Applied BI

It seems like Data Warehouses are hip again 09 February 2009

The last few years, I´ve often got the impression that quick ROI is everything. Even though we are facing tough economic times, many of the companies I talk to invest in centralized Data Warehouses now, not just more data marts. Some of the reasons for this development are:

1) Easy to use tools such as QlikView and Microsoft Office has made many more users develop analytical capabilities and exceptations to be able to act with facts. But, since different users working with different data sets have come to different versions of the truth, this has also been a problem.

2) The cost of maintaining data quality in many different installations and data marts starts to be significant, it is like maintaining point-to-point integration, instead of using an integration product. 

3) A centralized, quality assured and secure Data Warehouse lets users be even more creative and to be able to create their own analysis on quality data.

Of course there are downsides, since collecting data nobody uses or that data may be missing when it is actually nedded. BUT, the point is that this is a cycle, people start with personal or team solutions based on for instance QlikView, when the valuable data is known, it can be migrated to the corporate information model and then into the Data Warehouse. Not the other way around.