Banks often use "data quality" and "data certification" as if they meant the same thing. They do not. They answer different questions, and a bank needs both.
Data quality: how good is the data?
Data quality is about measurable characteristics of data, such as accuracy, completeness, consistency, timeliness and validity. Quality checks run rules against data and report results: how many records failed, which fields are missing, where values disagree between systems.
Quality tells you the state of the data at a point in time.
Data certification: can we rely on it, and prove it?
Data certification is a governance process. It confirms that a data asset has a clear definition, an accountable owner, mapped policies and controls, supporting evidence, including quality results, and a recorded approval.
Certification tells you the data is fit for its intended use and that someone has taken responsibility for saying so.
The key differences
- The question. Data quality asks "how good is the data?" Data certification asks "can it be relied on and proven?"
- The output. Quality produces scores, rule results and issues. Certification produces an approved, evidenced record.
- The owner. Quality is often run by data teams or stewards. Certification needs a named business owner plus an approver.
- The timing. Quality checks run continuously or on a schedule. Certification moves through gated stages with re-certification triggers.
- The audience. Quality serves data and operations teams. Certification serves auditors, examiners and the board.
How they work together
Quality results are evidence for certification. A Governed asset needs controls that operate, and quality monitoring is one of the most important. If quality results fall below thresholds, the certification should be reviewed.
Without certification, quality dashboards show numbers but nobody has formally accepted responsibility. Without quality monitoring, certification becomes a paper exercise that cannot detect deterioration. See our approach to data quality you can certify.
A simple model
- Define quality rules for each critical data element.
- Monitor them continuously.
- Link results to the asset's certification record.
- Trigger a review when thresholds are breached.
- Re-certify when the issue is resolved.
For the basics of certification itself, read what data certification is. To choose what to apply it to, start with critical data elements.
Why examiners care about the difference
Examiners rarely ask only whether data is accurate. They ask who is accountable, which controls apply and how you know they work. Certification answers those questions. Quality provides the numbers behind the answer.
To see both working together, explore our data certification software or book a demo.