How Neighborhood CU Is Leveling-Up Their Member Experience with AI

So, Kalyn, do you have a way to turn off your motion sensor lights, or is it is it too smart AI to learn from?

No. I I I wondered about that and that, I don’t think I have a way to turn it off. Let me just see, because it did go off while we were talking.

Yeah. I don’t I don’t think it was a way to turn the sensor off.

I didn’t didn’t see it. So, hopefully, maybe I’ll just need to move around in here a little bit.

Great. I think we are live, I guess. Yes, sweet. We are. Okay.

Good afternoon, everybody. Welcome for joining us. We are gonna start right at one o’clock. Give everybody a couple minutes to join in. So we will be back in just a minute to get started.

Alright. Once again, welcome for joining us, and thank you for taking time out of your days. I am Jenny Jackson. I work with CUNA strategic services. If you are not familiar with us, we are an organization that looks for vets and evaluates various vendors to determine what is the best of breed, what’s the most innovative, what is gonna do and work the best with our credit unions to help push them forward, compete with the banks, the neobanks that are always seeming to pop up everywhere.

And Interface AI was a recent addition to our portfolio in the beginning of twenty twenty two, I guess it is.

And we did a lot of re this this took us a long time. You there’s a lot of AI companies out there. AI is one of those buzzwords that doesn’t that is attached to things that aren’t necessarily, artificial intelligence.

So we we went through quite a bit. We selected, Interface for various reasons, such as a couple being their number of integrations, their feature set, continuously innovative, and the fact that their CEO and their company is very mission focused. And we do always try to find companies that really fit in and vibe well with the credit union industry.

So this afternoon, I’m I’m excited to have this session going. We’re gonna talk about credit unions, how they’ve been dealing with significant increase in call volume, abandonment rates, spikes in the range of twenty to thirty percent, longer wait times at ten minutes, sometimes up to thirty minutes, and higher levels of fraud. And as always, it’s hard to find staffing for call centers in various areas in credit union. To overcome these challenges, credit unions are turning to AI powered solutions of which Interface has a a very robust solution.

We’ll have Carolyn Jordan speaking with along with Sri.

Oh, here we’ve got the screen. Sri, who’s the CEO, and talk about the challenges that Neighborhood Credit Union was facing and how they’ve used Interface and AI to manage their service led levels, and provide twenty four seven support for their members.

Brief intros on both of them. Carolyn Jordan has over four decades of financial industry experience with forty three years of her professional career at Neighborhood Credit Union, where she is currently the chief growth officer.

She’s responsible for a lot, including digital, online, mobile channels, data analytics, special project projects, and new product development and marketing. She’s also a past, chair and current member of the CUNA operations and member experience council, which many of you may know as a national forum for credit union leaders.

Sri Sri NJ is the founder and CEO of Interface AI. Sri led a digital strategy from scratch for a financial institution to scale it to several billion dollars in assets. Decades of AI research and experience of leading digital strategies to scale financial institutions inspired Sree to start Interface AI with a mission of enabling financial wellness and leveraging AI to do so.

I believe, Sri, you are going to start us off.

Absolutely. Jenny, thank you for the introduction and the kind words, and we really value the partnership of, Cunanan strategic services. It really means a lot to us, and thank you for picking to face AI. And, Callan, we appreciate joining us today.

So, it’s been a fun journey, you know, launching AI and and and with your team, it’s it’s been excellent. So, today, I would love to start off with, talking about agenda, and we’ll, you know, kick things off here. You know, first thing first, we’re gonna spend couple of minutes introducing neighborhood creatinist Nick. And, and briefly, we’re gonna give you a a a sneak peek into interfaces intelligence self-service platform, and then we’re we’re gonna, come back and go through the entire journey of, a neighborhood credit union, here starting from, just finding the right vendor to implementing the project, going through the run and maintain as in the support and maintenance, what kind of impact the AI has already provided, some of the member feedback, and the road map.

What’s next for Nick? Right, and then we’ll open up for q and a. So let’s get started without further delay.

