One of the things I've heard repeatedly over the years is that organizations need to become more mature before they can properly address data quality. In practice, I had several discussions where somebody came to see Data Content Manager at our booth in Knowledge26 and concluded: great product but we’re not mature enough for it.
Sometimes it's governance maturity, process maturity, or organizational maturity. The wording changes, but the idea is usually the same: once the organization reaches a certain level of maturity, data quality will somehow improve. Or the opposite, data quality can’t be improved if the organization is not mature enough.
I completely disagree with that. In fact, after years of working with organizations trying to improve their ServiceNow data, I think that often it is data quality that increases maturity. Not the other way around.
What is maturity?
When people talk about mature organizations, they usually describe characteristics such as clear ownership, accountability, governance, standardization, and trusted data. What's interesting is that these are also the exact things organizations are forced to establish when they start taking data quality seriously.
Take ownership as an example. On paper, assigning an owner to an application, service, or business capability sounds straightforward. In reality, it often isn't. Different systems contain different answers. Teams disagree.
Responsibilities that seemed obvious turn out to be unclear. What starts as a simple data quality exercise quickly becomes a conversation about accountability and governance.
A company begins by trying to improve the quality of its CMDB data and ends up uncovering gaps in processes, decision-making, ownership models, and organizational structure. The data itself wasn't the root problem. The data simply made the underlying problems visible.
Organizations often treat maturity as a prerequisite for improving data quality when, in practice, many aspects of maturity are developed through the process of improving data quality itself. You don't become mature and then create accountability. You create accountability and become more mature as a result.
Accountability should be reflected in your CMDB and other foundational data sources, and it should be continuously enforced as people change roles and companies all the time.
AI is making this even more apparent. For years, organizations could tolerate poor data because people compensated for it. Teams knew which reports couldn't be trusted. Tickets related to SAP were critical because they had always been. Application owners maintained unofficial spreadsheets.
Operational knowledge lived in people's heads. The organization adapted.
Data volume is not maturity
The other thing I sometimes hear is that we don’t have a lot of data, so there’s no need to improve it. Or that we are not mature enough for DCM because we don’t have a lot of data.
Again, I disagree. The amount of data is not a measure of maturity. Often it is the opposite. It is easy to fill a CMDB with data. Just let Discovery run or integrate data from other systems, and you’ll have a lot of data. But whether it is useful or operationally trustworthy is a completely different question.
Why would you let your data deteriorate instead of implementing quality controls when you don’t yet have a data quality problem? To me, it makes no sense to fill a database with data and worry about quality afterward, when you could have possibly avoided the data quality problem in the first place
AI doesn’t work around bad data. It amplifies it.
AI changes the dynamic. AI doesn't know which spreadsheet contains the real answer. It doesn't necessarily know that a service owner is outdated or that a relationship hasn't been maintained for two years. It simply consumes the information it's given.
And the more data is given, the more critical it is that the data is operationally trustworthy. Correct, up to date, and backed up by a systematic process that involves the organization that knows what the correct data is.
As I said in my Knowledge26 Reflections post: AI doesn’t work around bad data. It amplifies it.
That's why so many conversations about AI eventually end up back at governance, ownership, CMDB, CSDM, and trust. The technology itself isn't usually the limiting factor. The quality and trustworthiness of the underlying data often is.
Perhaps that’s why data quality is receiving so much attention now. Not because organizations have suddenly become more mature, but because they are realizing that improving data quality may be one of the fastest ways to become more mature in the first place.
And in the age of AI, that maturity is no longer optional. The question is how to proceed.
Book a call with us, and we’ll show you.
