What Happens When Snowflake Meets Claude Inside a Community Bank?

This episode is really a Snowflake and Anthropic story. Roland Howard and Lee Easton trace how Identify became one of Snowflake's early banking partners back in 2022, years before the rest of the industry trusted the cloud, and how that same pattern is now repeating with Anthropic. Lee walks through Identify's new status as a Certified Anthropic Partner, part of the first wave certified when the Anthropic Partner Network opened up in June 2026, and explains why pairing Snowflake's infrastructure with Claude's ability to reason over financial data is what's finally putting real BI and AI tools within reach of small community banks that could never have afforded custom development.

They also dig into the provisioning risk nobody's talking about enough: employees spinning up personal Claude or Snowflake accounts outside of proper enterprise settings, and why that is the fastest way for a bank to end up in a data breach headline. The back half covers what AI readiness actually looks like (clean data, a real semantic model, and provisioning discipline), and why Lee sees his three-year moat as simply going deeper on Snowflake plus Anthropic than anyone else serving community banks.

Snowflake + Anthropic + Community Banks: The Core of This Episode

• Lee's Snowflake background predates Identify. He used it at ConocoPhillips to orchestrate oil and gas data before ever applying it to banking.
• Identify started using Snowflake with banks in 2022, when the industry still did not trust the cloud. That early bet is now paying off as FIS, Fiserv, and Jack Henry all roll out their own Snowflake and Google Cloud partnerships.
• Identify is a Snowflake Partner Network member and, as of June 2026, a Certified Anthropic Partner through the newly opened Anthropic Partner Network (Claude Partner Network).
• The stack is explicit: Snowflake handles data warehousing and infrastructure, Claude handles analytics, reporting, and natural language access to that data.
• Lee calls Claude "best in class for natural language querying right now, especially around financial services," and says in side-by-side testing against Copilot and GPT, banks consistently land on Claude.
• Provisioning matters as much as the technology. Lee walks through a real 2023 Snowflake breach caused by a bank skipping the Business Critical tier, and warns banks to lock down Claude the same way before employees start opening personal accounts.
• Lee's stated three-year moat is simply going deeper on Snowflake plus Anthropic than any other vendor serving community banks, packaged so it can be deployed with a single Terraform push.
• The clearest proof point: a sub-$300 million asset bank in Colorado with fewer than 50 employees now has data access that used to be reserved for banks in the $500 million to $5 billion range, entirely because of the Snowflake and Claude pairing.

Key Topics and Timestamps:

00:00 How did a mountain biker and computer engineer end up in oil and gas?

00:32 What mentor connection led Lee to Vast Bank?

01:41 What gap did legacy core systems reveal?

02:52 How does a sponsor bank reconcile crypto custody transactions?

03:35 What does Identify actually do?

04:37 Why does it matter that FIS, Fiserv, and Jack Henry are moving to Snowflake and Google Cloud?

06:05 Why did community banks distrust the cloud, and what changed?

09:23 How did core providers build trust with skeptical CIOs?

10:07 Why go after sponsor banks and BaaS players first?

11:44 What does Identify's new Certified Anthropic Partner status mean?

13:31 What happens when a bank skips proper provisioning?

14:47 What should a CIO do to get AI-ready?

17:37 Where is the gap widening between banks using AI well and those that aren't?

21:23 Clean data or taking action: where's the real bottleneck?

26:31 What is Lee's three-year moat?

28:14 How is a sub-$300 million community bank in Colorado using Snowflake and Claude?

32:04 Where can you find Lee and Identify?

