START TRANSCRIPT
Carolyn Fernald
Today you’re going to hear from industry data leaders including Robert Schoenfeld, the technical partner manager at Qlik. Eric Trusler the SVP of Cloud Partnerships at Equifax, Katherine Toll the Director of Customer Experience at GrowthLoop and James McGann, the Head of Banking, and Capital Markets at Snowflake. We’re going to talk a little bit about technology today. But mostly we’re going to talk about Troy’s vision and the outcome he is seeing at Renasant Bank. To set the stage, Renasant’s data estate modernization has been empowered by Data Rocket, Passerelle’s modern data stack built on Talend and Snowflake that creates a well-governed Data Cloud, ready for enrichment and activation by Equifax and GrowthLoop and introduces data observability and governance to every stage of the data lifecycle with pre-built dashboards and accelerators. That’s the biggest technology pitch you’re going to hear today. And I’m going to hand it off now to Bruce to get started on our conversation.
Bruce Ottomano
Thanks, Carolyn. And welcome, everybody. I’m Bruce, the Sales Director here at Passerelle. And I’m very excited to kick this webinar off with Troy. Troy, I’ve known you for a few years now. And I just wanted to put out there that I feel you’ve done a superior job at laying out a vision, a smart vision on how Renasant Bank can use data to support its goals. You’ve been able to work well across the business lines at Renasant to basically form a team to take advantage of data for institutional benefit, and you’ve executed. And I’d love to kick this off with one question. You know, can you describe the initiative that involves Passerelle, Snowflake Talend, Equifax and GrowthLoop? What is it? And how did you start it?
Troy Carmichael
Yeah, great. Like a question. And just for those of you who are joining who are not on the sales side, you had my email there at the very beginning, happy to even talk to you all later on. This is not really going to be salesy, though you may have just really our journey and really excited about what we’ve done. And it’s a team sport, I can talk a lot more about that I just get to be a representative and talk about it.
Years ago, our company was very, very low, late adopting technology, they couldn’t spell DW or EDW. I mean, they didn’t know what the data warehouse was. And I started a process. And then as I came in the company, the past couple of years, a lot of things have changed, we’d had something on site, we’ve made some migrations to Snowflake, we’ll talk about that. We’d had enough pain, and then enough experience that we knew where we wanted to go. And we were actually very good. I think it was really neat that we could get the way we’ve done it with our team. Part of it is we want to keep our data in-house. We wanted to be able to bring resources and we wanted to be able to develop an ecosystem – we’ve actually ended up calling it the Renasant Data Ecosystem, or RDE world, because there are so many different components of it.
Today we’re going to be talking about really some of the most critical pieces of it. We get our data in there on Snowflake and we chose Talend, and we’ll get into specifics on that why we migrated them -we were on some other systems before Snowflake and Talend. Those names will be mentioned here.
Then we started looking and Bruce, I just want to say a lot of what you have also helped us with I know we were using, so your IP and Data Rocket and with Talend. But also you have been a very key thought partner with us, with me, in growing where we are. And there’s a there’s a lot to be said about that.
Then you start taking a look at third-party data set, we’ll talk a little about Equifax and what that means. And then to tie it all together for some of our marketing and being able to quickly and reactively drill down to specific audiences we want to reach within the branch growth with his paint is really a paving a great road for that. But not only that, being able to measure performance over time. All right, in the center of all this is this philosophy that data is the most important asset we have. Second most important, first important is our people second is our data. And so you look at – how do we steward that? How do we monetize it? And part of it is – we load all of our data in and then from that loading of the data, we then have to go into what we call our Single Version of Truth, which is where we curate that data. This is probably a good point to talk about how we created our SVOT. How did we get there?
Bruce Ottomano
I know, you invested a lot of time and effort, creating that Single Version of Truth. So the bank can make the right decisions based on good data. You put a lot of time and energy on that, you know, what have you done to make sure that Esbat keeps up to date and stays up to date? Because data is always changing data is always flowing in, you know, what’s the magic, you’ve worked to have that SVOT be an SVOT?
Troy Carmichael
That Single Version of Truth is basically where everybody starts in our company when they come in first day. And, of course, we have a data warehouse. And so for analytics, and that is what is fantastic about it. But that Single Version of Truth is very important to pull all the data sources that we have, from our core banking, a lot of stuff to solve core, even systems that are not even related and pull it in, we had, we had very serious stability issues from in some different environments. So we were looking at a partner we felt would be long-term stable, with a significant time in the market that we felt had a good vision about the future of where they’re going.
