Today\u2019s blog has been inspired by a debate I saw on LinkedIn where somebody had suggested that we really don’t need data owners because they’re really hard to find and we could use process owners instead because they already exist at your organisation.<\/p>\n
I obviously got involved in this debate and felt it important to do a blog on to explain fully why, while I think process owners can be useful, they are not a substitute for data owners. They are not interchangeable and there are two reasons for this.<\/p>\n
The first one is that in my experience that some sectors and even some companies within sectors are not that mature when it comes to processes, so if your organisation hasn’t got your processes documented and mapped then it also doesn’t have the concept of process owners in place.<\/p>\n
In this case, you’re off to a non-starter before you even start\u2026<\/p>\n
Let’s consider the opposite of that and perhaps your organisation is really good at documenting its processes and has very well-embedded process owners as a role. These are people who understand that they get to have some responsibilities around that process and the inputs into it.<\/p>\n
When I work with clients like that I think \u2018this is great; this is going to make it easier for me to find my data owners\u2019, but I never think let’s use the process owners instead of data owners.<\/p>\n
What I do is to look at the process owners as being likely candidates to become a data owner for me and this works really well.<\/p>\n
They’re usually suitably senior. They’ve got the authority to make decisions about the process so then it’s an easy and logical next step to get them to take responsibility and have the authority to make decisions about data.<\/p>\n
Now my problem with just using process owners instead of data owners is that some data is only used in one process in one place in your organisation. \u00a0In this case, the process owner is likely to be the data owner. However, with data that is used in multiple processes, across your organisation that then gives you multiple data owners who are unlikely to think of consulting with each other to make sure that they’re making consistent decisions about the data.<\/p>\n
This in turn will mean that you’re actually no better off because you won’t have any common understanding of what the data means. You’ll have nobody making consistent decisions with that overview of what that data means across the whole organisation.<\/p>\n
By all means, if you have process owners, look at them as likely candidates to be your data owners, but please, I would encourage you not to think you can abandon trying to find data owners.<\/p>\n