What is data stewardship?
If you Google ‘data governance’ you will find that the terms ‘data stewardship’ and ‘data ownership’ appear alongside the many definitions of data governance. Over the years I have found that this tends to confuse people because they think that must mean governance, stewardship, and ownership are all different concepts.
However, they are not different at all. In fact, stewardship and ownership are roles and responsibilities which play a key part in successful data governance.
Data owners are the senior people in your organisation who own your data and are accountable for the quality of it. For data governance to work these data owners need to be reasonably senior people within your organisation and they need to be the people with the authority, budget, and resources.
I recently explained the data ownership role in more depth in another blog, which you can find here
The downside of this is that when you start telling those who are accountable for data that they might need to write things like data definitions and that they might need to investigate and fix data quality issues they can be resistant to this.
Unfortunately, it is also the case that even if they are invested in the data, they are probably not going to be in a position to understand it in granular enough detail to oversee it on a day-to-day basis. Also, given their senior roles within an organisation, they may not have the time to deal with the data in that sort of way. This is where data stewards come in.
In practice, when I’m implementing a new data governance initiative, I will identify the right data owners with an organisation, and once I have done that I will invite them to appoint one or more data stewards to help them in the delivery of their role.
The data owner remains accountable, but they will delegate the day-to-day responsibility to a data steward. In my experience data stewards often tend to be the subject matter experts but are still reasonably senior because they have to be trusted by their data owner.
If you took, for example, a finance department, it is likely that the finance director or his deputy would be the data owner for all the finance data. Then, in my experience, finance departments tend to be made up of a number of different specialist teams all focusing on different areas – and working with different data. This means it wouldn’t be practical to have just one data steward working under the data owner.
The data owner is likely to have to appoint the head of each of those sub-teams to be a data steward for them or that subtopic.
I hope that gives you an idea of what data stewardship is and how it fits in with data governance.