This is a continuation of a transcript of a Webinar hosted by InetSoft in September 2010 entitled "Designing a Good Dashboard." The speaker is Mark Flaherty, Vice President of Marketing at InetSoft.Mark Flaherty (MF): A good example is getting business intelligence on individual customers. You can understand their entire purchase behavior history. You can see what they purchase, and this is how drilling down can help close more deals, close them more quickly, generate cross selling opportunities and increase revenues. When you understand that this is what you can do, for the individual sales rep, first you summarize the information so it’s understandable to him or her. Then it should be intuitive where to drill down to help make the specific decision.
Read how InetSoft was rated as a leader for ease of use in Gleanster's agile BI report.
When you start offering a dashboard initially, you want to create a team that has a lot of expertise. Someone should be very experienced with dashboard intelligence, in the ideal world, in your organization. You also want to have a business analyst who responsible for collecting the requirements of the key performance indicators. And of course, you want the people who will be using the dashboard. Lastly you are going to have to work with the database team as well to make sure there are actual data behind these KPIs. With regard to the infrastructure, you need an IT team to work with, to make sure that database is ready to go, to establish the networking system, whatever the case may be.
And as with any project that cuts across functions in a large enterprise, you probably want project management involved to make sure everybody is communicating properly. Those are the technical aspects you would like to see in a dashboard development team. The first thing you start off with is figuring out the right KPIs, to make sure that the people who will be consuming the dashboards care about them. So generally speaking, you have to go to the stakeholders, ask questions, what are the corporate goals, what you are trying to improve in the organization.Next come questions about data update frequency. How often will people need to be using this dashboard? Do you use it in real time, or is the data being updated daily in a data warehouse? Finally, are there existing reports in the organization already that could be the basis for arriving at the right KPIs?
This Data Block technology is really made up of a back-end data mashup engine and a caching middle layer. There is a front-end we call the worksheet, which is a Web-based, spreadsheet-like user interface that a power user can easily use to combine and transform data blocks. The data blocks are typically started at the atomic data block level. That is typically created by an IT or a BI specialist. They choose these data blocks and performance tune those data blocks.
Then the power users can use those building blocks to build more complex, but more useful structures for answering the questions they have on hand. And all the data blocks you build, in turn become building blocks themselves. You can build on this architecture, building very sophisticated structures which you can use to answer a wide range of unanticipatible questions. " - Luke Liang, CEO, InetSoft