If you’re just starting out on your Data Governance journey, you may very well feel like you are playing the classic 80’s board game ‘Guess Who?’ when it comes to sorting out who should be doing what.
There are a number of different roles and responsibilities needed when you are designing and implementing a data governance framework and it is easy to get confused over which roles you really need and what you should be calling them.
But don’t worry ,in this blog I’m going to go through them all and tell you everything you need to know!
Firstly, don’t start feeling overwhelmed. When implementing data governance there will be many challenges and different mountains to climb; certainly, one of them will be the naming of the different roles. What if someone tells you that the term “Data Owner” won’t work at your organistion?
Well you certainly don’t need to panic if this does happen. In this article I’m going to share what I consider a starting point, but you don’t need to stick to the names of the different roles too rigidly. These are just my best practices, but go with the flow of the organisation you're working with. It’s much better to use terms that suit the culture of your organisation - rather than fighting to use role titles that you’ve read in this blog or found with the help of google. It’s what the roles actually do that is the important bit.
A Data Owner is a very important data governance role; in fact, so important that I have written a whole blog dedicated to explaining ‘What is Data Ownership?’, you can read it here.
A Data Owner is someone very senior in your organisation, who is accountable for the quality of a certain set of data. And I must emphasise that it really does need to be someone who is senior within the organisation because data is owned by the company.
You can expect to identify a number of Data Owners and they should all be based in the business side of things.
If you want more detail on how to find them read my blog ‘How to identify the right data owners?’ here. You can also read my ‘Can there be more than one Data Owner per data set?’.
Data Owners are key but we also need Data Stewards, as Data Owners will be able to approve things but, often, they won’t want to get their hands dirty and will be unlikely to understand the details of the data.
So what is a Data Steward?
Data Stewards act as support for the Data Owners; they are often a subject matter expert in a particular set or subset of the data and are in a good position to work through the different processes that get rolled out.
For a deeper dive into the difference between Data Owners and Data Stewards, read my this old blog post.
Having a Data Steward is key to the Data Governance implementation, as they are the people who do most of the work to make it happen. Because of how essential Data Stewards are, I've written another blog that gives a more in-depth look at the role of a Data Steward. You can read it here.
Often, a Data Producer is a more junior role (although, this all depends on the organisation). The Data Producer's main role is to enter the data into the system.
It’s really important that these people understand that they need to capture data in accordance with the requirements of the person consuming the data. This is often a missed step which can lead to bad data. In my experience, Data Producers don’t often deliberately capture data incorrectly. It’s just that no-one took the time to tell them what was needed and that leads nicely onto the next role that of Data Consumers...
I use this term to refer to anybody that is using or consuming data, and it really is these people that are responsible for defining what makes the data good enough for them to use.
These people should be using the Data Governance Framework to communicate what they need from the data and defining how they know whether the data is good enough to use (also known as data quality rules).
This role is related to those working in IT, and their main responsibility is maintaining the data on IT systems in accordance with business requirements.
I talk more about Data Custodians in a previous blog. Read ‘What’s the difference between Data Owners and Data Custodians?’ here.
Data Governance frameworks do not last very long when no one is responsible for them. So… bring in the Data Governance Manager!
This person should be responsible for evolving and embedding the framework within the organisation. This is the only new role you will need to fill when implementing Data Governance for the first time. Everyone else already exists - you just need to formalise their responsibilities regarding data. Depending on the size of your organisation (and the investment they are willing to make in Data Governance) you may be lucky enough to have a Data Governance Team headed up by the Data Governance Manager.
And once again, it doesn’t matter what you call this person, but there really needs to be someone responsible for the initial implementation and after that checking that the cogs are still turning, things are still running smoothly, policies are being reviewed regularly and processes are being updated when they need to be.
I strongly believe that you need executive sponsorship for your data governance initiative - it is the only way to make the initiative truly effective.
However, I don’t believe that there has to be a ‘Chief Data Officer’, so don’t panic if your organisation doesn’t have one. If your organisation has a Chief Data Officer then the Data Governance Manager should be reporting to them and they should be the executive sponsor of your Data Governance initiative. But don’t panic if you don’t already have a CDO, find an executive who understands the value of data and get them to be your sponsor. I have helped more organisations to successfully implement Data Governance that don’t have CDOs than those that do!