Latest / Leaders in Lending / From Ledgers to Algorithms: Revolutionizing Financial Data

Transcript

From Ledgers to Algorithms: Revolutionizing Financial Data · Leaders in Lending

0:02
Speaker 1: You are listening to Leaders in Lending from Upstart, a
0:05
Speaker 1: podcast dedicated to helping consumer lenders grow their programs and
0:09
Speaker 1: improve their product offerings. Each week, here, decision makers in
0:13
Speaker 1: the finance industry offer insights into the future of the
0:16
Speaker 1: lending industry, best practices around digital transformation, and more. Let's
0:22
Speaker 1: get into the show.
0:26
Speaker 2: Welcome to another episode of Leaders in Lending.
0:28
Speaker 3: I'm your host Matt Snow today joined by Chris Gifford,
0:31
Speaker 3: chief Data and Analytics Officer from USAA.
0:34
Speaker 2: Chris, Welcome, Thank you, Matt.
0:36
Speaker 4: Good to see you again. It's been a while.
0:38
Speaker 2: Yeah, good to see you. Always good to catch up.
0:40
Speaker 3: I know, you know, it's been a while since we've
0:42
Speaker 3: worked together, way back in the day. A lot's happened
0:44
Speaker 3: in data and analytics since then, so I'm really glad
0:47
Speaker 3: to be able to have some time to dig into
0:49
Speaker 3: that with you.
0:50
Speaker 4: Yeah, I'm back, we're working together, Chase. I don't think
0:53
Speaker 4: our hair was gray, and we might have had more
0:55
Speaker 4: of it. So it's been a while, but always good
0:57
Speaker 4: to check in. And what an exciting and field that
1:00
Speaker 4: we're talking about too, and so much has changed it's
1:04
Speaker 4: hard to even keep up.
1:05
Speaker 3: With It's true and you know, I was doing some
1:07
Speaker 3: research and preparation for our conversation.
1:10
Speaker 2: I wasn't sure i'd find anything.
1:11
Speaker 3: But do you know who and when the first chief
1:15
Speaker 3: data officer could have came about? No, well, you know,
1:19
Speaker 3: unsurprisingly it's in banking, so it was actually Capital One.
1:22
Speaker 3: Catherine clay daw seems to be referenced as the first
1:26
Speaker 3: Cheetah Chief Data Officer back in two thousand and two,
1:29
Speaker 3: so you know around that time we were together.
1:31
Speaker 4: Wow, that is interesting. I hadn't I hadn't looked that up,
1:34
Speaker 4: so I knew it's been around relatively short period of time,
1:38
Speaker 4: but very interesting. Cap One makes sense. They do a
1:42
Speaker 4: lot with data over there.
1:43
Speaker 2: Absolutely.
1:44
Speaker 3: Well, maybe we can start with that, maybe give a
1:46
Speaker 3: little bit of an overview of what, you know, the
1:48
Speaker 3: chief data analytics officer role looks like at USA.
1:51
Speaker 2: How did you come to that, and a little bit
1:52
Speaker 2: of your journey.
1:54
Speaker 4: Yeah, so I'll take you back, not all the way
1:57
Speaker 4: to the very beginning. But as I was finishing school
2:00
Speaker 4: and looking for my first job, I took a job
2:02
Speaker 4: at Gie Capital answering the phones. You know, I think
2:05
Speaker 4: you for calling Monogram Bank USA. This is Chris Gifford
2:08
Speaker 4: and so it's been a journey since then, but that
2:11
Speaker 4: was my very first job out of college. And while
2:13
Speaker 4: I was sitting there, I noticed that this terminal that
2:15
Speaker 4: was on the desk in front of me seemed to
2:17
Speaker 4: have capabilities. So I asked for the manual to it
2:20
Speaker 4: and started writing macros to do my job because I
2:25
Speaker 4: was lazy, but had the ability to try to figure
2:27
Speaker 4: out how to make the machine do it for me,
2:29
Speaker 4: and kind of launched into this journey of what we
2:31
Speaker 4: would today call robotic process automation. And I wrote a
2:35
Speaker 4: significant string of macros that did a bunch of my job,
2:38
Speaker 4: and my boss kind of tapped me on the shoulder
2:41
Speaker 4: and said, hey, why are your stats thirty five percent
2:45
Speaker 4: better than everyone else on the floor and you have
2:47
Speaker 4: like zero errors? And so I showed them what I
2:49
Speaker 4: was doing and they said, okay, fantastic, you do that.
2:52
Speaker 4: Now you're off the phones. And that was kind of
2:54
Speaker 4: the first launch into data and analytics, computers, computer aid
3:00
Speaker 4: or assisted even digital processing, robotic process automation, and just
3:04
Speaker 4: automation in general. So that started Capital. They have a
3:09
Speaker 4: great program that they identify people with you know, skills
3:12
Speaker 4: or the desire they will train them and so They
3:14
Speaker 4: put me through a lot of training, and I went
3:17
Speaker 4: from there into adaptive control systems, building models back on
3:21
Speaker 4: the mainframe. I didn't use punch cards. I'm not that old,
3:24
Speaker 4: but I know a lot of people that did. But no,
3:27
Speaker 4: we were writing code on the mainframe and a lot
3:30
Speaker 4: in SaaS, but building algorithms and using those in our
3:33
Speaker 4: adaptive control system strategies to drive you know, credit card
3:38
Speaker 4: line management, originations, authorizations, reissue, those sorts of things. From there,
3:44
Speaker 4: I went to Nation's Bank down in Charlotte, North Carolina,
3:48
Speaker 4: Carolina for a bit. This was prior to the b
3:51
Speaker 4: of a merger, doing more of the same but a
3:53
Speaker 4: bigger portfolio, and then moved to Bank one in Columbus, Ohio,
3:57
Speaker 4: and survived all of those mergers up through and included
4:00
Speaker 4: Chase at the end of it. I spent eighteen years
4:02
Speaker 4: at Chase doing more of the same. As I was
4:04
Speaker 4: leaving Chase, I was head of customer Data and Analytics
4:08
Speaker 4: and really building out that full customer three sixty bringing
4:12
Speaker 4: online data products for consumption in various use cases, and
4:16
Speaker 4: really trying to advance the use of modeling into all
4:19
Speaker 4: the decisions and optimization engines. So from a career standpoint.
