How to end up with the right model, right code, proper data dictionary, clear data lineage…
A quick note about a typical challenge, based on the real life experience.
Let’s imagine for a minute a computer game where we go from level to level until we get to the final winning goal.
One of the lower levels is where one can typically find themselves, and the challenges one is facing are marked below – illustrating where the organization is at the moment and where it wants to be.
- Level 1 is: we have data, developers, and they write code
Everyone surely has achieved this level
- Level 2 is: we can answer questions about data we collected and the code.
E.g. why we have system XYZ, what it does, who uses it, what data it collects and generates, etc.
Challenge: often, the organization will have only quantitative information.
- Level 3 is: we know the metrics and operational stats
Meaning we know how often our database system fails, how stable it is, performance metrics and SLAs, costs of operations, costs of making changes. How fast can we turn around a request?
Challenge: often, organizations don’t have qualitative information
- Level 4 is: we know what needs to be improved in our database, and who the benefactor is of the improvement
Challenge: Gap detection in a solution that failed, cause determination, solution planning
- Level 5 is: we possess knowledge that allows us to suggest a change that might benefit the business/owners of the database
Challenge: Information processing as means of change
- Level 6 is: execution. We implement the changes proposed.
The final note: Failure to understand the business needs and the questions that are being asked by them will result in a sound technological solution that doesn’t fit the organizational needs.