From Data Overload to Strategic Insights

[Music Playing]

Jim Marous (00:12):

Welcome to Banking Transformed, the top podcast in retail banking, I'm your host, Jim Marous. Despite having accumulated vast amounts of customer data for decades, many financial institutions still struggle to effectively harness its information for strategic advantage. Legacy systems, organizational silos and competing priorities often leave valuable data isolated and totally underutilized.

Jim Marous (00:39):

Meanwhile, the rapid advancement of artificial intelligence and machine learning technologies has created both immense opportunities and significant challenges for banks and credit unions trying to keep pace.

Jim Marous (00:51):

I'm joined in the Banking Transformed Podcast by Lee Wetherington, Senior Director of Corporate Strategy at Jack Henry & Associates. Lee has been at the forefront of helping financial institutions navigate these complex data and AI challenges. Lee also provides an insider's perspective on Jack Henry's recent alliance with Moov.

Jim Marous (01:16):

Financial institutions possess vast troves of customer data but often struggle to transform this information into actionable insights and personalized experiences. Our conversation will explore the critical barriers to effective data utilization, including legacy technology, organizational silos, and even cultural resistance, while offering practical frameworks for banks and credit unions to evolve their data capabilities.

Jim Marous (01:43):

So, Lee, before we dive into the specifics of data strategy, could you share a little bit about your background and your role at Jack Henry?

Lee Wetherington (01:52):

Yeah, Jim. I eat research and poop insights, that's about as short as I can get it. Practically, what I do every day as the generalist on our corporate strategy team is I ask three questions, and there's a never-ending static answer to any one of these three questions.

Lee Wetherington (02:11):

One is: what do we know now? What can we know definitively today? And then inside that, of those answers, we get to that question, which of those things are most durably predictive? And that helps set up answers to the second question, which is: what does the most probable future in a three to five-year frame look like?

Lee Wetherington (02:31):

And then once you get to a fairly concrete answer to that question, you get to the really juicy one which is: how do we want to be positioned in that most probable future to help our banks and credit unions win in that three to five-year frame?

Lee Wetherington (02:49):

So, given the topic today of data, data strategy, AI, et cetera, all of that is at the foundation of what the three to five-year future looks like and what you need to do to be positioned to win in that same frame.

Jim Marous (03:05):

Well, it's interesting, you've been in the financial services industry for quite some time. You've been at Jack Henry for 15 years, and you've seen a lot and we've both seen a lot. I've been in the banking world for my whole career in one way or the other. What is your perspective on the current state of data utilization and banking? Are most institutions effectively leveraging customer information, or aren't they?

Lee Wetherington (03:30):

No is the short answer to that question and that's been the answer for quite some time, and that is a function of a lot of things. First, I think most often what you hear in our circles where we travel is, "Hey, this is a function of infrastructure fragmentation, of data being siloed in this product, in this particular data silo or database over here, and then there's a separate silo with different data over there, and it doesn't all come together."

Lee Wetherington (04:05):

It's fractured, it's fragmented, it's not all together in one place, it's not cleaned, it's not normalized, you can't access it in real time, you can't therefore, feed it to either machine learning or different generative AI models for now is what the discussion is on different use cases there. So, the answer is an emphatic no.

Lee Wetherington (04:29):

The other reason, however, that I don't think data and data strategy has been – even a phrase I've heard mentioned, this is anecdotal, but I'm in a lot of boardrooms, year in year out on both the bank side and the credit union side – you rarely ever hear the phrase “data strategy.”

Lee Wetherington (04:50):

And I think that's because until AI has really forced the issue, really forced the industry to back itself into getting serious, sober and smart about data strategy, it's never been … even though we would say, "Yeah, we're kind of scattered and we're fragmented and we're fractured” there's never really been a time before this era that we're in now where if you don't have your data house together, if you don't have a clear data strategy – being behind in the era of algorithms and AI means that you're not just a little behind, you can be existentially behind, that is terminally behind.

Lee Wetherington (05:32):

And so, that has put pressure on every bank and credit union to really have a coherent approach to data, data strategy, and more specifically, what they want to accomplish with the use cases that data will drive in a meaningful timeframe.

Jim Marous (05:50):

It's really interesting from my perspective as well, and I agree with you totally that there's so much opportunity, especially in the digital world. When everything was in computers and on paper and on spreadsheets and all that, it really wasn't as usable as understandable.

Jim Marous (06:08):

My son's in data and analytics and actually, his biggest job is to make it so you can actually see the data work. But still the biggest challenge, the biggest barrier in my mind is while we may know everything about our customers, we don't necessarily let them know we know them.