4:23
Speaker 4: That's right at the last minute before I made the
4:27
Speaker 4: jump over to USAA.
4:29
Speaker 3: Gotcha, And so what are some of your top priorities
4:31
Speaker 3: at USAA, Like, what are the challenges there that you
4:34
Speaker 3: guys are trying to solve with data and analytics?
4:37
Speaker 4: Yeah? Sure, So when I joined USAA, they weren't as
4:41
Speaker 4: far along, I guess in their data journey as what
4:43
Speaker 4: Chase was. As you could imagine Chase being a very,
4:45
Speaker 4: very large financial institution. So step one really was build
4:49
Speaker 4: a foundation of data, and it is get all of
4:52
Speaker 4: the data, get it cleaned, hygiened available. We kind of
4:58
Speaker 4: skipped through some of the old are evolutions of data
5:02
Speaker 4: and data warehousing. We didn't go to kind of the
5:04
Speaker 4: big industry third normal form warehouses, went more to a
5:07
Speaker 4: modern lakehouse architecture, and here most recently rehosted the entire
5:12
Speaker 4: analytics environment up on the cloud, which where we can
5:16
Speaker 4: use entire lines of code to expand the universe, increase
5:21
Speaker 4: compute nodes, bring in other data sources. We're now working
5:26
Speaker 4: on implementing cloud native tools so we can do jobs
5:30
Speaker 4: end to end and stay in the cloud the entire time.
5:32
Speaker 4: Really trying to press for speed and ease in the use.
5:37
Speaker 4: I think the other part that we've been working on
5:39
Speaker 4: is managing risk. Right, with a lot of data comes
5:42
Speaker 4: a lot of risk, so you have to protect it.
5:44
Speaker 4: There is sensitive data p I I, pc I data
5:48
Speaker 4: and lots of evolving laws, rules and rights around how
5:52
Speaker 4: you manage it and protected data for your we call
5:54
Speaker 4: the members, but for your customers. I think the other
5:58
Speaker 4: thing is you need folks on your team that have
6:03
Speaker 4: the ability to understand how does the business work, how
6:05
Speaker 4: does the business make money, how does a bank make money?
6:08
Speaker 4: So it would be nice, you know, some days I
6:11
Speaker 4: wish I was just still sitting down in the basement
6:13
Speaker 4: of a bank somewhere in a corner, writing code and
6:16
Speaker 4: not talking to anyone. But really, for us to find
6:19
Speaker 4: the value in data, we have to understand the business,
6:22
Speaker 4: build those business relationships with our product owners, our process owners,
6:25
Speaker 4: our channel owners, and figure out if we bring data
6:29
Speaker 4: to the problems that they're experiencing, what kind of value
6:33
Speaker 4: and opportunities exist there. And so I think it's that piece.
6:39
Speaker 4: Building those bridges and being able to talk the language
6:42
Speaker 4: of business, educating your business partners on data in its
6:46
Speaker 4: value and how can you use it? Is kind of
6:49
Speaker 4: a key part of the chief data and analytics officers,
6:51
Speaker 4: probably not as well talked about as you know, the
6:54
Speaker 4: modeling and how many petabytes of data do you have
6:57
Speaker 4: and what platform are you on and what tools do
6:58
Speaker 4: you use, but building those just foundational relationships with the
7:03
Speaker 4: business to understand what their main problems are is really critical.
7:07
Speaker 4: One last piece kind of the secret sauce maybe to
7:09
Speaker 4: all of this is you have to constantly drive innovation
7:13
Speaker 4: this area, just like we talked about earlier, this area
7:16
Speaker 4: of data and algorithms and data science and even generative AI.
7:20
Speaker 4: I can't go through a podcast without saying generative AI, right,
7:24
Speaker 4: But even these areas that are evolving so fast, you
7:27
Speaker 4: have to constantly be pushing the envelope on innovation. And
7:31
Speaker 4: so some of the things that we drive here your
7:34
Speaker 4: standard hackathons. We've done hackathons what we call it like
7:37
Speaker 4: speed dating. We bring a business people in, we set
7:40
Speaker 4: them at a table, and we send a group of
7:42
Speaker 4: data scientists at them, and they've got two minutes to
7:44
Speaker 4: describe their business problem. And then the data scientists team
7:48
Speaker 4: go table to table to all the business people and
7:50
Speaker 4: they choose, Okay, which problem do I want to try
7:54
Speaker 4: to solve over the next two days, and we invest
7:57
Speaker 4: a lot into these. The teams absolutely love them, and
7:59
Speaker 4: I don't think we've had one yet where we didn't
8:01
Speaker 4: at least get multiple patents and you know, multiple millions
8:05
Speaker 4: of dollars of business value. So it just continues to
8:08
Speaker 4: press uh, press us forward to keep trying to stay
8:12
Speaker 4: on the cutting edge and UH and use all these
8:15
Speaker 4: tools and all this data for for real business value.
