Fiserv: Transforming Enterprise Data Capabilities to Lift CX

Fiserv: Transforming Enterprise Data Capabilities to Lift CX

Share
We speak to Marc Rind, CTO of Data at Fiserv, on the evolution of the fintech’s enterprise cloud data capabilities and what this means for improving CX

Marc Rind is the Chief Technology Officer (CTO) for Data at Fiserv. Completing an MA in Business Communications from Emerson College, his entry into the working world aligned with the internet boom of the mid-1990s, when Marc held a deep interest in what the internet could do for public relations. 

“This was all brand new when I was studying at Emerson College, and it set me on a path in the field of technology and software development,” he says. 

Data was a particular point of interest for Marc, which saw him work for a series of DotCom startups before joining Automatic Data Processing (ADP), where he stayed for 16 years. 

“I helped build ADP’s data cloud,” says Marc, “pulling data from across its systems into a common location for the 600,000 businesses to which they provided HR and payroll capabilities.

“Not only did this help serve data and analytics aims for ADP’s products, but helped generate actionable insights from across the company’s massive data sets, to provide things like ADP’s National Employment report.” 

Then in 2020, Rind moved to Fiserv, the fintech that combines data from over 1,800 financial institutions for everything from for core processing to providing cards and payment solutions, all the way down to merchant acquiring and card issuing. 

“Fiserv’s position means it has some of the most unique and fascinating financial data in the world,” says Marc. 

Since joining Fiserv, he has been working towards building something similar to ADP’s data cloud, by pulling Fiserv’s data into a common enterprise-grade ecosystem. 

The aim? To serve the various product needs for each of Fiserv’s business units, helping it better serve client needs and positioning it to connect data from across the organisation and deliver actionable insights for customers – insights they would not be able to get anywhere else in the world.

Fiserv’s four-year journey to cloud migration

‘Marc, what technology stack should we be looking at going forward, from a data perspective?’ This was one of the first questions Fiserv’s CTO of Data was asked when he joined four and a half years ago. 

Of course, no decision process is easy, and each organisation has its own idiosyncrasies that must be considered, but it became clear to Marc that Fiserv needed to move to the cloud.

Fiserv operates scalable compute, elastic compute and various other data science workloads. “When you have workloads like this, you see a lot of spikes and reductions,” says Marc. 

“So, to better manage data functions, we wanted to move these services into the cloud.”

What’s more, Fiserv operates multiple business units, as mentioned above – meaning historically its data sources have been siloed in these different units. “Each of these data sets has their own purpose, and their own clients that they're serving,” continues Marc. 

“We wanted to have something that we could federate development on top of without disturbing data owners within our individual units – because data owners know their data best.

“It was important to work with these individual units so they could build into this cloud platform, so they knew that their data wouldn’t be affected, but it could be used for more value-added purposes organisation-wide.”

Another important element for Fiserv’s cloud transition was partnering with Snowflake, because of its platform’s ability to separate storage from compute. 

“What that meant is that if one of our business units was inputting data for its own client reporting and analytics needs, other units could access that storage from within Fiserv without having to move the data,” Marc explains. 

This has helped Fiserv cut down on data copies, and with workloads coming in different forms, Fiserv can now create different sizes of compute, whether for a multi-year multi-customer or a single customer report.

“Thanks to Snowflake, we can scale up and down quickly without moving data into different formats.”

Today, Fiserv is also working towards data sharing not just across its own units, but in sharing data with its customers too. 

“Our customers’ own data journeys are picking up,” says Marc. “So we’re really working on removing the friction of getting data shared out to them as well. There’s a lot of excitement around that currently.”

Data sharing: The perks and challenges

Just as Fiserv is excited to expand its data-sharing capabilities with clients, so too is Marc glad to see that Fiserv clients are excited to have the fintech as their data solution provider.

“We have clients from across the financial services spectrum, from small banks and merchants to some of the largest FIs. But across our client base, the idea of us being able to quickly and easily democratise data out to them is exciting for them and it’s enjoyable for me to see,” Marc says.

The work Fiserv is doing – and the positive client response it is receiving in its efforts to move data in a similar direction, connecting it across the Fiserv network for data-leveraging product concepts – is also helping Fiserv take the next step. 

“You can have a great idea, but you need to get the data into shape to back up those ideas of product concepts – and that data likely comes from across the organisation,” notes Marc. 

This is one of the challenges Fiserv’s CTO of Data faces. “Getting data into shape, cataloguing it, getting it flowing in correctly and completing quality checks is a challenge because we have to make sure we’re doing it safely,” he says. 