Jim Marous (06:29):

In other words, we don't deploy it, we don't utilize it very well. We do that pretty well and have to on the risk and security side, but not at all when it comes to the experiences as we talk about or the engagements.

Jim Marous (06:44):

And it's frustrating because until you deploy that data, until you democratize the data across the organization and let other people see it, who had either never saw it or had to ask the IT department for it and it was always delayed or outdated, we really miss everything and you bring it up.

Jim Marous (07:05):

The fragmentation of data across different silos is a challenge, however, I'm now calling it more than anything else, an excuse because firms like Jack Henry and many, many others will work with you to bring these data elements together to make it so you can actually utilize that data and actually take it to market.

Jim Marous (07:28):

So, what practical steps can banks take to break down these silos, stop making them as an excuse, and create a more unified view of the customers? What are first steps?

Lee Wetherington (07:42):

So, let me mention something – we see data at the foundation of everything, and that's not a new revelation. And so, that's why we've bet the entire firm on modernizing the entire tech stack underneath the banks and credit unions that we support.

Lee Wetherington (08:01):

That includes, and actually is very much focused on modernizing the data infrastructure for the purpose of bringing data together for enhancing the ability to integrate anything at any time and make it available to anyone for any purpose or use case.

Lee Wetherington (08:22):

We worked hand in hand with Google and GCP to build something called the data hub. And the data hub is basically the liberation of data from all the silos, at least those that we have inside of Jack Henry.

Lee Wetherington (08:42):

So, if you think about we've got core data over here, payments data over there, digital platform data in another place, what we've done is we are now feeding all of that data into this data hub. So, the financial institution, the bank or the credit union has one place in which all of that data is being pipelined in real time.

Lee Wetherington (09:06):

And from there, you can do whatever you want or whatever the bank or credit union deems strategically important to do, including opening that data or subsets of that data up to partners of choice, fintech partners, et cetera, et cetera. And that, to us, was a very important first real building block for being able to play meaningfully, strategically, and in a timely way in this new data era.

Jim Marous (09:39):

So, we're talking as an industry a lot about personalization. I tend to see data, as I mentioned earlier, being used a lot more for efficiency and back-office effectiveness and risk and fraud more than really towards personalization.

Jim Marous (09:57):

What specific data capabilities do traditional financial institutions need to develop to remain competitive in an environment where fintech firms, big tech firms, my local grocery store, my local food market, my takeout food place, all are using personalization quite effectively?

Jim Marous (10:19):

What do traditional banks need to do to be competitive in an environment where the consumer’s expecting you to know them, understanding them, and reward them?

Lee Wetherington (10:30):

Well, this gets back to your first question, Jim, which is that not only is the data fragmented, but even if you put it all together, if you put together, if a given bank or credit union puts together in one place, all of the data that it has, let's say on a given account holder, they still do not have anything close to a full picture of that account holder's financial behaviors, preferences payment modalities, patterns, et cetera.

Lee Wetherington (10:58):

This is a really big blind spot, and I think this is the biggest challenge in putting together or one of the things that you come up against at the front end of really doing data strategy in earnest is that you realize that you only have a very small fraction of any given account holder's total financial data. And let me give you a data point on that.

Lee Wetherington (11:22):

We worked with Bank Director and Zynex Tech last year to ask this specific question, to ask banks and credit unions: how much of your existing account holders' financial data do you think you have? And the biggest segment of responses on both the bank side and the credit union side said, “We think we have between 25 and 75% of our account holders’ total financial data.”

Lee Wetherington (11:50):

And I will tell you, that’s delusional. They do not have anywhere near 25 to 75% of their existing account holders’ total financial data because of financial fragmentation, because of the fact that average American has relationships with 15 to 20 service providers, average smartphone has average 14 financial apps on it. And that is not fully understood, appreciated, much less measured by the average bank or credit union.

Lee Wetherington (12:19):

So, when you start getting into really beautiful, strategically sexy things that you want to do with AI, whether it's efficiency use cases, or fighting fraud in real time or even hyper-personalization, you can't get to meaningful personalization. If you're making assumptions about an account holder based on 20% or less of their total financial data, which is what I would tell you the average banker credit union actually has on hand. That data deficit is the biggest single impediment to meaningful personalization by banks or credit unions now in this era.

Jim Marous (12:56):

It's interesting, and I bring this up in a lot of my podcasts, is that I'll get in front of 200, 300 bankers, and I ask, "How many of you have closed a primary finance relationship in the last five years?" And almost nobody raised their hand. Mind you, it's a biased audience of bankers that work at the place where they probably bank, and so the people that have, have probably changed their jobs.