So Neighborhood Credit Union, has launched AI assistant on their call center. Right? So if you call, their, member support number that’s listed on their website on the contact us page, you will find a number. And if you call that, the AI assistant Nick is gonna pick up the call hundred percent of the incoming calls and, helps the members on various topics. I wanna take a, a minute here. And, Callan, do you wanna introduce us to Nick, what it does today, you know, how he’s been helping your members?

Well, a lot of the calls and thank you for that, Shrenda Boss, and thank you, Jenna Jenny. A lot of the calls that Nick gets are are routine calls, and that’s where he has been able to help us significantly.

He’s been able to take those balance requests, deposit history, transaction history, is the branch open? All of those routine requests that seems so simple, but that take up an agent’s time during the day. Nick can help with that. And what really has been cool about Nick is that he’s available, you know, twenty four seven. He doesn’t get sick. He doesn’t call in.

He’s always available, and and our members love having that access twenty four seven for, you know, whatever it is they may be doing wherever they’re at.

Got it. Thank you, Colin. And as there is a specific way Nick has been helping today, which really addresses some of the challenges, why you started this journey, do you wanna talk about what has been, some of the, benefits, from mix so far?

Well, yes. Absolutely.

So like many of, the folks that are out there in our audience today, we had been experiencing high call volume, actually, for for a while. And during the pandemic, that escalated.

Then you have the the thing of, low not being able to find staff, you know, short staffing. So Nick has been able to help us handle that call volume, reduce some of the wait times, reduce abandonment, and be able to give that member that round the clock experience, offering that consistent member experience. And and also being able to help them get their simple inquiries taken care of without having to wait. When our members use Nick, they don’t have to wait at all.

They get an instant answer, and then they can go through and ask as many questions as they desire that we have experiences for. And we do have quite a few experiences. And for the employees, what that does, since they’re not having to answer those routine inquiries, that frees them up, and that also allows them to have, you know, conversations, you know, for for more complex things. And that didn’t fill up our queue with calls that, you know, they don’t have to handle because, you know, members members love the call.

So those are some of the the ways that Nick has been helping us. Handling that volume, offering that twenty four seven experience, and then freeing our employees up to handle more complex calls.

Got it. Thank you, Caroline. And, the do employees now complain? They often get very complex calls.

Surprisingly, no. Because the the employees, they they really they really do want to be available to help the members. And, you know, when they get a call that, you know, could have been taken care of, you know, through using digital banking, you know, that’s that’s taking up time for another member. So they actually don’t complain about it because they really wanna be there to to help the members. And we know that there are some people that’s just not gonna use digital banking. So that’s where Nick is so helpful because he can allow us to automate that, you know, even if they don’t want to use digital banking.

Got it. And is is that brings up a very interesting aspect of the member behavior. Right, Callan? Like, you know, we have digital tools like online mobile banking, but some which many of the tasks that they would call in for could be accomplished through those digital channels, but they still call. Like, you know, there is a certain behavior of the member, and and, you know, there is always debate between, hey. Do we, change the member behavior or to have them adopt online mobile banking tools or we serve them wherever they’re reaching out to us?

You know, there’s always kind of debate there. So just curious, what has been the mindset? Like, you know, have you have you in the past, gotten a chance to, successfully change a member behavior, or you feel like it’s better to serve them how they usually bank?

It we we feel like it’s it’s it’s great to meet them where they’re at. We we like to offer as many self-service type of services that will appeal to them. So Yeah. Just to what you said, Shrinivas, you we’ve got some members that they will never use digital banking, but they love to call.

And that’s where Nick is so helpful. And then, you know, believe it or not, we’ve got some that they use both. They they use digital banking, but I guess depending on, you know, where they’re at at the time, they’ll they’ll want to call and use Nick. And so we just think that when it comes to service, if you can offer a multitude of self-service options and still have in person if that’s what they need.

We still have people available, obviously, in the contact center. But I think having the choice of Nick, our digital banking, and and they can do a lot of things with Nick. Before we implemented Nick, we had a limited feature set of, transaction history. And now they can get just as much transaction history through Nick if they wanna stay on the phone as they can in digital banking.

