Analytics community governance is the concept that those who know the data most – those who create, analyze, and draw conclusions from the data – should be the ones creating governance policies around that data. Also known as bottom-up data governance, this practice ensures buy in from key stakeholders and end users while also creating a distributed network of policy enforcers akin to open-source coding principles. It maintains the idea that data governance initiatives can be and in fact are already being created and used by those at the point of use in an organization. By interfering minimally with current processes, management or a data governance committee can instead focus on bringing to the forefront the most effective practices already in use and developing baseline policies to create standards of data stewardship throughout the organization.

Everyone within an organization these days interacts with data and therefore everyone needs to adhere to organizational data governance policies. However, data governance is often a large, unwieldy beast that is difficult to define and cumbersome to measure the impact of. Data governance is managing the affairs of an organization’s data that is critical to its well-being in accordance with established directives ensuring such data is readily accessible. Top-down policies built by a small but knowledgeable committee can be a great starting point to create a baseline of policies for an organization that accomplish the goals of data governance. Unfortunately, a small single group won’t have the complete knowledge needed for the task including all use cases, available data sets, data creation points, collection techniques, and technologies used along with a host of other variables. That is why we at CTI support the use of analytics community data governance strategy with necessary top-down baseline policies.

Involving all stakeholders that interact with the data when developing data governance ensures there are already champions of the policies using the identified processes in regular activities and effectively working as data stewards. Using policies developed by users also ensures that there is a significant amount of testing done on the process. With many eyes on the policies and processes of governance there is less likelihood of bugs or errors, and fewer needs will fall through the cracks. Conversely, the system has the potential to cause redundancy and version control issues while scaling if not managed well. This is where top-down policies become necessary as decisions are made around tool acquisition, standardization, data ownership, stewardship, and other key needs within the overall data governance strategy.

Analytics community governance can be adopted in a number of ways, but all maintain certain necessary elements. One key piece is a structured hub and spoke data stewardship program. This involves assigning one user within each business unit as the expert on that business units’ data. That user then remains in regular contact with the centralized data team, data governance committee, or other data stewards in the organization. Bottom-up governance policies coming from the centralized data group and top-down policies being implemented are communicated through this individual, and the data steward is responsible for understanding all datasets and governance used in that business unit. This role not only creates a resource for users across the organization about that units’ data but also begins to build a culture of data within that team.

Another key element of this strategy is ensuring end user engagement. Because bottom-up governance strategy is intrinsically driven by users, it is ideal for organizations to create a culture of using and understanding the data being used. Supporting users through trainings on data basics or providing access to learning resources can empower users to be more self-service in building analytics and more engaged in data governance. Additional trainings specific to the organizations’ data can further the return on investment for a data solution as users implement analytics to support their work. Workshops too can continually engage individuals in the data governance process while remaining agile and informed when implementing top-down policy.

Bottom-up data governance is the idea that policies and structure are widely created and enforced by users whenever possible. In an ideal hybrid model, some policies are still developed directly by management, but only where absolutely necessary. Instead management or a data governance committee should focus on standardization, documentation, and training to support user developed practices. Empowering employees in this way creates a culture of transparency and encourages greater adoption of data use overall. Data is after all a tool, and inclusive data governance within an organization can allow that tool to be used in ways a small group or single individual couldn’t have imagined.