8:18
Speaker 2: Yeah, awesome. I love a lot in there.
8:20
Speaker 3: And you did beat me to the AI Bingo card first,
8:23
Speaker 3: So maybe we can come back.
8:24
Speaker 2: To innovation as a topic. But you know, you hit
8:27
Speaker 2: on something.
8:27
Speaker 3: I want to get more of your perspective on the
8:29
Speaker 3: idea of getting value out of the data or aligning
8:33
Speaker 3: back to the business, because I think we are in
8:35
Speaker 3: a bit of a maybe a swing of the pendulum
8:37
Speaker 3: here where the focus is less on explainability or understanding
8:42
Speaker 3: how some of these technology work, like ll M or
8:45
Speaker 3: some of the generative AI. Like it just works and
8:47
Speaker 3: it's cool and it's fun right now, but do you
8:51
Speaker 3: see like different use cases where things should be a
8:53
Speaker 3: lot more rigorous around the explanation and like the tie
8:56
Speaker 3: to value from data to the analytics the insight back
9:00
Speaker 3: to the business, or is there is there a use
9:04
Speaker 3: in a world where it's like, hey, it just works,
9:06
Speaker 3: so let's roll with it and onto the next thing,
9:07
Speaker 3: like how do you think about.
9:09
Speaker 4: Yeah, there's there's definitely both an evolution and two categories
9:13
Speaker 4: of kinds of models and problems that you kind of solve.
9:16
Speaker 4: So so the start with the evolution of it. I
9:19
Speaker 4: can remember within my career having a I won't say
9:22
Speaker 4: which company, even though I've already named the wall, but
9:25
Speaker 4: somewhere back along that line, I had a boss that said,
9:28
Speaker 4: I don't believe in models. I think, you know, we
9:31
Speaker 4: should underwrite loans by looking at their income and having
9:34
Speaker 4: a person judgmentally decide this or that. And so that
9:40
Speaker 4: that was again not that that's within my career event horizon, right,
9:44
Speaker 4: so we're not that far removed from it, but we
9:47
Speaker 4: have come a long way now into saying no, we
9:50
Speaker 4: can use models to help us make better decisions, even
9:54
Speaker 4: help us avoid some of the maybe subconscious biases that
9:57
Speaker 4: we as humans have when we look at data or
9:59
Speaker 4: make decisions like that. And I think that's where you
10:02
Speaker 4: have to pair it up with the laws, rules and
10:04
Speaker 4: REGs in this space, which really are put in place
10:06
Speaker 4: to help us not put in or allow those biases,
10:11
Speaker 4: or you know, have people being impacted disparately across these
10:15
Speaker 4: different decision areas. So to kind of put a finer
10:17
Speaker 4: point on that, like a credit risk decision, should I
10:20
Speaker 4: extend credit to someone? Yes or no? Is one where yes,
10:24
Speaker 4: there are opportunities for models, but it's probably not going
10:28
Speaker 4: to be a oh it's a black box. The algorithms
10:31
Speaker 4: always right, just trust it, right, We have to be
10:34
Speaker 4: able to show what's going on inside of the model.
10:37
Speaker 4: Once you get beyond you know, two or three way
10:40
Speaker 4: interactions of data variables, the brain starts going like WHOA,
10:43
Speaker 4: I don't understand this anymore. And some of these algorithms
10:46
Speaker 4: can have sixteen, eighteen twenty way interactions of variables where
10:50
Speaker 4: we have no Our brain has no ability to comprehend
10:53
Speaker 4: how is it getting to its answer? But yet we
10:56
Speaker 4: can show through additional add on analysis, are is it
10:59
Speaker 4: doing the right thing? Is it making the right decisions?
11:02
Speaker 4: Are there disparate impacts or their unintended consequences of the model,
11:05
Speaker 4: and those things need to be layered on and so again,
11:09
Speaker 4: the credit risk decision is a key one Obviously, fair
11:12
Speaker 4: Credit Reporting Act and available availability of credit to all
11:17
Speaker 4: are key reasons why and necessary so that we have
11:21
Speaker 4: a good, solid and fair financial services industry. Some of
11:27
Speaker 4: the other areas, though, like who should we send a
11:30
Speaker 4: marketing mailer to? Actually, you and I used to do
11:34
Speaker 4: mailers back in the day, right, I don't even think
11:35
Speaker 4: they send out mail anymore. But maybe it's emails, right,
11:38
Speaker 4: or alerts or those kinds of things. So if you
11:41
Speaker 4: send an alert, it's like, hey, are you interested in
11:44
Speaker 4: applying for a whatever? Or Hey, we've got a partnership
11:47
Speaker 4: with this other company, would you like to see some
11:49
Speaker 4: of their things? Certainly it can be annoying to a member,
11:54
Speaker 4: but it doesn't really put them at risk. It doesn't
11:56
Speaker 4: put the company at risk. So there are some more
11:59
Speaker 4: innocuous of things that we've can be doing with with
12:03
Speaker 4: the models and the AI to target and optimize in
12:08
Speaker 4: ways that we can't really see or understand. Maybe another
12:10
Speaker 4: one would be trying to figure out why is this
12:14
Speaker 4: customer calling and automatically wrap them to a customer service
12:19
Speaker 4: rep that is best suited to handle that kind of
12:21
Speaker 4: a problem, so again not really harming the customer, trying
12:24
Speaker 4: to drive efficiency some of just those back office operational
12:28
Speaker 4: you know fish where the fish are kinds of things
12:32
Speaker 4: are certainly there. And then when you start thinking about
12:34
Speaker 4: like the cool new kinds of generative AI solutions, man,
12:38
Speaker 4: I get excited. I want to be able to ask
12:41
Speaker 4: my supercharged digital assistance, like hey, go get me new
12:45
Speaker 4: homeowners insurance. And I wanted to like to be able
12:48
Speaker 4: to go do that, right, And so you start getting
12:51
Speaker 4: into could could I could we as a bank? Could
12:54
Speaker 4: we as a country? Could the government allow us to
12:56
Speaker 4: get to a point where we have digital assistance like
12:59
Speaker 4: that that have the free reign to go out and
13:02
Speaker 4: solve these kinds of problems for us? Certainly it's easy.