So so we think service is about meeting them where they’re at and giving, various self-service options that that appeal to the members.

Oh, that that’s really helpful. Carolyn, thank you. I just wanna take a quick moment to, introduce the big picture, you know, here. And Nick is powered by Interface AI. And as a company, our vision here is to help you leapfrog from digital to intelligent self-service.

You know, and and, you know, what we mean by, those intelligent self-service, and I kind of explained that in a bit. But, really, the biggest challenge we all face, today, is being able to create meaningful and experiences for our, customers and members. Right? So being able to have that seamless experience, but there are challenges to addressing that.

You know, if you look at, our, you know, digital, architecture, you know, we have system of engagement at the top. You have online banking, mobile banking systems, and then you have call center systems and live chat, co browsing systems. Those those become system of engagement. And the system record are like this core banking, like, you know, Episys or a DNA and then a LOA system like MeridianLink and then debit card, credit card processing system.

In order to create any decent banking experience, all those arrow marks you’re looking at needs to be connected.

Right? So it’s impossible project, to have all of this connected. And many times, you have chosen a call center or a live chat system, that really industry agnostic. These vendors serve multiple industries. But when it comes to online banking, you choose people who have served banking in the past. The reason is, you like to have a lot of integration. So, you know, that’s that’s kind of the really, the key which requires a lot of integration.

But in order to solve this problem, enable a great member experience, what we are doing and what we mean by intelligent self-service is really introduce this layer between your system engagement, system record. That’s where we sit architecturally as a system. This is a system of intelligence, right, that connects with your system engagement, system record, and powers each and every system engagement, either your call center or live chat or an online mobile banking system. It powers each of that with a, you know, banking brain.

Right? So, they could move between the channels and get that really efficient, experience. Right? So, you know, today in the journey, we’re gonna be looking at how Nick has has, achieved one of the pillars, which is automation.

But there are other pillars here that would, make would continue to evolve in the future, like being able to do upsell, cross sell, being able to provide personalized service. Like, if you’re calling back again and again, repeatedly, there’s ability to recognize that and proactively help you.

It also will have ability to, you know, provide financial insights and also do voice verification caller ID. Right? Just one of the pillars is live, which is really the starting point, and you build everything else on top of that.

We’re gonna see more about that in a bit as well about Nick and the journey.

And briefly, I wanna take a moment to, demonstrate the, you know, next specific ability today, which is what we call AI powered phone banking. AI powered phone banking is one of the products in this intelligent self-service suite of products that we are offering. Right? So and, specifically, this is what, you know, you know, neighborhood creating is using as well.

And the way it works is, Nick has a ability to pick up hundred percent of the calls and respond to everything that it can. Otherwise, it’ll seamlessly hand it off to a staff. That’s how Nick works today.

And this has its own crawl, walk, and run journey. Right? You start with, you know, Nick answering just informational questions where automation could be just five to ten percent, and then you transition to stage two where Nick has a way to answer transactional inquiries, like checking balances, fund transfers, where the automation could go up to, you know, somewhere between forty to sixty percent depending on the integrations. And the stage three is upsell cross sell where AI itself has the ability to, introduce new products and provide financial insights during the call.

Right? Today, Enabler Credit Union’s maker has gone as as far as stage one and stage two, and stage three is is something that’ll be coming up very soon. So we wanna take a brief moment to show a demo of stage one and stage two and three.

So keep in mind, Nick is, yet to launch stage three, and we’re gonna go through this journey right after this demo.

For this demo real quick, I would like to introduce Bruce, who will be, doing a brief three to four minutes demo to show you the, AI for phone banking in action. Bruce, would you do the honors?

Yes. I will, Sri. Thank you very much. Folks, while I’m talking to the demo, to the bot, his name is Gary, I want you to notice a few things.

I’m gonna change speeds. I’m gonna change accents. I’m gonna add a whole bunch of verbiage and vernacular. And, let’s see how Gary handles that.

I think he’s gonna do a pretty good job. You can listen in. You can observe those things on your screen.