Notable Quotes

"We're a big Snowflake partner, so we do a lot of the infrastructure and warehousing through Snowflake. And then we're now an Anthropic partner, so we do a lot of the analytics, reporting, and usage of data through Claude." — Lee Easton

"If the cores do it, the banks are eventually going to have to, or they're going to start trusting it." — Lee Easton

"Employees are going to be setting up Anthropic accounts, and your job is just to make sure they do it through an enterprise. If they go set up personal accounts and run it on their machine, you're at risk." — Lee Easton

"Our anthropic relationship is the same thing we have with Snowflake. We started this path with Snowflake in 2022, we got a big client, we did a big Snowflake build. The same thing is happening with Claude." — Lee Easton

"Claude is really good at financial data, and that's what these banks have. Why even try to compete when you have something like Claude, best in class right now for natural language querying." — Lee Easton

"The moat is just Anthropic. We want to be a strong Anthropic partner, packaging Anthropic with Snowflake and Azure or AWS. That tech stack, we've built it, we developed it in Terraform, we can roll it out by pushing a button." — Lee Easton

"The CEO of a small community bank can explore a data set about his entire business. That's a game changer, because community banks are small businesses." — Lee Easton

Companies and People Mentioned

Snowflake, Anthropic, and Claude are the throughline of this episode. Also mentioned: Identify (goidentify.com), Vast Bank, ConocoPhillips, FIS, Fiserv, Jack Henry, Google Cloud, BigQuery, Databricks, Chime, Cash App, Q2/Banno, Eric Sprink, Verafin, Sardine, Virtus AI, Tyler Brantley, All-In Podcast

Links

• Identify: goidentify.com
• Lee Easton on LinkedIn: https://www.linkedin.com/in/lee-easton-45073926/
• YouTube shorts
  • https://www.youtube.com/shorts/WS9pHgQh2c4
  • https://www.youtube.com/shorts/-yqooD34A_g

Want to work with Propagate Fintech? Fill out a contact form at www.propagatefintech.com

Transcript

SPEAKER_00 0:00

Lee, welcome to the show. I would love to understand how a mountain biker, computer engineer, ends up in banking data. Walk us through that story.

SPEAKER_01 0:10

Yeah, definitely less of a computer engineer than when I graduated college, much more of a mountain biker. But yeah, I I think a series of fortunate events led me into banking. My uh my career started in oil and gas. And I always knew I was gonna run my own business. It was just a matter of time and kind of building some financial stability in my personal life was important for me after college. But the sideline, you know, nights and weekends for five years, read the four-hour work week and read a lot of other entrepreneur books. When I finally left, five years of uh into my career at Conico Phillips, I got connected with a guy here in Oklahoma who was out on his own. He was consulting, he was an oil and gas guy, and he was a friend of mine's dad. And we worked together for maybe six months once I had left my full-time job to start my own business. And then he eventually became a CIO of Vast Bank, which is a it's a bank here in Tulsa. But they're at the time they were doing a lot of crazy cool productive stuff. Uh they had raised up raised a lot of capital, they were building their own core, so they were developing a side core. Yeah, I mean, so that got me in because he was a mentor. He brought us into the relationship with a bank. We got to develop a lot of technology with that bank and learn a lot with that.

SPEAKER_00 1:41

That's awesome. It's wild how far one relationship can take you down a path that turns into so many, so many different things. So I would imagine having that kind of front row seat to the happenings of a bank helped you identify a gap in the market. What did that gap look like to you?

SPEAKER_01 1:59

Yeah, it was it was uh a similar gap that I was seeing in oil and gas. It was, you know, big industry that have been around for a long time built on technology that just had not advanced. So legacy technology, community banks, community banks run on cores, right? Like in that those cores have been around for years and years. And so decades and decades. Decades. I mean, as long as I've been around some of these core systems, same technology. For sure. That was that was the when we started working with Bass Bank, that was the first and most challenging thing we experienced. And then that ended up being the gap that we today still aim to address.

SPEAKER_00 2:39

And so when I think about cores, I think of just like big, huge accounting systems. So I'm guessing this is was like a data play here helping Bass Bank turn all this big data into information.