Jim Marous (13:18):

I then ask, "How many of you have opened a brand-new financial relationship at a non-traditional or an alternative financial institution last two years?" And everybody raised their hands. I said, “Look around the room, this is the silent attrition that you're being lulled into where you think you know everything about your best customers, when the reality is, they may be simply as Ron Shevlin would say, simply using as a money warehouse.”

Lee Wetherington (13:45):

Basically, it's just a storage place. It's not really where the transactions go. I also look at the fact that few institutions still look at money movement to see where money's flowing in and outside the organization, understand their behaviors.

Lee Wetherington (14:02):

How does Jack Henry, how do you as a person help your clients manage data better and build a better data strategy?

Lee Wetherington (14:14):

Well, that very question is what drove us, beginning about seven years ago to get really aggressive and to be the first major technology service provider in the financial services space to plumb fully into all of the major financial data exchanges – Finicity, Plaid, Akoya, Yodlee, MX, Intuit, Stripe even.

Lee Wetherington (14:42):

And so, we were the first to get fully plumbed into the open banking ecosystem in the United States, because we knew the only way to plug the data deficits we were just talking about that banks and credit unions have on their own account holders, is to fully lever and lean into standardized API-based open banking rails to set up the infrastructure by which those banks and credit unions can ask their account holders for the permission to bring this fragmented data back to the bank or the credit union. That is in our parlance what we would call getting to first app status.

Lee Wetherington (15:18):

So, we are not of the opinion that banks and credit unions are suddenly going to convince their account holders to get rid of the 13 financial apps on their phone and only have one, but first app is the app that helps make sense of what's going on across and in between all those other apps, and giving you one pane of glass at the bank or credit union preferably to make sense, to make meaning of that. And from that, to know what to do next or how to do better, and this gets to the hyper-personalization you were talking about.

Lee Wetherington (15:51):

It's only when they get a preponderance, at least, a preponderance of my total financial data from all of those disparate financial apps and providers back to bank or credit union, that they then can begin to analyze or run models or feed into models a behavior, my fuller set of data to be able to recommend to me something that's meaningful.

Lee Wetherington (16:14):

I mean, I tell banks and credit unions don't dare try to begin having some kind of automated capability for recommending next best product or service to your account holders if you don't have the full picture of what they have and don't have.

Lee Wetherington (16:31):

If I'm recommending to Jim that, "Hey, Jim, I noticed that you don't have an emergency savings account," and Jim's like, "What are you talking about? I do, I just don't have it here, and therefore, you don't see it and you don't know about it,” now you're just wasting my time. And not only that, you're revealing that you truly, you're verifying that you don't know who I am, which now impacts trust, reputation, we could go on down the line.

Lee Wetherington (16:57):

So, to answer your question, what we did is we plumbed in early to the open banking ecosystem, we negotiated first of its kind pricing with Finicity to be able to make it cost-effective for an average bank or an average credit union to do API-based aggregation back to bank or credit union.

Lee Wetherington (17:16):

Because by the way, before that pricing, it was not doable from a budget standpoint for an average bank or credit union to offer that. Now, we're getting into the era of 1033 of the personal financial data rights rule.

Lee Wetherington (17:30):

And I think the art (and not the science so much) and compliance of permissioning ALA 1033 means that the winners of these data wars that we're in – by the way, in case our listeners don't understand it, we are in full out data wars right now.

Jim Marous (17:48):

Oh gosh, yes. Oh yes.

Lee Wetherington (17:50):

The entities that ask first and ask best, and those entities that can lever their trust if they have that in place to get yeses when they're asking first and best, to aggregate data back to bank or back to credit union, so that that is home base for all of this otherwise fragmented financial behavior, those are going to be the winners. And that to me, is really as interesting as all of the other technology questions and issues we can talk about in terms of AI, infrastructure, use cases, et cetera.

Jim Marous (18:25):

So, Lee, you have all this major, all these major tools available for your clients and for prospects in the marketplace. You and I both know that even though you can show everything you can do on behalf of the financial institution, not everybody says yes. I'd even go further to say the biggest obstacle to implementing an effective data strategy goes beyond the technology.

Jim Marous (18:55):

In other words, behind the scenes, and more important than even the machines that make this happen, there's other things that get in the way. From your perspective, why do organizations hesitate from moving forward when you have the tools to help them move forward? What are the things that get in the way from your perspective?

Lee Wetherington (19:14):

Well, I mean, historically, we've always had the baseline, hesitation and constraint around regulation. Until 1033 was finalized by the CFPB just last October, we did not have a clear set of coherent rules around open banking, about financial data exchange, et cetera.

Lee Wetherington (19:36):

We also, by the way, still don't have a clear and comprehensive set of coherent rules or regulations or guidance even around what's okay and what's not in terms of AI, different kinds of models, data ingestion, data output, et cetera, et cetera.