13:05
Speaker 4: Now It's like, hey, here's the entire year end financial
13:09
Speaker 4: report for whatever company. Can you distill this down to
13:12
Speaker 4: the key points and compare to you know, their competitors.
13:15
Speaker 4: That kind of stuff is available, these ministryvia kinds of
13:19
Speaker 4: things I want to be handled by AI and m
13:21
Speaker 4: L And so I think in our personal lives and
13:24
Speaker 4: areas outside heavily regulated businesses, there's probably more opportunities for
13:28
Speaker 4: it to serve us and create experiences that are very cool.
13:33
Speaker 4: Within financial services obviously, and necessarily a highly regulated, regulated business.
13:39
Speaker 4: There there's a lot more concern and caution and controls
13:43
Speaker 4: we need to rep around it. Sorry, that was a
13:45
Speaker 4: long answer, did I did? I covered a lot.
13:48
Speaker 2: No, I like it.
13:49
Speaker 3: I think, you know, one of the other things that
13:51
Speaker 3: there's again lots of like similarities and being aware of
13:54
Speaker 3: the history of like you know, even with the more
13:57
Speaker 3: complicated models, maybe it's even more true like the garbage
13:59
Speaker 3: and guard the job right, So you talked about generative
14:02
Speaker 3: AI and some of those, like you know, understanding the
14:04
Speaker 3: data that these algorithms and machines are using to base
14:08
Speaker 3: their recommendations or insights on is important. Even some of
14:13
Speaker 3: the changes in innovation, I'm seeing more of being able
14:17
Speaker 3: to provide references or links like here's why I'm giving
14:20
Speaker 3: you this answer, here's you know, maybe a study I've
14:22
Speaker 3: seen or other sites. So I do, like, like you said,
14:27
Speaker 3: really excited about the space and the rate of change
14:29
Speaker 3: and the ability to add you know, more insight and
14:34
Speaker 3: speed to I think call back to what you were saying,
14:37
Speaker 3: like you know, the the usefulness of these technologies really
14:42
Speaker 3: comes back to how individuals or members are going to
14:45
Speaker 3: get value out of them that we provide. So in
14:49
Speaker 3: terms of other innovations, the hack of phone, I wanted
14:52
Speaker 3: to come back to that because I really like the
14:53
Speaker 3: idea of combining the.
14:55
Speaker 2: Business in that.
14:56
Speaker 3: You know, a lot of the hackathons I see kind
14:58
Speaker 3: of come more ground up, you know, engineers or programmers
15:01
Speaker 3: thinking about ways to make maybe their lives easier, or
15:04
Speaker 3: some consumer focus, but usually generated from within those units.
15:09
Speaker 3: Was that something that was always a part of your
15:11
Speaker 3: guys program to include the business or I like that combination?
15:15
Speaker 4: Yeah, and it's something that's strilled into me. I had
15:17
Speaker 4: a really, really good manager at Chase, and again I
15:20
Speaker 4: won't name names because I've had a lot of good
15:22
Speaker 4: managers through the years, but I can still hear his
15:24
Speaker 4: voice saying, solve business problems, right. That's That's what the
15:28
Speaker 4: data analytics team was there to do. It wasn't to advance,
15:31
Speaker 4: you know, the capabilities of science. We weren't doing science
15:35
Speaker 4: fare projects. We were there to solve real business problems.
15:38
Speaker 4: And that's something I've carried forward in my career and
15:41
Speaker 4: I've really kind of preached that into my teams, Like
15:44
Speaker 4: I will do hackathons I'm doing monthly, I'll do them quarterly.
15:47
Speaker 4: I'll give out large prizes if you win them, but
15:50
Speaker 4: we're going to work on solving business problems if we're
15:53
Speaker 4: going to take them on. Not to say that there
15:55
Speaker 4: aren't opportunities in those back office roles in the data engineering,
16:00
Speaker 4: especially with some of the generative AI auto code kinds
16:06
Speaker 4: of things like you can think about in Get Up
16:08
Speaker 4: or you know, I've watched some of my team go
16:11
Speaker 4: from you know, writing code long hand, all of a
16:13
Speaker 4: sudden it's just like they're just hitting their hitting tap right,
16:15
Speaker 4: and it's just it's like writing a line of code
16:17
Speaker 4: at a time and they're just kind of overseeing it
16:19
Speaker 4: to make sure it all works. So certainly there are
16:22
Speaker 4: a large number of opportunities in the back office spaces
16:26
Speaker 4: for these kinds of things to have value. But you know, again,
16:30
Speaker 4: at the end of the day, I think it is
16:32
Speaker 4: at USAA and kind of our focus on our mission
16:35
Speaker 4: and our in serving our members, it's really about what's
16:38
Speaker 4: one more thing that we can do to take problems
16:41
Speaker 4: off of our members plate and help them, you know,
16:44
Speaker 4: achieve financial stability in their lives.