SPEAKER_01 2:52

Yeah, that that specific use case was around payments. So that so technically they're they're coined as a sponsor bank, you know, in the world of Bass or banking as a service, they would they were looking to sponsor some programs. They had uh, I think a custodial relationship with one of the largest, you know, crypto wallets uh at the time. And so a lot of a lot of files needed to be transferred that contained transactional data. That was the primary challenge, was just reconciling which to which ledger for which system. And it was just a lot of legacy systems that did not speak in the same format. Okay.

SPEAKER_00 3:27

So would you be able to break down for those who may not be familiar with Identify what you guys do, the place that you guys sit at in the market? Yeah. Yeah.

SPEAKER_01 3:36

Identify started professional services to help banks leverage their data. We were building dashboards and Power BI and Tableau and all these different tools. That lasted about six to twelve months. The problem we kept running into was you know, there's no standard around the data, and every system accessing that data became challenging. So our mission shifted. You know, today our mission is just help banks solve fragmented data and help banks access their own data. Wild to say that because we know it's theirs, but they just have challenges getting to their own data. So we provide data warehouse, and now we provide BI tools and AI tools for the bank. We package that together. So identify is kind of a one-stop shop for a bank to get access to all their own data and then start using it. We're a big Snowflake partner, so we do a lot of the infrastructure warehousing through Snowflake, and then we're now an anthropic partner, so we do a lot of the like analytics reporting and usage of data through cloud.

SPEAKER_00 4:38

Got it. So you guys are partnering with Snowflake, doing a lot of work with them. I see that FIS, you know, Pfizer, they're also moving to Snowflake. What does that signal to the market from a wider perspective? What's happening here? What's the bigger narrative? Yeah.

SPEAKER_01 4:54

The big narrative is that all the banks are moving to the cloud. And so Snowflake was the platform we used when I was at Conco Phillips. So my my background in oil and gas is is, you know, it gave me some insight on a really great tool to orchestrate data. When I started this business and started working with banks, I felt like Snowflake was a bit overpowered for what we needed. What's wild is I, you know, I think once we got to a large enough client, large enough bank with large enough transactional data sets, we started using Snowflake with banks in about 2022. At that time, it was like the industry still it still didn't trust cloud. Like the one big bank we had brought on that year that we had this huge Snowflake deal with, they had never done anything in cloud. And a lot of the the concerns were around the security of cloud. And I'd go to Acquire Be Acquired conference every single year. And every year for about three or four years, it seemed like conversations around moving to the cloud got more productive, more productive, and more productive. To me, it was just mind-blowing that this whole industry still didn't trust it, like coming from that background.

SPEAKER_00 6:05

And when you say that they didn't they don't trust it, some people like to host it, some people like you to host it in the cloud. Yeah. Do you find that you know people just like they wanted to take it on, managing it themselves? I mean, it seems like such a such a hassle for community institutions to deal with keeping it on site and having the expertise to to handle it.

SPEAKER_01 6:26

Yeah. For those that were managing data to any extent, they preferred to have it locally hosted in a SQL environment with their own servers. For those that had maybe grown a bit at scale, they would host in a in a data center maybe locally or you know in their region. Um and then they would have an IT provider maybe that supported that data center, supported their servers in that data center. But for whatever reason, like to them, that was way more secure and way more efficient than leveraging Azure or AWS. That has completely changed. There are still banks that have a lot of on-prem infrastructure and data center-related infrastructure, but it's moving and it's I think there's a lot of momentum. I think what had to drive it was ultimately all these tools that are in the cloud. You know, I think the usage of, you know, new products, digital banking and new tools drove banks to look at cloud solutions so that they could integrate with those new tools and bring customers better products.

SPEAKER_00 7:28

How do you feel like the the perspective on on-prem versus in the cloud has changed for, let's say, community banks over the last five to seven years? Since I've only been in about five, I would say that the perspective has gotten a lot better.

