AI Early Delinquency 1-29 DPD Gap

Alrighty. I will start the webinar for you in three, two, one.

Hello. Good morning, everybody. Thank you for joining. I know we’re a minute before the hour here, so we are gonna give a couple minutes for everybody to log in, and, we’ll get this webinar kicked off.

Really, you know, talking about how Interface here is is challenging and pushing the narrative forward on on closing delinquencies and supporting you and your teams and your consumers. So, again, just a couple minutes. My name is Michael Azevedo. I know some of you might have been expecting Sri, and I’m sure he’s watching somewhere, but I’m excited to be here and take you through this today. So, again, couple minutes, and we’ll get this kicked off.

For those of you who’ve joined over the last couple seconds and you’re wondering why am I sitting there so quietly, we’re just giving everybody a couple extra minutes here to get logged in before we kick off the webinar and, really, you know, the one to twenty nine days past due gap and how we can, you know, close those early delinquencies and help support you and your consumers in a meaningful way. So another thirty seconds or so, and I’ll I’ll continue to I’ll go ahead and push forward Just another couple seconds here.

Alright. Getting the bat signal. Says, I think we’re settled. We’re ready to start moving. So my name, Michael Azabile, and I’ll do an introduction to myself here in just a couple minutes and and talk a little bit about myself.

You were probably expecting Shri. I know, again, I mentioned this earlier. He’s probably watching.

But I get the the honor and privilege here to be with you this morning and and really talk about how Interface is challenging the existing processes in place and putting, you know, tools and resources specifically around AI agents and and and and supporting your your teams to ultimately support your consumers and improve their well-being and their lives. And so the one to twenty nine days past due gap and how AI closes early delinquency when your team can’t, that’s what we’re here to discuss today. We’ve got a lot to cover.

I will take questions and answers if we’ve got time at the end. My ask, please throw it into the chat. I’ve got a team in the background who’s promised me they’re gonna be on the spot, and they’re gonna be either answering it within the chat for you or at the end. With time, I’ll try and do as many live as I can. But, of course, as always, the point of these, if you are at any point ever interested or we don’t get to your questions, we will do an outreach after the fact. And I’ve got a team of people who would be more than excited, right, to reach out and and discuss this in in great depth and really see if this is something that can support you and yours.

So let’s let’s talk about what we’re gonna do today. Couple things. So a little bit about interface. Some of you are probably aware of us. Maybe you’re existing customers. Maybe you’re exploring a relationship with us, or maybe you’ve never heard of us. But, anyways, I’m gonna take you through who we are, right, and, you know, the value proposition that we bring to the table and really what we’re driving the industry forward here.

The main point of today, why collections for AI for early delinquency are important. What are those business cases? How are we operating? What does that look like?

What does it feel like? What will it do for you? So we’ll cover that in great depth. We’re actually gonna show you this thing.

So I’ve got a wonderful team in the background. They’re gonna hop on. We’re gonna get into, you know, what the platform looks like from the employee standpoint, the tools and resources you have.

But, also, what does it what does it feel like for the consumer? I think it’s important. Right? Where has AI taken us in this journey of how consumers can interact, and and what’s the value proposition, the efficiency that creates?

And, of course, as mentioned, if we’ve got some time, we’re gonna do some QA, talk about next steps, and, ultimately, we would love to have a deeper conversation with our teams. So with that, we’re gonna do a little shock and awe. We’re gonna start with a poll, and this is gonna help me really frame today’s conversations based on where you’re all at. And there’s these questions, you know, progress through your journey around AI.

So you’re seeing a couple pop up. You know, the first one is, you know, what is the single most what’s the biggest pain point you’re feeling today when it comes to collections? Right? Let’s set the tone there. Where where are you trying to solve?

Right? I’ll go ahead and read the second one because I think it’s prompted on your screen as well, but we’re gonna give you a minute or two here to answer these. I know they’re lengthy, and you’ve got lots of options. We love engagement. It helps me. It’s gonna help us put this into context for you. But that second question, what core platform are you using to service, you know, these collections at this point?

So, again, we’re gonna give you a minute there to answer these, and then we’ll progress. It’s got two more, I promise, and then we’ll get into some of the other content.

Alright, team. I think we’ve got little roughly about a minute. So at some point, let’s trigger the last two questions here for them.

So some interesting results here.

The single’s biggest pain point right now, great responses, rising delinquencies, contact rates. We got some interesting stats for you here that we’re gonna address. My apologies. I realized I just didn’t scroll down. So you probably saw questions three and four, and we’ll talk about them here in just a second.

What are the collection systems?

Kubo, Temenos, pretty standard correlation, probably related with core activity. Makes a ton of sense.

Do you currently use SMS in texting? Yes. UltraP, big product out there. Right?

No. Also a response there. And texting’s interesting, and we’re gonna cover this as we get into that portion of it. Right?

Like, why is texting important? Should I be texting? These are probably questions you’re asking yourself. I will kind of relate it and correlate it to generational conversation.

Right? Some generations prefer texting, and we’ll get into that a little bit more. And then how soon do you plan to deploy AI into your collection workflow? So we’ve only got one who already has.

We’ve got people who are thinking the next six to twelve months and then a good portion who’s not planning for it. And, you know, that’s curious.

And maybe through today, gets you a little bit more excited, but we’d love to understand that thought process too. So even if it’s not, hey. I don’t wanna I’m not interested, but, like, maybe you wanna talk with someone on the team in a little bit more depth about why and why you’re thinking that way and why we’re positioning the way we are, we’d love to have those conversations as well.

So I appreciate this context. It really helps me frame up how, you know, the today’s conversation’s gonna go. A little bit about me. I am not Sri.

I am Michael. I am your presenter today. I’m actually the VP head of sales here at Interface. I am relatively new.