16:47
Speaker 2: Yeah, totally agree. I like that approach.
16:51
Speaker 5: Hey, there, former host Jeff calviner here to let you
16:54
Speaker 5: know about an exciting opportunity to strengthen your understanding of
16:57
Speaker 5: artificial intelligence and financial services. If you want to better
17:01
Speaker 5: understand the fundamentals of how AI models work and how
17:03
Speaker 5: they will impact the financial services industry, your luck. Upstart
17:07
Speaker 5: is launched an online course and certification designed to do
17:10
Speaker 5: just that. This course will give you the knowledge you
17:13
Speaker 5: need to leave the conversation about how your organization can
17:16
Speaker 5: best leverage and manage this revolutionary technology. If that sounds interesting,
17:21
Speaker 5: check out the course as it's currently free for a
17:23
Speaker 5: limited time. You can learn more at upstart dot com
17:26
Speaker 5: slash AI Certification. Again, that's upstart dot Com slash AI Certification.
17:31
Speaker 4: Thanks, and now back to the show.
17:35
Speaker 3: What other innovations have you seen in the past, you know,
17:37
Speaker 3: a couple of decades in terms of analytics, data insights,
17:43
Speaker 3: just that whole evolution within in that space.
17:46
Speaker 4: Yeah, the I can remember kind of in my career.
17:50
Speaker 4: Again I won't say where, but one day we first
17:54
Speaker 4: built the data warehouse, right and before that, I think
17:58
Speaker 4: everybody in the world was using what different names across
18:02
Speaker 4: different companies. But it's like some kind of a marketing
18:04
Speaker 4: extract file em for you know, So the hey, we
18:08
Speaker 4: pulled a file and you guys can start using it.
18:10
Speaker 4: And we were doing analysis and we were creating marketing
18:13
Speaker 4: files and those kinds of things, and then we started
18:15
Speaker 4: loading those into a warehouse. And I can remember we
18:19
Speaker 4: loaded the very first month of data into the warehouse
18:21
Speaker 4: and was like, all right, fantastic. It took four months
18:25
Speaker 4: to get the second month of data into the warehouse.
18:27
Speaker 4: But all of a sudden, we had like time series data,
18:29
Speaker 4: and there's a whole new era of analyses that we
18:33
Speaker 4: could start doing. And when you start loading you know,
18:35
Speaker 4: those and constantly well defined structured data with quality, start
18:40
Speaker 4: putting that time series together, all of a sudden, you
18:43
Speaker 4: can just bring data to problems like here's how many
18:46
Speaker 4: widgets you do per time period, and how many have errors,
18:51
Speaker 4: and just this entire amount of analysis that starts just
18:55
Speaker 4: bringing insights into problems. From there you go onto building
18:59
Speaker 4: models and optimizing and then auto optimizing these different things.
19:04
Speaker 4: And so that evolution has been very interesting to watch.
19:07
Speaker 4: I think in the modeling space, you know, started out
19:10
Speaker 4: just with straightforward old linear regression, went to logistic regression,
19:15
Speaker 4: and then kind of at a point in time, it
19:18
Speaker 4: just seemed like this entire world of models and algorithms
19:21
Speaker 4: exploded on what was available to us. And next thing
19:24
Speaker 4: you know, we've got random forest and genetic models and
19:29
Speaker 4: just all kinds of things that increased the predictive power
19:32
Speaker 4: in these spaces. Now again that's where we started getting
19:35
Speaker 4: into But can you explain it? She had to be
19:38
Speaker 4: careful which algorithm you used, But certainly you know, over
19:41
Speaker 4: the last man twenty five thirty years, we've seen an
19:44
Speaker 4: explosion of those kinds of models and algorithms available to us.
19:49
Speaker 4: The other thing I would say, I'm not a statistician.
19:54
Speaker 4: I don't have a degree in applied math and stats.
19:58
Speaker 4: But what I did was I learned that the tools
20:01
Speaker 4: that these algorithms come in make it easy for a
20:04
Speaker 4: person even like myself, to be able to build models
20:07
Speaker 4: and deploy them and add value into the business. So
20:10
Speaker 4: the enablement of tools and just a significant increase of data,
20:15
Speaker 4: data storage, and compute horsepower has made modeling accessible to
20:20
Speaker 4: a much broader group of people. Certainly, I have tremendous
20:24
Speaker 4: respect and for true applied math, stats PhDs and majors,
20:29
Speaker 4: and I hire a ton of them. But I'm just
20:32
Speaker 4: saying there more and more people can play in the space.
20:34
Speaker 4: Given the advent of those tools as well.
20:38
Speaker 2: Yeah, I couldn't agree more.
20:39
Speaker 3: And I haven't gotten to the point where like I'm
20:41
Speaker 3: just tabbing through code yet, but I have found like
20:43
Speaker 3: some of these things do help. You know, there's like
20:45
Speaker 3: an AI assistant debugging my code too. I can remember
20:48
Speaker 3: again like getting started in the mainframe days, like you know,
20:52
Speaker 3: you had the manual, you had your best attempt, and
20:55
Speaker 3: you had to debug it yourself. There was no like
20:57
Speaker 3: going out in Google or looking on for RUMs. And
21:00
Speaker 3: now to today where if I've got something formatted weird
21:03
Speaker 3: or written slightly and correctly, you know, I can get
21:07
Speaker 3: help really on demand if it doesn't run and get that.
