Getting to the root of a data quality issue
I’ve recently been going over with a client the pros and cons of utilising their IT incident management process to handle data quality issues.
IT incident management involves the creation of a log to help record and manage any issues that crop up in relation to IT supported systems, the goal being to reduce any adverse impact on your business operations. Since most organisations have such a formal process in place, it is logical to think that the obvious thing to do is to add the handling of data quality issues to your existing IT incident management process.
However, it is not always a straightforward decision and while there are undoubtedly some advantages from combining the two processes, there are also some downsides.
This is a much debated topic with many experts taking different stances over the viability of combining the two processes. I think the answer on whether it is right or not will depend on your organisation, it’s set up and culture. So, I thought it would be useful to summarise the key points to help you make your mind up about whether it is the right option for you and your organisation.
Many of these points arose from a LinkedIn discussion I was involved in some time ago during which many felt that the use of an IT incident management process for data quality issues does come with some notable advantages, chiefly:
Providing users a single central location to log any potential data issues. They don’t have to think whether it is a data or system issue – they just have to report it to one central place.
You can usually reuse the available workflows, tracking and reporting for data quality incidents.
It can encourage more efficient meetings concerning data, as details recorded may inform broader, fleshed out reports.
But will this work in practice? With a quick resolution often prioritised over all else, temporary fixes often result when it comes to addressing data quality issues using an IT Incident Management process.
When we are looking to fix data quality issues tactical fixes are not ideal, particularly, if it can be prevented from occurring in the first place. That’s where the implementation of a data governance framework comes in. Data Governance is about encouraging more proactive management of data quality, seeking sustainable improvements and identifying the root cause of issues.
Implementing tactical fixes instead of addressing the source of the issue is the most common issue from using your IT Incident Management Process to handle data quality issues, but other downsides include:
Some of the tools utilised within IT incident management aren’t necessarily connected to business processes, so any data quality resolutions which require heavy business involvement to correct, can be difficult.
The burden of having to tag or identify which particular issues are data quality specific.
Data quality issues can be unnecessarily escalated if SLAs haven’t been changed to reflect the differing timescales for such incidents in comparison with normal IT issues.
As helpful as it can be to reuse an existing process it can promote a culture of ‘fixing’ rather than ‘solving’. This is why it’s so fundamental for organisations to invest in a proper data governance approach to ensure that the best decisions are being made on how best to tackle data quality issues.
However, an IT incident management process can be a viable starting point. It may be equitable to trimming weeds rather than ripping them out by the roots, though, it does nonetheless keep your organisation aware of data quality issues as and when they pop up.
That being said, it’s always worth keeping in mind the end goal of sustainable, long term improvements to your data quality and the continued management of it. An IT incident management process can be a brilliant short-term fix but it can’t compete with the confidence and reassurance a data governance framework provides over the long-term.
There’s just the small task then of convincing your organisation to opt for the cultural change necessary to reap such long-term rewards!
Originally published on nicolaaskham.com