Nearly every data leader I talk to is in the midst of a data transformation. And it’s no surprise, really. As businesses look for ways to increase sales, improve customer experience, and stay ahead of the competition, they are realizing that data is their competitive advantage and the key to achieving their goals.
But as businesses work to transform their data to drive greater value from it, they often encounter challenges that derail their efforts. At many organizations, data is siloed, living in a myriad of disparate systems, making it difficult to gain holistic insights across the business. While at others, the data lives in a data lake, polluted by dirty, inconsistent, and incomplete data that is difficult to trust.
Many others are grappling with the pressures caused by the ever-changing regulations and the need to comply with them. The list goes on.
To address these issues, organizations are embracing data governance as a way to support their data transformation efforts and bring order to the data chaos. While data governance provides the necessary policies and ownership needed to standardize the data, many data leaders are finding that data governance alone is simply not enough. Instead, they must establish a data governance framework that goes beyond simply defining data standards and policies by including the tools and processes needed to democratize knowledge, effect change, and improve data quality.
Establishing a Data Governance Framework
For many organizations, data governance begins and ends with the definition of standards and policies. And while governance standards and policies are important, they are only one element of a data governance framework. Change management, data knowledge, and data quality are also crucial for your success, along with the tools and processes needed to support them.
Take SwissLife as an example. A leading provider of life, pensions, and financial solutions in Europe, SwissLife developed a data governance framework that enabled them to accelerate the implementation of their data strategy. Their data governance framework included:
- Standards and policies – Roles and responsibilities, data protection, classification, retention, archiving policies, and compliance audits/diagnostics
- Data change management – Communication, training plans, coaching, and support for key stakeholders and HR assessment processes
- Data knowledge –Data catalog (business glossary + data dictionary + usages) and data lineage flow maps
- Data quality – Methodology, controls definition and implementation, connected KPIs, and remediation plans
This data governance framework served as a foundation for their data transformation, enabling SwissLife to maximize the value of their data across insurance management, regulatory compliance, customer focus and innovation, and performance and optimization. Implementing a data catalog was a critical step in their efforts to democratize knowledge, and effective change management helped to fuel their success.
Democratizing Knowledge with a Data Catalog
Data catalogs are an important element of successful data transformation and governance initiatives. They give business users, many of whom are not data experts, clarity on data definitions, synonyms, and essential business attributes so they can understand and use their data more effectively. They show who owns the data, allowing for greater collaboration across the business. And they provide a self-service way for everyone in the organization to find the data they need and turn what used to be tribal knowledge into useful and accessible information that they can use to make better business decisions.
The data catalog was a critical tool within SwissLife’s data governance framework. And the team recognized the importance of engaging with the right stakeholders to get the functional and technical requirements for it right. Their first priority was to enlist the data owners. However, because the data owners are not necessarily data specialists, the team organized a series of workshops with them to demonstrate how a data catalog, specifically the business glossary, could help them reach their objectives and optimize collaboration with the data stewards and their peers.
The team also worked closely with data scientists, business data analysts, and actuaries to understand their specific needs and expectations for the data catalog, including whether or not they should replace or enrich local data dictionaries for specific usage.
Lastly, the data team collaborated with the Information Systems (IS) architecture team to align on the approach for the data catalog. The teams quickly agreed that the solution should not only be business-oriented but also include capabilities that presented an integrated and unique vision for their data.
They determined that the goal was not to eliminate all existing data dictionaries but rather create a unique consolidation point that they could consider as the master data for metadata. This approach also required strong collaboration with the IT teams, both operational and functional, to understand their needs and ease the work required to maintain the IS or manage the projects.
By taking a collaborative approach to the requirements definition, SwissLife earned the trust of the data owners, data scientists, business data analysts, actuaries, and the IS architecture teams. Their approach demonstrated how the data organization would democratize knowledge to make it useful for each business stakeholder. Moreover, because each team provided input into the process, the data team had confidence that the data catalog would meet the broad and diverse needs of the organization.
The Importance of Change Management
Change management is often the last thing an organization wants to think about. Why? Because it is hard work. But incorporating change management principles into your data transformation and governance initiatives pays off, as does securing buy-in from executives and business leaders.
Effective communication and change management fosters greater collaboration and trust across the organization. It keeps executives, business leaders, business users, and IT informed on the progress and helps to prepare them for the policy, process, and technology changes that are coming.
The data team at SwissLife embraced change management from day one. In fact, they credit change management as a significant contributor to their success. They acknowledged that many of their data users were not data experts. So instead of delivering 100% of their scope, all at once, they took a Minimal Viable Product (MVP) approach to their data catalog initiative by selecting a key use case as their starting point.
The data team selected the reverse engineering of a cross-domain use case based on data science processes that involved data, contributors, and IT systems. They focused on understanding what was available to build algorithms, locating where the data was stored and how it was transformed, and building new data sets that could be shared and reused across the data science team.
To ensure a successful implementation and broad adoption, the data team also spent a significant amount of time with their architecture team to position the data catalog alongside other platforms, such as their enterprise architecture and business process modeling.
In partnership with our company, they prepared workshops to clarify the vocabulary and understand the various capabilities of the solution so they could present a strong MVP to the various stakeholders that was simple, stable, and easy for beginners to understand. The goal was to deliver a capability-rich data catalog for the data users to test while allowing room for feedback and adjustments so that the catalog aligned with their needs.
The data team also focused on communicating early and often. They hosted a series of roadshows designed to evangelize the data catalog and promote the adoption of the solution. They also shared their ambitions for the project and the objective of their data catalog MVP. Because many of the stakeholders were beginners, the data team also provided educational information on what a data catalog is — and what it is not.
This step was important because they needed users to realize that a data catalog will not solve all their data problems. The team also explained the various roles and responsibilities of each stakeholder, and they listened to and valorized the benefits for all contributors and consumers of the solution.
The Secrets to a Successful Data Transformation
If your organization is embarking on or is in the midst of a data transformation, I urge you to consider two things.
First, think about data governance as a framework and not just a collection of policies and standards. While policies and standards are important, you need a data catalog that integrates your data and democratizes knowledge, making it easier for business users to find the data they need and use it to achieve business goals.
Second, embrace change management from day one. Secure buy-in with key leaders and stakeholders. Collaborate with your business partners by involving them in the process. And communicate early and often so that everyone is aware of what is coming and how they can support its success.