Data Innovation Summit turns five next March. Along the way, we have had fantastic speakers unselfishly sharing their knowledge on stage with their peers. Without them, this journey would be impossible.
This interview is part of an interview series dedicated to humanising Data and AI innovation and celebrating speakers who have presented at the Data Innovation Summit. The emphasis lies on the Data/AI people/practitioners, their professional journey and their stories.
There has been a great shift in how we perceive data management. With increasing data volumes, sources and regulations, data processing and managing are becoming more complex than ever. The financial industry can particularly boast with its complex data landscape.
To get a better understanding of what has changed in the financial industry regarding data management, we sat down with Martin Walker who spoke at Data Innovation Summit 2018.
Hyperight: Hi Martin, it’s great to talk again! You were a speaker at Data Innovation Summit 2018. To refresh our memories and introduce yourself to our readers, please tell us a bit about yourself and the company you are coming from.
Martin Walker: When I presented at the Data Innovation summit in 2018, I was working at Nordea as Chief Information Architect. I have since left Nordea and am now CIO / CTO for the Nordic KYC Utility, a new joint venture between the banks aimed at improving the effectiveness of KYC (Know your customer). We are building a new business that is focused on managing high-quality customer data.
Hyperight: Next year we are celebrating our 5th anniversary. A lot has changed with data and advanced analytics during these 5 years. From your point of view, where do we see the biggest changes and advancements we have had, with a focus on the financial services industry?
1. How do we manage high-quality data? The data landscape is highly complex (within organisations and within society), the challenge we have is threefold:
- How to ensure data is entered correctly at source – in the KYC world, this means ensuring corporate registers have the correct information submitted by corporates
- How to aggregate (and match) data from multiple sources, each with different data models, languages, standards
2. Legal Ground for processing – including GDPR & US Cloud Act. For all data processing activities we need to think about data ownership, data classification, legal grounds for the processing, consent, where the data is stored, which vendors are we using etc.
Hyperight: Presenting at Data Innovation Summit 2018, you stated that in order to have a hold of our data landscape, we need to know what is our data, where is our data and how it’s being used. Nearly 2 years after and in light of growing data regulations, how much are companies aware of the importance of understanding their data landscape, are we moving in the right direction?
Martin Walker: I think that awareness is growing, especially as we are now seeing the first GDRP fines issued. We are also seeing either fines or share-price movement resulting from poor control of data. Does this mean that organisations are rushing to buy data catalogues? Not necessarily.
Hyperight: The use of AI and machine learning in data management, or augmented data management as Garner names it, is becoming a more prevalent trend. What are the benefits on one hand, and the challenges on the other?
Martin Walker: I believe that AI and machine learning can help in a few areas, see it as the pinnacle of the pyramid. I think most companies could find a few useful areas where AI & machine learning could help. For example, we are looking at using AI to verify the authenticity of National ID documents and checking they belong to the owner. In the scheme of things, this is a relatively small area vs our wider business. The heart of our business is effective data management, using business rules engines, databases, merging and matching techniques that have existed for years. Once we have this working we will look to use AI/ML in a number of edge cases to really optimise our processes.
But there are challenges with AI / ML:
1) Poor data quality – if you feed the algorithms with poor data quality you get poor results.
2) On a social side – If you use historic data to train the AI /ML algorithms, you will reinforce human behaviours of the past, e.g. sexism, racism etc. I also see dangers when AI/ML is put in the hands of the ignorant, e.g. Using algorithms that have been trained using western data sets on a use case in the developing world.
3) Hype and expectation – Some of the hype coming out is mind-blowing, I believe self-driving cars were meant to have arrived by now – I cannot see how the current generation of technology will be able to handle all of the edge cases. E.g. How a self-driving car can figure out a car park, with children walking around, people with trolleys, etc.
The heart of our business is effective data management, using business rules engines, databases, merging and matching techniques that have existed for years.
Hyperight: Taking about the decade to come, what are your future outlooks for data management and analytics for 2030?
- Data Management – More of the same, data is always going to be fragmented, this space is always going to be complex and full of challenges. Hopefully, awareness will rise and more effective data governance will result in higher quality data at the source
- We will have a lot more data to analyse or drown in
- The legal situation is going to be interesting. Although GDPR set out with good intent, I think that it is causing confusion, it needs to be refined. In the meantime the US will introduce a US equivalent – will be interesting to see what it has in store
- Blockchain will not be successfully used for any data management use cases e.g. KYC, Identity management