However, I am not new to industry, and and hopefully, some of you recognize me. If not, I have been spent the last five years really within a platform that engages consumers using credit and credit data. And, really, my passion and my background is from banking industry. So I spent a good portion of my first part of the career really in different banking and lending groups, supporting product as a lender, running teams.

And I felt a passion to challenge the status quo. I always sat in rooms and it you know, the answers we would always get were, that’s just how we do it. And that really always irked me in a way. And so I jumped into fintech about five years ago and and helped scale a company.

I then met Sri, and Sri convinced me that this was the future. And I’m here, and I’m excited to really challenge how the world will look, you know, over the next, you know, one year, three years, five years, and it’s moving fast. With the with the way that progression of technology is happening, it’s almost impossible to plan for three years. We’re thinking in weeks, months at this point. And so that’s where my background lies. I’m excited to talk a little bit about smart collections. I think this is an interesting and unique entry point for a lot of financial institutions to accelerate how they’re using AI, how to improve concern consumer journeys, and I hope you can see that through today’s presentation, of course.

Little background. Right? And, again, probably got a range of people who have experience with Interface or maybe you haven’t and you just got our marketing team is just gurus and got you onto this. And I’m excited to talk a little bit about why I’m so excited of where this company is going.

Right? Interface, you know, been around since twenty nineteen, but a little before that really is is a a a pilot, as a as a concept, and really founded by two brilliant minds. Right? You’ve got Sri, who is our founder and CEO.

He’s built world class products. You know, he’s an ex banker. Microsoft has a has a real drive to improve the financial institution’s connection to their consumer. And in doing so, leveraging technology to keep that human interaction, right, to keep that that cohesiveness, that connectedness with your consumer, but deliver world class solutions that keep you up with today’s modern banks and credit unions.

Right? Really driving effectiveness, efficiency, and deploying products that keep you competitive in today’s market. I’m sure a lot of you are thinking about the market and, you know, where you sit from a generational standpoint in your book and where you wanna sit and how do I get there. And these are the tools and resources that we’re developing here that help you do that.

And then alternatively, we have Bruce, and Bruce is that mastermind man behind the computer. Some of us think he may be real, may may not be real, but he is the brains behind the technology and and has scaled multiple companies using enterprise level artificial intelligence, and we’re excited about the future of where this is going. Right? We’ve got over a hundred credit unions and banks today who trust us in a in a capacity of ways, and I’ll talk about that.

So we’re not new to the market. You know? We are unique in how we’ve approached the market.

And it’s bold, but true. We are the most capable AI bank GPT platform with agentic solutions for credit unions and community banks. You know, if you talk, and I’m sure you have with Sri in the past, right, about his vision, it’s solving a simple problem and doing it within a vertical that absolutely needs it. So we’re not looking to expand into health care or retail or or or or or, you know, our breadth of reach, we wanna do this right in this specific industry, which needs it drastically.

Our our, you know, our community banks and credit union partners, you’re trying to solve a lot of problems. And you’re playing against an almost an unfair field, right, with some of the largest financial institutions in this country having almost what seems endless budgets to develop technologies And solutions like Interface where we’re challenging that, where we’re developing on your behalf are the tools and resources that will keep you at the forefront of your consumers’ minds and really driving those relationships forward for today and tomorrow. Right? We gotta start thinking about those generations coming up.

How are we gonna interact with them?

A lot of you probably know us through the tools and resources that are already in in market. Right? We’ve got a voice AI. We’ve got chat AI.

We’ve got employee AI. We’ve got an agentic smart collection, which we’re gonna talk about today. But really where, you know, my heartstrings were pulled when I talked to this company and where we’re going, it’s beyond that. It’s where does the future take us.

Right? And we’ve developed a native agentic platform to really be able to scale quickly and effectively. Right? We believe we’ve got an edge to be able to pursue this in the near future and really challenge how consumers interact with their financial institutions.

So things like an AgenTic online banking or an agentic contact center are not far off. They’re actually in the works. Right? We’re moving in this direction.

And what does that mean? It means it uses, obviously, fraud protection AI with the with BankGPT, our platform at the base of it, you layer on fraud prevention and intelligent banking and and the connections to the systems that are important today, right, in order to understand consumers and then place these agents in a in a way which they interact.

And then as we talk about this next, with an agentic model behind it, it’s self learning and it’s self developing. And what it’s doing is creating better and better interactions every single time.

And so as we look at AI and and everybody’s in a different place with how they’re using AI today. Our financial institution partners, I hear a range of, yeah. We’re dabbling. Yeah.

We use some LLMs. I some controls in place, and some people are masterminds already. They’ve created agents that are operating within their organizations. Right?

But it it could mean so many things to so many different people. Right? At the base, you had l NLP bots where these simple, you know, limited language understanding, highly scripted bots that that are in the market and people will call AI. Right?

And that scales up from partial generative to fully generative where, you know, capability comes a little bit more. But, really, where today, the news cycle keeps us focused and what most people are now thinking AI is is the AgenTik AI. Right? This is a self learning, self developing agent that is autonomous, that it it’s it’s superior in language.

It understands, but not only does it understand what it’s being asked, it can do something about it and render in ways that engage that consumer and drive them in further. Right? Like and it give you the ability to really enhance that interaction and the efficiency behind it. So as I say those two things, right, you’re talking about a solution that drives results.

Right? So revenue driving results, but also operational efficiency. So it’s saving and and and driving in in both directions, and it creates such a massive tool that can be so important in your journey and where you wanna take your organizations. There are a couple things, though, as you talk about AgenTic AI, and and, you know, I hear this in the market.

