@work Toolbox

Business Intelligence Reports: make your figures decisions enablers


Historically IT has been introduce into legacy industries to improve task automation like the machines did during Industrial Revolution. Like machineries, computer can provide cadence information, but can also keep historical data which can be then reviewed and analyzed to provide a business related information. This process is called Business Intelligence.

Business Intelligence concept was not actually born with Computing Era but later in Industrial Revolution around mid-1860s. However Business Intelligence has been introduce back in 1950s by Hans Peter Luhn (IBM Research) with following definition:

the ability to apprehend the interrelationships of presented facts in such a way as to guide action towards a desired goal

During Mainframe Era, computing performance and data storage grown so fast that we started to talk about Data Warehouse to define volume of data available. However initially, analyzes where limited to common operational indicators for operational performance purpose. Those Key Performance Indicators (KPI) used to be reported on monthly basis to proof how well (or not) a department/group/production line/etc. is doing.

Here are KPI examples which can be used for IT Department:

  • IT Systems availability (can be split into towers: applications, hosting, network, etc.)
  • First level incident resolution
  • Incident resolution on time
  • Number of crisis
  • Volume of data (stored or transiting on WAN links)
  • Etc.

Basically KPI are giving us a recent picture of our scope of responsibilities: next step will be to use them to improve our self or be more pro-active on what’s may happen.

Indeed, everybody can do Business Intelligence as its own level as long as you are taking time to read KPI in a historical way to extract a trend, pattern or behaviors. But as you may expect you need it will take some time at any stage of such process, this is why you need to make sure that your BI project is following 3U rule: useful, usable, used.

  • Useful: There is a formal need to a specific information request/goal. It will give you sponsorship of the final BI report customers
  • Usable: It has to be easy to read, understand and reuse the information provided. If it is too complicated, data won’t be used.
  • Used: Is the resulted information really used? If not let’s stop wasting your time on delivering it and come back to your internal customer to better understand what he really needs.

Once you get it right then you can enter into the SQARA wheel:

  • Select data sources needed
  • Query data as per your need
  • Analyze by extracting trends and behaviors
  • Report to customer with proposed action items (this is your real added value)
  • Action agreed activities to interact with current situation (this is where you confirm your « Used« )

Then run the process once again in order to see our trends and behaviors has been impacted.

Following such process won’t replace a Business Intelligence expert nor a Data Scientist but it is simple enough to  get some quick wins and then bit of satisfactions.