And then a very important thing for us we’ll get into is just training, we had such a hard time on some previous systems of really being able to stay up to date and stay innovative. But I might just take the time to bring in Robert with Qlik Talend. We’re really happy earlier this year, there was a Talend and Qlik partnered up now if the Talend is a part of Qlik, but Robert can tell us a little bit about you know, where you see within the Talend world, Qlik world and tell the team a little bit about what was going on with us and y’all and how y’all helped us?
Robert Schoenfeld
I’s looking at the cost of having bad data in your ecosystem to be able to make, you know, good decisions. It really is a garbage in garbage out scenario. The data quality capabilities within Talend not only enable data engineers, but really assets across your organization to take data quality to the next level from being able to build data quality, doing deduplication capabilities within Studio, having some self-service applications and cloud to do profiling and take action on data, and even collaborating between assets. Not to mention, we also have a trust score in our data inventory solution where you can harvest data and find potential issues either at rest, or at data in flights. And the ability to do this either on prem or in your VPC, to make sure that you’re ingesting data in a way that’s compliant with an organization is also I think, very important to Qlik in Talend, and taking those assets and adding them to the broader Qlik. Product ecosystem, I think only makes that kind of story even more compelling and stronger.
Troy Carmichael
Yeah, so you know, one thing that kind of made me sit at what that looks like to me at our level. Now, one of the things that we’ve benefitted from has been some of the parallel parallelism that we’ve had previous toolset is really hard to do things in parallel. And our business as our data has grown the company and people using it, we’re getting more and more demands to get that data, right. And ready and early. And we just recently, I know, when we were scaling out, we actually had our Network Operations guy say, Hey, move the stuff, you know, maybe early the morning now because we’re able to, with that parallelism, do things that we’ve not been able to do before. There’s some clustering. Bruce, you’ve been around a whole lot listening to us as we’ve been developing our toolset, maybe some things you want to comment too about, about Talend and how you’ve seen us using it, and maybe a picture of what you’ve seen from other people as well.
Bruce Ottomano
At Passerelle, we’ve been in the data business for many, many years, and we landed on Snowflake and Talend as an ideal combination to best serve our clients. We became Talend’s first US VAR over six years ago, and SI reseller. Same with Snowflake, we became a Snowflake partner over six years ago. And one reason we selected Talend over all others. They were the first to make a highly performant Snowflake connector, two years ahead of anyone else. They were a cloud-first company. We’ve yet to meet a system we can’t connect to with Talend and have a superior connector.
Troy Carmichael
Sorry to interrupt you just because this is something exciting- something I just encountered yesterday. Our team was talking we’ve had to do some things in our previous system, which we’re migrating now the last little final bits. We built this new system, we did it with Talend. And it’s faster than we’re doing it natively. So I hadn’t really thought about that until this moment. But yeah, so there was that connectivity to Snowflake that I’ve noticed, and just recently, the team was talking about that pointing out to me, and I’m still learning more and more, which is what like about it.
What’s also interesting is the training, I just can’t overemphasize that right now. Like even though there’s a lot of changes going on at Talend. Our team, I’ve grown over 300% in the past year and a half, staff-wise. Being able to just send people to training and get online self-help is a really big help.
Bruce Ottomano
Thanks, Troy. Yeah, just to add to that, yeah, the Talend Academy is phenomenal On Demand training, all our consultants went through it for their certification. The consultants will testify, it’s phenomenal training to get quickly enabled. And what we at Passerelle – some clients were looking for some unique accelerators, you know, our, we built a product called Data Rocket, running on top of Talend and Snowflake. You know, for example, we have what’s called a dynamic, well-governed managed ingestion engine. It’s a metadata framework that does dynamic schema on read with Change Data Capture, batch ingestion process, that helps us to pull data even faster in a smart way to reduce the cost of connecting and flowing data and ensure I know we’re using that for your FIS connectivity, correct?
Troy Carmichael
Yeah. So that yeah, we should talk about that, too, that that there’s two things I just realized, I probably want to make sure we talk here and this little bit of segment. One is that meta-data driven piece that you’re doing with Data Rocket. We’d look at some of the vendors out there, and there were some problems with it, especially in our environment, the way that FIS is, and other similar models. And I wanted something that was driven by data in a database to load data into our system Change Data Capture, but then can also easily report and audit on it. That’s why I love that first question that you asked on your on your little survey there earlier, Carolyn, because that’s when it’s really important to me is to get the people oversight out of my business because it’s so transparent. I’ve got to say one thing before we move on – is Talend’s connector with Salesforce. The Talend connector with Salesforce has done more things than we ever could without it – we’ve been really impressed with the connector. We’re about to do some other cool things with it. It’s been a lot more than the time we have allocated. But you know, Robert, thank you so much for being here. And being part of this.