SPEAKER_01 7:44

You you asked earlier about, you know, FIS and Pfizer moving to Snowflake. You know, I I I was placing a bet in 2022 that that banks would start moving to the cloud. The I think the first announcement was Jack Henry doing Google Cloud, and that was around 2022. And that was a good sign. There was a very slow adoption, and you know, I don't think it rolled out nearly as fast as they thought it would, but I think they started some stuff with Bano and Google Cloud and Google BigQuery. That was a great sign. Then we saw FiServe announce a relationship with Snowflake late 2022, that took a little bit of time for them to kind of turn that into a product and name it. And but I started seeing these signs that if the big core providers are making initiatives to move to these large cloud partners like a Snowflake or Google Cloud, then I was like, I was right on the money. I knew it would happen. If the cores do it, the banks are eventually going to have to or they're they're gonna start trusting it. And now FIS and FIServe are Snowflake partnered. FIS has a lot more detail in like they're using DPT and I think they're using some Databricks pieces, but they're you know, they're rolling this out to customers as like data management as a service, which I love. FiveServe's been working on it a little bit longer directly with Snowflake, Google Cloud, and Jack Henry. They've been doing it, I think, the longest. And so now every bank is aware of those those cloud providers. They're trusting of those cloud providers because their core providers also using it. So I think that we're past the hump, and now we're now we're actually trusting it. So here we had people six.

SPEAKER_00 9:25

So you had CIOs, let's say, who were like, you mean to tell me that we're gonna send our private NPI offshore to who knows where and trust, you know, like that whole skeptical narrative has been replaced with, well, I don't want to hire people here to manage it when I can just offload it to the cloud, the trust factor is there. Cool.

SPEAKER_01 9:47

Yeah, and that's what's been great for our business. We were a bit early in what we were doing, and I knew it would just take a little bit of time. So I was trying to find those those banks that were already showing signs of I don't know what to say, a lot of innovation. If they were doing a lot of innovation, punching above their weight, maybe. Punching above their weight. So most of the banks we started working with early on were sponsor banks. So these were banks that were partnered with like China and Cash App.

SPEAKER_00 10:14

You're talking like Challenger banks.

SPEAKER_01 10:15

Yeah. Well, they're they're they're banks, they were working with Challenger banks. They were working like Got it.

SPEAKER_00 10:21

I see the coastal communities of the world. Bingo.

SPEAKER_01 10:24

Yes.

SPEAKER_00 10:24

Okay, that was that was one of our early clients as well. Oh, cool. Eric Sprink, man, what a guy. Yeah, yeah. He's a great dude.

SPEAKER_01 10:33

So those are the banks we went after first because we knew one, we knew they had a much larger data set, so they had a data problem. We knew that you know that data problem likely couldn't be solved with locally hosted services and infrastructure. So it was a it was a great starting point, 2022, 2023. Now the rest of the market has kind of caught up, and your traditional, you know, $350 million bank in small town of Oklahoma is actually gonna trust the cloud and use cloud services.

SPEAKER_00 11:02

Awesome. Well, I don't I I don't think the pendulum is gonna be swinging the other direction any anytime soon here. So you talked about Anthropic. I'd love to unpack what you have going on with anthropic here. I know that one of the big concerns around utilizing AI is that you are, you know, going to be exposing sensitive information to who knows who, where, how it's gonna be. You know, like the whole AI world is still so, so bleeding edge that there is a lot of hesitation around leaning into the technology. What's going on with you guys and anthropic, and how does that tie into what you guys are doing for banks? Yeah.