There’s, you know, there’s the mistrust, the fear that gets thrown out there. There’s the uncertainty of it. But, really, it’s there’s a way to do it appropriately. And when you have people like Sri and Bruce in the background who’ve been in this for so long and have started putting guardrails in place to make sure that it’s operating effectively, there are ways to make it safe, and there are ways that it can be effective and efficient for you.

Right? At the core of what we’re trying to solve, though, is member experience, customer experience.

We need to provide efficient tools that meet the consumer where they’re at and deliver results in that forum. Right? Not only understand the pain point or the ask or the need, but go beyond that. Deliver advice. Deliver solution. Deliver a way in they can achieve that. And so that’s where that experience is critical as we think about AgenTic AM.

The most fearful tactic that you hear in the market today is this hallucination. Right? Like, it’ll run off. It’ll do its own thing if you don’t keep your your arms around it.

It’s scary. Right? And while you if you were to just connect Claude into your digital site and let your consumers interact with it, I would agree with that. That’s not how the world works, and that’s not how this world works.

Right? And so there’s a lot of infrastructure in place that’s been developed over the years that we continue to focus and prioritize, right, to make sure that the guardrails, the checkpoints, everything is in place for your consumers to have a safe and intelligent interaction. Right? We we only source the information from your source points, which allows us to look at policies, decision criteria.

Whatever the whatever we’re trying to facilitate in that interaction can be very much guard gated and and controlled, and that gives us a zero hallucination, you know, policy and striving point.

The other thing that, you know and I mentioned this, it’s it’s ease of training and maintenance. So as we think about the consumer interaction, on the opposite side of it, there’s probably someone at your on your team, right, who needs this resource and needs to understand, you know, how can I operate with an AI to make this efficient? So while, you know, AI today can probably solve a lot of the interactions, it can solve everything, especially with the guardrails in place. There are certain things in which you’ll say, like, I can’t have an agent make this decision.

A human does. And it’s self learning and smart enough to know when those are and to pull in humans. But at the same time, as we develop the consumer interaction, we have to develop the employee side of it. Right?

When do we pull them in? What information do they need when they get pulled in? How can they then leverage that information and act quickly? Right?

Creating an efficiency and a trust with that consumer at each touch point.

And so a little bit about how the platform operates today. Right? We have our members, and a lot of what I’ve just spoke about is gonna kind of be reconfirmed over the next slide or two as we get into the smart collections component. And I promise we’re getting and I promise you’re gonna see it. And I promise we’re gonna try and answer questions. So just level with me for a second. I’m I’ve got a lot of passion for where we’re going with this.

Our members have different ways in which they interact with us.

You know? And and where we sit in the digital world, like, the obvious is. Right? Like, you got phone.

You’ve got website. You’ve got mobile app. You got text. You got online banking and mobile banking.

Right? Like, these are the ways in which consumers are touching you digitally. And we’ve created a layer, essentially, right, a system of interactive intelligence, which is communicating with these consumers, these individuals, gathering information, automating, driving value where possible, being personalized, and obviously with security behind it. Right?

In the center of that interaction, right, where we’re attached is your systems. Right? It’s what you use today. Right?

Like, this is the world in which we operate today. We could get into a concept of what the future looks like, and maybe this changes, and I Sri could go on for hours here too. But for today, this is how this world looks. Right?

We’re connected to your LOS. However, your your CRM, your knowledge systems, your core, your debit cards, right, your RPA systems. Like, these are the systems of record. And so the way we really gatekeep it and we keep that zero hallucination concept in check is this is the systems in which we’re gathering the information.

If someone asks us about the weather, we would have no idea what the the the agent would have no idea how to respond to it because it’s not within the realm of intelligence that it has. Right? So at the center is your information. And then on the other side of it, right, to drive the efficiency, the operational efficiency behind using tools and resources is that same level of attention and interaction with your employees.

Making sure they have a tool that that allows them to access everything that they possibly need. They don’t have to engage five systems to answer a question when when it comes through, we’re prompting them with, here’s the questions. Here’s the thread. Here’s the answer from your policy, and here’s the exception you need to make.

Do you wanna make that? And engaging our tool to be able to do so.

Right? Again, going back to this concept of driving value, so driving outcomes for the consumer, but on the back end saving from an operational standpoint and efficiency standpoint.

Now some differentiation. Right? Because we talked about the different levels of of agents and bots and and systems in which AI works. Right?

But how does how are we uniquely different, and why do we feel like we’ve got this this exciting approach to the market? Right? A typical vendor today sits just on top of. Right?

They’re not orchestrating. They’re not creating. They’re not embedded. It’s not AI native. Right? It’s sitting on top, and it’s doing its best.

And it’s probably, on average, you know, using AI to about on thirty percent of member interactions. Right? And it’s it’s it’s the the way of how it was, right, where interface has now taken a different approach and created this accelerated solution is it’s AI native, built from the ground up using an orchestration layer, right, attached to all those core systems, being able to drive almost a hundred percent of member interactions. Right?

Not just when the consumer’s interacting with whatever forum they’ve entered, whether that be chat or voice or digital. Right? Like but also if a con if a human on the back end on your side is involved, it’s also supporting that human to be effective and efficient and support them in a way that drives an outcome that’s beneficial. Right?

Beneficial for the consumer and beneficial for you.

And the world is ever changing. As mentioned previously, like, you can’t think a year out anymore. You can’t think three years out anymore. You gotta think about next quarter.

You gotta think in in two months. Right? There’s new versions of Cloud being released almost every thirty days at this point, it feels like. Right?

And and the world of AI is is moving fast. Right? Today, what does a unified approach looks like? It’s one AI brain.

So it’s this concept of having all the consumer interaction that we have, right, using the orchestration layers, so those connections, and delivering on outcomes, right, or creating efficiency. It can upsell, cross sell. It can, you know, be a computer. Right?

