I cannot tell you how many times I have had customers that do not have a specific vision for their data.  What I mean by this is that there is no data strategy.  A data strategy would consist of the following:

1. A diagram defining and depicting all aspects of data stores across the enterprise.  This enables us to view the enterprise as a whole, and not ‘piecemeal’ by each system.
2. A road map for elimination, upgrade or replacement of the data stores (and their resulting programs).
3. A system inventory of information for each of the components (platform, software, business users, data structures, etc.).  You may want to consider including business owners for each system that is assessed during this strategy.
4. An interface map to show where (as well as how and with what tool) data is moved between systems, such as the data warehouse or you master data environment, etc.  You may want to include frequency of data movement.
5. Note and understand the data quality requirements across the data strategy.  For example, is it OK to have 75 states listed in the state table?

What I see at these companies is a tremendous amount of redundant data, stored multiple times across the enterprise.  This not only causes increased inefficiency for the programs and interface, but it creates complexity when it's time to perform upgrades of the databases that contain the data stores.  Even though space is real cheap, it still adds more need for disk space.

Overall, I think most of these companies simply do not have the resources, nor the time, to create a data strategy.  So, maybe we need to consider contributing to the data strategy when we do a new project. Thoughts?