Why do most data warehousing efforts fail? I hear the same old reasons over and over again – lack of sponsorship, lack of alignment with objectives, slow time to market, etc. etc. etc.
I think the primary reason why data warehousing efforts fail is because they fail to clearly demonstrate value. If this is an IT driven project, the likelihood is that the team’s focus is on looking at what data sources exist in the organization, and how to bring them together into one huge database. The debate is on whether to use dimensional modeling or a normalized approach. It doesn’t matter!
If you are very lucky, some of the team members’ focus may be on the reports and dashboards that will be provided to the organization. At least this looks at it from a business or requirements perspective. This will work better, but the chances are that, unless the organization has reconciled to spend the million dollars on what they know will be a costly endeavor, this will fail also.
Why? Because the focus is not on clearly showing, in monetary terms, or non-monetary metrics, that the million dollars spent resulted in 5m dollars of additional revenue, or 3m in cost reductions, or 2m in efficiencies in your sales organization. So why bother?
This is why I thought of the phrase, “ROI Driven Data Warehousing”, to describe a methodology, a way of thinking, that would guarantee success one way or the other, even if it means cancelling the project. In the next posts, I intend to focus my thinking on the “how to” part of the process. The focus will be on coming up with metrics that would apply in certain scenarios, in my experience and in others. Stay tuned!