Standing Up a Data Management Capability
By Wil van der Walt, Vice President Enterprise Data Governance, Live Oak Bank
Managing data has been around for a long time, from the pre-computer era of paper copies to early computers storing electronic data. It is only in the recent past that data has been managed formally and taken on a boardroom presence. Standing up a data management capability is a common discussion point at all levels within organizations.
Data Management Maturity Model
A good starting point when looking to implement a data management capability is to determine the level of maturity. A maturity model provides insight into the current level of maturity and can drive the roadmap. There are many maturity models available, and consultants can provide guidance on implementing them.
The benefit of the maturity model is that it baselines the level of maturity and allows the organization to track progress. It provides a comprehensive view of data management capabilities and what needs to be done. It must be noted that most models are fairly generic and will need to be customized to the organization’s needs.
Strategy Alignment
Aligning the data management strategy to the organization’s strategy is key to delivering the right value at the right time. Strategic projects need to be reviewed and determine the need for data management capabilities. For instance, with digital transformation, the need for quality data is paramount to the success of the project.
Being able to identify data management capabilities that impact revenue generation or operational expenditure is key to demonstrating value-added activities. Understanding customer behavior using clean, accessible and trustworthy data supports customer retention and growth. Embedding data teams in the business process provides the opportunity to address data issues at source.
Data Architecture
Data Architecture provides the blueprint for managing data. It is a conceptual framework of all the systems/applications used by the organization and the data within. Systems / applications are categorized by purpose throughout the technology stack, for instance, Salesforce as a CRM tool and Databricks as a Data Lakehouse. Data is categorized by domain throughout the data lifecycle, for instance, customer and product domains.
Data flow diagrams provide visual illustrations of the data within applications and the movement across applications. Strategic projects can make use of these diagrams to determine their technology and data needs. If it exists, use it, if not, ensure new technology integrates with it!
Data Inventory
A data inventory is the systematic cataloging of an organization’s data assets. It includes data element definitions, purpose, and associated business rules. It should also extend to regular terms used by the organization to provide a shared understanding.
The data catalog provides a connection between the physical data elements in a system/application and the logical or business friendly data elements. This supports data lineage initiatives in that the provenance of data can be verified through the definition and purpose of each of the data points.
Data Quality and Accessibility
Data quality and accessibility capabilities are independent activities that provide trust and integrity in the data. Data quality is managed through the Data Quality Dimensions. Data must be complete, unique, valid, accurate, consistent, and timely.
Data accessibility is the ability to access and use data. It relates to the right people having the right access at the right time to the right data to perform their duties adequately. Managing user roles and permission sets is key to delivering accessibility and supporting the overall cybersecurity strategy.
Artificial Intelligence
It is never too early to start the artificial intelligence (AI) journey. AI is the science of making machines think like humans and is a support function of Data Management. It requires selection and planning and many attempts to refine models to the point of value add. Getting started early allows for fast fails and matures in line with the overall data management capabilities.
Data Governance
Data Governance is the exercise of control and shared decision-making over data as an asset. It is a key component of any data management capability. Good governance provides the trust and integrity of data across the organization.
Establishing a Data Governance Council with policy approval and prioritization accountabilities drives the data management capability. A stewardship program assigns custodial responsibilities to people within the organization who have the subject matter expertise.
Policies are drafted from standards that have been set by best practices applied to specific data-related issues. This drives the resolution of data issues and the adoption of governance practices. Risk based governance can be applied once this process has been established.
Conclusion
Data Management is key to delivering a competitive edge in a digital world. It is owned by the Chief Data Officer and must be represented at the board level. It is not a generic function where one size fits all but a carefully crafted capability that forms part of every organization’s “DNA”. Investing early and committing time and effort to delivering a world-class Data Management capability will positively impact the organization’s overall performance.