Bruce Ottomano
Thanks, Troy. Yeah, I think the next chapter was what once you assemble that SVOT you basically landed all your first-party data powering Renasant Bank, you turn towards third party data. Troy you know, what was his criteria and reasoning on why Equifax and what are you hoping to gain from a benefit perspective using the Equifax data and the Equifax relationship?
Troy Carmichael
Yeah, so let me just talk about partnership you know – you opened our eyes to some of the things that hadn’t considered, and one of these was third-party data. It’s been a great partnership but with Equifax there’s a lot of things I could say about it. But one thing for me that’s just the cost of ownership is you know, Equifax has so much data, but we can narrow it down to the specific regions that we need to be in. We’re not in all 50 states, but we can narrow down to just what we need. And once we signed that contract, I kid you not I sent the paperwork in Wednesday afternoon. Thursday afternoon, I had all that I needed to light it up in Snowflake. If I can get that load loaded straight up and Snowflake, then that reduces my load. Oh, it’s just, it’s just crazy. But let me before I get into any more of my thoughts, pass it off to Eric, to talk about what you’re seeing with Eric Equifax. You know, we’re looking at your commercial dataset right now and consumer datasets.
Erik Trusler
Thank you for that Troy. And actually, I want to start back up at maybe the 50,000 foot level, because you led a really important journey for the bank. And I think it’s a place where our audience can learn from when you think about doing something impactful for an institution. It’s about the people and culture. It’s about the technology to serve as a business outcome. And you have led that journey internally. And what does that sort of physically mean? What does that look like? It means that you have brought us to your business leaders and the data owners. So that we could talk side by side about what you’re trying to accomplish, and ensure that there’s debt on alignment to the things that are important to the bank. And that step is the facilitation of success going forward. So then, when you talk about the data, yeah, the data is really important. We’re talking about the commercial data at the moment, as you alluded to are 180 million business locations across the globe, but Renasant didn’t need that – it needed five states. For now, maybe seven states tomorrow, who knows. And inside of Snowflake, just a tremendous technology, we’re able to target just what you need. And then as you pointed out, sort of when you need it. And so as our conversations slide over into the world of consumers, households, that same concept is right there, meeting with your leaders, talking about business outcomes, and configuring that data just precisely to what you’re looking for.
Troy Carmichael
Yeah, might be a good time to let people know because if they don’t know me, they might think I’m an extrovert or something. But really, I am a I am a data guy, I’ve been in there slinging SQL. I got discovered while at the bank, and that I could do some other things. But I’ve been hiding out there today. So I want to say that if you’re listening to this call, what I think Eric just said is really important for your success. And I hate you just you know, take up some of this time, we’ll make it up on the other end for you, Eric, but it is getting just get some of these leaders in front of your other leaders in your business. Because you know, the technology, if you’re a technologist on this, you know, enough of it to guide them through the fear they have. And if you’re on the sales side to bring your technologists, they can help you with your fear. So now, going back to the data and the relationship, and what y’all have done has been very collaborative. You know, Bruce, you see people using this data, let’s talk about maybe some of the benefits, you’ve seen maybe combining this data with other data, kind of, let’s walk us through some of the some of that process, too.
Bruce Ottomano
From our perspective, Troy, we worked hard to build a smart financial services ecosystem revolving around Snowflake. And so we teamed up closely with Snowflake’s Data Marketplace team, there was a lead in charge of the Financial Services datasets. And through our discovery, we want to lean in on and form a contractual partnership with the top dataset providers in the world that could bring the greatest value to banks to drive deposits, drive lending, understand the wealth of your own core customers at a household direct household level, to know who to target to drive deposit cash inflow, to know what core members own a house, and do they have equity in the house to know who to promote a home equity line of credit to Equifax rose to the top over all other large dataset providers to help to basically help banks accomplish their mission, which is drive lending, drive cash inflows and reduce risk. So we formed a partnership with Equifax for that.
Troy Carmichael
With Equifax, we have the customers that we see in our markets, we’re able to you’re able to identify show us information about our customers – there is some proprietary information, monthly and ongoing factors that impact credit worthiness. Let’s talk about some of the things that people might see when they use some your data sets.
