From almost any angle, the adoption of an enterprise data warehouse (EDW) makes so much sense. After all, an EDW - designed to produce a unified and comprehensive view of data that an organization accumulates - would logically provide a myriad of business benefits. A companys business knowledge resides in one central location, making it easier for users to access. Profitability at all levels of a company can be quickly reviewed. But the ultimate benefit is that the EDW provides organizations with a single view of the enterprise data across disparate business units and IT systems - a single version of the corporate truth, if you will - to assist management in making sounder and better-informed business decisions. Yet, despite the promise of increased business efficiencies and greater overall profitability, many companies find that the buy-in process for an EDW can be an overwhelming challenge. There are a number of reasons for this reality, yet in the end, they all tend to fall into one of two categories:
Before examining in greater depth the reasons why it can be so difficult to rally universal approval for an EDW, it is necessary to define the term.
As mentioned, a true EDW is a single version of the organizations business data. Given the variety of IT systems and repositories of data that reside in most business organizations, particularly larger ones, creating such an entity can be daunting. Still, unless it consolidates all data elements into one cohesive entity, it cannot truly be labeled an EDW.
The EDW must incorporate many departmental disciplines, including but not limited to accounting, sales, R&D, human resources, etc. It should be flexible enough to incorporate new disciplines in the future as necessary.
While single-subject data marts have traditionally been built using denormalized models as their underlying architecture, EDWs tend to utilize a more normalized model. User friendliness and orientation should be key priorities for the proper construction of the EDW in order to encourage widespread use of the system as well as allow users to operate solely at the business level; that is, no one should have to be overly technologically savvy to share in the systems benefits.
With this definition as a backdrop, it is easier to comprehend the challenges required to gain approval for the EDW concept. These challenges are not insurmountable, as long as they are correctly and quickly identified.
One of the primary questions is, what does each group within an organization require to ensure their initial and ultimate acceptance of the EDW project? Often, unless an organization takes the initiative to develop a center of excellence (CoE) - an internal group that provides consulting services and oversight to the various data warehousing development groups within an organization - the company just flails around trying to make the project work.
For the end-user community, approval is a byproduct of knowing that they will have access to a tool that is extremely intuitive and that does not require them to learn a foreign language. For the IT support staff, it is being clear on the exact nature of their domain responsibilities, as IT is increasingly becoming the gatekeeper of a companys information infrastructure. For executive leadership, it is having the confidence that they are going to be able to get the trusted information they need to make intelligent business decisions as well as understand the specific system ROI - if not in actual dollars, at least in strategic and productivity improvement. Compliance officers have to be assured that the system will operate in accordance with all of the appropriate rules and regulations of their industry.
Ultimately, everyone has their own agendas and priorities when it comes to the EDW project. What must happen is that all of these agendas and priorities must align seamlessly with the overarching strategic goals of the organization. Once this occurs, the clarity of the buy-in can begin to take shape, with each person realizing that he is just one cog - albeit an important cog - in the overall machinery.
This begs the question: where do these overarching goals originate? When youre talking about a medium-sized or large enterprise, these goals are primarily established by the organizations executive leadership team. Typical high-level strategic objectives include cutting service response time by 20 percent; adopting a just-in-time operating policy where no more than seven days worth of supplies are kept on hand; growing a certain percent per year for the next five years with a hybrid organic/inorganic strategy; opening up new markets; or increasing net profits by x number of dollars.
Once these objectives are clearly and succinctly laid out by management and filtered down to the individual departments, the best scenario is that each department not only embraces them but immediately sees the obstacles to achieving them with the companys current technological resources. The ideal response would be, Gee, Id love to increase customer service, but we have one database here, another one over there and a manual process here, so theres just no way our current systems can help us meet this goal. Consequently, the introduction of an EDW that can help knock down these obstacles would ostensibly be a welcome addition.
Sadly, personnel from different departments often harbor certain fears about new technology as well as being generally defiant about embracing change. Take IT, for example. Perhaps they simply dont buy in to the project right from the beginning. The support staff thinks, We dont really need this, and it will just be more work for us. Things are fine the way they are.
Defining the EDW
The Challenges - and Overcoming Them
Who Sets the Goals?
Fear of Technology and Change
By
NOV 30, 2007 11:54am ET
Enterprise Data Warehouse Challenges
Advertisement
Add Your Comments:
You must be registered to post a comment.
Not Registered?
You must be registered to post a comment. Click here to register.
Already registered? Log in here
Most Read
Most Emailed










Be the first to comment on this post using the section below.