Like, it it can replace online and mobile banking in a way where it’s interacting with your consumers. But the future and really the value that this world has started moving towards is is autonomy. The ability for agents to become self learning to to to improve. Right?

And and when I say that, it it’s in a lot a lot of different layers. Like, every time a consumer interacts, if you have an in a powerful agentic agent in place, it’s learning from the previous interaction. And then the next time it’s personalizing and curating and doing better because it’s learned from that. And it’s once that comes into play and and this world unfolds, right, every time you engage with that consumer, you’re gonna drive meaningful outcome for them to help drive them through financial goals and have a tool, right, that allows you to do so.

And if we go back to the, you know, the the sentiment of how do we keep up with the banks of today, like the the major players that are spending billions of dollars a year to to to, a, gather consumers and, b, be top of tech, it’s using solutions like this. It’s being at the the forefront of how and what is possible with now how technology works.

And so I appreciate you letting me express my my excitement and enthusiasm for where, you know, the world lives today in which we can serve consumers tomorrow. Right? Like, how do we bridge that gap? And I go back to my own story of, like, just how we’ve always done it.

Well, that that’s not gonna work anymore because now the world’s changed. And we have to change with it because our consumers are gonna demand that. Especially, again, going back to the comment of you’re probably sitting there thinking, how do I attract new generations? And and this is the the expectation.

We’ve already surpassed the gap or the the we’ve eclipsed the the point where Gen z and and even millennials are now using chat and Claude for financial advice.

More are doing that than not.

So how do we now play in that space? That’s that’s critical. And for today, we’re gonna focus on a on a very simple, right, and problematic use case. Right?

So the collection crisis. Everybody’s probably feeling this. It was in the responses we got from the poll. Right?

Delinquencies are up. Credit cards obviously being the highest. Charge offs are up. Our thirty day cure windows are shrinking what it feels like, and we’re just not being effective and efficient enough.

Maybe we don’t have enough team members, right, in place to be able to do these outreaches, and that sources a problem in itself. Maybe we just have manual processes still in place. How do we automate those processes? How do we create an efficiency here?

Right? This is this is a problem statement, and it’s true for a lot of institutions. So don’t feel like if this is you and you’re sitting here thinking about it, you’re alone. You’re not.

It’s something that similarly we’ve narrowed in on and said, this is something we can help solve with the infrastructure we have in place.

And we believe, you know, uniquely, twenty twenty six is kind of an inflection point for this. Right? So we know the problem is at a tipping point. It continues to be compounding on itself.

Agentic AI is ready. It’s the world we live in. Right? Like, you can’t open up whatever your source of media is, right, and not see something about AI.

Right? And so the world’s there. And now our consumers, at least the consumers that engage with us digitally, are are wanting that. Millennials and Gen z, they’re the fastest growing segment for anyone, and we’re losing, I mean, a lot of cases to technology platforms that we can’t keep up with because of their stacks, because of how fast and how much money they’re spending.

But that’s what they want. That’s what the that’s what the data point tells us. Right?

And so I’m excited to introduce the Smart Collections AI. Right? And it’s our first agentic multichannel collections AI. It’s purpose built for credit unions and community banks.

It coordinates different agents. And I’m gonna go into these agents in in some depth, and we’re gonna look at the platform, and we’re gonna look at the capabilities, and you’re gonna actually hear me engage with this here in just a couple minutes as well. Right? But it really orchestrates between different channels in which you can engage your consumer, and it takes a huge burden lift off your teams.

And what that will allow you for do is to reorganize your teams to be more effective and efficient with escalations, with priority, with strategy, with different ways in in doing outreach and supporting the organization. This isn’t a rip and replace. It’s a supportment tool. It’s something that will drive value for your consumer and efficiency for you.

And with agentic modeling, and these are kind of concepts that are true across the board. Right? There’s reasons and actions. Right?

It’s it’s not rule based. It’s it’s the way the agentic modeling works is it’s reasonable. It it it will understand the language it’s been given, the the the prompt it’s been given. It will execute in real time based on the parameters set by you and your organization.

That could be policies, whether that be federal policies or your own policies, you know, cadences, and and you’ll see all this within our journey capabilities of how you wanna interact. What’s what is your specific institution style, if you will, in talking to your consumers?

It has built in compliance. Right? Like, it is regulatoriously safe. Right? It it checks all the boxes, and you can provide audit logging.

Everything is there. Right? It’s native core integration. So it’s talking to your systems of record.

It’s not storing information. You’re not fielding it, you know, files to try and keep it up. It knows when and who it needs to talk to by just integrating in. It’s becomes, again, part of that orchestration layer, something that sits on top of and that starts engaging as your teams do today.

And it’s proven to scale.

It’s true not only for us here with one and a half million, you know, conversations processed across our hundred institutions who are using BangGPT, but in the world.

AI has scaled. It is where we live today. And so as we dig a little bit deeper in now, what does that mean? What are these agents?

What do they do? How do they interact with those? Right? There’s really three agents that are all coordinated into a single strategy.

So once someone meets the criteria in which they need to be engaged, and we’ll talk more about the criteria when we look at the platform here, right, there’s really three three different journeys that you could put into the market. First is gonna be a voice agent, which is placing outbound calls. Pretty straightforward. Right?

The difference here is it’s not just an agent that’s gonna go out and say, hey. You’ve got hey. You owe me money. Call me.

Right? It’s an agent that goes out and says, hey. You’re behind. Let us help, and here are your ways to help, And then actually execute an action on that in real time.

Not a, hey. I can help, but you’re gonna need to call me back between nine and five PM, between Monday and Friday. No. At any point, it can reach out, and it can can and it can help this consumer.

Right? If you called me at eleven thirty on Thursday, May twenty eighth, yeah, I’m not thinking of that call. Right? Because I’m here with you.