21:10
Speaker 3: So it has really democratized, really the ability for folks
21:14
Speaker 3: without that deep training to maybe do more deep analysis
21:20
Speaker 3: or interesting things that previously unable to. Are you seeing
21:25
Speaker 3: because of that our data is data becoming more consolidated
21:30
Speaker 3: or are there still silos that exist between you know,
21:32
Speaker 3: like credit and marketing and like the use case kind
21:35
Speaker 3: of dictates granularity and the types of data or you know,
21:39
Speaker 3: how do you see those things?
21:41
Speaker 2: Kind of?
21:41
Speaker 4: So, yes, there are still definitely silos of data that exist,
21:45
Speaker 4: and so even within a company like a bank. There's
21:50
Speaker 4: historically been silos right across the different subject areas, and
21:53
Speaker 4: the advent of the warehouse or even a cloud based
21:56
Speaker 4: architecture is allowing us to break down those barriers and
22:00
Speaker 4: build out environments where it's less out, you know, looking
22:04
Speaker 4: for data in the stata availability instantaneously. I think the
22:09
Speaker 4: other thing is this is again a little bit of
22:11
Speaker 4: an older buzzword, but that API architecture right, so everything
22:16
Speaker 4: should be published and available via API. I think the
22:19
Speaker 4: biggest bridge, or the biggest silo historically that is starting
22:23
Speaker 4: to get broken down is we take all those files
22:25
Speaker 4: as marketing analytics files we talked about are different feeds
22:28
Speaker 4: from different systems of record, and we build out this
22:31
Speaker 4: awesome analytical environment and we you know, we put tools
22:35
Speaker 4: on it, and we have time series and we can analyze, analyze, analyze.
22:38
Speaker 4: But the rubber hits the road when what you find
22:42
Speaker 4: in your analytics data you can pump over into your
22:44
Speaker 4: operational space and actually impact what your member or your
22:48
Speaker 4: customer is experiencing. And that bridge analytical to operational to
22:53
Speaker 4: me is the biggest silo of all and one that
22:55
Speaker 4: we all need to keep thinking about. How can we
22:59
Speaker 4: build bridges analytical to operational to make new data widgets, algorithms, alerts,
23:06
Speaker 4: et cetera rapidly available. You know, you can just publish
23:09
Speaker 4: it via API to be consumed into the operational optimization engines,
23:14
Speaker 4: next best engines, those sorts of things. And so, yes,
23:19
Speaker 4: silos definitely do still exist, and we are silo busters
23:23
Speaker 4: right and in that data analytics space and data scientists
23:25
Speaker 4: space and need to continue to do so. I think
23:28
Speaker 4: when you start looking across industry there are opportunities there
23:32
Speaker 4: as well, because there are very very large silos across
23:37
Speaker 4: industries that probably need to be attacked or at least
23:41
Speaker 4: rationalized at some point in the future to think about
23:43
Speaker 4: what kind of experiences and value can we bring to consumer.
23:47
Speaker 2: Yeah, that's good.
23:48
Speaker 3: I never thought of analytics, you know, being a silo
23:51
Speaker 3: buster type of a job. I'm sure there's someone out
23:54
Speaker 3: on LinkedIn with.
23:55
Speaker 2: That in their headline right now.
23:56
Speaker 4: Maybe we had that let's get a patent on that
24:02
Speaker 4: real quick, exactly.
24:04
Speaker 2: But you know you've mentioned a cloud too.
24:06
Speaker 3: I'm assuming it's not your own cloud that USA develops.
24:10
Speaker 3: So it makes me think and this has come up
24:11
Speaker 3: with some other guests as well, like the idea of
24:14
Speaker 3: bill buy or partner like as a as you're in
24:17
Speaker 3: this space and it's constantly changing and growing, Like, how
24:20
Speaker 3: do you think about this is a core competency USAA
24:23
Speaker 3: should own. We got to develop this in house versus uh,
24:27
Speaker 3: you know there are people out there that just do
24:28
Speaker 3: it better cheaper than us, and and let's leverage their
24:31
Speaker 3: their technology. Is there like a guiding principle you use
24:35
Speaker 3: or how do you evaluate those situations?
24:37
Speaker 4: Yeah, I think you have to be open to kind
24:40
Speaker 4: of all of the above, right, And I think it
24:42
Speaker 4: also goes by you know, the amount of maybe revenue
24:46
Speaker 4: your company has with the economies of scale you've achieved,
24:49
Speaker 4: do you have enough dollars to invest into yourself finding
24:53
Speaker 4: these new technologies or do you need help? And so
24:57
Speaker 4: we think of it here as at least kind My
25:00
Speaker 4: strategy has always been for the knowledge workers, the people
25:03
Speaker 4: that really understand the data and the algorithms and how
25:06
Speaker 4: to draw value from our particular population. People that really
25:10
Speaker 4: want to draw value from our particular membership, those related
25:14
Speaker 4: to the military or their families. I want to keep
25:19
Speaker 4: that knowledge on site. I want to have them be employees.
25:24
Speaker 4: Now if we're talking though, architecture or platform or warehouse environment.