SPEAKER_01 11:45

Our our anthropic relationship is very it's the same thing we have with Snowflake. They've got a partner network. Theirs is just getting rolled out. I think this month, June 2026, was like the floodgates opened for anthropic certified partners. So we we started this path with Snowflake in 2022. We had got a big client, we did a big Snowflake build. So we we explored, you know, what's called SPN Snowflake Partner Network. We got certified, you have to do a series of training and uh testing, and then you get certified, you pay a fee, you get into the partner program. And that's been awesome. So they send us a lot of leads, which is great. Same thing is happening with yeah, same thing happening with Cloud. Um, they've got CPN, the Cloud Partner Network. You get certified, you go through, uh, I think at least 10 architects have to be fully trained. And once you get certified, they're same thing. You're gonna qualify in like an industry segment. So for us, it'll be financial services, and any bank that goes to set up Cloud on their own that needs a certified partner could come to us. On the provisioning side, I think that is probably the biggest call out and the biggest concern that I have. If if banks aren't at least setting up Anthropic with a enterprise edition and all the default security, then we're gonna have problems. And it's gonna happen. I think it's a matter of time until there's some sort of bank that that hits the headlines for data breach or whatever, because they just didn't provision it right. This actually happened with Snowflake in 2023. So Snowflake has a pricing tier called business critical. Business critical is what we tell every bank to sign up for. It's $10,000 a year minimum, but by default, all the security settings are correctly provisioned, access settings are correctly provisioned. Well, some bank had an IT employee that was dabbling with Snowflake, and instead of they didn't want to, they didn't want to go in and spend $10,000 and do business critical, so they signed up on the website using a what's called an on-demand or a paper usage account, which is the same thing if you go set up CLOD and use a personal account. Oh no.

SPEAKER_00 13:59

Nothing, nothing is part of personal account at the enterprise level. What can go wrong? What can go wrong? And a password got leaked.

SPEAKER_01 14:07

Oh no. Oh gosh. So that I'm fearful. I think you know, IT leaders and banks need to get ahead of this right now. It's already late if you haven't. Um, employees are going to be setting up anthropic accounts, and your job is just to make sure they do it through an enterprise. If it because if they go set up personal accounts and run it on desktop, on their machine, you're at risk. What was that?

SPEAKER_00 14:31

Let me uh try out on a topic that I think is interesting right now, which is AI readiness. If you've got, let's say, 30 seconds with a CIO, what are the top three things you would tell them to do to be AI ready at their institution? Yeah, that's a cool question.

SPEAKER_01 14:48

AI readiness for from my perspective, being a data provider, is to have clean data. Now, I think from a like if I were a security company, it'd be a whole different step one. But from our perspective, to make any of these tools really useful, you have to have a good data model. They call it a semantic model, but you need to have you need to have a semantic model developed. That way, when you're querying through the tool, like ChatGPT or Cloud, you're getting accurate results and you're not getting results that are maybe pieced together.

SPEAKER_00 15:23

And so you're talking about hey, what one was how many uh accounts were opened by people walking in the door at the branch on Main Street? Exactly.

SPEAKER_01 15:32

Yeah, what's our deposit runoff, you know, across all of our digital banking customers? Stuff like that. You know, like it will try, and if it has access to the data, it will try harder. But you you want to develop that model first and then produce that model. I will say, I think from a CIO's perspective, though, probably the most important thing is the secure designs. Like to get AI ready, your organization, whoever's gonna use a tool, probably needs to go through some basic training. I would say work work on, and this is I would say this to not just a bank. This is just to a business. Like as a business, using AI, provisioning, and yeah, provisioning is probably the most important thing because if you're not careful, you will have mountains of technical debt. People will be creating chats and projects and agents and skills in 20 in 20 different silos. And then the problem that we aim to solve with the tool only gets worse. Like we only deepen our problem around silos, and that needs to be addressed out the gate. We've learned we've learned that lesson in my own company, and now we have standards like how do what do we name projects? All the projects are named with a naming convention, skills, skills are named with naming conventions, organization-wide settings for how we use skills, all that stuff needs to be defined.

SPEAKER_00 16:55

Usage and consumption. Everything. It is such a huge topic, and we're all new at it. Everybody's new at it right now. Yes. Like when to use Opus versus Sonic.

SPEAKER_01 17:04

Like those things have to happen, and then you create the data model on the back end with your data warehouse, and then you start using it.