Lee Wetherington (19:51):

Now we have other rules and regulations that impact and obviously inform and are explicit about what is okay and what is not okay in terms of personally identifiable information, et cetera, et cetera, et cetera, but regulation has been a constraint.

Lee Wetherington (20:05):

Now, the interesting thing to think about where we are, is we are in the middle of the current administration having a full out assault on the regulatory infrastructure in financial services that went after the CFPB first.

Lee Wetherington (20:20):

Now, there's talks of trying to consolidate the FDIC, even maybe the NCUA into the auspices of the OCC, et cetera. So, there's not only deregulation right now, there's regulatory upheaval and chaos. So, the question I would pose strategically – all of that is another way of just saying disruption.

Lee Wetherington (20:46):

We're in a window of disruption in financial services, and it's good to remind ourselves that in every window of disruption, our inclination is to focus on downside risk and forget about the proportionate upside potential, and that's what we're looking at.

Lee Wetherington (21:02):

A lot of people right now, seeing what happened to the CFPB, a lot of folks said, "You know what, I don't know what's happening with the CFPB, I don't even know if there's going to be a CFPB," and therefore, they think maybe all of the CFB’s rules are also on ice or go away.

Lee Wetherington (21:17):

No, if a CFPB rule like 1033 has already gone into effect (which it did by the way on January 17), it carries the weight of law. It would take an act of Congress literally to rescind that rule outright. Some people say, "Well, hey, Lee, that might happen." No, that's not going to happen because 1033 actually enjoys fairly broad, bipartisan and bicameral support. I say all of that to say one way of reacting to chaos and upheaval is to sit on your hands and say, "I'm just going to wait to see how this plays out."

Lee Wetherington (21:52):

Meanwhile, the winners of the data wars are going, "Oh, goody, goody, you are all distracted. We're going to lean in, fine tune our open banking plumbing, get serious about asking permission, not only for the narrow set of consumer financial transaction data within scope for the CFB’s 1033 rule, but for data way outside the scope of that rule.”

Lee Wetherington (22:19):

Because if we go ahead and build the infrastructure and the permissioning and the aggregation to open banking rails and all of that to make that a reality, that's exactly how we win the data wars, is by bringing the most data back to bank or credit union with which to do any number of things – improve real time fraud detection and prevention, yes, but also, to get to those more juicier or hyper-personalization use cases that you were talking about just a moment ago.

Jim Marous (22:48):

This is music to my ears, it gets so frustrating for me. I've been in the banking world forever, I've heard larger financial institutions say in many cases, “We'd rather beg for forgiveness than ask for permission when we know that we're doing things within the confines of what the regulations meant to say.” It doesn't always specifically say what we would love it to say, but bankers get in their own way because they make the excuse, "Well, we haven't been given the specific authority to do this."

Jim Marous (23:20):

At the same time as you brought up all the data driven, non-traditional finance institutions and even some traditional ones are saying, "No, we're well within our — I've used the example of the signature cards. We thought we had to have signature cards, and all of a sudden somebody came up with the idea, no, we just have to know our customer, which is a whole different realm you're looking at.

Jim Marous (23:41):

And in the digital world, especially when you're in many cases, getting permission-based data usage. So, if the client says, "Yes, I want to bring these other accounts into the realm," you are not doing anything illegally, you're doing everything within the confines of what Uber does, what Amazon does, what all these other companies do.

Jim Marous (24:04):

I get frustrated because a lot of legacy leadership really wants a complete go ahead and they'll never be comfortable, which is why we have people starting new account opening processes still with driver's licenses as opposed to all the digital information that's out there, because that can help the customer do things faster and easier.

Jim Marous (24:24):

We get in our own way, way too often. And you put it so eloquently in the way you described it, because we can't keep on waiting for permission if we know what we're doing is not pushing the edges yet, we got to do the basics right.

Lee Wetherington (24:41):

That's exactly right. Let me point out another really interesting irony about where we are, and this goes under the category of careful what you wish for or more specifically, careful what you vote for. So, nobody loves the CFPB in our circles, at least in the credit union banks, nobody likes the CFPB. There's all kinds of examples of overreach and we won't get into that.

Lee Wetherington (25:10):

But careful. The CFPB was the only federal regulator beginning to really bring hard scrutiny to bear to non-chartered financial service providers. The ones that you're talking about that have been behind the silent attrition and eating the lunches of banks and credit unions, often without the banks and credit unions fully realizing that's going on – the CFPB set up shop in a lot of those biggest providers.

Lee Wetherington (25:44):

And then we had the change in administration, and now, the attack and the gutting of the CFPB. Okay, great, guess what that's doing – that is making the playing field even less level now between chartered providers and these non-chartered financial service providers that they're competing against.

