- Executive engagement. Comprehensive data governance programs require executive-level support and resources.
- Progressive framework. Implementing data governance incrementally and leveraging AI technologies can help organizations optimize their decision-making processes.
- Consistent approach. Establishing well-defined policies and procedures, as well as data lifecycle management, helps organizations maintain data integrity, security and compliance.
As critical as data is for organizations to flourish in today’s economy, simply collecting and storing data without a data governance policy leaves an organization without a structure to handle the information.
Only 49% of enterprise decisions are driven by data and only 10% are powered by insights, said Arnab Sen, Tredence, Inc. vice president of data engineering. Yet it’s the data that will provide more benefits. “C-level leaders and teams are developing strong data management programs that can evolve to meet the latest business, customer, operational and regulatory requirements.”
Comprehensive data governance programs require executive-level support and resources; engaging the C-suite is a critical first step, added Christine Frohlich, Verisk Marketing Solutions head of data governance. “Once support is secured a good next step is to research technology tools that help keep the information accurate and updated. Understand not all companies have the resources to manage emerging privacy regulations and data governance complexities on their own. It is okay to look for external support through trustworthy partners that understand your business and specialize in consumer privacy. However, data governance should be a meaningful focus in any organization, it is not a side-task, nor can it be relegated to a part-time job.”
Once C-suite support is established, data governance models should include the following five elements:
1. A Progressive Data Governance Framework
"Using this framework, organizations can implement data governance incrementally at their own pace and based on specific needs, requirements, and use cases," explained Mahesh Ramichetty, OvalEdge vice president of special projects. Ramichetty highlights the flexible nature of the framework, which allows organizations to tailor their data governance strategies to their unique circumstances and objectives.
"The progressive approach utilizes the latest technologies, such as AI, and focuses on core areas like data literacy, data access, and data quality improvement," Ramichetty continued. By leveraging cutting-edge artificial intelligence tools, this forward-thinking framework empowers organizations to streamline their data management processes, foster a more data-literate workforce and ensure that users have timely, secure access to the information they need.
With an emphasis on continual improvement and adaptability, this framework aims to help organizations not only comply with data governance regulations but also derive maximum value from their data assets. By implementing data governance step by step, companies can tackle challenges more effectively, reduce risks associated with data breaches or misuse and ultimately optimize their decision-making processes based on accurate, high-quality data.
Related Article: Data Governance for Dummies
2. A Taxonomy & Labeling System to Identify Sensitive Information
This is among the first things to consider in a data governance guide, according to Sisi Zhang, Razorfish executive vice president of data science and analytics. “Organized, clean data that’s free of duplicates and redundancies is key to developing and maintaining strong customer data governance models. A taxonomy should be applied to all aspects of customer data, including data gathered through marketing touchpoints. This categorization will make organizations more efficient, reduce confusion and improve overall data ecosystem connectivity.”
A taxonomy and labeling system helps organizations identify the sensitivity of different data types, such as personal identifiable information (PII), intellectual property, and financial and health data, added Raymond Velez, Publicis Sapient global chief technology officer. “By clearly labeling data, organizations can ensure that the appropriate level of security and access controls are applied to different data types, helping to mitigate the risk of data breaches. With the proliferation of complex and different regulations by state it’s critical that organizations are prepared to comply to avoid penalties and fines.”
3. A Rigorous Data Quality Framework
Poor data quality will pose a challenge for organizations that have not applied a rigorous data quality framework to identify, fix and monitor for data quality issues, said Peggy Tsai, BigID chief data officer. “If there are any data values that deviate from standards, data users should adhere to a process for fixing the data issue at the authoritative data source. A process should include ongoing monitoring for missed data quality issues.
Tsai added that duplicate data is a major challenge for data integration because it can represent multiple copies that should not exist. In many integration projects, duplicated data needs to be removed as it causes noise and clutter in an integration environment. Merging duplicate data records together can help identify missing data to form complete "golden records."
Consistency of data is crucial, though especially challenging for global and regional organizations, as various geographies and markets have specific nuances, Zhang added. “Organizations with a large global or regional footprint need to ensure they have governance models in place to address these variables as data is accessed, shared and stored in different places.”
Related Article: Data Governance: Offense or Defense in the Digital Age?
4. Assignment of Data Governance Responsibilities
"At the heart of any data governance guide should be a set of policies and procedures that establish the roles and responsibilities of different stakeholders within the organization," said Velez. These policies and procedures should act as a foundation for a robust data governance framework, including guidelines for data ownership, access control, data quality and privacy protection.
Without clear guidelines on these critical aspects, Velez warned that data governance can become fragmented, leading to a disjointed and inefficient data management process. Different teams within an organization might operate in silos, resulting in inconsistency across various data-related tasks and objectives. This lack of cohesion can hinder the organization's ability to leverage its data effectively and may even pose potential risks in terms of compliance, security and overall decision-making.
By establishing well-defined policies and procedures, organizations can ensure that all stakeholders understand their respective roles and responsibilities, fostering a more collaborative and streamlined approach to data governance. A strong foundation in this regard will help prevent fragmentation and promote consistency, ultimately enabling organizations to harness the full potential of their data assets while maintaining compliance with relevant regulations and protecting the privacy of their users.
5. Establish Data Retention Policy & Rules
A policy for data retention should be established to ensure that data is kept only for as long as it is necessary and disposed of securely once it is no longer needed, advised Ani Chaudhuri, Dasera CEO. This approach helps organizations maintain compliance with various regulations and reduce the risk of data breaches.
In addition to policies and procedures, data governance guides should also include guidelines for data lifecycle management, Velez agreed. "This includes guidelines for data retention, archiving, and disposal, which are crucial for ensuring that organizations comply with legal and regulatory requirements," Velez emphasized. By covering these critical aspects, organizations can effectively manage their data throughout its entire lifecycle, from creation to eventual disposal.
Final Thoughts on Data Governance Models
Having a comprehensive data governance guide that encompasses both policies and procedures, as well as data lifecycle management, will help organizations maintain a high level of data integrity, security and compliance. In doing so, they can minimize potential risks associated with mishandling data and better leverage their data assets for improved decision-making and overall business performance.
Comprehensive data governance models should be “living” policies that get revised as necessary to meet the needs of an evolving organization. But even as those models are updated, they should still include the five elements above to ensure that they continue to address the needs of collecting and managing data.