Erik Trusler
Yeah, mine is a simple mind. So I like to keep it simple. And you know, you have your client base, your first party data, and there’s lots of things that you know about your customers. But they’re also things that you’re blind to, they might be a small customer of yours, but that, frankly, have a lot more capability or capacity that can be consumer or commercial. So it’s sort of illuminating when you are shining the light into those blind spots. It can help you determine best prospects that can be consumer or commercial, and then trying to go find, look alikes who else is out there. And so that age-old effort of prospecting, and we like to bring sort of those insights back to you. I did want to touch just one quick thing I know, we need to move and you touched on CRA. But this has been an important part of our, partnership together, which is the ability once we understand the business needs to talk about what data can be used for what purpose, you don’t need to be an expert in every aspect. But together, we can talk about business outcomes. And then we can talk about what limitations there are or aren’t on the data. And making sure you’re staying on the on the right side of the rules and regulations that that we all deal with in the FIA space.
Bruce Ottomano
Try again, hats off to you that you’ve worked hard to form that SVOT, add in third party data to give you a richer view of your customers and prospects. You know, we were excited to introduce an additional layer to our ecosystem, our friends at GrowthLoop. You know, Troy, I know your title is EVP Enterprise Data Management. But I think of you as a chief monetization Officer, you know, how do we make money off of the data you’ve worked hard to make trusted? And I think GrowthLoop plays a role in that. Why don’t you share your vision around the art of the possible to monetize that data?
Troy Carmichael
No matter where we sit, any organization needs to be thinking about the revenue that we’re bringing to the bank or the expense reduction, right, it’s all the same thing. And so where I sit on data is – I believe data is currency. Data is a liquidity segment, so everything I’ve got is going to be focused on how we use that now that there are a lot of ways we can use that data. But one of the ones that becomes interesting, we start putting Equifax with our first party is – how can we take that data together and make it easy so they don’t have to come to IT or to the CTO, but they can actually themselves figure out based upon whatever your primary who they need to talk to. So I’ll just kind of pass this over to just Katherine about GrowthLoop and then you know, and how cool y’all are with what you do.
Katherine Toll
Awesome, more than happy to talk about how cool we are. Um, but so let’s quickly level set like what is GrowthLoop. So GrowthLoop is basically a no-code interface that sits on top of your data warehouse and allows anybody, no matter how technical they are, to essentially build super precise audiences and then export them to destination platforms. So start there, that’s what it is. And like, why does that matter? And I think why did that resonate with Renasant? So at the risk of getting a little bit philosophical, I think they’re kind of two big bets that we made as a company. And Troy can, you know, validate how he feels these are in his, you know, in the context of his experience working with us, but number one was kind of the idea of like this rationalization of your marketing technology around the data cloud. So the idea that you’ve made this massive investment in your data cloud. And historically, the paradigm has then been that a lot of data ends up getting copied to different platforms to then be used for activation. And so our premise was that at the end of the day, once you’ve invested in this data warehouse, like in your Snowflake, everything should sit right on top of that, it should be a super thin layer, it should be very plug and play with your tools.
So you really have to take like able to take advantage of all that security, all that scalability and maintain the integrity of that data model in one place. So that was number one on again, that kind of philosophical take and number two is in terms of who’s actually able to use that data. Our Second big bet was that it makes sense for the strategic decision makers to actually be able to do hands on the data and be quickly activating it to those destinations. Instead of relying on an analytics or a data team to make pretty simple audience polls – those back-to-back communications of iterating on audience definitions between those teams, causes a lot of lag time and prevents marketers from acting in as agile way as you would want. So those were our really our two big bets. in Troy, I’d love to hear a little bit about how you feel like that’s been playing out in a relationship so far.
Troy Carmichael
Yeah, so um, I tell you, one of the values was just initially it was y’all knowing helping us bring back bring together some of the things that we needed. Once it was there, you know, you hit on a concept – bring your apps to the data, don’t keep on sending your data out. So we had a challenge, we have vendors who would love to receive all of our data and charges for it. And then the complexity because they are creating a swamp and let’s say, your CRM, or a swamp in your social media analytics.
So we’re able to generate those audiences, push it into, let’s say, our Salesforce or into our core system into social media, all of them. There’s this concept of journeys, I think about like these emails used to get like email campaigns and early days, when I think of a journey that people come in and hit trigger points, maybe in an ad, and I’ll get them into social media, or maybe they don’t get a call, they will walk into the into the branch of the teller, there’s a journey that we start putting together with the whole customer experience, that are pretty much automated that data, one of the things that happened recently is we’re developing and if you haven’t paying attention, mortgage rates are going up, right.