Hi. I’m Gary. Our conversation here will be recorded to help me get better. How can I help you today?

Hey, Gary. I’m out and about driving around. I need to make a deposit. Are the branches open?

Yes. We are open. Our main street branch only has drive through capability today. Do you want me to send you driving directions to the nearest branch?

Yeah. That’d be cool.

Great. Please provide me the ZIP code or city name.

I’m in San Mateo, California.

Thanks.

Looks like there are a few branches that are open in your location. The branches are open from nine AM to five PM, Monday to Friday, and ten AM to two PM on Saturday. I just sent you a text with driving directions, and you can always talk to me whenever you need help. I am available twenty four by seven. Is there anything else I can help you with today?

Yeah. Hey, Gary. I started a business, and I don’t think I’m gonna be able to make payroll. Can you help me skip a loan payment?

I can surely help you with skipping a loan payment. I have sent you a text with a link to the application to skip a loan payment along with the instructions.

Please click on the link and follow the instructions.

Please note that the application should be filled at least five days before the payment due date. To assist small businesses like yours, we have removed the thirty five dollar fee associated with skipping a payment. Do you wanna know which loans are eligible to skip a payment? No.

I appreciate that. But can you tell me what the balance is in my checking account?

Sure. I will help you with your account balance.

I see that you already have registered for voice verification.

Would you like to verify using your voice?

Yeah. Okay.

Great. Make sure there is no background noise and go ahead and say, my voice is my password. Please verify me.

My voice is my password. Please verify me.

Thank you.

We are able to successfully verify that you are a customer of Interface Bank. The balance in your checking account ending with two two three four is two thousand three hundred and thirty seven dollars and fifty five cents.

Also, I wanna remind you that if you maintain a minimum balance of three thousand dollars, you will be eligible for an additional one percent cashback. Do you wanna know the balance of your savings account?

No. I appreciate that. But listen, I’m gonna be traveling to London. Can you put a travel notice on my account I mean, my card?

I can help you set a travel notice. I’ve noted that you’re traveling, and I have successfully set a travel notice on your card. We’ll make sure your transactions run smoothly. By the way, you’re traveling a lot, and I recommend you switch to our new credit card with no foreign transaction fee and many more benefits. Do you wanna know more?

No. No. I appreciate that. But listen, I deposited a check-in an ATM in San Diego, and I haven’t seen the money in my account yet. Can you tell me the status of the deposit?

Sure. Happy to do that. Give me a moment. I will transfer the call to our specialist, John.

Okay. That was Gary. Back to you, Sri.

Thank you, Bruce.

So that was the experience there. So, you know, this is AI port phone banking. You could see how AI was able to help with just general inquiries as well as transactional inquiries and in the process to upsell, cross sell, provide financial insight. The last part of it, which is the stage three upsell, cross sell financial insights, that is coming to the Nick, but everything else Nick could do today.

So just, to start, Caroline, from from your end, just to understand your journey, in launching Nick, and and whatnot. So I have a few questions to get started.

You know, one of the first questions I I wanted to ask you is, how did this journey begin? Right. So I understand you did evaluate AI a couple of years before you met us, and you, pulled it off and then reevaluate it.

And and then, this time, you you went ahead, starting the AI journey. So just curious, how did it start? What was the vision? If you could share, that would be great.

Yes. Of course, Shraddha. So like you said, we had started that journey.

We we we had the higher call volume that we had been looking to try to figure out how we could manage.

And then the pandemic hit. We we were already looking at solutions just because we could see we’re having more call volume. We can’t keep staffing consistent.

How do we, you know, manage this this staffing issue and still be able to give that consistent experience? So that’s what started us on the road.

Then the pandemic hit, and we did pull back, initially, but what we saw as many others saw that, you know, volume increased probably for us thirty to fifty percent.

And because we closed our our branch lobbies like everybody else. So the problem that we had been trying to solve, you know, handle call volume better, handle staffing, manage staffing better. The pandemic escalated that and and made that even worse.

And so we thought, well, you know, this AI thing that we have been looking at, you know, even though we are in a pandemic and there is some uncertainty, maybe we need to pull that back in and and look at that again.

