The Imperative for Reimagining Data Governance
Simply having well-governed data assets does not guarantee that you're generating business value. You need a holistic view across different components and activities.
- By Mark Peco
- October 26, 2018
During the mid-2000s, enterprises began to recognize the promise of data analytics and how it could help them compete. A perfect digital storm was brewing. Society was in the midst of a transformation driven by smartphones, social media, digital entertainment, real-time communications, and consumer-based e-commerce. Publications such as the Economist and the Harvard Business Review were presenting these opportunities to executives as strategic opportunities that should be harnessed.
Emerging from the financial crisis in 2009, enterprises began to embrace their new data analytics opportunities at a strategic level. The belief in 2010 was that data -- in all of its new and diverse forms -- would become the fuel needed to drive innovation and competitive advantage. It became obvious that enterprise business success was highly dependent on the quality of the input data used as the fuel.
The discipline of data governance began to take shape a few years earlier thanks to stringent government reporting obligations created in response to major corporate accounting scandals including Enron and WorldCom. Data governance was maturing to include the processes, organizational structures, and accountabilities needed to transform data into a trusted and valuable corporate asset.
Effective data governance has quickly become a fundamental enabler of data analytics success. The scope of data governance defines which data assets, policies, life cycle phases, people, processes, and analytics use cases will be included. Accountabilities are assigned and policies are implemented around data that exists within the carefully defined scope.
For discussion purposes, assume the first phase of data analytics adoption was from 2008 to 2018. During this phase, data governance evolved into a corporate leadership role. It became focused on oversight and accountability for policy compliance related to defined data assets. Policies were developed and implemented to improve data availability, understandability, accessibility, security, relevance, and quality. Roles such as chief data officer and chief analytics officer were appointed to provide organizations with an enterprise focus to their data governance challenges. Data governance was elevated from departmental perspectives to an organizational view.
Today, a new phase of analytics adoption is beginning. A higher standard for demonstrating measurable business value from investments in data analytics is quickly emerging and will extend beyond 2018. As an analytics leader, you have to deal with new challenges: understanding the mechanisms of value creation and demonstrating measurable analytics results to company leaders. These new demands are the natural result of a developing analytics maturity and increased expectations.
Taking a step back, we can ask pointed questions about how successful companies have been in creating value during the first phase of analytics adoption. Based strictly on a data perspective, there is anecdotal evidence that companies are acquiring, storing, processing, and delivering larger data volumes now than a decade ago. The big data wave continues to bring more data into organizations. However, more data stored and processed does not equate to generating more value. Gradually, enterprises are beginning to treat data as a governed corporate asset due to the maturing data governance programs of the first phase.
However, simply having well-governed data assets does not guarantee that you are generating business value. Recent research published in the MIT Sloan Management Review (S. Ransbotham, March 2016) shows that the actual business value being generated from analytics is no longer rising. Generating business value -- in the form of cost reductions, revenue growth, or risk reductions -- from analytics requires changes in how you make business decisions and the actions you take. The promise of data analytics, dating back into the mid-2000s, was to create incremental and measurable business value. To fulfill this promise, you need to reimagine data governance.
Addressing Business Value
Generating value from data requires a holistic view across different components and activities. You can model this view as a value chain, sometimes called a value stream. A value chain shows the dependencies between different components and activities that all must contribute to the ultimate goal of value creation. Value chains span organizational units and require collaboration and trust to be successful.
A simplified value chain can be described by these dependencies:
- Value is created by taking actions
- Actions are determined by decisions
- Decisions are identified and evaluated with analytics
- Analytics is enabled by basic information
- Basic information is created by integrating data elements
This means that value, actions, decisions, analytics, information and data represent how value flows from data to data usage to business actions and results. This chain includes data, processes, technology, analytics models, and people.
Effectively creating value from data analytics investments requires multiple components (people, data, technology, and processes) to work together. This implies that coordination, planning, and execution of data analytics must take place at the value chain level and not simply at the data component level.
Data Governance Reimagined
Data governance needs to be reimagined and transformed into data value governance, which has a much broader focus than the traditional view of governing data only. Data value governance includes all the components of data governance as well as a broader set of "things" to be governed. These "things" are defined by defining the data value chains that must be successful if an enterprise is to create the desired business results.
Lessons learned from the earlier iteration of data governance -- such as virtual teams, accountability assignments, policy development, procedural implementation, and enforcement -- need to be expanded and integrated with an overall business strategy that describes a data-driven business model that depends on analytics for success.
This expanded form of governance needs to be integrated with your corporate strategy and must deal with difficult issues including change management, skills development, organizational structures, politics, communications, and compensation models.
Data value governance needs to be holistic across the value chains that collectively create competitive positioning for your firm. It should be viewed as the engine for transforming your company into a data-driven organization enabled by advances in analytics, machine learning, and artificial intelligence.
[Editor's note: Mark Peco is a TDWI instructor who specializes in helping companies develop strategies, architectures, and programs that accelerate data-driven business success. He is a TDWI faculty member and teaches a broad range of courses offered through TDWI's Onsite Education Program, which brings tailored BI and analytics courses directly to an organization's conference room. Mark works directly with clients to understand their training needs, develop custom curriculum, and deliver content specifically aligned to corporate objectives. A library of onsite courses is available here or contact [email protected] for more information.]