25:31
Speaker 4: I find that that can be a little bit more outsourced, right,
25:34
Speaker 4: And so that's where we have been able to leverage,
25:38
Speaker 4: to your point, off prem public cloud and we have
25:41
Speaker 4: built pipes, large pipes. We've enhanced those pipes with AI
25:46
Speaker 4: and mL to test data that's going through those pipes
25:49
Speaker 4: out to the cloud, tested for sensitive data or for drift,
25:53
Speaker 4: or for anomalies, if values of different metrics or things
25:57
Speaker 4: are migrating beyond and acceptable range. But yeah, it really
26:03
Speaker 4: is that opportunity that the people that have built the
26:06
Speaker 4: clouds and built the tools to get the data to
26:08
Speaker 4: the clouds are way better at it than we are.
26:11
Speaker 4: That's kind of their single focus. They optimize everything. And
26:14
Speaker 4: so taking advantage of that was a pretty easy, easy decision,
26:18
Speaker 4: I believe for USAA. And then as far as like
26:22
Speaker 4: tools and solutions, I think that kind of goes back
26:26
Speaker 4: to the if you're big enough to generate and can
26:29
Speaker 4: afford hundreds of millions of dollars of investment into you know,
26:33
Speaker 4: building out your own solution. You know, there are a
26:35
Speaker 4: few banks that can do that. There's a few financial
26:37
Speaker 4: services companies that can do that. For the next tier
26:40
Speaker 4: it probably is one of those consortiums or finding a
26:44
Speaker 4: company that does it and buying them or using their products.
26:48
Speaker 4: But if you think about some of those problems that
26:50
Speaker 4: it would make sense for a consortium to tackle, like
26:53
Speaker 4: AML or fraud detection where we're all fighting a common enemy,
26:58
Speaker 4: those kinds of things make a lot of sense. You know,
27:00
Speaker 4: how to target matt for our product versus someone else's.
27:03
Speaker 4: We're not going to probably share notes on that or collaborate,
27:07
Speaker 4: but some of the problems I think do make sense
27:09
Speaker 4: for the consortium approach to get after advancing your capabilities
27:14
Speaker 4: using these solutions.
27:16
Speaker 3: Yeah, I think those are great examples, Chris, And it's
27:18
Speaker 3: been fun, like maybe thinking backwards a little bit to
27:21
Speaker 3: the things that have changed, but as you're looking to
27:23
Speaker 3: the future, like what are the things you've got your
27:26
Speaker 3: eye on for, you know, potential innovation that's continuing to happen,
27:30
Speaker 3: things that maybe aren't quite at a commercial scale yet,
27:33
Speaker 3: but you know you could see the potential like what's
27:35
Speaker 3: what's the future of data analytics?
27:36
Speaker 2: Got you're excited?
27:38
Speaker 4: Yeah, thanks. So there's a couple of things that we've
27:41
Speaker 4: been playing around with again in the ai mL space.
27:45
Speaker 4: Think about edge computing, So how do we put algorithms
27:49
Speaker 4: and identification and predictive capabilities out onto the edges of
27:54
Speaker 4: this wide network, whether that be on you know, your laptop,
27:57
Speaker 4: your iPad, your iPhone, or yourself phone of any manufacturer
28:01
Speaker 4: operating system. But you know, how do we put things
28:05
Speaker 4: there to observe and help our members, you know, figure
28:09
Speaker 4: out how to navigate our website or how to navigate
28:12
Speaker 4: our app So there's some some ideas and opportunities we
28:16
Speaker 4: think we have to play in that space and then
28:19
Speaker 4: connect that back into our systems and applications as the
28:24
Speaker 4: member is navigating through our processes. I think, like, if
28:28
Speaker 4: you think about what gets me most excited, especially working
28:31
Speaker 4: for USAA and for our membership, think about a scenario
28:34
Speaker 4: we're let's say we're we have all the data from
28:38
Speaker 4: Matt and his entire all of his transactions and all
28:41
Speaker 4: of his accounts, and all of a sudden, we see
28:44
Speaker 4: a sixty thousand dollars remote deposit capture from his cell
28:49
Speaker 4: phone of a check and then we have probably I
28:53
Speaker 4: don't know on average, four seconds before Matt logs off,
28:58
Speaker 4: and we miss the op opportunity to say something to
29:01
Speaker 4: Matt about that check so what is that check? Is
29:05
Speaker 4: it normal for Matt to deposit sixty thousand dollars or
29:09
Speaker 4: is it an anomaly outlier? Does it happen quarterly, annually,
29:14
Speaker 4: it's never happened before. All of those answers to this
29:18
Speaker 4: questions would help us give Matt advice on what he
29:21
Speaker 4: could do with that money, right, and we think that
29:24
Speaker 4: advice is one of our best products, and we want
29:26
Speaker 4: it to be one of our best products given the
29:28
Speaker 4: military that we serve. And so if you think about
29:31
Speaker 4: the data needed, So what I just described was transaction
29:33
Speaker 4: time series data that's been analyzed and categorized and bucketed
29:37
Speaker 4: into chunks where we can say is this an anomaly
29:42
Speaker 4: or not. But then also it's like, well it's quarterly.
29:46
Speaker 4: Maybe it's an investment or dividends from you know, coming
29:49
Speaker 4: from stocks, it's a one time thing. Maybe he just
29:53
Speaker 4: got a settlement for a lawsuit. Or maybe it is
30:00
Speaker 4: severance package. Sorry, maybe it's a severance, right, and there's
30:03
Speaker 4: a large and so there's a job change coming. Maybe
30:07
Speaker 4: this is an inheritance, and so there's again a lot
30:11
Speaker 4: of different things. Every one of those things I've described
30:13
Speaker 4: would require a very different response in those four seconds
30:17
Speaker 4: to talk to Matt about what he could do about it.