And so you know, we had a whole group that was planning, they wanted to go one approach towards how they’re going to reach customers. And they said, we need this tool, now more than ever, because now we can start brainstorming on the fly – the decision makers who know how to do it on the fly, they can generate those audiences. It’s like the whole liquidity of cash flow, the more that the more liquid, it is more liquid, you can make this knowledge in them, the easier it is for them to do that. So I think that’s really helped. And then, for me, a big passion is being able to measure results over time, by default, and you can adjust about default, there’s a certain amount that gets held back from that treatment of marketing. With GrowthLoop, we can see how the actual control group went versus the results. And sometimes, I mean, guys in marketing, like admitted, but sometimes you spend $1,000, to make 10. And then sometimes you spend $10, to make 100. Well, you’d much rather be based on spending $10, make 100 and $1,000 make 10. So I like that. And, and I think that is empowering.
Katherine Toll
I think what you’re talking about with the flexibility and agility and like how that plays out in a few different ways. So first of all, it’s just being able, again, for a marketer to really easily iterate on criteria, they’re getting a live audience size back from the data warehouse. So they’re figuring out how many people actually meet this criteria. They’re getting breakdowns across key characteristics, so they can better understand who is actually in this audience, how does that inform the messaging, the creative, so they’re able to iterate you know, in real-time, the second piece, like you said, is measurement.
So even beyond just kind of those descriptive characteristics and building an audience based on that feedback from the data warehouse, you’re also able to optimize those programs post launch, so you can figure out what’s working, what’s not, what do I lean into what we need to adjust a bit. And then I think the third element of flexibility is similar to what I talked about earlier, in terms of where does it actually make sense to have your activation platform sitting in your segmentation platform, it makes a lot of sense to have it on the data warehouse, because as Troy was saying, you can now build build these really dynamic cross channel journeys that are based on any data point in your data warehouse. And really, the only way to do that is to make sure you’re locating again, that segmentation and that journey building as far upstream as you can see, you then have access to all the downstream platform so you can build that like super precise, super custom and response journeys for customers. So I think that idea of flexibility and I promise I’ll stop talking now. And that is one that I feel like you see, like play out in a few different ways across the spectrum of, you know, our partnership with Renasant and GrowthLoops, kind of, you know, philosophy more broadly.
Bruce Ottomano
And I’ll just add, you know, we wake up every day, thinking about how can we add more value for our clients? How can we stay one step ahead our thinking and bring something to the table? What solutions are out there now that support banks and credit unions are trying to accomplish? And we’ve worked really hard to build this circle of trust with a really smart financial services echo system. We call that Data Rocket that, again, it revolves around Snowflake. We can’t be friends with everybody, we’ve got to pick a winner, go deep, be an expert at it, to implement it and help customers maximize the value out of the technology. Again, we’re contractual partners with GrowthLoop, like we are with Equifax and Talend and Snowflake and other players, because we felt GrowthLoop reduces data sprawl. You know, to your point, Katherine, you know, why spend millions of dollars trying to build up as a, I guess you’d call a dual version of truth. It’s hard not building up a Single Version of Truth. Keep it that way, monetize the heck out of it with a technology like GrowthLoop and move on. Moving on to Snowfalke, you know, Troy, you know how it Snowflake empowered Renasant Bank to realize your vision to use data to support the bank’s goals?
Troy Carmichael
Yeah, so I will say early on in our journey, this is when I was still hiding out. And people didn’t know. Something I was just I was just kind of listening in and some of the conversations that we were migrating off the existing system. I’ll tell you that when they first proposed Snowflake and Talend, that the board rejected it. And they said why? And they’re like, why is this? Because it can’t be this good that that combination can’t be this good. Look at what we’re doing with these other systems. I kid you not, it was delayed for nine months, they go back and we actually the only answer you had was go in and add a bunch of stuff to get a get a closer to what the board expects.
So I’d say it was revolutionary compared to what we were experiencing. I’ve been around data my whole life trying to get out of it. I tried to get out of finance, I just kept on getting sucked back in. So here I am. But what I can say about it is that they’ve done a lot of innovative things, even since we’ve been there. That gives me confidence that they’re gonna continue innovating. And even where we’re at now – we had the ability to come into this not being not a whole lot of technical debt, because our company had been slow adopting data warehouses. When we migrated early systems of Snowflake, were able to take advantage of some of the technologies they were doing, such as role-based masking. So now anytime I get compliance coming into see who I can see PII information on, it’s like, here’s a list, and everybody else is masked. Snowflake’s got just little things here and there that seemed little when you talk about him. But when you put it in architecture, it minimizes the time I spend on compliance, and taking care of auditing to confirm is so minimal, because we were using Snowflake infrastructure.
Alright, the other channel was Snowflake, before I hand it off to James starts talking is that, you know, most banks use one, primarily three large core systems. Snowflake to us is about as big as that, because we’ve invested so much of our energy on there, you know, I’ve hitched my horse to it. It’d be painful to leave – if we had to, we could, but there’s no way I ever want to because it’s like, to me it’s a long-term vision, what they’re doing is fast. There’s a lot of pieces about it. James, why don’t you talk a little bit about Snowflake, some of the things that you’re seeing in the banking sector.