Lee Wetherington (26:07):

So, do you think that those non-chartered providers, given the new latitude and freedom that they're having, because the CFPB is being attacked by the current administration, do you think that they're going to let up on the gas or they're going to press the pedal?

Lee Wetherington (26:22):

They're going to press the pedal, and so you've got to zoom all the way out to see these unintended consequences and effects of things you probably hoped for, dreamed for, and voted for, and now you've got again, these downstream effects that you didn't necessarily anticipate.

Lee Wetherington (26:38):

By the way, we also thought that we're not only going to have a strong economy, a lowering interest rate, continuing lower inflation, we're getting now the opposite of that. And so, you've got to take all of this stuff into account, and then strategically ask, “What opportunities does this create? Instead of sitting on your hands and saying, “Let's just wait until the dust settles.”

Jim Marous (27:02):

Well, you and I have both been in the banking industry long enough to know that being a fast follower or waiting to see what happens is not very good in a marketplace that is disrupting on a daily basis.

Jim Marous (27:14):

The reality is the consumer wants personalized experiences, they want their financial institution to show empathy in helping them manage their money better, and they're willing to partner with that if you give them value in exchange.

Jim Marous (27:29):

The problem with the industry is we continually – I remember the old stockbrokers that used to take this long, 18-page diary of what my life was like, and then put in their side drawer and never have access to it. The reality is the digitalization of data is made so anytime we collect information, the client knows and expects us to use it for our benefit. If we don't, then we lose that leverage.

Jim Marous (27:55):

I use the example that my personal financial institution knows more about me than I probably know about myself, but they never show it. And therefore, they never take any action toward the fact that I have a car loan, a mortgage payment, an investment services account, a savings plan with Acorns that takes money and makes it flow in and out of my traditional financial institution for the benefit of my financial wellness.

Jim Marous (28:24):

Doggone it, help me with that, my financial institution, you have that information and the client now knows you do. When Uber can provide a very customized relationship with me on my way to a hotel, tell me where I can buy food, where I can go out to dinner, what things I may want to do outside of the hotel, and yet my financial institution can't even tell me how to save my money better or how to get a better interest rate than I'm getting currently because they have all the interest rates I'm paying and everything else, you're breaking that trust.

Jim Marous (28:57):

One of the things that Jack Henry's done over time, and I've been watching you quite closely, is you continually make these partnerships to make it so that you can bring more to your client's relationship and that, in effect, bring more to their customers and members.

Jim Marous (29:12):

One of the things you've done recently is you build an alliance with Moov, which really made a lot of ways with its open-source payment processing capabilities. Can you explain a little bit of behind the strategic thinking behind these lines, and how it fits into your broader vision of Jack Henry?

Lee Wetherington (29:30):

Yeah, this is a data strategy play. I’ll just tell you straight up, this is a data strategy play. And I mean that from the vantage of the bank or the credit union. We talked about earlier how banks and credit unions don't have all of the financial data on their account holders, but they're not, to your point, as you just were talking about, they're not even necessarily acting on the data that they do have.

Lee Wetherington (29:56):

They're not using it, they're not leveraging it, they're not figuring out how it can bring value proactively to the account holder. And so, it turns out that sitting in the court of every bank and credit union in the country is all the data that you need to automate approval of what historically we would have called a merchant account application.

Lee Wetherington (30:23):

So, just to give a little history on this for everybody who doesn't necessarily track this closely, but 10, 15, 20 years ago, that process of applying for a merchant account with a bank and by the way, there were only a subset of banks that would provide merchant services depending on how far back you go …

Lee Wetherington (30:43):

That process would take maybe two weeks and you had to provide a lot of data, you had to provide transaction history and balance histories and credit worthiness tests, and all this other stuff to get approved for a merchant services account with the bank and provided that.

Lee Wetherington (31:00):

Then we got to the era of the pay fax. So, we talk about square, the dongles, et cetera, and so, what did they do? They said, "You know what, we'll have a master merchant services account with a bank, and then we'll just do little sub accounts off of that internally. We don't have to ask the business for as much information, and we can do that online, and now we'll get you that approval in two to four days rather than two weeks."

Lee Wetherington (31:24):

What we realized is that the banks and credit unions have the data to automate instantaneous approval for merchant account services for existing customers that they have in their banks and in their credit unions. So, why not do that?

Lee Wetherington (31:41):

Why not now go into an era where, forget the pay fax era of two to four days, you can get approved right now at a click, and because your smartphone is now or an app on your smartphone, specifically the bank or credit union's mobile banking app can be the point of sale device for accepting payments, you can now click to begin accepting payments, and literally begin accepting payments instantaneously.

