When a company begins to explore data governance, its attention eventually turns to how much time the data governance program will demand. The truth is that the company is already governing its data, regardless of whether it has a formal program in place or not. The real question is how efficient and effective it is in governing that data.
For example, if a chief marketing officer (CMO) issues the directive that the marketing department will now use a different lexicon in doing business – i.e., marketing campaigns will now be viewed as initiatives and marketing media will be called tactics – that CMO is in effect triggering data governance. This directive not only impacts how the department will communicate, but also how it stores, measures and reports on its operations. Marketing media versus marketing tactics has subtle and sometimes glaring differences, as does the data that is used to track their costs and results.
When the CMO makes such a decision, it falls to the responsibility of the data steward to implement the change. Data stewardship is the implementation arm of data governance, and data stewards are the individuals who effect the policies invoked by the data governor(s). The data steward could be defining terms, implementing a rule, or collaborating with peers on data issues. Just like the CMO, a data steward spends a portion of his/her time on data governance issues.
When a client asks me how much time the data stewards will need to do their tasks, we must first estimate how much time they are currently spending on stewardship-like tasks. Keep in mind that a person who backs into a task and performs it piecemeal – without a true appreciation or understanding of what he or she is doing – will be considerably less effective than a person who is taught and equipped to do it.
I liken it to digging a flower bed. Someone may buy one flower and scoop out the dirt with his or her bare hands to plant the single flower. An experienced gardener, however, will know to have a plan for where and when to plant the flower, have a spade and gloves ready, have some peat moss and water at hand, be prepared for weeding, and know how to protect the plantings from dogs and kids. Can both the casual person and gardener plant a flower? Sure, but who will be more efficient and effective in the long run? Whose flowers are most likely to survive the season?
A typical stewardship task is finding and fixing data quality errors. The person performing this work can estimate how much time he or she spends doing it. In the case of business data stewardship, this type of activity is part of the functional responsibilities, regardless of whether the person is an account manager or a loan officer. When a formal data governance program is launched, light is shed on the time and effort managers should be spending on data stewardship activities – once they are proficient in those tasks.
Figure 1: Activity Percentages at the Start of a Data Stewardship Initiative
Figure 1 shows a sample time allocation of data stewardship tasks that we used on a client engagement. The tasks total 100% of the time the average, formally designated data steward spent on those types of activities. Granted, each data steward already had a day job; most of the managers had marketing or sales operations titles. One of them handled marketing campaign planning and execution for their EMEA division. It was estimated that because of the poor state of their data, the EMEA staff spent 80% of their time on data steward activities and only 20% of their time on true marketing tasks.
The quality of the data and maturity of the data management capabilities will have a direct bearing on how much time a functional manager will need to spend on stewardship tasks. For example, an organization that is early in its program will spend a significant portion of available stewardship time on defining standards, documenting business rules and establishing data capture guidelines, whereas a company that has a mature data governance culture will have most of those guidelines and standards documented. The evolution or change in the stewardship mix can be seen in Figure 2.
Figure 2: Data Stewardship Group Activities (averaged over 3 months)
At the start of an initiative, the balance of work will be tilted toward foundational tasks such as defining standards and performing initial data quality assessments. Over time, as the company’s data management capabilities mature, work efforts will shift toward more collaboration, designing improvement processes and responding to requests for information about the data. Additionally, staff will gain the time (through the completion of foundational tasks) and knowledge to document metadata that was otherwise not available at the start. One reason for this progression is because the data becomes better documented, the quality is elevated, and the data stewards begin to publicize its applicability and usage. Demand for the data grows, as does the proportion of time the stewards work with others planning its delivery.
Bottom line, when someone wants to know the time commitment for data stewardship, two factors need to be understood: (1) Data governance activities are already occurring; it’s just a matter of how effectively they’re being performed, and (2) the time each steward allocates to stewardship tasks will not only be dependent on his/her functional role, but also on the organization’s data governance maturity. Keep these two factors in mind when projecting staff headcounts.