I recently took a very touristy outing in Boston when we went on the “Codzilla” boat ride in Boston Harbor. On a hot day, it is great. The boat goes out of the harbor, gains speed, and then tries to get everyone wet by turning sharply and quickly, stopping short and diving forward, and generally stirring up the water until it is falling over the decks and spraying the crowd. On the day we went there was at least one tanker moored in the distance – quite the opposite of the fast and agile boat we were on.
I recalled this experience as I read a nice blog post Boris Evelson of Forrester published recently that succinctly captures much of the essence of agile BI and data warehousing, which is something we here are Kalido (now Magnitude Software) have been discussing for years.
Boris notes that the battle over customer versus internal business processes requirements and priorities has been won by customer, and with this are two priorities that fall out of what Forrester calls the “age of the customer” – business agility and information agility. One quote from the post sticks out for me:
“However, while organizations of all sizes made significant headway over the last several decades in their enterprise BI accomplishments, many organizations still struggle with making their data and information management, BI, and analytics environments agile…Alas, enterprise grade BI platforms are often anything but agile.”
A lot of the hype and discussion around BI tools and platforms centers around the visualization capabilities, support for mobile, collaboration among users and the overall slickness of the user interface and experience. In-memory has become all the rage, but that is really about faster delivery and doesn’t necessarily address agility. On the BI tool side of the equation, few talk about what has to happen to make the data more agile. “What’s behind your BI” really does matter – this was in fact a marketing concept we discussed 6 years ago, but it still rings true today!
Being able to deploy and manage agility in the data is the heart of the problem. Even with the most agile of BI interfaces, if the underlying data structure behind it is not also agile enough to support change, the BI interface is not going to be able to compensate for it. Many of these tools rely on a data warehouse behind the BI, yet if your data warehouse is constructed using traditional ETL-tool-based techniques, you really aren’t going to be in a position to change direction or turn on a dime like the Codzilla to deliver a new business requirement or perspective. Rather, your supertanker of a data warehouse is going to be slow to turn, slow to change, and difficult to alter its momentum to support new requirements.
With the new BI reality Boris discusses, as enterprises attempt to keep pace with changing customer business process requirements and priorities, you are going to have to think about how to make sure the information infrastructure behind your BI is going to become more agile. This is the benefit of, and at the heart of, our model-driven and automated data warehouse solution. We’ll be talking more about this concept of data warehouse automation and the benefits it brings to making agile BI a reality through a more flexible and agile underlying information infrastructure.