This article was co-authored with Duke Dyksterhouse, an Associate at Metis Strategy.
IoT devices, wearables, SaaS applications, and social media channels are but a few of the sources from which data enters organizations today. When thoughtfully combined and analyzed, data from across those channels can deliver new insights and unlock new opportunities. Organizations that institutionalize and scale those insights across the enterprise can make informed decisions faster and ensure no lesson needs to be learned twice.
Converting siloed information into enterprise-wide insights requires a commitment to data governance, and doing it right is more than a passing endeavor. At its best, data governance can adapt and scale as a company’s strategy evolves, accommodate growing troves of data and, not least of all, provide a common nomenclature and trust that eases communication across business units and functions.
If data is the new oil and speed is the currency of business, then data governance is the link that fuses the two. It’s the set of systems, policies, and procedures an organization uses to ensure teams have the right data at the right time in order to enhance and automate processes, products, and experiences. It’s an exciting and valuable function in today’s competitive landscape, but getting there takes significant work. In this article, we lay out a three-step process to develop and mobilize a data governance program that moves at the speed of business.
Step 1: Establish foundational components
In many organizations, data governance is often limited to compliance, privacy, and security. These are critical domains, no doubt, but widening the scope and diversifying the representatives who oversee it can usher in more business value through faster, more informed decision-making and operational efficiency. Any data governance program should include four primary components: a data-governance steering committee, data owners, data stewards, and a data management team.
First, take stock of your data governance steering committee. If you don’t have one, assemble one. Include leaders from all business units and functions. If you have one, but it lacks cross-functional representation, augment it. Every business unit and function should have a representative on the committee. Depending on the size and scope of the committee, this could be a C-level executive or someone who works closely with the BU’s core data and IT systems.
Representatives must first articulate the committee’s objectives, which should include a set of both business- and compliance-driven objectives. Articulating these objectives will help illuminate the data governance objectives the steering committee is best suited to carry out. As an example, consider a healthcare organization that manages administrative processes on behalf of large hospital systems. The steering committee identified an objective to drive more automation into reporting processes. To accomplish this objective, they determined they would first have to drive common data definitions across their client base.
Once you’ve assembled a steering committee and defined its objectives, it is time to assign roles. Each BU and function represented should have a data owner, who will establish and uphold the policies and procedures that will, through an iterative process, alleviate the worst data-quality issues in their respective domains. Sticking with the healthcare example, consider that each business unit defined claim denials slightly differently, which impeded the organization’s broader adoption of solutions that would allow them to automate claims reporting. Recognizing the need to reconcile these varying definitions, the steering committee aligned on a common definition of claim-denials that would enable data aggregation and automated reporting. The committee then assigned data owners to take this common definition and manage the alignment of data to it within their respective business units or functions.
Next, you’ll need to assign data stewards. Stewards are functionally aligned and tactical. They serve the data owners in driving policy adherence, leading domain-specific change management, and reporting data-quality issues. For example, a steward aligned to the marketing department in a B2B software company might be responsible for encouraging the practice of classifying leads by region using common nomenclature (let’s say North, South, East, West) in the company’s CRM tool. The data steward would be expected to teach this practice to the sales representatives who use the tool, monitor its adoption and suggest how the policies that underpin it might be improved.
Finally, it is important to establish a data management team. This team, typically composed of technical IT resources, is the backbone of your data governance initiative. It works to enable and monitor established policies and procedures. To this end, it conducts audits to ensure adherence to privacy and security policies; evaluates data for accuracy, relevance, and completeness; and drives your data lifecycle strategy—from the creation and initial storage of your data to its expiration and destruction.
Step 2: Build the muscles you need to introduce new data into your ecosystem quickly and accurately
Once the data governance organization has been built and its initial policies defined, you can begin to build the muscles that will make data governance a source of nimbleness that will help you anticipate issues, seize opportunities, and pivot quickly as the business environment changes and new sources of data become available.
Your data governance capability is responsible for identifying, classifying, and integrating these new and changing data sources, which may come in through milestone events such as mergers or via the deployment of new technologies within your organization. It does so by defining and applying a repeatable set of policies, processes, and supporting tools, the application of which you can think of as a gated process, a sequence of checkpoints new data must pass through to ensure its quality.
The first step of the process is to determine what needs to be done to introduce the new data harmoniously. Take, for example, one of our B2B software clients that acquired a complementary company and sought to consolidate the firm’s customer data. The data governance team determined that each organization had a different way of managing customer-entity hierarchies, which define relationships between customers that appear to be different but roll up to the same parent organization. The steering committee determined that the acquired company should inherit the customer-entity hierarchy of the acquiring company to protect key Wall Street metrics. To accomplish this, the organization had to pull the following levers:
- Data Modeling & Design: Map the acquired company’s customer hierarchy to the incumbent hierarchy and update data-modeling artifacts (e.g., entity relationship diagrams) and tools accordingly.
- Data Dictionary: Update the data dictionary and master data management tool with historical context to specify how customer data from the acquired company was mapped to the incumbent customer hierarchy.
- Data Compliance and Access: Assess whether the existing compliance posture suits the new customer data and decide whether to deploy additional access or security provisions.
- Data-Quality Design and Implementation: Build controls into key applications to prevent users on the sales team from creating duplicate records or entering free-form text (instead of searching existing records).
- Communication and Change Management: Data Stewards communicate the changes to impacted users and manage subsequent changes to people, processes, and technology
Managing the introduction of new data is challenging, but resist the temptation to pursue one-off solutions that offer speed at the expense of long-term scale and reusability. Invest in the process and use it to create the foundation for outsized returns on your data assets. Think of the example above as the beginning of a J-curve: a thorough analysis and implementation, while perhaps leading to a short-term “loss,” can lead to dramatic, and scalable, long-term gains.
Step 3: Formalize operational data management practices for continuous data quality
The final step is to codify data management tools and practices that preserve the quality of your existing data and support target business outcomes. Best-in-class data management programs typically have defined procedures, cadences, and tools that support the following:
- Master Data Management: The systems and processes that support the creation of a singular master reference source for all critical business data (e.g., customer, product), and that in turn reduce the number of errors and redundancies in business processes
- Data-Quality Auditing and Monitoring: The deployment of tools and automated processes that help identify data that do not align to defined business or compliance rules and therefore do not meet defined quality thresholds
- Data-Quality Reporting: The practice of defining data quality metrics or KPIs, routinely reviewing their progress, and determining plans of action for to improve them
- Data Storage Operations: The practice of defining where and how to store various types of data across the data lifecycle – from introduction to destruction – while accounting for unique privacy and compliance considerations
- Data Stewardship: The practice of allocating resources across key business units and functions in service of data quality policies—and of managing changes related to the introduction of new data into the environment.
Codifying these practices can lead to higher data quality in terms of accuracy, completeness, consistency, timelines, validity, and uniqueness. High-quality data can be the difference between a happy customer and a disgruntled one. For one healthcare client, investing heavily in controls and monitoring technologies helped ensure data quality for information in motion and at rest allows organizations to provide customers with real-time information and deliver consistent experiences across physical and digital channels.
A good data governance program can improve the performance of individual business functions and units. A great one leverages the organization’s data to fuel enterprise-wide transformation. Make yours a great one. You’ll enjoy more speed, agility, and, ultimately, better business outcomes.