“What’s more, when we aggregate data and build models, we have to make sure that any client sensitive information, and the sensitive information of our clients’ customers, is kept anonymous while being presented in an aggregated form – ensuring we’re delivering insights without exposing anything we shouldn’t be.”

This is an area, understandably, of high scrutiny, and something Marc and the team simply have to get right. They do, though, and the benefits are just as exciting for Fiserv’s clients as they are for Fiserv itself. 

Meeting specific client needs

Of course, not every client of Fiserv’s is the same, and it is important to differentiate between them to deliver the tailored services they require. 

Marc has three broad, but often interchangeable categories to define client needs and expectations: ‘Do-it-myself-ers’, ‘Do-it-with-me’s’ and ‘Do-it-for-me’s’.

“What I like to call ‘Do-it-myself-ers’ often have their own data warehouses, they just need access to the data.” 

That doesn’t mean it’s a walk in the park for Fiserv, though. Not only does it need to remove friction for banking clients in accessing their data, it must also help them connect the data so they can make sense of their Fiserv data, simplifying the process of incorporating data into whatever they are working on.

“The ‘Do-it-for-me’s’ sit at the other end of the spectrum,” adds Marc. “They need support in building the dashboards and getting access to the insights. They want to know where they should be focusing their attention, so for us, it’s about guiding them through that. 

“Then, the ‘Do-it-with-me’s’ sit somewhere in the middle of that. They may have dreams and aspirations for their data strategy but may not know how to get there, which is what we help them do. Whether that’s for building data visualisations, providing the insights or even just helping build those data warehouses for them. 

“The expectations of what they want may shift as we continue working with them too,” Marc explains. “So, there’s never really one bucket in this case, and in any case with any client, it’s about understanding their individual needs. As I’m sure we’ve all heard, no two clients are the same.”

Scaling an enterprise cloud data platform for the future

Indeed, the services Fiserv can offer its clients evolve alongside the rapid pace of technological innovation. 

And, just like Snowflake’s platform can help Fiserv rescue data copies today, in the near future, Marc is excited by the prospect of being able to apply new models on top of this data, without having to move it. 

“Traditionally – if you have a service, model or function – you would bring your data to that function or whatever service that is. With Snowflake, it’s the opposite.

“Because of the way Snowflake is engineered, you can bring models through the data. That’s where most of the activity is, as close to the data as possible.”

Snowflake’s technology, which enables Fiserv to effectively bring the mountain to Mohammed, means the scope for Fiserv’s future innovation is extensive. 

“This encompasses everything ranging from getting data cleaned and harmonised to looking at say, loan information to identify the loans which might be at risk of defaulting, or highlight customers who might potentially be at risk of leaving the bank,” Marc says.

“We’ve also been able to pinpoint relationships between different account holders and identify different banking customers that live in the same household. 

“Once we’ve done that, we’ve been able to deliver banks’ insights into the different products they can offer these clients, using data from all the banks in our partner network.

“We have so many banks as clients, so we can leverage data from across the client portfolio more and more in the future.”

Of course, it isn’t possible to build the right products and services without the right talent onboard. To achieve this, Fiserv leverages its long-standing partnership with Apexon to get top talent quickly enabled with the knowledge base required to work at Fiserv. 

“When we were modernising workloads I needed to get talent in quickly,” notes Marc. “Apexon has been a great partner to support us in getting our growing workforce up to scratch as quickly as possible.

“As we go forward, we will continue to work tightly with Apexon as we continue working on many different projects. They’ve been joined at the hip with us for a long time and long may it continue.”

Data fueling product concepts 

For now, the main aim for Marc at Fiserv is to continue working data into a position where it can keep fuelling new product concepts.

Marc concludes: “My aspiration is getting Fiserv data in shape so that anyone who has a product idea can quickly vet it and start to build on it. This can only happen as we continue to remove friction and provide those insights to our customers based on that data, which is so important because Fiserv has one of the most unique databases in the world.

“I want our customers to ultimately gain the advantages of leveraging our intelligence to feed their own. And, whatever our data feeds, I want the data ecosystem to be fuelling all the great product concepts we’re seeing our clients produce.”

Streamlining secure transactions through advanced data management technologies
Engaging with innovative fintech solutions for seamless banking experiences
Empowering client relationships with actionable insights and personalised data strategies
Exploring new horizons in data analytics and financial technology landscapes

**************

Make sure you check out the latest edition of AI Magazine and also sign up to our global conference series - Tech & AI LIVE 2024

**************

AI Magazine is a BizClik brand 
​​​​​​​

Share