SPEAKER_00 17:13

Yeah. People using me methos to see what the weather's gonna be. Exactly. Exactly. It will happen. It will happen.

SPEAKER_01 17:19

You give people a tool with no guidance and just ready, just be ready for the worst.

SPEAKER_00 17:25

Where where are you seeing the delta opening right now between banks who are doing these things and going through the motions here and those that are not or not ready to? Man, I see. Are they like are they bagging more new accounts? Are they are they better at capturing Gen Z customers? Are they underwriting better? Like what I'm curious what you're seeing as you're because you guys are working with such a big cross-section of the market.

SPEAKER_01 17:51

Yeah, I think the problem being solved first and foremost, that's been a an attractive data use case for the last few years is customer insights. And in more specific, like marketing related strategy. So out of the gate with any AI tool, banks are just trying to understand like what is the breakdown of our customer portfolio? How many of them are checking savings, cards, CDs, loans? That analysis alone takes a lot of work with all the different tools and systems a bank might be using. Being able to do that in a quick query and break all that out, so out the gate. That's typically where we see it go. What they do with that information is kind of different. Like culturally, if they're a if they're a very proactive organization, they can start assigning out marketing tasks and sales tasks and relationship building tasks. They can take, you know, branch level activity and focus on like changing roles or leadership. Some of the banks look at it as like an operating efficiency thing. Like, can we hire less in certain areas because now we don't need to spend as much time handholding a customer for onboarding, whatever it might be.

SPEAKER_00 19:05

Yeah, I heard an interesting conversation with the all-in podcast. There's a lot of critiques of this podcast. But uh, you know, they were talking about how there's so much fear-mongering about the disruption to employment from AI. These guys are like classic kind of capitalists. They don't see that being the case. They really see it as being an enabler of further extraction of great ideas from human beings. Really like an imagination problem that we have in terms of what AI can do to support what they're doing in their in their day-to-day. How do you see AI supercharging the output of what a bank offers to its customers?

SPEAKER_01 19:43

Man, it's a big transitional season for banking when AI is adopted. All the all the potential use cases and outputs that they had for their data become realized almost within that first week, maybe that second week when they start using an AI tool. It's wild to watch. You know, every uh every bank wants to do to know more about their customer, and it's just always been so locked down. And this is like, this is the thing that kind of like freezing, right? And now it's and now they don't know they don't have to know SQL, they don't have to know how to like get in there and code to understand more detail or like clicking.

SPEAKER_00 20:20

That's the semantic component now, is it enabling more people to join the party. Yes. It's kind of wild.

SPEAKER_01 20:27

Like the CEO of a small community bank can explore a data set at you know, about his entire business, his entire organization. Amazing. And that that's a game changer for because you know, community banks are small businesses. And that's that's big.

SPEAKER_00 20:46

So AI allows such a greater uptake of information into the institution. And it's one thing to be able to streamline all that data, but it really is only powerful if there is a commensurate output of activities, right? Sending out pre-approved offers, whatever that may be. Where do you see the bigger challenge being bringing the data in in a cleaner, smarter way? Or once that's done, the output, the getting the pre-approved messages out the door, right? Like doing something with the information, the tactical happening.

