There’s no doubt that we live in a data and analytics driven day and age. Businesses are gathering data about their customers, transactions and a plethora of other items at an amazing rate. Those organizations that are able to capitalize on that data by implementing analytics are increasing their profits by substantial amounts and are providing services and goods specialized to their customer’s specific needs from that gathered data using predictive analytics.
However, for both small and large businesses alike there has to be a guiding force, if you will, that provides the blueprint needed to help guide and inform any analytics initiative. Enter data governance.
One Size Fits All or Not?
Simply put, data governance is about providing high quality and consistent data as well as maintaining and addressing your company’s data in terms of data privacy, security and usage in a way that meets the needs of the business. The fact is, data is one of the most important assets an organization owns. It’s true that no matter what the size of the organization data needs to have policies and procedures in place that govern and protect it. However, the problem that most organizations face is how to structure those policies and procedures in a way that fits business needs and requirements that are unique to that organization. What works for one company may not work for another company when implemented in whole. As such, each organization’s effort to build an effective data governance program should be considered carefully and an individualized framework created where the organization can build policies and procedures tailor made to their requirements, unique and common, that aids in developing and designing an effective data governance strategy and program.
DQM or Data Quality Management focuses on the life cycle of high quality data and starts with providing a locus of accountability for each domain within DQM. Used with data governance, data governance provides organizational wide accountability from both the various business units and IT allowing for functions that address both business and IT perspectives. Weber et al., (2009) suggest that there are contingencies within organizations that make the one size fits all approach impossible. Additionally, “All sources postulate a universal data governance approach, namely one that should fit all organizations alike. At the same time, they fall short of analyzing the interrelation of the distribution of accountabilities for DQM and contingency factors. Moreover, they do not come up with more than one data governance design”.
Since Weber et al., (2009) present the problem that the one size fits all approach to data governance doesn’t work they suggest a more “flexible” model that includes two parameters to allow for contingencies. Their model for data governance, as with others, provides roles, responsibility assignments and decision areas. However, when applying the model organizations will need to configure the model based on “their individual needs by elaborating the distribution of interaction types, and by involving additional roles and decision areas in response to specific requirements.”
Nwabude, Begg, and McRobbie (2014) also studied this one size fits all issue and noted that based on their review of existing data governance frameworks there are no frameworks that are specifically designed to take into account the contingencies and specific limitations of small businesses. One such limitation was not having an IT department to implement roles and processes as defined in the framework given by Khatri and Brown (2010) or to be able to interpret technical language.
A Suggested Guide for a Data Governance Program for Small Business
Nwabude, Begg, and McRobbie (2014) suggest that when building a data governance framework for a small business to keep the following in mind.
1. Easy to Implement – It should be easy for the owner or employees to understand and learn the solution early on so that the solution will be adopted easily. The amount of effort needed to implement the solution should be minimal.
2. Low Cost – The cost of the solution, including implementation, maintenance and training should not give the owner sticker shock. The resources needed should not be the same as that of a Fortune 500 company!
3. Consider Management and Internal Structure of Small Business – Simply put, most of the decisions are going to be made quickly by either an individual owner or a few managers and not a Board of Directors.
4. Have Fewer Roles – You don’t have to have a lot of roles! Take into consideration the structure of the organization and how decisions are made and communicated across the organization. Typically, communications within a small business are faster and more direct than in a larger business. This removes the need for certain roles within a data governance framework.
5. Require No IT Staff or Experience – Most small businesses don’t have an IT staff and typically any IT issues are handled by the owner or some other non-technical staff. Technical jargon and terms should be avoided and the framework should be able to be understood by any non-technical staff.