And and we did. And and, obviously, we were we were glad we we did that because we we really felt like that it was gonna be the answer for the consistency, managing that staff that, you know, when you just can’t get folks hired, but you still need to make sure you have a good experience for the the members. So that’s what, you know, started us on that journey was, you know, our ongoing problem and then having the pandemic escalate that.

Got it. What what were some of the other options, or to, deal with increased call volume, after the pandemic, telling you? Was there any other options that you evaluated or you know? Because you had knowledge about AI with the prior evaluation. You you knew AI is the right path to go forward.

Do you mind shedding some light on that?

No. I don’t mind at all. We we did explore some other options. We we thought about maybe looking at a outsourced contact center.

A lot of credit unions do that, and we started down that road. But as we looked at our members and and looked at what we really needed, we thought, well, we don’t think that’s really gonna help us because at the end of the day, they’re employing people, and, you know, they’re they’re probably can’t really automate like we needed to, so those calls will probably come back to us. And that was the main concern. You you employ a a, outsource center, and then the calls still come back to you.

And and what happens with that, your your members still have not, gotten assisted. They haven’t got their problem taken care of. They haven’t got their answer to their inquiry. So, we decided that the outsource, solution was not going to work for us with the call center, And that’s where we thought, okay.

Well, AI is is very viable because we can automate, you know, we we can hook into our systems.

And we also saw that, you know, we had a a very, you know, kind of basic audio response system. We saw the opportunity to enhance that through AI and through interface. So we thought, you know, we can probably get, you know, you know, real bang for our bucks by being able to take care of informational inquiries and then, upgrade, you know, some of these transactional inquiries from our more basic, smart teller system.

Was there any other, apprehensions and concerns you had while you’re thinking about launching AI? Like, for example, in terms of, members, being not happy, who probably earlier had a choice to talk to a staff member now being, directed to AI. Of course, if AI cannot answer, they always can get to staff. But just curious, you know, was there any, any concerns at all that you were, you were also evaluating as you think about the AI, journey to begin?

We had some concerns. And, you know, people don’t get real excited about talking to digital assistants and chatbots.

But what we figured out was that if we could choose a virtual assistant that was really good, that could really take care of the member’s answers, then they wouldn’t mind as much. Because at the end of the day, you know, I think what people really want is to have their question answered, to have their their problem resolved.

And if that’s a person, great.

If if it’s a chatbot or a virtual assistant, that’s great too. Because their time is what they’re looking to maximize. They don’t wanna spend a lot of time. So while we did have that concern, initially, we we felt like if we get the right partner, and that was key for us, if we get the right partner with the the right digital assistant, and and the right platform, then we’ll be able to to manage the member concerns.

Got it. Thank you, Calvin. One of the, next question I had is now that it was clear AI is is the path that you wanna choose to solve some of these problems.

How did your, vendor selection process start? Like, you know, would you mind giving us some, some of the details around that?

Well, yes. Yes. Actually, when we first started, I started looking at some of the presentations from Venovate, to be honest. And because there there were lots of AI vendors just as Jenny had mentioned earlier, there’s a lot of them out there. So we started looking at some of those presentations to kinda get a feel of, you know, what was available.

And then, you know, as a credit union, we we get a lot of, solicitations from various vendors, that are, you know, selling us stuff, so to speak. And so we started looking at I started looking at some of the communications that I were getting from from AI vendors and, you know, what they had to offer.

And, it just so happens that, you know, for you guys, I was, I seen you at Innovate, and then, I saw that we, received the some email communication from you all. So that really made it convenient for us because I remembered that I had seen Interface at Innovate, and then you guys have reached out to us.

Got it. No. I appreciate that. And, so, you know, one of the good things there is, Interface also won best to show, or or only other AI vendor, that have happened.

So, probably it surfaced, your shortlist, pretty easily there. So, this is one of the following follow-up question is, in in terms of the, vendor selection, you know, you had a shortlist of different events. You’ve, you’ve you’ve, shortlisted a vendor. And in terms of the implementation process that began right after that, so just curious, like, what was the, implementation process look like?

