Ready to learn Big Data? Browse Big Data Training and Certification Courses developed by industry thought leaders and Experfy in Harvard Innovation Lab.
A company can put master data management (MDM) in place and make a tremendous effort to ensure its data quality. It can take pains to develop good data governance procedures and invest in leading-edge analytics technologies to make the most of its data. But any errors in its reference data can undermine all of these initiatives.
Reference data is a non-volatile and slow-moving subset of enterprise data. It is often standardized by external bodies and businesses generally use the same reference data throughout their operations. Examples include country codes, SIC codes, currencies and measurement units.
Typically, companies make use of several applications that feature drop-down lists of reference data, and may have various forms that are prefilled with certain reference data. To keep this data synchronized, some companies rely on reference data management or RDM.
Any inconsistencies in the reference data that a business relies on can result invalid transactions, incorrect reporting, and may require manual fixes that result in higher operational costs. In a recent survey by Accenture, 70 percent of the companies polled acknowledged that reference data quality issues had increased their costs. The study put the total impact at more than $6 Billion for 2015.
These are some of the issues that can result from poorly managed reference data:
1. Higher maintenance costs: Inconsistent reference data across applications results in invalid mapping, which can cause data corruption. Rectifying the problem is costly—especially if transcoding is required during ETL to handle the discrepancies. This may require creating new data objects within MDM or creating custom code.
If mapping procedures aren’t standard and automated, corrective action will be even more expensive. Reference data discrepancies can also affect system stability and extend the time it takes to roll out new applications.
2. Operational Inefficiencies: When reference data such as ZIP codes or country codes don’t match across different applications, transactions may rejected, leading to system-wide alarms and the need for extensive manual intervention. And as companies expand into different international markets, it becomes more challenging to ensure that the business’ refence data remains consistent.
3. Inaccurate analytics and poor data governance: Reference data discrepancies degrade a company’s master data, which affects all of its operational and analytical systems. This will also lead to inaccurate reporting and analytics. And since good data governance depends on standardization to facilitate data mapping, auditing and management, any reference data inconsistencies will significantly undermine these efforts.
4. Non-compliance: Regulatory initiatives such as GDPR require companies to manage all of their internal and external data in an integrated fashion. If a business’ reference data is inconsistent, the requisite data management transparencies are difficult to achieve, exposing the company to a variety of the risk of regulatory and non-compliance risks.
To avoid or correct these issues, a company should make RDM part of its master data management and an integral part of its data strategy.
Managing reference data successfully
To manage its reference data effectively, an organization needs a comprehensive data strategy and a complete RDM solution. A good RDM tool should support locale-specific values plus internationalization for full global compliance. The inclusion of a central, business-user friendly repository can simplify the complex mapping of different data hierarchies, while complete transcoding of incoming data across all operational systems without the use of custom code can cut down on master data maintenance costs.
Other features should include:
- The ability to define and manage reference data across multiple functional areas
- Easy integration with existing MDM, applications and third-party data sources
- Cloud-based deployment for greater scalability, zero downtime and future-readiness
- An easy to use data steward interface for creating, mapping managing and storing reference data
The software should also offer robust change management, audit controls and workflow functions for reviews, approvals and audit trails.
Reference data anchors a company’s data initiatives, and maintaining consistent, high-quality reference data is essential to successful data management. A sound RDM solution provides enterprise-wide benefits, including lower maintenance costs, greater operational efficiencies, more accurate analytics, reliable data governance and full compliance.
First appeared in Information Management