Lee Wetherington (32:08):

Now, moreover, because of advances both in technology infrastructure, and in real-time payment, that is real-time money movement, if you put all of this together, banks and credit unions have the ability to offer micro and small business owners the ability to begin accepting payments from scratch, doing it right away when they accept those payments, getting the funds for those payments in near real time or at a minimum same day, which is also a revolutionary instead of taking payments by card or whatever.

Lee Wetherington (32:45):

And then three days later or seven days later, you get a wire for one amount for all of that, and you've got to reconcile that back to all of your individual card-based payments that you accepted. Well, we also realize, "Wait a minute, what are you reconciling to?"

Lee Wetherington (33:00):

You're reconciling to your bank information, and I mean to your accounting system, whatever you're using, that maybe QuickBooks or Zero or whatever the case. Well, we've got all the APIs for all that, so why not just also automate reconciliation so that you get to continuous real time reconciliation so that the average business owner is no longer spending 10 to 12 hours a week trying to reconcile a batch wire for transactions that happened three to seven days ago.

Lee Wetherington (33:31):

Now, you're cooking with gas, this is what I call a great, a really powerful example of data strategy. How can you lever the data you already have in your bank or credit union to do something truly new, different, and material in terms of value, both in saving time, saving headache, and allowing the average micro or small business to focus on their business rather than trying to play accountant when none of them have the time nor the expertise to do that.

Lee Wetherington (34:01):

We realized in working with Moov, we put all of that together. Here's another little thing Jim that we don't talk about very often. Moov is built on public cloud architecture in Google and GCP. Everything that we're building the entire tech stack from the core up that we've been rebuilding inside of a public cloud infrastructure on GCP, basically, we're both in GCP, so I want you to think about this for a second.

Lee Wetherington (34:32):

So, we can do things, we can get to speeds of money movement by both being in GCP that are otherwise, not possible, and second, don't require us making hops on the public internet to make those money movements between accounts. So, now, we're not just fast, we're the fastest. And so, that's a result of having modern infrastructure and using it in strategic ways to be able to bring new value in this case, to small and micro business owners.

Lee Wetherington (35:06):

Here's the other thing; in terms of you mentioned payment flows and your son being in analytics, the first thing any bank or credit union should do is they should understand who exactly they're serving right now inside of their bank or their credit union, and when they do that using basic payment flow analytics (this is not AI or machine learning, it's basic analytics), they would find out that 13 to 35% of their retail checking account holders are actually micro and small business owners.

Lee Wetherington (35:35):

A lot of these are Gen Z with side hustles, independent contracting, all the rest of it. People I don't think realize we're now at 5 million new businesses being formed every year in the United States. A lot of that is driven from Gen Z and their understanding of what it means to have a side hustle or to be a small business.

Lee Wetherington (35:58):

And banks and credit unions can serve that kind of client, that kind of account holder customer or member all day long using data that they already have if they just wake up to what they have, how to leverage it, and by the way, once you do that, why not ask that small business – by the way, this is an interesting question.

Lee Wetherington (36:18):

If you've got a small business owner who is presenting as a retail checking account holder in your bank or credit union, you can ask them permission to aggregate back to that bank of credit union, much more again than just consumer financial data. You can say, "Hey, why not permission this API to your accounting system?"

Lee Wetherington (36:40):

And through that one permission, get access to everything that particular business is doing, and also you will see through that what accounts they have with other service providers at the same time. It just goes on and on and on, so I'm going to stop here, but that's why we did the move with the Moov without trying to play with the language.

Jim Marous (37:00):

Well, it's interesting also looking at new data sources, new utilization of data, it's again, the actual deployment of solutions. Jack Henry, for years, since the beginning of my time has really specifically served community banks and credit unions. That's been your sweet spot, at least that's what I've known you as. You also move on a dime and you understand probably better than anyone else in the marketplace, the value and what community banks and credit unions can do now more than ever.

Jim Marous (37:37):

Five years ago, a community bank and credit union would have almost impossible time keeping up technology, innovation, new products, new services, new experiences with the big guys, be it … it doesn't even have to be that big because it costs so much to keep up. But because the solution providers across the board, yourselves being included, but so many others that now have compartmentalized the ability to innovate at speed and scale, not only are the small financial institutions able to change things almost instantaneously.

Jim Marous (38:13):

I mean, back in both of our days, you had an annual plan because it took a year to do anything and you'd never get there. There's a major advantage now, I believe, between community banks and credit unions, and competing in the marketplace where they can actually over compete, over deliver what I'm going to call the mid-range financial institutions, those of number 10 to number 3010, because they can move so quickly.

Jim Marous (38:42):

What other advantages from a data strategy standpoint to community banks and credit unions have that you really answer to, but really puts them at a place where they can compete eye to eye with the Chases, the Wells, the Bank of Americas and the U.S. banks?

