Saturday, June 28, 2014

Interactive Data Visualization

Recently, "interactive report" is becoming a hot topic in data visualization.  I believe it is becoming the next generation UI paradigm for KPI reports.

Interactive report is sitting somewhere in between static report and BI tools …

Executive KPI report today

Today most executive reports are "static report" provided by financial experts by pulling data from various ERP system on the regular basis, summarize these raw data in a highly condensed and simplified form, then generate a static report for the execs.  When the exec gets the report, it is already in a summarized form that is customized based on his/her prior requirement.  There is no way to ask any other question that the report is not already showing.  Of course, the exec can ask for a separate report which requires additional development time and effort on his/her staff, but also need to wait for the new report to be developed.

This is a suboptimal situation.  In order to survive or maintain leadership in today's highly competitive business environment, execs not just need a much broader perspective (from wide variety of operation data) to make his/her decision, but also he/she has to make the decision fast.  Static report cannot fulfill this need.

Business Intelligence Tools

On the other hand, BI tools (such as Tableau) or OLAP tools can do very detail analysis in wide range of data sources.  However, using these tools to perform more detail analysis (such as slice/dice/rollup/drilldown) typically requires specially trained data analysis skills.  In reality, very few execs use these tools directly.  What they do is to ask their data analyst to prepare a static report for them using these BI tools.  The exec still get a "static report" although it is provided by the BI tools.  Whenever they need to ask a different question, they need to go back to the data analyst and ask to prepare a separate report.

There is a gap between the static report and BI tool.

Interactive Report

"Interactive Report" provides a new paradigm to fill this gap.  It has the following characteristics …
  • Like a static report, "Interactive Report" is still based on "static data", which is a fixed set of data generated in a periodic batch fashion.
  • Unlike static report, this pre-generated "static data" is much larger and wider that covers a broader scope of questions that the execs may ask.
  • Because the "static data" is large and wide, it is impossible to visualize all aspects in the report.  Therefore, only one perspective of the static data (based on the exec's pre-specified requirement) is shown in the report.
  • However, if the exec wants to ask a different question, he/she can switch to a different perspective of the same "static data".

By providing a much large volume of static data, "interactive report" provides a more dynamic data navigation experience to the execs to find out the answer of their ad/hoc unplanned questions.

There are many open source technologies (such as Googlevis...) to support interactive data visualization from which the "interactive report" can be built.  And many of them provides a programmatic interface with R so now data scientist without much Javascript experience can produce highly interactive web pages.