James McGeehan
Thank you, Troy It’s a difficult task to be left to go last after this crowd. But I’m thrilled to be doing it. I think it shows the contextualization of the collaboration amongst teams to come to a common outcome. So just very quickly – what is Snowflake? It’s a secure and scalable data platform and ecosystem of content that is really driven to solve business challenges and really empowers data-driven decision making. And really why we see Snowflake internally is we just are an operating system to deliver all of the fantastic data and intellectual property of the both, the Renasant Banks of the world and the GrowthLoops, the Equifax and the Talends. And if you really heard, just to contextualize what we’ve heard over the last, you know, 20 minutes or so, we’re Snowflake gets the honor and pleasure to simplify access to high-quality data through Talend, with focused, domain specific content of relevance from Equifax, and then really, to drive a business outcome for that hyper personalized client outreach through GrowthLoop. Right. And so I think, Troy, it’s just sort of a testament to your vision.
And I think, given the macro backdrop today, right, we’re in a serious time of transition with a dramatic rate of change, right? So the amount of impossible outcomes today is just so significant that people need to have that high quality data to really inform decisions in a quality manner. So, you know, let’s, let’s take a step back into the financial services industry. It’s predicated on physical data movement. And I think some of the magic that maybe your board was Resident about originally, it’s just, it’s a simple concept of let’s just stop moving the data, right, let’s bring the work to the data, simplify access to with those role-based access controls for the high quality, secure governed data state to make those best business decisions. And as I get to talk across banks, payment companies, capital markets, firms, what are their key business priorities? It’s really risk reduction, client and revenue retention. But all underpinning this is profitability. Because, as Troy aptly mentioned, we had the fastest increase in interest rates in four decades. So the cost of capital is higher. What am I doing from a technological event to connect to clients while protecting the bank and bringing myself that much closer to my clients to protect my revenue streams? And I think this unification of simplification of tech architecture, and administrative effort is enabling that possibility of the profitability outcomes that we think not just from a tech simplification standpoint, but we have to talk about the words of Gen AI, that expectation from board levels and investors that really is thinking about profitability, and profit, and productivity amongst our teams. So it’s a pleasure to be here, and I’ll kick it back to you, Bruce, and Troy.
Bruce Ottomano
So I’ve seen a lot of technologies in my 40-year career, I’ve seen nothing like Snowflake. That day one, when we became a Snowflake partner, our consultants that were used to using a different technology, SQL Server, and Oracle and other platforms. They almost didn’t believe it, you know, the experience in using Snowflake is like, this can’t be true. If we don’t have to do you know active passive log shipping mirroring, and all the management, there’s none of that you’re using Cloud as a service. And the platform is so flexible, easy scale up, auto scale out, time travel, you know, what did the data look like, you know, 43 days, 10 hours in two seconds ago, me the capability Snowflake has was five years ahead six years ago. And I remain confident that they’re still years ahead of anyone else in the market. And we’re blessed to have chosen Snowflake, and made the right choice six years ago, and we would make the same choice today.
James McGeehan
Thank you for the kind words, Bruce. But I also think something that Troy mentioned, it’s the banking core data that sits inside the walls of many firms. How do you unlock that value to bring together to what Katherine’s really talking about connecting to your clients. We have connected to Salesforce, but we also can help you get access to the banking Core Data, who are also clients of Snowflake to then hopefully, bring a much faster time to value around that client experience, understand your client footprint. And even if it’s one of those pieces now post SVP the deposit stickiness. How are you measuring the liquidity of your underlying business? To really reduce risk of your firm and scale out and connect your clients while protecting the bank? I think this is where I think some of the tidbits that have been said here just to unify it again, simplifying access to that really core banking data to then have the client outcome that you want to have. And I think this is where we’re thrilled to be part of that connective tissue.
Q+A
Carolyn Fernald
How are you approaching Gen AI at Renasant, Troy and then maybe we can get everyone to weigh in on how a banks and credit unions can be looking at generative AI and what their approach should be?
Troy Carmichael
Yeah, I’ll tell you where we are, is, we’re carefully looking at it. You know, I think we’ll talk to some of the partners here to say how they’re using it. There are some concerns about data, you know, leaking out of our system, just you know, Google, Samsung and AI, you’ll see from stuff that happened not too long ago, and, and also to the value compared to other things that we might be able to just buy out of the box. So we’re doing that analysis, are putting some R&D there. And there’s one more thing I was about to save, I can’t remember if I might interrupt someone else in the end, but maybe I’ll pass it around. I’ll just pick up maybe, maybe James, or one of you guys to go, how are you guys looking at AI in your world? Oh, I just want to say it’s neat Snowflake has come up with something recently that we liked, it’s going to allow us perhaps to use AI our own stuff within their own world and keep data from leaking out. That’s a huge thing. So we’re gonna explore that.