Lee Wetherington (39:01):

Well, this is what we would call enabling banks and credit unions to punch above their weight, and it organizes everything we do. First, we realize practically speaking that a lot of the banks and credit unions that we serve are never going to have a data science team, they're never going to build out a team of AI engineers, be able to attract that, et cetera. So, they're really relying on their primary sort of tech stack provider to give them these capabilities.

Lee Wetherington (39:33):

And we are absolutely convinced that with the new tools, whether it's machine learning that's been around a long time it's the new generative AI models that are coming – by the way, I want to talk to you about what's coming on AI because I think that also informs this question.

Lee Wetherington (39:53):

But we think that we can have a new golden era of relationship-based banking based on efficiencies that free up banks and credit unions to do more with the same, that is by specifically the same people. In other words, we see AI generally as a tool set that basically gives an iron. This is Keith Fulton, our new Chief Data Officer says it this way, "It's giving everybody an Iron Man suit."

Lee Wetherington (40:26):

So, that now, in terms of productivity, in terms of efficiency, you can do orders of magnitude more at the same time that you can focus more qualitatively on the things that AI and machine learning and nothing else is really good at, which is relationship building and relationship deepening. And so, that's the way we see AI. If you look at what's coming, so I want to go ahead, if it's okay for me to-

Jim Marous (40:58):

Actually, that was my final question, was to ask you to look ahead and … it appears you had no lack of passion. I was going to ask what excites you about banking and what's going to be coming down the road? It looks like everything does, but from your perspective, what's exciting on the horizon?

Lee Wetherington (41:16):

Okay. First of all, what's exciting for me is AI is the UI. So, the ability we talk about back office as if it's like this one thing and it's not, it's this terrible patchwork of the backends of 15, 20, 30 different systems. And sometimes, you've got to be logged into seven or eight of them just to track down and resolve one moment of need for a customer.

Lee Wetherington (41:39):

Think about all of that going away with a unified back office that ties together all of the administrative backends of those systems and providers and partners that you rely on so that you can just literally with a ChatGPT interface, just ask the system a question, and it gives you a definitive answer. Now, eventually, it's not only going to be here, but it will be a hundred percent correct, a hundred percent of the time. This leads to what's coming up in AI.

Lee Wetherington (42:08):

What I'm most excited about and what doesn't get talked a lot about in our industry is the growing size of context windows and AI. A lot of us are still focusing on the sheer size, the parameter size of models, and you see these different models topping each other in terms of parameter size, that is not where it's at.

Lee Wetherington (42:26):

The models with the biggest context windows – when I say context windows, that's literally what you're typing in when you're typing in something to ChatGPT. But instead of thinking about that, "Well, I can just tell ChatGPT what I want or ask it a question," think about copying and pasting into that window, that AI an entire novel and to say, "Now, tell me everything I need to know about this novel with the focus on this or that," and it can do that. Think about that now.

Lee Wetherington (42:58):

So, you start saying, "Wait a minute, wait a minute, what are we talking about here?" So, what you need to be focused on going forward is the growing size of context windows, Google's latest Gemini public, I should say, Gemini model, has a 2 million token context window, which is to say, if I'm doing my math right, yeah, that's 20 novels, that's 20 novels that you can just put in that window, and that is in its short-term memory, and it can be definitively accurate about anything you ask it about those 20 novels, okay.

Lee Wetherington (43:30):

There is also a model out there that now has 100 million token context windows. That Jim is 1000 novels. Now, when you start getting to those kind of scale points, we're very quickly going to get to the point where the average bank or average credit union could put all data it has into a context window.

Lee Wetherington (43:56):

Now, if you don't know what that means or what that implies, I would tell you, you're not going to get it in the next few minutes from me, but just think about dumping every single piece of data you have into a short-term memory of a model that can definitively answer with 100% accuracy, 100% of the time anything about the dataset that you just put into that context window.

Lee Wetherington (44:19):

We're going to be there pretty soon and it's really, really exciting. We talk a lot about Agentic AI, so I think that's being covered pretty well in the space, it's really exciting being able to have something that's proactive, can take action.

Jim Marous (44:35):

It's a version of what you're talking about, because the reality is that contextual insight that you're talking about, that the ability to do that for each customer in a way that is highly personalized but adjust based on the person's life, adjusting from yesterday to today to tomorrow, makes it so, without taking many risks, a financial institution can say-

Lee Wetherington (44:59):

Which is exactly what we want to zero out.