Right? And that’s the world we live in. Humans create capacity points, and we’ll talk a little bit about about more in a second. So a voice agent that’s really in the market driving inbound or driving outbound, right, trying to engage in that forum.

And this speaks to specific individuals whether that’s generational or not. Not everyone’s a voice person.

My preference is not to talk on the phone. And if you were to ask my wife, she would be so annoyed with me because I get off the phone in, like, thirty seconds or a minute every chance I get. I don’t it’s I don’t know what it is about picking up the phone and talking. I’m a texter.

So the second agent is an s s SMS agent. And and this really is a powerful tool. If you’re not doing it today, there’s a huge population of people where this is their preference. This is how they would want to deal with this.

It’s less personal. Right? Like, they don’t feel judgment. Right? And we’ll talk more about that concept of judgment and and how, you know, using tools like this kinda solve that problem statement as well, but it allows a different form in which you can engage.

And the last piece is email agent. And truthfully, I heart I struggle. I understand why we have an agent that does this, and I’m sure a lot of people communicate this way. Right?

But I struggle with this being the future. Right? If I were to think about it, it’s really engaging in those digital tools like SMS or potentially apps in the future. Right?

Like, that will be the future of this. But today, these are the forms in which actually we see good result. Right? We drive meaningful change in those delinquency periods and get consumers on the right path for you.

So quickly to just talk about, you know, the traditional methodology versus what we’re doing here today. Right? And I mentioned this in the previous slide or at least gave it some context. Right?

Today, we’ve got manual dialing by humans in a lot of instances. And it may not be everyone here, but some of you, you gotta raise your hand saying, okay. I got a lot of team members. She’s doing a lot of phone calls right now.

Right? That’s limiting.

And it’s single channel. Right? Like, it it’s phone number. And, again, going back to my statement, if you called me today at eleven thirty, I’m probably not picking up that phone. Right? And it’s and it’s costly.

You know, a, the connection point, b, the human behind it, and then, you know, c, the the amount of time it probably takes to actually get this done because you’re probably getting hung up on quite frequently or told to call back at, you know, after five PM when I get off work, that type of thing. Right? There’s risk behind human error. Right?

There’s no guardrails for humans. You can train someone as best you possibly can, but, you know, at certain points, like, know, there’s not enough training in the world. Right? And so it it it there’s risk in that, and it’s routine.

Right? You’re simplifying a process that can be managed, right, where humans could be so much more involved in in solutioning on things that are not not that simple. Right? Like, they they can take escalations.

They can handle the relationship piece, can give you a tool in a different forum that, right, really drives engagement with the consumer in a different way. And then, obviously, restricting. Right? You probably got hours in which you can work people versus not work people.

Right? And, typically, these aren’t the greatest hours to call. We’re a smart collection system like we’re proposing today is is twenty four seven. There’s no there’s no dinner to make it home for or kids to put to bed at night.

Right? Like, this is something that can reach out at the consumer’s preference when they’re ready to talk, when they’re ready to address the issue. Right? It can coordinate between multiple channels in that outreach.

Hey. I don’t wanna talk on the phone. Text me. And you’ll see some of this here in just a second.

Right? It’s dramatically lower in cost.

It’s it’s built with all the guardrails in place to be compliant, and it can handle the early stage, you know, human focus and allow your humans to focus on the complex cases. Right? Like, the the zero to thirty is where this really escalates and and really drives meaningful results. And then what that’ll do down the line is just move reduce the bottleneck. Allow your team members to really handle those sticky cases in which you need to have nuance and human touch and empathy in order to address, in order to maintain your consumer base.

And so one of the biggest, you know, things that comes up when you talk about collections is obviously trust, and control.

It can’t go your compliance teams are gonna be all over this as soon as you say, I wanna use AI, right, to handle collections. Okay. Hold on a second. Let’s talk through that.

Right? So the way the system works, much like our voice chat and employee, right, the the infrastructure layer that that that is the piece in which we connect in to your policy engines. Right? We have, you know, only available to us what rules you’ve already put in place, and you don’t need to go recreate the wheel.

We’ll just attach to where those policies live today. Right? We have an automated QA center, right, that’s detecting gaps and violations and and surfacing that for you. Hey.

We’re noticing this. You should you should pay attention. Someone needs to look at this. Right?

We have a dashboard, which you’re gonna see here in just a couple minutes too. Right? It it’s surfacing, you know, volumes and success and and escalation, and and we’ll get into that a little bit better. Right?

And and a compliance studio. Right? So self learning and and automated to roll back when required, right, based on policies set by both federal and and institution. Right?

Like, it provides, you know, this this this layer of trust in which you can have a system that knows when and how to act and when not to act. Right? And then when to surface human involvement.

Okay.

That was, you know, lengthy. I’m actually gonna stop sharing my screen, and I have a a wonderful human in the background who’s going to pull up her screen, and she’s going to show us the platform on the back end. And we love to start here, and then I promise I’m gonna pop these earbuds out. You’re gonna hear me actually talk to this agent in real time. But first, I wanna show you what it is that you’re seeing and what it is that you can do with this platform.

Right? So a dashboard by nature, anybody uses Tableau or Salesforce or anything else, the first thing you’re gonna get to is is your home screen. And this is really gonna share your your higher level stats. Right?

Like, how many delinquent accounts are there? How much is past due? What’s your cure rate? Right?

You know, what by portfolio? What’s the state by portfolio? What policies are being allowed or versus denied? What’s being bubbled up? Right? You can orchestrate and execute actions. It’s giving you trends.

It’s showing you the distribution of each of those states, the activity that’s occurring within the states.

Right? So this is that kinda home page, that ability for you to have your quick snapshot in. Right? But let’s get in a little bit deeper.

