Data Governance - the key to your success as a data professional

Disclaimer: This article is in partnership with Amy Elliott, MBA and Data Mentorship Program

Data Governance is an important part of the data cycle for any data organization. Maintaining strong data governance process is not only the leadership's responsibility but also every data professional's responsibility. Amy shares some of the ways data professionals - of all levels - can start their data governance journey.

--------

Continuing the #DataGovernance discussion this week, our mentor and advisor to the Data Mentorship Program , Amy Elliott, MBA ?talks about some #bestpractices in data governance. These best practices should be followed by all #dataprofessionals.?

With the fast changing data world, it is crucial to follow specific data governance rules to establish and build your credibility as a #dataprofessional?

Good data documentation is both something organizations should prioritize and a continuous process.?Oftentimes documentation is skipped (for various reasons), but the most common I’ve seen is related to priority.?If you make documentation a priority you can save time in the long run, something many organizations overlook.?Three key things you might want to consider (outside of making it a priority), start somewhere and evolve over time, don’t hesitate to leverage tools, and organization is key to usage!

Start Somewhere - Evolve Often

Rome wasn’t built in a day, neither should implementing good data documentation.?Start by identifying a few things you can start (or stop) doing now that can improve your current data documentation.?You should keep in mind your audience, who is the documentation for??Is it for your technical team, power users within the business, or executives.

Here are a few ideas (just to name a few):

Comments in Code - If the person who wrote the code left the organization, would someone be able to look at the code and understand what was done and why??If you share code with your power users, do they understand how to modify it?

Data Lineage - Can you tell the full data flow for how your dashboards and reports are populated??If one of your sources needed to be archived, do you know what the downstream impact would be?

Data Dictionary / Metadata - Do your users know what the fields are??What is the source of record??Who owns data quality??Are they visible on reports or dashboards??And if so, is it set up to update the content in one place and push it out to all dashboards/reports?

Are Calculated Fields well documented? - Who owns the definition of the calculated field??What is the calculation??Are there different departments/teams that calculate it differently??Is that calculation documented on dashboards/reports?

Data Quality - Do you have dashboards or reports showing the quality of your data? Do you know what processes failed and have missing data??Do you know if your transformations are causing data to be lost??Can you track errors in data quality for manually entered data (CRM data updated by a contact center)?

Chart Formatting, labels, and notes - Are your reports or dashboards miss-leading??Are there good labels on your charts??Do you have notes or annotations helping explain what’s being visualized?

Don’t Hesitate to Leverage Tools!

There are so many different tools out there that can help with data documentation.?Have you grown out of your home grown system or prefer your team to focus development efforts elsewhere??There are a lot of different data governance tools out there designed to help with data documentation.?I’m not going to go into detail of any specific one, but as I mentioned before make sure to pick one that has the features you need.?Just because a tool is highly rated doesn’t mean it will work well for what you need.

When looking at the tools you currently have, you might want to look into what features they offer.?There are some BI (Business Intelligence) tools used for creating dashboards that have built in capabilities to document what fields are and show data lineage.?These can be really helpful when getting started; however, they may not always be the best solution as they don’t typically do everything.?

When thinking of data documentation one area to also consider setting up your tools to tie our project management into your code deployments.?This works great because it gives more visibility into what was done when, why it was needed, and who requested it.?You can even use that data to highlight areas for improvement!

Organization is Key to Usage

Chances are you will end up with documentation in different places.?Try to make sure you aren’t duplicating efforts.?If you use your BI tool to document fields and data lineage, can your data governance tool dynamically pull that information??

What about any documentation your team does??You know, those spreadsheets, word documents, and visual diagrams??If you don’t have it organized and a clear process then you could put a lot of effort into the documentation only to be never seen again.?These types of files can be easily organized in a data governance tool, company intranet, or even in a team calibration tool.?If you're just getting started, sometimes a quick reference sheet can help point people to the right places.

Think about how your audience will use the documentation, ask them questions about how they would use the documentation.?The best way to approach this is to understand what problem you are trying to solve.?If you have a lot of great data but no one knows what you even have, they may spend more time trying to make things more complicated.?Organization and how you leverage the tools together is critical to the overall usage of your documentation that often times results in a lot of saved time for everyone involved.

There is no one size fits all solution to good data documentation, keep in mind that what works well for one organization or team may not work for another.