SPEAKER_01 21:23

Yeah, the the the clean and unified data, that's the problem we aim to solve. We think is going to get easier and easier because of the AI tools. You know, we can we can start building skills for every single data set that we bring in. And then those skills are repeatable and reusable. And like I expect banks to be pretty good at eventually self serving on their own data sets. We tell banks that if they're exploring AI tools to do their own testing, do an extract of card transactional data or ledger data for 30 days, you know, delete all the rows and columns that might indicate. PII, so customer names and addresses and stuff. So scrub your data, but then load that file of 30-day activity into Copilot, load it into GPT, load it into Claude, use the same prompt side by side, look at your outputs, determine what you know what looks best to you, what tool brings the most value. It almost always points to Claude. But in that exercise, we're asking them to take an extract of however they can get it without any cleaning. And I think it's only just going to get better and better and better at recognizing patterns and trends and types of data sets that, like, I'm not going to say we're going to identify we'll be out a job. We will just provide the tools in a package format so the banks can you know run these kind of exercises themselves. My goal is to focus more on more on getting banks to becoming self-servable. Like building versus buying, I want to I want to equip them so they can start building. On the other end, like the action items, the things that they need to go do once they get the result from the query, that's outside of scope for us. We can share experience and feedback and we can talk about what other clients of ours are doing and mo help motivate maybe ideas and provoke thoughts. But like I think there's even a space there for consultants to come in, like maybe even more opportunities for consultants to come in and guide some of that strategic work if the bank isn't capable of doing that themselves.

SPEAKER_00 23:28

That makes sense. I was a former partner executive in fintech, and so my brain is kind of wired like this, but you know, you guys are at one end of the spectrum where you're cleaning the data, you're getting the data prepped and ready for kind of bigger, bigger and better things. You also then have fintech players who are at the other end, right? That are like ready for you to hand them the ball to run down the field and open up those accounts, drive adoption, increase deposits, you know, whatever it may be, like the Virtus AIs of the world. You know, we had Tyler Brantley on the podcast a while back. And, you know, these guys would love to have clean streamlined data because it allows them to do all of their kind of business development growth-related actions with the clean data. Man, yeah.

SPEAKER_01 24:14

So you're you're you're hitting on a couple things that we are moving into that we're we're like excited about, I guess. Optimistic, I think, is our focus right now. Just partnerships and distribution channels. If we can, if we can provide the platform for banks to then start querying their own data and using their own data, they can also become more nimble. So they can they can go find best in class vendors. They don't necessarily have to worry as much about core integration anymore. They've got a data warehouse, they've got an infrastructure that they can build on top of, which would hopefully empower partners that they bring in too. Because then the partner might be on Snowflake as well, the partner might be on AWS, and you've got cloud connectivity. Kind of that that's back to like the cloud conversation. If all the banks start moving to the cloud, we can start taking advantage of cloud connected services, and we don't have to use APIs. Like we will probably ship files for as long as we have to in this industry. We'll skip APIs. Like there are companies that use APIs, but the natural evolution is to move to cloud connected services where you're privatizing and sharing like an AWS bucket and or a, you know, and Snowflake, Snowflake has that technology, Google has a technology, they they're all doing it. It's just not a lot of people have moved to it yet because you still have parties that are on print. But once we get cloud native, all these partners, all these vendors, we would like to provide some sort of marketplace for our banks. We would need to go build out those relationships with those partners. But we do this a lot right now with card issuers and fraud companies. Those are like the two, the two biggest vendors in the mix that we work with is fraud, like transaction monitoring tools. And if we if they have a file format that they need to ingest from a bank, we typically build that into our mapping. When we start with a bank, let's say they're on Verafin or Sardini or whatever, we will provide that mapping so they can go partner with that fraud vendor and easily connect and pass data along.

SPEAKER_00 26:22

Nice. That's awesome. Feel free to plead the fifth on this question, Lee. But as a founder, how are you thinking about your moat with the advent of AI here? Yeah.

SPEAKER_01 26:32

I think like three-year plan for us, the moat is just anthropic. We we want to be a strong anthropic partner, packaging anthropic with Snowflake in Azure, AWS. Like that tech stack is we've built it, we developed it in Terraform, we can roll it out by pushing a button, and then we have a bunch of other buttons we can press for DCI or CSI or COCC or F FIS, whatever. So all the like our goal is to create emotes around accessibility and like ease of integration for the data stack that a a bank, a community bank would have. Sure, we'll we'll have competition in that space. And of course, I hope to see banks building this themselves over time. But when a company like ours, the every bank we work with, we typically come across one or two new data sources, that becomes a skill. And that skill we then train into our model, and then that that creates value for the next bank we work with. So I kind of I see the flywheel starting to gain momentum. Every relationship we add creates value for the next two or three, and that becomes our mode.