Like, okay. Great. You can show me what’s happening. But, like, how does the tool work, right, from a human level?

So now we’re on accounts page. Right? So, again, this probably mirrors up to what we just saw on the dashboard because that’s where it’s drilling this information for, but you’ve got accounts, and these are humans. Right?

We gotta remember that. These are humans behind this. So you’ve got Steve and Jennifer and Michael, right, all with amounts due and past due and their current state and how long they’ve been and what kind of product. I’ll summarize for you.

But let’s say, hey. I wanna drill in. Let’s take a look at Steve. What’s Steve’s situation?

Right? Well, Steve here, we have his information, and this is coming from your core or the repository in which we’re pulling this information. Right? We have his account details.

So what are we looking at? What are we solving for? We have his contact preferences that he set so we know how to engage him. Right?

We know what journey he’s on set by the parameters, and we’ll talk about these here in just a minute. Right? And then even further, let’s say I’m auditing this thing, and I wanna see what’s happening. What interactions have taken place?

Your teams can come in and see exactly. Right? Like, if he’s engaged with us, right, like, you can see what those interactions are. Right?

You know, other things here, and he doesn’t have any policy decisions there. But, like, if there was a decision made at a certain point based on interaction, you’re gonna be able to see that. Right? Or if we made an offer of some kind.

Right? Like, you’d be able to see what that offer was.

And so as we, you know, take the next step here, right, like, this is total members. Right? And this gives us a different look at these. Mostly, what I wanna pay attention to is is language.

Right? Like, this is a big ask for a lot of our financial institutions. Hey. Can you address multilanguage?

And you’re seeing here that it’s possible. And we know what language to then reach out in whether regardless of the form, right, the channel in which we’re gonna interact with this consumer. So all of this is very important. Right?

And it gives you, obviously, member ID so you can drill in. Let’s take a quick look at Melissa. It, you know, should be a pretty, you know, similar view here in in the types of accounts she has. Right?

Her total past due, her language of preference, the consent she’s given us.

You know, again, essential repository. So this is a system replacement. It could be, at least, depending on what you’re using today. You could be operating out of this for your collections.

Right? Because you have the insight. It’s connected to your core most likely or another repository in which this information is being gathered. And then your ability now is really to set up, okay.

Well, I’ve got a system. I know who needs to be talked to. Well, how do I do that? And let’s go ahead and click on products because, you know, not all products are created equally.

Right? Like, you have different rules based on the products in which someone could go delinquent. So the rules, right, in how often to engage, what to engage with, all can be broken down by product. Right?

And so we’re seeing auto, credit card, etcetera. You’re seeing here the buckets, right, days past due, and and you can set these buckets differently again by product type. Right? So we know, you know, hey.

Grace periods one through five, and then you’re in your early and your mid and your late.

You know what disclosure blocks to be including, you know, what workflows and statistics, right, prohibited topics. Right? Don’t say these things. Right?

The agent should never address, and that’s where it’s reading its source of truth. Right? It knows how to handle escalations. Right?

And, again, being defined at the product level and then based on the stage within the product level. Right? Like, the delinquency state and the self-service options available to these consumers. Right?

And then, again, the compliance notes. What where are we reading this from? How are we driving results? How are we trying to automate as best as we possibly can?

I’m being mindful of time. I see that clock up there. It’s ticking, and I promised marketing I would be on time today. Known to be a little worried.

Apologize. Alright. So let’s get into the journeys real quick because this is the marketing that speaks behind me. Right?

So you’ve got different journeys based on stages. Right? So you’ve got late stage versus dental versus standard. These are all things your teams are gonna have access to and and can customize.

It’s gonna sound and feel like you. That’s the purpose of it. Right? It’s meant to be personalized at the consumer level.

It’s meant to be reactive, but it’s also meant to be you. Right? And so as we go in and we take a look at this, right, these are all the things, right, like, that are most important in this stage from a collection standpoint.

So you’re setting those rules. You’re setting the flows, of course.

When do you want text to go out versus phone call to go out versus reminder emails? Right? All of this customized based on your team’s preferences and and, again, the outreach methodology in which you operate your institution from.

Alright. Let’s jump now. Yeah. Let’s jump into policies. I think that’s an important one too because we’re setting the rules.

Right? So, obviously, you’ve got federal based rules. Right? Like, these are layers in which we have to follow law.

Right? Like, there’s things you can and can’t do, so we’ll follow those, obviously. But then you’ll notice at the top there, the all later is federal and then institution. We know our partners have different rules too.

Like, you have institutional rules that differ from some of these baseline policies that are in place. And so those rules, you can set up within your institution instance as well. So, again, the agent only interacts with the consumer and provides results based on the guardrails set in place, and these are guardrails. Can’t hallucinate.

It can’t go beyond what it knows. Right? It’s set to believe within these policies and respond within those policies itself.

Yeah. Let’s show them templates real quick because I think that’s know, it’s the interaction piece. It’s how are we addressing it. And so, you know, obviously, we can templatize.

Right? There’s some standard language that’s pretty consistent among the industry, but you have this ability to go in and really make it sound and feel like you. Right? So this is a sophisticated template.

Right? And, again, we build everything with with the end to getting a modeling behind it. So even when you’re setting these templates up, ease of use, operational efficiency, like, this isn’t hours and hours and hours and hours of work. Right?

You’ve got a team obviously here at Interface that could support you, but you’ve also got a powerful tool that’s driving these for you and and quickly and effectively.

I know be mindful. I wanna do the phone call here in just a second. I’m gonna try and leave time for questions. Let’s go ahead and look at those So the truth of the matter is is we’re not in an industry where we’re ready to say, hey.

AI handle everything. I don’t need to touch this. I don’t need to see anything. Right?

