Expert Interview (Part 1): Nicola Askham on Implementing a Data Governance Strategy
At the recent Collibra Data Citizens event in Jersey City, NJ, Paige Roberts of Syncsort had some time to sit and speak with Nicola Askham, The Data Governance Coach. In the first of this two-part interview series, Roberts and Askham speak about some of the advantages of implementing data governance as an overall strategy, rather than tactically addressing each new government regulation requirement.
Roberts: Can you tell the readers a little about yourself before we begin?
Askham: I’m known as The Data Governance Coach, and I am very lucky actually, because I have my dream job of helping lots of people do data governance. I do believe that data governance is done more successfully when it’s implemented by people within the organization but, I also understand that they might not have the expertise to do it alone. I love working with people to help them do it themselves, to give them the skills and knowledge they need to do data governance successfully.
So talk to me a little about the last project you worked on?
I’m working on a mixture actually but, I think it’s fair to say, they’re all data governance. They’re all related with either designing and implementing or improving and extending it. Although, it’s not the primary reason for all the projects I’ve been working on; GDPR has been a common thread amongst all of them. Even if we’ve been doing data governance beforehand, we’ve had to extend the data governance project to support GDPR. That doesn’t seem to matter what industry it is, it’s pretty universal.
What kind of changes are you having to make to a data governance program, in the first place, versus one where you have to extend it to cover GDPR?
If you’ve got a good basic foundation in, then that’s the whole joy of it, because you can just tweak the bits you need to do. There’s very little to do. You need to probably put a few more fields or columns in your data glossary. With GDPR, they talk about data maps, which is really just data lineage. So, if we haven’t got all of the personal data in it, we then go back and do data lineage. It really feels like an extension of something you’ve already done rather than recreating it.
I’ve been doing a lot of work that’s been regulation focused for the last few years, particularly in the financial services industry. The one thing I’ve always preached is, “Don’t do the bare minimum to meet the regulation, because when the next regulation comes along, you will need to start from scratch all over again. And there’s always going to be another regulation coming.” One approach is tactical: “Oh, we’ve got a regulation. Quick run around, fix it. Get ready for it.” The other is more strategic: “There’s a new regulation, but there’s also other regulations. And more new regulations are bound to be coming – what can be put in place that is a good foundation to support all of them?”
Going back a few years, at one client we were doing data governance for regulatory reasons. The person who’d brought me to the company said, “I want to do data governance properly. Obviously, we need to prioritize what needs to be done for this program but, bear in mind, I want this to be something that is scalable and extends to everybody.” And I went, “absolutely.” That was going really well, but we had a change when a new program manager came in, and started saying, “I don’t want to gold plate this. I don’t want to do more than the minimum.” He ignored my comments that it was short-termism and going to cost more money in the long term.
Well, that has to be rough when the project changes right in the middle.
Yes. Luckily, at that one, we got far enough with the foundation that it didn’t derail us at that stage. At the wrong time that kind of thing just leaves you with something that’s not sustainable. It doesn’t deliver the business benefits that having improved data quality can give you. All that it’s done is ticking the box for that regulation. If you already had a data governance framework in place, instead of needing a huge project team running around trying to document everything, you’d already have in-place data owners, data stewards, people who know what their data is. You’re asking the right people, and documenting it in a way that they can use it, and it’s useful going forward.
Then you get benefits beyond just compliance.
It’s true. Everyone asks for the benefits of good data quality. They don’t say, “We want good data quality,” but they want accuracy, integrity. Well, that would be data quality. [Chuckles]
And that would be data quality. [Laughter]
I love the quote, Felix said this morning, “artificial intelligence on bad data turns to artificial stupidity”. [Laughter]
That was very well put. Also, “Artificial intelligence without data governance is unethical.” That is one thing that I’ve been hearing more and more.
Yes, that was a really good point. A few years ago at a conference, somebody asked me, while I was on the panel, a question about the ethics of data governance. That was the first time anybody said anything like that and they just totally threw me. I actually think doing data governance is ethical, because you make sure that the right data is managed in the right way.
Come back for part two when Askham speaks about what goes into planning and starting a data governance initiative as well as what’s next for her in coaching.
Download our white paper to find out more about the keys to data governance success.