A common challenge when setting up data governance and data stewardship capabilities in any organisation is the perception, real or imagined, that this is adding additional work to already stretched staff.
This perception can mean that key stakeholders resist engaging in governance and stewardship roles. Instead, stewardship roles may be delegated or assigned to the wrong people
This sets the governance program up for failure.
So, does data governance really create more work?
What is data governance?
In many organisations, data governance is poorly understood.
Data governance is an oversight function – intended to support the planning for and prioritisation of data management activities that are already, in most cases, taking place.
At a planning level data governance serves the collective, enterprise need by combining and prioritising the individual, business unit needs.
As with most planning functions, governance shifts labour from business as usual firefighting and troubleshooting, to an upfront effort.
If the governance team is too junior they cannot make decisions and cannot allocate budgets. In these instances, planning sessions are additional work for these staff that will probably not see a return.
If the team is empowered to make decisions and decide data management priorities it is likely that they will collectively save substantial time and effort. Work can be done once to meet the enterprise need. And the end result will deliver more value.
In most cases, however, governance is confused with stewardship.
Stewards perform day to day tasks to support data management activities.
In many cases, these tasks are already being performed.
For example, a systems analyst may be tasked with assessing the impact of a planned change to a database e.g. reusing a field to meet a new purpose.
Typically, to perform his task he will start by setting up workshops with identified stakeholders that may have some knowledge of how that field is used.
These users may include business subject matter experts – power users that understand existing and planned uses of the system – and systems experts – technical users that may understand where the data is stored, relationships to other systems, and so forth.
Over a number of weeks and months the analyst will document a specification – probably in a document of a spreadsheet that will be forwarded to decision makers for signoff.
Finally, a go / no go decision will be made and, in many cases the documentation will be archived on a file server somewhere, never to be seen again.
This is an example of an inefficient data stewardship process.
The systems analysts, the subject matter experts, the systems experts and even the decision makers are all playing stewardship roles.
As discussed last week, data steward should not be thought of as a job tile, but rather as a job function.
Data governance brings structure to the team of people performing these ad hoc functions, hopefully with improvements in efficiency and effectiveness.
As with any change process, this can bring stress and fear.
But ultimately, this should reduce everybody’s work load.
What do you think?