So escalations are still a thing. And so this is how escalations are bubbled up. Right? And so you can have very good insights to the interactions that are happening, the priority level that’s being set by that, the queue that which is in, and then assignee.

So you’re seeing a dashboard here. So these all have been bubbled up where the agent said, hey. A human needs to interact.

Pay attention to these. So now as you think about how your systems work today, your teams are very focused on now the critical stage priority deal, like collection deals. Right? They’re no longer dealing with the simple, like, hey.

You just forgot. You’re six days in. Pay this. They’re dealing with the the hardship. I I need it.

I need you to be my institution of choice. I need to trust you. Can you help me? That’s how you get to interact with your consumer now because you allow technology and an agent to handle the the simplicity, the process oriented piece of it.

Let’s jump into callbacks. So the beauty of a system that’s twenty four seven is, yeah, we could take a callback. So if they don’t answer the phone and they call back at a later time, even if it’s after hours, we’re gonna be able to surface that. We’re gonna be able to try and resolve, right, on your behalf.

Right? So you’re seeing all the consumers that called back in after the fax. Right? And then again, what happened in those interactions much like the screens we’ve seen before.

So, hopefully, you’re getting a full view of, you know, how does the system operate, what capabilities do I have within my fingertips.

I think the the other two most important just to notate on, and then I’ll pull up these earbuds and and we’ll get this going, is let’s take a look at the QA center just real quickly. I think that’s it it’s it’s valuable here. Right? So the system is QA ing itself.

Like, it’s it’s looking at and reviewing and scoring and and allowing you access to that. We’re not hiding behind some mysterious wall of how’s the AI working. Is it solving my needs? You have interact you you have an ability to see that.

Right? You have an ability to actually, you know, curate and and improve. Again, agentic modeling, it improves upon itself, but with with responsible and knowledgeable prompts from humans, it gets better faster. Right?

And that’s that’s the system in which we’re creating for you.

Last piece, reports. Right? Obviously, you gotta have audit logs and reporting, and it’s all available here for you. We put it in a simple place.

And, again, I go back to, could this be a whole system collection replacement for you? It could if that’s how you choose to use it, or it could just be an orchestration layer tied into your existing system. Maybe you’ve got an amazing relationship with someone and you wanna maintain that. Absolutely.

We can just manage the engagement piece and pass this information through. But, you know, obviously, you’re seeing a tool here that’s extremely robust and and can be an operational and cost saver for you, especially if you’re using our other tools. Right? Like, think about the one brain concept.

It learns upon every interaction with every consumer and every engagement point.

Alright. I’m gonna pull out my earbuds, and then we’re gonna actually just I think you should be Now and I’m gonna turn on my volume so you can see this in real time. You’re seeing here I’m getting a call. Again, it’s eleven forty five on a Tuesday.

Don’t these people know? I’m just kidding. Let’s go ahead. Let me put this on speaker.

This is Alexis’ agent calling from First Community Credit Union.

Oh, no. Yeah. This is him.

Thank you, Michael. To verify your identity before we proceed, could you please provide the last four digits of your Social Security number or your date of birth?

One zero eight one.

Thank you for verifying your information, Michael. This is an attempt to collect a debt.

Any information obtained will be used for that purpose.

I’m calling regarding your credit card account ending in six zero three zero.

It appears that the account

I tried talking.

I was gonna make a joke and say, knew I shouldn’t take that in your job. My wife was gonna get revenge on me. But let me go ahead and do this. So, agent, what options do I have?

I understand, Michael, and I appreciate your willingness to address this.

Your credit card account ending in six zero three zero has a past due amount of three thousand dollars. We can explore a few options to help you resolve this.

Okay. What what are my options?

You have a few options, Michael.

One, you can make a full payment of the past due amount of three thousand dollars. I can send you a secure payment link via SMS or email to complete the payment.

Two, if you’re unable to pay the full amount right now, we can discuss setting up a payment plan to help you manage the payments over time. Three, if you’re experiencing financial hardship, we can connect you with our assistance team to explore other potential solutions.

Which option would you like to proceed with?

Alright. Send me an SMS with my options to pay now.

Certainly, Michael. I’ll send you a secure payment link via SMS so you can make the payment at your convenience.

If you have any questions or need further assistance, please feel free to reach out. Thank you for addressing this matter, and have a great day.

Awesome. Give me one second. I’m a plug my headphones back in, make sure we keep everything silent.

So as you see there, I kinda threw a couple curveballs, started talking, right, in mid I’m sorry. Multitasking. There we are. We’re back on the right screen now.

Okay. Through a couple, you know, mid statements in there that kinda threw her off for second and had to think about it, but the the agent in the background is thinking about that, and it’s self solving. And it realizes the conversation’s not over. It’s giving me a second to finish my thought, in which then it took me down the path of of resolution still.

Right? So a simple interaction, again, that’s voice. There’s text. There’s email, right, capabilities. Again, engage the consumer in the forum in which they would like to address the issue at the time that they wanna address the issue is the key concept that this allows for you to really facilitate in a meaningful way.

And so, you know, we’re getting to a point, and I I’m gonna hit the right on the head with q and a. I got I got ten minutes. So I did it even though sometimes I go a little long, but I got there today. Right?

So, the future of collections is here. Right? So this allows you to cure early, which which should really improve the consumer experience more than anything else. Giving them their options, talking to them in the form that they wanna talk, giving them the ability, you know, to address it on their time and the way they prefer.

Right? And you can scale. So you don’t have to add FTE, right, to handle the larger delinquency volumes coming in. You’ve got a tool that’s gonna manage quickly and effectively.

Right? And then allow your team to handle those those nuance, those those escalation situations that allow you to contain and keep a member or a customer happy and with you, most importantly. Right?

