Unifying data quality with data governance ensures trustworthy data, stronger compliance, and effective AI-driven decision-making in financial services.
Those in financial services can have the most sophisticated data governance framework in existence, but if the data is unreliable, the whole process is a waste of time.
Data governance without strong data quality is bureaucracy masquerading as governance. And those handling data quality without governance processes to guide them could end up potentially not meeting accuracy rates or complying with industry-wide regulations.
Smart financial institutions understand that data quality is a vital metric by which data governance should be judged.
Example of data governance without data quality
A global financial services firm spends eighteen months building a data governance program, with roles defined, policies documented, and a data catalog deployed. It is then declared a success by the senior team.
However, six months later, a regulatory audit reveals that customer risk scores used for compliance reporting were based on incomplete and inconsistent data. This leaves the firm facing fines and realizing that building a strong governance framework on unreliable data was not a good idea.
The roles of data governance and data quality
Data governance frameworks can create a misleading semblance of control while leaving the actual quality of data unaddressed. In the age of AI, with data being ingested at scale, this issue is more important than ever as those in financial services look to deliver a competitive advantage that can only be achieved with access to accurate customer data.
Financial institutions need to realize that the role of data governance is to provide the structure, roles, definitions, and accountability when it comes to data, which is important when regulators are increasingly asking not just for policies, but verifiable accuracy and proof of effective controls.
By contrast, data quality provides the evidence—the proof that the structure is producing trustworthy data. Its quality monitoring produces an audit trail that demonstrates compliance is real, not theoretical.
Without quality metrics, data governance is an act of faith, and with them, it becomes an act of management.
Data quality must be a continuous process
One of the most common failures in data governance programs is treating data quality as a project, when delivering it needs to be an ongoing process. Too often, a team is assembled, the data is profiled with issues identified, and an effort is made to clean the datasets. Reports are generated, and then the project ends. Not surprisingly, six months later, the data has degraded again.
Data quality needs to be embedded into operational workflows—to take place on an ongoing basis in real time through batch processing of customer records, and also on a consistent basis at the customer onboarding stage.
This shift to delivering continuous data quality is what separates financial institutions that trust their data from those that merely hope it is usable.
The cost of separating data governance and quality
Too often, data quality and data governance sit in two separate departments within financial institutions. Data governance often sits within a central office, focused on policy and compliance, while data quality is handled by the IT or data engineering departments.
This causes issues if the data governance team defines policies that are technically impractical, while data quality teams remediate issues without understanding the business context or priorities. With this approach, issues commonly arise, leading to problems with accountability and the senior team possibly losing confidence in both functions.
Single accountability model is key
Data governance and data quality must be unified under a single accountability model, with shared tooling, metrics, and incentives.
Data governance needs to exist to enable data quality at scale. In practice, this means governance must define the rules and the acceptable accuracy rates, required completeness thresholds, and timeline requirements. Those in charge of data quality should then follow these rules, ideally by automating quality practices.
This way, governance establishes ownership and data quality enables accountability in the process.
In summary
Having accurate customer data is the foundation of a competitive advantage, operational efficiency, regulatory compliance, and AI-driven innovation. The issue is that financial institutions that treat data governance and quality as separate initiatives will find themselves with beautiful frameworks, but untrustworthy data. They will experience the worst of both worlds—the overhead of data governance without the benefits of data quality.
The question is no longer whether you have data governance—it’s whether you can prove that your data is trustworthy.
In fact, financial institutions that unify data governance and data quality, using governance to enable quality at scale, will achieve something more valuable than compliance or documentation—trust, both internally and with customers. And in a world where decisions are increasingly made by algorithms and AI, trust is the only currency that matters.
Barley Laing has 27 years of experience in the technology and data sector. As the UK Managing Director of Melissa, he is dedicated to addressing the data quality and ID/compliance requirements of organizations in the UK and globally. He also oversees a team that offers data consultancy, sales, and technical support for their extensive range of leading web services, applications, SaaS, and on-premise software solutions that improve customer engagement and adherence to KYC, KYB, and AML regulations. Contact Barley at barley.laing@melissa.com or connect on LinkedIn.