Sustainable data quality management is a long-term, ongoing approach rather than a one-time event. Here’s what it looks like in practice:
Key Characteristics:
- Continuous Process: It’s not a point-in-time data cleanse activity, but an embedded business-as-usual process that delivers high-quality data forever.
- Three-Stage Approach:
- Find it: Discover and identify poor quality data using tools like infoboss
- Fix it: Correct data at source and address root causes
- Monitor it: Continuously monitor data quality to ensure it stays fixed
- Core Components:
- Data Ownership: Assign clear ownership to data assets with accountability
- Quality Rules & Validation: Establish and maintain data validation rules to prevent poor quality data from entering your systems
- Automated Monitoring: Implement continuous, automated monitoring and alerting when quality issues occur
- Performance Tracking: Monitor data quality improvement indicators over time
- Stakeholder Views: Provide tailored data views for different stakeholders
- Cultural Aspects:
- Build a sustainable data quality improvement culture within the business
- Support data democratisation – empowering those who can improve data quality with the necessary tools
- Ensure data owners are alerted when poor quality data enters the data estate
- Preventative Focus:
- “Switch off the dirty data tap” by identifying and fixing root causes
- Implement processes and tools that ensure business data rules are maintained and nurtured
- Put validation and control measures in place at the point of data collection
Why Sustainability Matters:
Research from the University of Denmark shows that data quality degrades at approximately 2% per month if left unmanaged. This means that without sustainable practices, data quality improvement initiatives will fail and need constant repetition.
Implementation Approach:
Sustainable data quality management requires:
- Senior leadership sponsorship
- Dedicated resources (project manager, data quality excellence center, data stewards)
- Appropriate tools for discovery, monitoring, and alerting
- Integration into business processes and systems
- Long-term commitment beyond initial “fix” projects
The goal is to embed data quality management into the operating model so that high-quality data becomes a natural outcome of business processes rather than requiring periodic cleanup initiatives.