The IT service management (ITSM) industry is usually pretty good at identifying where things are going wrong. There are some exceptions of course – for instance, in consistently meeting employee expectations of IT service delivery and support (where there can be an expectation, or perception, gap) – and situations where the root cause (of the issue) isn’t truly understood, i.e. the wrong factor is blamed.
But there is, however, a significant service management issue that’s too-often overlooked and perhaps not even considered – that of data quality. Or, if spun around for greater impact – the results of bad service management data.
This article digs a little deeper into the problem of bad service management data, outlining the most-common data-quality management issues we see when offering data-quality-improving capabilities to our customers.
Not Knowing (About the Data Quality Issues)
Hopefully, this first issue requires little explanation. For whatever reason, many organizations have never challenged the quality of their service management data or have deliberately chosen to blissfully ignore any potential data-quality issues.
As a result, the service management data is assumed to be accurate (or it might be massaged by those who know better) as are decisions and actions that leverage it.
It’s Seen as a Tough Nut to Crack
Organizations and their people might find it too difficult to tackle the issues posed by data-quality issues (but they at least know there’s a need to improve their data quality).
This might be due to the belief that it’s one of those “too hard to tackle” problems (and so nothing gets done), a lack of knowledge related to potential solutions, competing demands for improvement attention and resources (for instance, the focus on system development and the delivery of new features is deemed more important than data quality), ability to track progress or a method for tackling the issue in smaller pieces to name a few.
Companies Don’t Invest Enough Resources
This somewhat overlaps the previous issue. For instance, there’s more focus on system development, and the delivery of new features, rather than on ensuring that what’s already in place is fit-for-purpose (in terms of data quality).
Or it might be that data-quality management plans are created and put in place, but then not followed (after go-live at least). Or that there’s insufficient investment in the process for maintaining the data that’s required for systems to work, especially automation.
The known data quality issues thus continue to adversely affect decision making, operations, outcomes and both job and customer satisfaction.
The Lack of Data Ownership and Usage Clarity
This point raises an important question: Who owns which data? When helping our customers we see lots of data without ownership, or without a proper definition on how and where it should be used, and by whom.
This vagueness can only add to data-quality issues and their unfortunate outcomes.
An Inability to Focus on What’s Wrong
This is where organizations don’t really know what needs to be fixed. They know, or feel, that things aren’t right and that there are issues, but they can’t easily point to where the main problems are (such that they can be addressed and ideally mitigated). For instance, it’s commonly hard to report “what’s missing.”
A Lack of Suitable Skills, Tools, and Techniques
This is when organizations lack data-modelling capabilities and/or an understanding of the information architecture. This might be caused by a number of things – for instance, the aforementioned “missing the need,” not having funding for (or access to) suitably skilled people, or not having a fit-for-purpose data-quality management tool.
Long-Held Excuses for Why Not to Start
We probably all do this is life – a task seems difficult, so we keep putting it off. Perhaps ignoring it in the hope that it will eventually go away. Sadly, poor data quality will never simply correct itself.
There are of course other common reasons for not improving data quality. For instance, not knowing where to start – especially when you’re overwhelmed by a number of issues that need to be addressed. Or there’s the “We need to sort out X, before we can truly address our data-quality issues” excuse. This is commonly articulated as: “Our process/service/CMDB/etc. is not quite there yet, so we can’t get started with any data-quality-improvement initiatives.”
So, these are the seven most-common data-quality management issues we see in ITSM. There are of course more, please feel free to add any that you’ve experienced in the comments.