James McGeehan
And Troy, thank you for the setup. I’ll try to keep this very brief. This is obviously a topic that we’d love to talk about ad nauseam. But really, you can’t really have an AI strategy without a secure governed foundational data strategy. And I think, many times we see the expectation of this profitability and productivity miracle, but we also forget, what is the one thing we don’t want to have happen? We have 100 things we want to do. What’s the one thing that Troy just mentioned, you don’t want to have happened? You don’t want to have data and IP exfiltration. Right. So how do we then taking that as the governing North Star, then work backwards. So truthfully, this is where we’re standing together to help people keep the data at rest, in a secure governed data state, and then bring the modeling to it with choice of models, whether it’s open source models, such as a llama to, or it’s a proprietary model that you develop on your own, and you can use containerized services, we’re embedding services we call Snowflake cortex that were announced last week, that will be embedded within native Snowflake to be called as a function within the platform on that secure governed data to then solve your underlying business problem. Without that risk of data exfiltration and IP theft?
Troy Carmichael
What about others on the panel? Tell us about AI in your world.
Erik Trusler
Yeah, I’ll take a stab at it in the Equifax world. And certainly for financial institutions large and small. And, and regardless of vertical. There’s lots that AI can be doing in the moment, right from your customer service and your knowledge, your knowledge graph and other tools that help enable efficient businesses. I’ll take an angle for Equifax for the moment. So one of the things that we do, of course, is we bring models and scores into credit decisioning, consumer or commercial. And frankly, the the net of signals keeps the broadening from not just traditional lending activity and payment activity, but also payment of rents and payment of utility bills and, and what have you. So that net is wide. And AI is in Gemini is an incredible tool to help create even more predictive models. But importantly, in that process, a couple things haven’t changed. We all still hold a responsibility Equifax holder responsibility to ensure that there’s no disparate impact in the modeling. Equifax as a responsibility that we have helped you with to ensure that there’s, if there’s adverse action, someone isn’t eligible for the best offer, that there’s transparency as to you know, what in that individual’s background, caused that adverse action. And so Equifax has some IP or are around Explainable AI, that’s really important. So I think the takeaway here is, AI can be used a lot of different places in ways. As you look at sophisticated modeling. There’s tools and technology that Snowflake has helped support that when we bring our models, we carry forward those obligations of transparency, and we’ll, we’ll bring those to you.
Bruce Ottomano
You know, at Passerelle I’m excited about how generative AI could really drive the return on investment in scale. You know, we’ve been, you know, we use the garden leaders for Visual Analytics dashboards, you know, enable employees across a bank or credit union to get insights, make that fact based decisions. That’s all well and fine, but if you look at most institutions, there’s so many, only so many people that know how to build a dashboard, and only so many dashboards that get built. What I think Generative AI is going to bring to the table. Every employee knows how to ask a question. Every employee has something on their mind. Some questions they’d like to just ask, have the data and have a dashboard, have a report just be automatically built for him or her. That’s the promise of Gen AI, to drive democratization of data out to every employee at an institution versus some number less than that. The bottleneck being the effort it takes to create a dashboard, create a report to then share out that all goes away with Gen AI.
Katherine Toll
I think GrowthLoop when we think about Gen, AI, and AI in general, there’s like these three categories, where we see you know, it applying to our software. So number one is I think what Bruce is referring to, which is really enablement. So it’s like, how are you better enabling again, those strategic decision makers to better access, understand and activate the data. Because right now, obviously, we are, again, like a thin layer, it sits on top of the data interface, or on top of the data warehouse, there’s still some level of, you know, understanding of the data that has to go into actually activating it for marker to segment that data. And I think the goal is, and this is a feature that we’re actually we just rolled out on, is how can we allow a marketer to just, you know, type out a natural language, the kind of audience they want, and have that just generated, you know, into a SQL query based on the underlying data, instead of them having to comb through filters and figuring out figure out like, exactly which one seems to best match their use case. And so enablement, right, in terms of just better enabling folks to build those audiences really easily. And intuitively. And the other two buckets that we also think about, it’s going to be insights driven targeting, this is a bit more to what Eric was chatting through, which is how are we you know, enabling marketers to essentially activate like, propensity models and propensity models, affinity models, risk of churn, you know, lifetime value predictions, etc, without actually having to force a data team to build those. And so that’s number two is that insights driven targeting. And then the third piece is going to be actually the generative AI content. So we have a partnership with a cool company called typeface that actually allows you to, you know, actually generate the content in a creative that ends up removing a bottleneck we sometimes see folks run into, which is that you can now build audiences super quickly activate them super quickly. But your creative and your content, like doesn’t necessarily run at the same speed. And so I think those are really those three buckets again, so enablement, and then that more insights driven targeting, and then the generative content as well.
