Gary Barr is the chief data officer at BNY Mellon. We caught up with him to discuss changes in data culture, the organisation’s data management strategy, and different approaches to fintech partnerships.
There has been a lot of talk around central storage and data cleansing. What is BNY Mellon’s approach to managing the rising level of data?
Maybe we just think differently to everybody else but you are describing two things. You are describing a data lake and then you are describing the governance around the data to make sure that it is fit for purpose. While that is very sensible, one of the challenges of that processes is that it slows the data down. If you are a user and you are trying to use data, but you have to wait for it to go into a lake, and then you have to wait for some central group to cleanse the data, that causes friction. I think that as the volume of data is increasing and the velocity of data is increasing, those things are starting to be challenged.
As we think about that 2025 vision, we are really starting to question whether or not we should centrally store our data and whether or not we should cleanse our data. If you think about Amazon, Ebay, Airbnb, any of these companies, they don’t cleanse their data. They provide the user with sufficient information and feedback loops to enable the users to make smart choices. And so one of the things that we are starting to look at is how do we bring some of those concepts that are used outside of financial services into our ecosystem so that we can start to provide an Amazon-like experience for data shoppers.
Part of me is sort of like while we have to build a lake, while we have to clean data, and while we have to do these things, when you look long term, I don’t think that is probably the way that we will do it. I think that we will enter a very different environment where we will need to start putting data out there and allowing people who want to use the data to make smart intelligent choices. I think that is a really interesting question, and I think it something we are spending a lot of time thinking about.
The traditional way of cleaning data takes time, and the reality is that there is so much data now, and unstructured data of course, and everybody wants it instantaneously. I would argue that clean data that is laked is not very helpful. Raw data that is dirty is probably not very helpful either. So, you want something in the middle, and it is finding that balance that we are being to really think about.
What do see as some of the crucial elements of data management that are potentially being overlooked in the industry?
The question of data culture is really significant. Meaning banks and financial services in general are laggards when it comes to the adoption of data decision making. So, you’ll hear someone say, ‘yeah we took a whole bunch of data, we developed a model, we took it to the business and the business didn’t trust the data,’ and it is because financial services has for 50 years been a products driven, relationships driven business. So, when some come and say, ‘look I’ve got this new way of thinking about that,’ you get a lot of resistance.
Why is there such resistance?
If you talk to any CDs, and in fact I asked the question in a meeting recently, what is the biggest challenge you face in creating a data driven business? Culture was the number one answer.
You can deploy great technology and you can throw data at almost any problem, but the outcome has to be value and the outcome has to be accepted.
Do you see the industry moving to a more collaborative approach to developing fintech solutions with client input?
If you talk to Wells Fargo or JP Morgan, most of these organisations will have recognised that they have to start to explore, they have to invest in, they have to incubate, they have to innovate increasingly around the digital ecosystem.
If I’m honest, I see a lot of banks investing, but I don’t see a lot of banks striking these big collaborative partnerships.