SPEAKER_00 27:41

Okay. So essentially deepening the productization of what you do through your through your partnerships, through distribution channels, making it kind of more plug and play for your clients. Yep, exactly. Awesome. Uh so Lee, I understand that you guys are working with Anthropic, the newly minted kind of winner of the AI race, it seems, surpassing open AI for an institution who, let's say, is a little guy looking for a strategy to punch above their weight class. How do they work with you and Anthropic to do this?

SPEAKER_01 28:14

Yeah, we we just recently signed this exact bank. We've got a customer we just brought on out of Colorado. This bank is less than 300 million in assets and less than 50 employees. Like, never before would this bank be able to interact with all of their core data, all their LMS data in such a way. Like financially, they wouldn't really be able to hire, identify to build them anything custom in Power BI or Tableau just because of the cost, uh, engineering hours. But with a with a tool like Cloud and our ability to get the data cleaned and moved to Cloud and securely connected, we're gonna let them do a lot of the development around what they want to see out of that data set. So we're kind of packaging this and empowering them. What what this means is, I mean, this is novel. This is this is call it, you know, 2,000 FIs in the US are probably right in the same category. They're on a really small core provider, they're not on one of the big three. And there's a lot of banks out there that that just want to unlock that core data. And and I get that these smaller cores probably want to roll out some AI tools. Like it makes sense that they would, but why even try to compete when you have something like Claude and best in class, right, for natural language querying right now, especially around financial services. Like Claude is really good at financial data, and that's what these banks have. And so this is gonna be a huge win for them because it's gonna put them way above their peers in how they use that data and what they can use that data for. We're still in implementation, so a month or so from now, we're gonna look back on you know what use cases they were able to uncover with this data, like comparing local markets, local uh lending peers, customer segment groups that they maybe didn't know they had. All that insight uh they're gonna work through over the next month or two, and then we're gonna re-evaluate. But that level of research would never have been possible. One, because of the finances of this smaller bank, you know, hiring uh an agency at professional service rates would just be too costly. And then also the limitations just from the core provider with no, you know, deep research or analytics tools. And it's all unlocked because of Claude.

SPEAKER_00 30:42

Amazing. So your collaboration with Claude has essentially brought, you know, tier one capabilities way down market to true community institutions. Absolutely.

SPEAKER_01 30:55

And that's that's I think the most pivotal thing for identifying 2026 is we've opened up our total addressable market where we were normally working with banks from 500 million to about 5 billion in asset sizes. Now we can tap into these banks that are 50 million in assets to 500 million in assets with 10 employees. Like we have a we have a really strong prospect right now that has 15 employees. It's a little, you know, $50 million bank. And uh, you know, they're on one of the smaller core providers that doesn't have a ton of reporting and analytics tools, but we know we can get a data extract. And so our partnership, Snowflake and Cloud, enables us to clean that data, stage that data, securely connect all the tools, and then roll that interface directly through Cloud to the bank.

SPEAKER_00 31:44

That's awesome. I love that. Well, I would love to stay in touch with you guys after this client goes live and to just see like what are you doing now with this new superpowered capability come, let's say, you know, end of July, maybe early mid-August. Yeah. Superpower is exactly what it feels like. That's awesome. Well, Lee, thanks so much for coming on the show, man. Where can people follow what you guys are doing? Yeah. Um we post a lot on LinkedIn.

SPEAKER_01 32:07

So look up my my own LinkedIn page for Lee Easton. Look up identify on LinkedIn. Um, and then our website, goidentify.com. There's there's a lot of content we like to put on that. And excited to be here. Thank you for your your time and invitation today, Roland. Absolutely. Thanks for being here.