Benefits of Effective Data Management | KPMG

8 Strategies to Increase Data Accessibility and Accountability

When organizations achieve effective data management, they break down data silos, enhance collaboration increase efficiency, optimize resource allocation, and mitigate compliance risks. These improvements ensure that business objectives are supported by an effective data strategy, maximizing the value of data and technology investments.

Here are 8 strategies to increase data management effectiveness:

1 | Establish Data Owners

A clear data ownership structure across the organization can establish responsibility for data in a particular domain (i.e., finance, supply chain, manufacturing). Data owners and accompanying stewards are responsible for data quality, accessibility and the provision of metadata. They can achieve this by implementing processes such as:

• Marketplace Management: Owners are responsible for the data marketplace, overseeing data definitions and ensuring adherence to governance policies.
• Review Process: These data owners should establish a review process to maintain data quality.

2 | Invest in Data Management Tools

Data management is a complex task which requires the right tools. Emerging capabilities in this technology category can dramatically improve an organization’s ability to manage and improve data quality. These capabilities include the use of Generative AI to automatically pre-populate definitions, track lineage and where data is used, and anomaly detection engines to find data quality errors proactively.

3 | Ensure the Data Owner is Close to the Data Users

It is critical that the data owners understand the needs of data users. This means this task of managing data must be driven by the business, with IT and the data management organizations providing automation, engineering, and process support. This is a difficult cultural shift for many organizations, and some strategies to achieve this include:

• Ensuring data owners know what data users need, why they need it, and how they use it. The data owner should ask if they are providing the data in the best format so the user can employ it productively. For example, a user might need a specific granularity at a specific time or by SKU.
• Ensuring data owners commit to providing the necessary data, even if availability is limited initially.

4 | Proceed with an Agile Mindset

Foster a culture of experimentation and learning, where learnings are valued, and the organization recognizes that progress, and iterative releases, are more important than a complete data model. Techniques important to support this mindset include:

• Agile Teams: Setting up cross-functional agile teams that include data scientists, analysts, and business users to refine data periodically.
• Continuous Improvement: Recognizing that data access will improve progressively with the help of data stewards who expand its access in a methodical, governed, and trusted way.

5 | Create Data Products

The data owner is a key resource whose responsibility is to provide data for their domain for users within their domain and across the organization. The emerging way to achieve this is for the data owner to create ‘Data Products’: a set of curated data which provides data for all users. The data owner should maintain that product in an agile manner, continuously releasing enhancements to the product to meet the emerging needs of the organization.

6 | Create a Data Marketplace

The data owner is a key resource whose responsibility is to provide data for their domain for users within their domain and across the organization. The emerging way to achieve this is for the data owner to create ‘Data Products’: a set of curated data which provides data for all users. The data owner should maintain that product in an agile manner, continuously releasing enhancements to the product to meet the emerging needs of the organiA data marketplace enables users to understand the data is available in an organization. In addition to user-friendly search and filtering capabilities to help users find relevant data quickly, the marketplace needs to provide certain other functionality to be effective:

• Metadata: Data must have definitions and lineage information so users can understand the data in context.
• Access Control: Data access must be effectively managed to maintain security and compliance.
• Data Quality: Checks on the data as it flows through the data supply chain need to be in place to ensure data is of high quality.

7 | Track Usage to Show Value

Use detailed data usage analytics and reporting tools to gain in-depth insights into data quality, utilization, and ultimately the value from data management investments. Some techniques to support this strategy include:

• Usage Monitoring: The data owner should monitor how many people use data products, how often, and whether they meet their intended purpose to determine value. It’s a question of accountability on the data owner’s part and the data user’s part. If not, then the development wasn’t worth the investment.
• Feedback Mechanisms: Providing mechanisms for users to submit qualitative feedback on data products.

8 | Be Cross-Functional

Empower each business function to oversee its own data domain while ensuring data sharing across the organization. Protocols, policies, and structures must be in place to ensure that those domains can be shared throughout the organization. Data stewards should be engaged to allow for close collaboration. Effective strategies include:

• Centralized Governance: Creating a centralized data governance team to oversee data-sharing policies.
• Data Integration Platforms: Using platforms to enable seamless data flow between different systems and departments.