It protects those relationships. And, truthfully, this is a this is a tool that we can be up and running and and and quickly.

Right? Like, this is something that if you’re if you’re feeling the pain today, there’s a quick resolution. So, again, I mentioned this earlier. I got a team of people who would be more than happy to engage with you to go a lot deeper, talk a little bit more about your specific instances, of course, address significant amount more questions than I’ll probably get to today.

But, hopefully, this gives you some insight to, hey. Interface is actually a unique and very interesting spot to start solving some of these pain points now using intelligence. Right? And models and agents that allow for us to be effective and efficient.

So I I saw a ping. Let me just go ahead. I’m on my computer laptop, I gotta address some questions here. I’m getting pinged on the background.

But let me also pull up real quickly. And there’s a nice QR code there, so don’t hesitate. Go ahead and take out those cell phones, scan that. We’d love to talk to you a little bit more.

But let me go ahead and pull up the chat and also some of the questions.

Q and a. K.

Does this product integrate with Okuvo so the collection activities are noted in the collections platform in lieu of needing to log in to your separate system?

Absolutely. So I mentioned this, and I know sometimes I’m a little quick to words. So let me restate that. So, yes, it it operates two different ways.

Right? We could be the engagement tool that’s connected to your system of record today, whether that’s a Kuvo or some other collection system you’re using today, potentially, it’s your core. Right? Like, you could operate out of that.

And in fact, even if you were to take our platform and use, you know, that dashboard, the the landing page concept, it would still provide the information to your sources of record. Right? Like, that’s that’s critical.

It just provides a different option should you not have a system that you’re you’re you’re preferring. Right? But either way, yes, you can use your current system, attach us on top, allow us to run the interactions, improve the situation, and or you could say, hey. Like, that was a really cool tool.

I don’t have all the data and reporting. I’d love to get something like that. Either of those options are available. So, hopefully, that answers it.

Another planted okay.

Where you’re currently currently using for texting, how is smart collections different?

Well, text needs to be replied to manually by a person. So that that’s a key differentiator here. The smart collection system is truly agentic and automated. Right?

It allows without human interaction or interference, right, responses based on your policies and procedures. So it would probably be a pretty massive operational savings or an efficiency savings for you. However, you know, if you’ve got systems in place, just kinda like the last question, let’s talk about let’s digest that. We’ve got a very smart team here of engineers and architects who will who can kinda structure this in a way that makes most sense for you and your organization.

More questions coming in. I’m trying to read on the fly here.

We can escalate all interactions that need human assistance to your current collection platform. Yes. Accurate. That is true.

We can escalate into your current platform. You can anything you saw can be routed into your current platforms. Or, like, that data just gets passed into that repository, that location you want. Those are all just API connections and integrations.

Absolutely can operate that way. Today, you saw just the full breadth of the solution, which includes, obviously, the channel in which you can access that information.

Can AI switch to from Spanish and English? Yes. It’s bilingual. So it it it will pick it up. So even if maybe your system potentially didn’t register the consumer as a Spanish speaking or Spanish preference, if they start speaking Spanish to us, agent’s gonna pick that up and and revert. Right? Like, it’s gonna change and start talking back to that consumer in their preferred language, obviously.

I’m seeing a slowdown in questions. I think we tackled them. Maybe some of you just want some personal time. I love that too.

Let’s set that up. Again, QR code is on the screen for you. Please please feel free to reach out. It’s yeah.

Of the things, if you ever worked with me, and hopefully I’ve worked with some of you in the past, is we’re here to help.

My teams the teams here were consultants. We we wanna make sure that you’re understanding the current landscape of technology, the capabilities in which that technology allow you, we’re and here to help provide and paint a picture of how we can connect that.

Oh, one more just popped through. Hold on. Integration with Fiserv Portico, which is a core. Do you currently upload into it? Fiserv is notorious expensive to integrate with.

We’ve integrated with Fiserv in the past for AI voice.

So let me let Dan you know what? I’m gonna have someone reach out to you. We’re gonna get a little bit more in the weeds there. I’m leaning towards, like, shouldn’t be an issue, but, like, let us get you someone and a firm response there. And, again, no commitment. You don’t have to take a a demo and bring your c suite or whatever you we’re just gonna get you a good response there so that you have the information you need to to make decisions here. But, again, yes, we’re integrated with Fiserv on other products, so it should be a yes.

How are call recording stored? Call recordings are stored in our cloud securely, available to you to listen for playback. So we have that portal in which you can download. You could see the the transcription of it and or download the video and and watch the video should you choose. We also surface major points within the video for you, so it’s an intelligent look back. You’re not watching potentially, like, a forty five minute video type of thing.

Are there different AI voices that can be used? I have heard some that sound very similar to a live person and not as robotic. Yep. Absolutely. You heard just our standardized demo version. There’s a library, or you can create your own custom voice. So if you’ve got someone on the team who’s lovely and soothing, like, absolutely, let’s make that person the voice, right, like, and and have that interaction while they’re firing in now.

How are call recording stored? Oh, I think we answered that one. Okay.

All done. I got the big thumbs up. We got five minutes here earlier. You know?

That saying of, like, I’m gonna give you five minutes back always kinda like but I hope whatever you do in the next five minutes is both world today or worthwhile. And I hope today was educational, important, valuable. You don’t feel, like you you walked away with less information, but hopefully more information. And, again, the team here at Interface, we’re excited to engage with you.

Let us help you. Let us be a consultant. Let us let us tell you what the the art of possible is and really drive consumer interactions. It’s important to me.

It’s important to everyone here, you know, including Shri, which is our core mission. So I appreciate you all for spending fifty six some odd minutes with me. Please enjoy your day, and I I hope to see you in the very, very near future. Have a wonderful day, everybody.