Managing Your Data Using Common Sense
Today we are happy to present our first guest article by Bryan Rockoff from Bespoke Analytics LLC. Bryan is an extremely experienced analyst in the BI space and will share some of his insights in the field of data management. We look forward to working with Bryan more in future.
- SyncHub Team
When do I need to think about data management?
Let us start with some guidelines about when you need to embrace these concepts required to manage your data. You do not need to worry about managing data (outside traditional housekeeping activities) when you have data that is:
only used within a single application
not shared between apps
not merged with nor integrated with data from another application
ERP systems are great examples of doing a fine job of managing data within the application.
The world gets more interesting when two systems need to share data. When data sharing is required, the people responsible for the applications get together and agree who does what, what the data elements mean, how to merge the data, which application is responsible for creating and maintaining data elements, that sort of thing. This could easily be handled in a couple of meetings.
And what do I need to think about?
In today’s complex business environment, companies may have hundreds of systems that exchange data. Consider the daunting task of getting hundreds of application owners to agree on a single set of principles. Consistent methods and processes, paired with standardized tools, are needed for efficiency and effectiveness.
Three foundational steps are required to manage your data assets effectively:
Leadership must provide the direction and the high-level support to guide the utilization of your data. I like to think of that as Why. It could be as simple as stating the value an organization will get in return for improving the management of its information assets.
Next you need to agree on the rules or policies that implement the Why’s. This describes What you need to do. Each rule or policy must tie back to a Why. Be wary of creating rules and policies that look like the US Federal Tax Code with so many regulations it is impossible to follow. The K.I.S.S. principle works well here.
Finally, How. Here is where you describe the way you are actually going to do the work, transitioning from aspirational objectives to practical implementation. It may take one or more processes or tools to implement the What. The How must tie back to a What which ties to a Why.
How does this work in real life?
Here is a real-world example from a client engagement:
The client has a vast library of financial measures. Unfortunately, the names were not meaningful, they were minimally documented, and the nuances were shared verbally. The phrase “tribal knowledge” was taken to a new level. All prior attempts to clarify and standardize institutional knowledge around these metrics had failed due to a lack of executive sponsorship.
We came in with an executive mandate and documented each item down to the formula and approved uses. We also documented the data sources, targets and transformations - bringing all this knowledge into a data glossary, and making it broadly available across the company. This standardized the use of metrics, and ensured leadership got consistent results and better decisions.
So it’s not as complicated as I thought…
You may have noticed I have not used terms like “data governance”, ” information strategy”, and “data management”. While there are standard definitions of these terms, there are some differences across models and strategies in how the functions are grouped and interoperate. I have avoided them on purpose so you, the reader, can focus on what is needed to be successful, rather than the terms.
You do not need to be an expert in all of the different terms and the underlying disciplines. Just apply common sense. Stay focused on helping the business. At the end of the day, it is not about a mastery of any methodology or approach, it is about helping business-people use data.