30:19
Speaker 4: Some of the options like does he have a cardlan
30:22
Speaker 4: that he could pay off? Does he have an investment property,
30:25
Speaker 4: does he have an investment portfolio that he's saving for
30:28
Speaker 4: in the future. So, again, all of those data elements
30:33
Speaker 4: coming together bring you more opportunity to say the right
30:37
Speaker 4: thing in those four seconds, the most meaningful thing to
30:40
Speaker 4: Matt giving him the very best advice I think of,
30:44
Speaker 4: especially again serving the military. If we've got say a
30:48
Speaker 4: Navy nuke, so someone that is in the Navy, they
30:52
Speaker 4: are becoming a nuclear specialist and they re enlist. The
30:57
Speaker 4: reenlistment bonus is quite sizable, and it's a very very
31:00
Speaker 4: important role. And so if let's say Matt was Hey,
31:04
Speaker 4: I just re enlisted. This was my reenlistment bonus, that
31:08
Speaker 4: next conversation should be, Hey, Matt, we know you're probably
31:11
Speaker 4: gonna go out and you buy a new Forward Lightning
31:16
Speaker 4: or some big, you know, expensive car. But what if
31:20
Speaker 4: you bought a normal car and put half of that
31:23
Speaker 4: into retirement right now at your age, that is going
31:26
Speaker 4: to set you up down the road. So imagine being
31:28
Speaker 4: able to have that kind of impact on a young
31:31
Speaker 4: service member's life because all the data was in the
31:35
Speaker 4: right place at the right time, the algorithms recognized it
31:38
Speaker 4: and put forth the best advice that they could in
31:42
Speaker 4: those four seconds. Right. So to me, that is the
31:45
Speaker 4: ultimate use case of are we doing this right yet?
31:49
Speaker 4: And do we have all of the data coming together
31:51
Speaker 4: in a way where we can even have a chance
31:54
Speaker 4: at putting the best advice out to one of our
31:57
Speaker 4: members within that very short window of time that we have.
32:01
Speaker 3: Yeah, that's really impactful, and I think you're right, Especially
32:03
Speaker 3: with a business and a membership like USAA, that advice
32:07
Speaker 3: or that part of the relationship that you play so
32:09
Speaker 3: critical and maybe brings it back to some of the
32:13
Speaker 3: foundations of banking and financial services, where you know, every
32:16
Speaker 3: transaction did involve a human right there.
32:18
Speaker 2: There was always someone involved.
32:20
Speaker 3: Usually someone you knew who had that personal connection and
32:24
Speaker 3: a different form of database so I just through interactions
32:27
Speaker 3: with the person and knowing their situation. So how do
32:30
Speaker 3: you take the transactional data and make that personal again?
32:32
Speaker 3: I think is really could be powerful.
32:35
Speaker 2: Cool.
32:35
Speaker 3: Well, Chris has been great catching up. I think we've
32:38
Speaker 3: covered a lot today. Is there anything maybe we missed
32:40
Speaker 3: that you want to cover before we wrap.
32:43
Speaker 4: No, it's great. I get excited getting together with the
32:47
Speaker 4: other people in the industry that have thought about these
32:50
Speaker 4: kinds of things. I get excited about what the future
32:53
Speaker 4: can be. I wonder how much of this can come
32:56
Speaker 4: to fruition within our career lifetimes, or how much are
33:00
Speaker 4: we going to have to hand off to our kids
33:02
Speaker 4: to you know, finish this work. I keep telling my kids, man,
33:06
Speaker 4: data and analytics is still the opportunity of this generation,
33:10
Speaker 4: and they do not listen. But I hope maybe yours will,
33:14
Speaker 4: or are other people that are listening to this podcast
33:17
Speaker 4: will have more success than I have had convincing people
33:20
Speaker 4: that this is still an exciting, highly evolving, highly impactful
33:25
Speaker 4: field that is going to continue to grow and move forward. Yeah.
33:29
Speaker 2: I don't know, Chris, I'm over three.
33:31
Speaker 3: I've got one who wants to be a lawyer, wan
33:33
Speaker 3: who wants to be an actress, and one who wants to.
33:34
Speaker 2: Be a YouTube star.
33:35
Speaker 3: So not really sorry following and dad's footsteps there, But yeah,
33:42
Speaker 3: a lot to be excited about.
33:44
Speaker 2: And thanks again, Chris, hope to catch your chick again soon.
33:47
Speaker 4: Yeah, of course, great man. Thanks.
33:49
Speaker 6: I'll start partners with banks and credit unions to grow
33:51
Speaker 6: households and expand consumer lending. Through its leading AI lending platform,
33:56
Speaker 6: up Start powered banks and credit unions leverage AI to
33:59
Speaker 6: offer higher approval rates and experience lower loss rates while
34:03
Speaker 6: simultaneously delivering the exceptional digital first lending experience that consumers demand.
34:08
Speaker 6: Whether you're looking to grow and enhance your existing personal
34:11
Speaker 6: and auto lending programs or you're just getting started, Upstart
34:15
Speaker 6: can help. Upstart offers an end to end solution that
34:18
Speaker 6: can help you find more credit worthy borrowers within your
34:21
Speaker 6: risk profile. With all digital underwriting, verification, loan closing, and servicing.
34:27
Speaker 6: It's all possible with Upstart in your corner. Learn more
34:30
Speaker 6: about finding new borrowers, enhancing your credit decisioning process, and
34:33
Speaker 6: growing your business by visiting upstart dot com slash lenders.
34:38
Speaker 6: That's upstart dot com slash lenders