Databases Reference
In-Depth Information
Figure 1-7.
Analytics organization models
In a “decentralized” model, each business or function will have its own analytics team:
for example, sales and marketing will have their own team, finance will have their own
team, etc. On the one hand, this enables rapid analysis and execution outcomes, but on
the other hand the insights generated are narrow and restrictive to that business function
only, and you will not reap the benefit of a broader, game-changing idea. In addition, the
focus and drive for analytics is not driven top down from the highest level of sponsorship;
as a result, most analytics activities happen in bursts with little to no strategic planning or
organizational commitments.
The “shared services” model addresses a few of the shortcomings of the
decentralized model by bringing the analytics groups into a centralized model. These
“services” were initially governed by bygone systems, existing functions or business units,
but with a clear goal to serve the entire organization. While these were standardized
processes, the ability to share best practices and organization-wide analytics culture
is what makes the shared services model superior to the decentralized model. Insight
generation and decision making could easily become a slow process: the reason is
that there was no clear owner of this group, and it is quite common to see conflicting
requirements, business cases, etc.
The “independent” model is similar to the “shared services” model but exists
outside organizational entities or functions. It has direct executive-level reporting and
elevates analytics to a vital core competency rather than an enabling capability. Due to
the highest level of sponsorship, this group can quickly streamline requirements, assign
prioritizations and continue on their insight generation goals.
