I Dream of Data

For those of you that aren't certified "data geeks", you would probably not understand dreaming of data.  For some of us this love affair is new, and for some of us this has been a long standing obsession.

I clearly remember the first time I got the data bug.  It was an entry level database programming course and the lab assignment was to build and execute SQL scripts (inserts, deletes, updates, selects) against a provided data model using an Oracle (v5 I think) database.  It just seemed so natural to examine the data model and the problem and just pick what you wanted and join tables to build "grids" of data.  The career path in IT was set from the 2nd year in university!

Throughout my career, I have always been fascinated by manipulating data to gain insight into what is happening in the business of my organizations.  This has fostered a deep appreciation for the value of Business Intelligence, but on a larger scale an appreciation for how overall data management is critical to the success of an organization.  The Business Intelligence discipline must be supported by other components of an Enterprise Information Management (EIM) strategy such as Master Data Management (MDM), Data Governance, Data Architecture, and Data Quality Management to provide maximum value to an organization.  To forge ahead without engaging the rest of your IT department in treating data as an asset is a recipe for frustrated teams and ineffective solutions.

So when I fall asleep, I dream of a perfect environment where all pieces of an Enterprise Information Management strategy are in place.  Sometimes I feel like I am drawn to a big light at the end of a long corridor filled with data.  When I finally get to the end I see beautiful visualizations of information presented by my local Business Intelligence portal, harnessing all the inherent value in my beautifully managed data sources.  This is a big dream, but what dream worth pursuing is easy?

Good night, sleep tight!

Ensuring BI Projects don't Fail

Presentation from Timo Elliott, Business Objects.  Great slides!

Why Business Intelligence Projects Fail -- And What To Do About It

Mark

Translating Back

Interesting article on how BI applications have to be traced back through the Systems Development Life Cycle to become business information again.  Here is a diagram from the article:

http://www.b-eye-network.com/view/index.php?cid=7566

Mark

Article on Pushing the BI Limits

This is a link to a great presentation by Timo Elliot a Senior Director with Business Objects. This is one of the best presentations I have seen on pushing BI to the limit!

http://www.timoelliott.com/blog/2008/06/delivering_results_with_busine.html

Mark

Squeezing the value out of what you have

Ran across this interesting post about how you can replace just the pieces of your mainframe application that require the most change, and doing it with an external business rules management system (BRMS).

http://smartenoughsystems.com/wp/2008/05/28/making-meals-from-your-mainframe-leftovers/

If you think about this, you can apply it to the BI spaces as well.  I can think of a couple places where this would apply.

1. Data marts, if you build your data marts properly, they are much less costly to enhance than to replace.  Rather than spinning up another data mart to get the slice of data you are looking for, you could leverage an existing data mart and add the data required to it.  This may require some retooling of the front end reports / dashboards if the existing data structure changes, but this is much less costly than having to build, support and store another data mart.
2. Tools, if your toolset doesn't support your current requirements, don't lose heart and buy yet another tool.  Work with your senior technical folks to come up with a way to extend the platform, there may be exposed services that can be recombined into something new.  Also, reach out to your vendor, explain your requirements and you would be amazed at what they can dig up from other customers.  Lobby your vendor to include new functionality in their next release.  The whole idea of crowd-sourcing is picking up steam, so the majority of vendors are working hard to include customer requests in their products.

Do less with more!

Mark