How do I improve inconsistent and incomplete data documentation? What has worked for your team?

In my experience, tackling inconsistent and incomplete data documentation requires a two-pronged approach: standardization and collaboration.

Standardization - We implemented a template-based system for data documentation. This template outlines key details like data source, format, definitions, and update frequency. This ensures consistency across all documented data elements.

Collaboration - We created a centralized knowledge base where everyone can access and update data documentation.  This fosters a collaborative environment where data owners (e.g., different departments) can contribute their expertise and keep documentation current.

For example, if your marketing team initially documented "customer clicks" inconsistently.  Through collaboration and the standardized template, you can now have a clear definition that includes click type (link, button, etc.) and platform (website, email).

The value proposition is that you not only identify inconsistencies but also facilitate the creation of a system that encourages ongoing data stewardship and collaboration. This ensures your data documentation is not just improved but remains valuable and up to date over time.

Inconsistent and Incomplete Data documentation impacts data accuracy and reliability. It is important to implement a document management system as to implement the required data documentation controls. This has worked very well for our team

Technical people never enjoy doing documentation and often time is not allocated or is insufficient in the project plan to allow them to do it properly. So build it into the project plan and review the developed documentation in the post project review to ensure it has been well done. By making this part of the signoff for the project completion it is not forgotten.

Asking your data team to do the documentation without ensuring they have the skills needed to do it well is setting them up for failure. So define the documentation standard required and train them to meet that standard. Consistency is key if you want to get consistently good documentation. So spell out what data documentation is required for every project, every change etc.

When issues with the documentation are discovered, the appropriate person is notified and the documentation is corrected.   The documentation is done as part of the project  or task  so this rarely happens however it is possible that mistakes can be spotted later or updates to a process are required.  

Inconsistent and incomplete data documentation causes lot of issues in data insights. It is important to have right data with right documentation in an organized form may be through document management system. This is important to deliver data driven business outcomes