- The Datazed of Data Excellence
- Posts
- Responding to the data audit
When speaking with Data Governance and Quality practitioners, I often find a couple of themes:
Lack of sponsorship and/or resource to get things done
Being saddled with responsibility for data problems that they can’t fix due to issue 1.
The text below comes from a fictional story I wrote last year, which followed the journey of a new data leader. In this section, she had left the firm, and a new person had been promoted up from an MI role. He now had to manage the actions from the Internal Audit review of the Data Management team.
Have you gone through an audit like this?
Did any of the findings occur in your organisation as well?
What have I missed?
Email readers hit reply, and web readers add a comment below.
Thanks!
From: [Group COO]
To: [me]
Subject: Review of data governance
I've had the draft review through from IA. Pls take a look and put some time in my diary to discuss.
Thx
-----------------
From: [Internal Audit]
To: [COO]; [CFO]; [CRO]; [Head of Data]
Subject: Review of data governance
Please find below our preliminary findings from our data review. Let me know if you want to discuss any of them, otherwise we'll submit to ExCo on Friday for the meeting at the end of this month.
Approach
We've carried out our review based on the Data Governance Framework supplied to us by specialist consultancy, Datazed Associates.
Data Ownership
We reviewed the list of data owners and data stewards which you provided to us. We noted:
Two data stewards and one data owner are no longer employed by the company, and those roles need to be reassigned.
One steward is from our outsourced IT provider and they should be replaced by an inhouse employee.
We interviewed two data owners and two data stewards.
One steward exhibited a good understanding of the importance of data quality and their role in it.
One owner and one steward were aware of their role but sketchy in what it entailed;
The other data owners seemed to be unaware of why their name was included and suggested that we speak to IT as they are responsible for the systems.
Actions: Data team to update the list of data owners and stewards and ensure that they have been sufficiently trained.
Data Quality
We reviewed the data quality KPIs as submitted to OpsCo each quarter, and noted that they trend green; with the occasional amber being added to the Data Quality Log for management.
We did not find any evidence of challenge on the KPIs, such as whether we are looking at the right data sets and whether there are new sources of data which need to be added to scope.
Other data quality requirements are managed through manual controls. Some teams were able to share a tracker of the control activities being carried out; but others were unable to evidence other than through anecdotal emails.
Actions: Consider if DQ KPIs are fit for purpose; Increase monitoring (or automate) manual controls.
Data Lineage
We were provided with a large-format Visio diagram which was last reviewed in August 2021 but the change control indicated that there had been few updates since November 2016. In particular:
The NEPTUNE platform was not shown in the lineage; and
PLUTO was still shown, although it has since been descoped.
Action: Data team to conduct a full review of the end-to-end lineage diagrams.
Data Issue Management
We looked at the issue management log and sampled a few items to validate the activity. We were happy that issues and concerns are added to the log and progress is tracked.
We note that some of the earlier items are still shown as "in progress" after more than three years, and would like to see these items closed in an appropriate manner.
Action: All items which have been open for more than 12 months to be resolved, mitigated, escalated or closed.
Thank you for your time and your co-operation during this review.
Kind regards,
Internal Audit
You read through the email. Infernal Audit you heard someone call it. They always seemed so nice in the meetings, then produced a document that made you sound like you were useless. This time wasn't so bad as most of the shortcomings pre-dated you. Still, you’re going to have to get this resolved.
The COO sounds reassuring.
You don't need to worry too much. IA always have to find something to talk about and nothing in this email is going to bring the company down.
What do you think of the proposed actions? Can you take them on?
They all seem reasonable, but there's a lot of work to do.
The main problem is that I'm going to need a lot of input from the data owners and stewards, and I can't get the time with them. Every other meeting is a no-show, or they come late and "have to" leave early.
For data lineage, I did some Googling and found a tool that can do it for us, but it's really pricey.
Can't our current tool do that? It costs enough!
That's in the new module, but we can't get that without moving to the latest version, which means we'd have to upgrade our licence and migrate to their cloud product.
No can do my friend. The new broking platform is late and way over budget so I'm done for until the end of the financial year.
What about that course you did; and that chunky DMBOK book which sits on your desk?
That covered the theory, but it doesn't get me the time I need with the business.
I see. Maybe we should get someone external in to help you out - give you a bit of air cover...
If this resembles your experience at all, you’ll want to click here.
Before I go, I have some things to promote:
dataZED Get Together on 6 September.
dataZED of Data Events calendar.
dataZED Fantasy Premier League, raising funds for Data and Tech Aid.
After the summer break, I'll be restarting the dataZED Get Togethers with an event on Wednesday, 6 September, titled Navigating the AI Wave: Reimagining Your Role in Data Governance & Quality.
Whether you are a CDO, Head of Data, or work in a Data Governance or Data Quality role, then AI is going to change it massively.
On the one hand, many of your processes will be supported by AI and ML tech.
On the other hand, that AI and ML data needs governing as well!
Join me and Alex Plenty, CDO at Lloyd's of London, where we'll be talking about how AI is changing data governance and quality, and what it means for you and your role.
London EC3 location, with the session from 4pm-5pm followed by drinks nearby.
dataZED Get Togethers are tailored for data governance and data quality practitioners, and offer a platform to connect, learn, and share experiences. They are designed to foster a spirit of collaboration and learning among industry professionals.
This session will have an insurance focus, but will be of interest to all data governance and quality professionals.
Are you hosting a data event, webinar or get together?
The dataZED of Data Events is here for the data community to easily find out what's coming up.
1. It's not just for big events, but is also open to small get-togethers, webinars and anything else you can imagine.
2. There are no restrictions on geography. A webinar on the other side of the world could be ideally timed for someone to join early morning or late evening.
3. These aren't just events for senior execs either. Let's also share the events that are open to those on their career journey.
Finally, it's time for the dataZED Fantasy Premier League!
This season, we'll be focusing on the data side of the competition and will be asking our team managers to share how they are using data and analytics to help them find those extra points.
The first Q&A is with last year’s runner up, Ben Dugdale, can be found here.
And to tie in with the focus on data, we're going to be supporting D.A.T.A. (Data and Tech Aid) who are mobilising the data & analytics and technology communities to unite in supporting social and humanitarian causes.
Register here
Have a wonderful week,
Charles
Read a few previous newsletters online:
Join the conversation