Robert Schoenfeld
Maybe I can jump into – thank you, Katherine. When we talk about data, I mean, you really want to go back to the basics. I know I’ve mentioned this before, but you really want to make sure that the data going into these models that’s being processed is at the highest quality possible. I think that’s where Talend really excels from that perspective, being able to create ecosystems that monitor your data assets, like choice, put it the date your data ecosystem, make sure you know where the data is coming from, make sure you can identify issues. And you can take action on that data as quickly as possible before getting into those models. And maybe having issues with that there’s been blunders with several companies over the years as far as compliance or you know, revenue, impact or reputation. So data quality, I think, is a foundational thing, when you get into AI in generative AI. From a Qlik standpoint, they’re at the forefront of leveraging generative AI to be able to ask a lot of questions and do that type of analytics. They’ve also been a player in AI via auto ml for a long time as well from that asset.
Carolyn
What data trend are you most looking forward to, as you look ahead to 2024? And I’ll start with you, Bruce.
Bruce Ottomano
I think general AI is number one, the promise that it can bring to the table is truly unlocking the value of the data, you know, knocking the knocking out of the park. And I think 2024 is a year that’s going to happen.
Troy Carmichael
I’m focusing a lot on making sure that the people who are overseeing me can see it quickly. So it doesn’t slow me down on fulfilling these things that people want. We do research in Gen AI and monetizing the data. So some of the things that I’m taking some time on right now is to just some of these basic kind of like, you know, gentlemen, this is a football type basics like here, it’s basically get the basics done. So we actually can use our creative energy on these things are coming next year.
Erik Trusler
I’m gonna echo back some things we’ve talked about. You know, Snowflake is an amazing technology and that phrase that we that was used here, don’t take your data to the apps, bring your apps to the to the data, has material implications for how businesses run and how quickly things can happen. So that that’s, that’s going to be a game changer continue to be a game changer in 2024.
Robert Schoenfeld
I might be drinking the Kool Aid a bit, but I’m really excited to see where Talend and Qlik are going from a roadmap and integration standpoint, I think it’s going to be a very compelling offering next year.
Katherine Toll
I want to talk about one buzzword that actually hasn’t come up so far. But as Eric was really alluding to it is the idea of composability. So we’re seeing a major movement away from having you know, these like, let’s call them mega solution, customer data platforms, and making your solutions more modular and plug and play in the context of plugging into the data warehouse in Snowflake. And I think that’s when I’m very excited to see how that continues to develop in, you know, helps, I think all companies kind of continue, continue to invest in data integrity, and that single source of truth, while still getting all the benefits of those composable platforms.
James McGeehan
So I want to go back to one word that Eric mentioned, that I think is crucial. So we know automation. He said it was number one, right. So to enable automation, especially in the financial services industry, which is highly regulated, the word is explainability. The ability to observe, go back audit that information, and then… LLM’s could be magic 100%. But someone getting turned down for a loan, and saying the LLM turned it down without the understanding the observability and explainability of the outcome – it isn’t possible. So I think this is why I’m excited about this panel, because all of the data quality that’s been discussed the quality of the domain expertise of the content, and then the activation of it is exactly why we’re collaborating this type of manner, with the foundation to understand that financial institutions have to be able to explain each and every step of the process and the outcome, not just to there to maintain trust of clients, but also to maintain trust of regulators. So I think that’s, that’s my view on this. And I think automation can and will happen. It may not happen at the speed at which people believe it has. But at the end of the day, this is where I think we’re accelerating very quickly to and looking forward to it.
Carolyn
Do you have any parting words you can leave with us as far as what your secret sauce is? How, how you have managed your data estate modernization and words of wisdom?
Troy Carmichael
Yeah, I think a lot of it was referred to throughout this conversation, but I think you need to think of yourself as the secret sauce in your company. The greatest asset that a company has is the culture you create around you. There’s a lot of stuff on this call that you can pick up from a technology standpoint. But I pay attention to all the stuff that was said here that was not technology based, because that’s what’s going to make a difference on your competitiveness to the people in your marketplace. So that that’s my secret sauce, I think – it’s going to be you.