Jim Marous (45:02):

Based on what you've told us, based on what we know about you and your relationships, based on what's expected to happen tomorrow, we can act like that GPS system that says, “Get off the road right now, take the alternative route. It doesn't seem faster, but there's something coming up that you want to avoid, and I'm going to get you to your destination faster than the traditional way of doing things,” that’s all the customer wants.

Jim Marous (45:26):

The customer wants their day to be easier, they want to take some of the burden off. They don't expect you to be perfect. Yes, so we all know that heck, I was in the brand system and we were complaining about something, and main office would shut down a marketing program because of seven complaints among 70,000 customers we are serving. That day is going to be going away but the power of these tools, the power of gen AI, the power of learning models, the power of just the dialogue.

Jim Marous (45:58):

When ChatGPT first came out, everybody said, "Oh, this is dumb, oh, this is not being very smart at all." And then you realized after playing with it a little bit, no, I was the person who wasn't smart, I was asking it the wrong things.

Jim Marous (46:11):

It's the prompting tools, but the ability to adjust over time is the real power. And I think AI moving from the back office to the front experience, now, the biggest question though is, and this will be your last question Lee, will be, will financial institutions be willing to take that step?

Jim Marous (46:31):

Because just because we can, doesn't mean the industry does or will, because we find excuses not to change the way we've done banking forever, because none of us are going out of business. We may not have great years all the time, but when things are going okay, you don't want to get in the mix of saying, “Why change things.” The bottom line is, will they actually take the ability and do something with it?

Lee Wetherington (46:57):

I think they will because when they see that this reinforces and turbocharges and amplifies the humans in the loop. It doesn't take the humans out of the loop, it gives superpowers to the humans that are in the loop. And that is another way of saying this is relationship banking on steroids is what this is.

Jim Marous (47:21):

Oh, and by the way, given to everyone from the call center employee to the branch employee, to the private banker, to yourself, so if I don't use anybody-

Lee Wetherington (47:31):

I think once they make that connection that this is not replacing them, that it is giving them superpowers they've never had before to do relationship banking at a level of quality and scale, never before possible they will not hesitate because they know it's ultimately finally curated and arbitered by the human beings serving the human beings, I think your community banks and your credit unions get on board.

Lee Wetherington (47:58):

And the exciting thing is that as we see in the next three years, the convergence of these context windows that become functionally unlimited in size with Agentic AI, which is the ability to have an AI drop somewhere, understand its environment, context and act on that, but you've got to have integrations to do that and then that all gets us to both text to action and speech to action – I mean, holy cow, we're in a whole different world.

Lee Wetherington (48:31):

I'm trying to figure out what my job means in that context, Jim, I know you're trying to figure out what your job's going to mean. I mean, by the way, I'll leave everybody with this; if you want to get a taste for why at least I'm excited about this – Gemini just released for free, by the way, it's deep research engine.

Lee Wetherington (48:52):

Go out there and just try it once, and then give yourself three weeks, and then I'd love to hear from you to see if you can make sense of what's going to happen.

Jim Marous (49:02):

I go the same direction with perplexity. I'm amazed, and then every day it gets to be something different. Like ChatGPT now is going to be able to adjust pictures in different styles, and it's crazy but you can't stop learning this stuff.

Jim Marous (49:19):

I leave everybody on my side saying, I am an advocate of continuous learning, of ongoing learning because what is coming up, you need to take learning as part of your day every day because it's happening that fast. And if you get this much, it's that much more than the person next to you, and sometimes, it's not if you can outrun the bear, it's if you can outrun the person who's being chased also along with you.

[Music Playing]

Lee Wetherington (49:42):

That's exactly it. And you can respond with fear or excitement and excitement is what will drive that day-to-day learning. Get excited about it, it's crazy, it's fun, and lean in.

Jim Marous (49:52):

Lee, it is always a pleasure to have a guest on who has the enthusiasm, energy, and passion for what they do the way you do. I will guarantee we're going to be doing this again. This is a blast. I appreciate you sharing with our audience what's possible, not to put things in your own way that are fictitious or at least, in our own minds, not fictitious, they're in our own minds. We need to embrace the change. Lee, thank you so much for being on the show today.

Lee Wetherington (50:23):

Thank you, Jim.

Jim Marous (50:26):

Thanks for listening to Banking Transformed, the winner of three international awards for podcast excellence. If you enjoy our work, please give us a positive review. Also, check out my recent articles from The Financial Brand and our fantastic research for the Digital Banking Report.

Jim Marous (50:43):

This has been a production of Evergreen Podcasts. A special thank you to our senior producer, Leah Haslage, audio engineer, Chris Fafalios, and video producer, Will Pritts.

Jim Marous (50:53):

If you have not already done so, remember to subscribe to Banking Transformed on both your favorite podcast app and YouTube for more thought-provoking discussions on the intersection of finance, technology and leadership.