearly defined.)
      
      
2. Who is the target audience for this dashboard?
   
   1. CEO
   2. Department heads
   3. Foremen monitoring factory floor production
   4. Downstream knowledge workers
      
      
3. Should there be customized views of the same report for different members of the audience?
   1. Marketing head does not need to view HR Manager's reports
   2. CEOs would require consolidated reports with access to different views of the data.
      
      
4. What are we measuring?
   
   
   1. Is it a process?
      1. Is this process clearly defined?
      2. Does this process takes into account “business nuances” that could skew the conclusion stemming from this data?
         (i.e., If a customer repeatedly contacts Tech Support is that due to high-quality customer service or a testament to an "un"-user friendly product or system.)
   2. Is this process/system transparent from top to bottom with a clear break-down of tasks, events and responsibilities?
      
      
5. What actions do I want to drive out of each individual report?
   
   
   1. Should we automate the request for more info or status checks when certain thresholds are reached?
   2. Do I want to generate an alert for any process that goes out of control?
      
      
6. Who is accountable for managing or monitoring the task, events and processes?
   
   
   1. What action is taken by the Operational Manager in response to a given event?
   2. Who has the ownership of the business process and what is he/she doing about it?
      
      
7. Is each view/report complemented with comparative and historical data?
   
   1. How do we fare against our leading competitors?
   2. What were our sales for the same period last year?
      
      
8. Do we have our measures clearly defined?
   
   1. Pre-aggregated calculated values, also known as KSI (key success indicators) and KPI (key performance indicators) could be used as measures.
   2. Dependency/relationship among different measures should be well understood before commencing any dashboard venture.
      
      
9. What type of views does this dashboard have?
   
   1. Analytical (using statistical and financial formulas to provide perspective);
   2. Strategic (making sense of where we stand in respect to our corporate vision);
   3. Monitoring (Typically real-time data gets channeled to this view. The ability to drill down to the molecular level of transaction data is vital);
   4. Consolidated view of all that's mentioned above (more typical for CEOs).
      
      
10. What technology best fits our requirement?
    
    1. ASP.Net / Java.
       
    2. Reporting Services (best when a high level of user interaction is not required
       
       
11. Should process data sequence also be reflected in report KPI viewing sequence?
    
    1. For a person familiar with a certain flow of data in a process-chain an unfamiliar viewing sequence could alter the perspective of data presented.
       
       
12. Does the dashboard paint the complete picture?
    
1. Missing relevant business information from a dashboard leaves the reader/viewer with more questions than answers.
   
   
13. What are the key “takeaways” from each report/view in the dashboard?
    
    1. Each report must have some key driving conclusion.
    2. A report without a conclusion is much like a meeting without an agenda.
    3. Takeaways could be either visual cues, or textual information
       
       
14. Have I selected the right display medium to create a visually dazzling dashboard?
    
    1. Use of the right chart type prevents readers from making a visual perspective error
       (Note: There is a fine balance to maintain when adding graphical content to the report. A graphical look and feel that does not enhance the user’s understanding of data is merely visual fluff.)
       
       
15. Have we decided upon a look-and-feel theme that will be consistently applied across all reports?
    
    1. Using different colors to represent the same process across different reports could send a wrong message to the user that it's another sub-group of data.

Copyright 2008 - Dashboard Insight - All rights reserved.