By David Loshin

The three steps that I suggested in my last post about where to begin with data quality are truly meant to help determine where to begin, but also to guide the development of a longer term strategy and plan. Let me recall the three steps, but this time put them into the long-term perspective:

  1. Solicit data quality requirements from the business users. The objective of this task is to understand how information contributes to value creation, the users’ perceptions of data usability, with the intent of identifying the key value drivers for data quality and consequently determining key data quality measures as well as the levels of acceptability in meeting the business users’ needs.
  2. Perform a data quality assessment. The assessment serves the purpose of evaluating the current state in relation to the defined data quality measures.
  3. Establish a process and repository for data quality incident reporting and tracking. This puts practices and capabilities into place for addressing data issues in a way that continually improves the environment to achieve levels of acceptability for all data quality measures.

The user data quality requirements frame a desired end-state for data usability. Matching the measures collected during the current state assessment against the levels desired by the users provides a gap analysis that will highlight issues in areas need of specific improvement.

Employing a prioritization strategy to the list of documented issues provides a tactical roadmap for evolving toward the proposed end-state by incrementally addressing the root causes and eliminating the sources of data quality issues or by instituting methods for early validation and error detection.

In other words, the starting steps not only provide immediate value, they also help the data quality practitioner to craft a medium- and long-term plan for data quality improvement.





Leave a Reply

Your email address will not be published. Required fields are marked *