Can can you walk us through that?

Yes. Well, you’re right. The it’s shortlisting was easy because, for for one thing, there there were not a lot of the vendors that were doing voice banking. We were we were specifically looking for a solution for the contact center in voice banking. And so that, you know, that that gave us a a short list of vendors. But beyond that, the other things we were looking for once we, you know, had a list of vendors that could do voice, virtual assistance, You know, we we wanted to be able to see time and tenure in the AI space. That was critical for us because as I mentioned earlier, we wanted to make sure we got we had the right partner to get the best digital assistant that we could, you know, for our resources.

We looked at the quality of the digital assistant, our chatbot. And then the next thing was the integration capabilities with our, data processing core system. If we were going to be able to have those transactional experiences and upgrade the, audio response experience that we had, we had to have some, integrations for me familiarity with our core processor, which is Jack Henry.

The next thing was compatibility with our phone system. We wanted it to to be easy, and we wanted a ease of implementation.

I started, my IT partners with me in the process early because I wanted to make sure that they were comfortable with the vendors that we were looking at, and they could ask all of their questions and and and feel that, they had everything that they needed to to make a smooth implementation.

And then the last thing was the service support and maintenance of our partner.

You know, what did that look like? Because this is a new space for us. And, we, you know, we don’t have AI capabilities. So being able to have that firm service support and maintenance was a key thing also that I looked at when I was evaluating the vendors from my shortlist.

Got it. Got it. No. I appreciate, expanding more details on the vendor selection itself before we move on to the implementation. A couple of questions there, Kellen, if you don’t mind, on the vendor selection process.

There are different options there. Like, for example, you you know, there’s people who would only provide, AI on the chat. There are people who only provide AI on the voice or people who do both. And seems like voice was very important because of the pandemic call increase and things like that.

And you were specifically looking for vendors who provide the a a AI assistant on the contact center, which is a voice aspect. Right? So, that that that that makes sense. And, just curious, one of the aspects you touched upon is the integrations and and, you know, the you know, especially you are a Jack Henry shop, so business integration was very key.

You know, so post implementation, and we’ll go go to that, during your selection, process, what are the things that you cared about post implementation? I know you’re a a a kind of a big fan of, like, the run and maintain process that interface run, but there are other vendors who give you self-service tools where you can manage AI learning. But in Interface’s case, there there’s fully managed service. So can you specifically talk about that selection criteria, during the vendor selection process, how much it meant to you and why?

Oh, absolutely.

You know, two things. First of all, with, it it it was important to us to have somebody that could really work with our processor without us having to get too involved.

And and that was the key thing for us with with the interface because, you guys are VIP vendor.

And then the other key thing was the managed services. You know, a a lot of times when you implement a new program like, a digital system, something that’s brand new, you know, you get excited about it, but then you can’t really leverage it and and improve and enhance it, because you’re having to do everything yourself. What really spoke to us, about Interface was the fact that with the service and maintenance and support, you all kind of led us through that, you know, doing everything.

If if you get a new system and it’s kinda left up to you to learn it and do everything, Sometimes it it takes a long time, you know, to get up to speed, to be able to, you know, figure out what you need to do and and figure out how you can improve it. The the managed services and support and maintenance, you know, takes care of all of that.

You guys, you know, kind of foresee what, you know, is out there that we need to add and experiences.

And, you know, we can go about, you know, looking at how we can serve our members. So that was just key for us to make sure that we could continually improve the program and we would stall in our progress of continually improving our digital assistant.

Got it. Got it. Thank you, Caroline. So, I mean, one one of the new aspect of AI, it’s a learning software.

Right? And, you know, there’s a continuous investment required to continue to make sure, the AI learns. And there is a a continued ROI that is more benefits that you know, more automation is possible, but it requires in continuous investment. It’s not like online banking.

You set up and forget. It’s an AI that continues to learn. There is, some involvement, to continue to, reap more benefits out of that. Right?

So, and your team was very visionary in that aspect. Like, you know, you saw that, and you chose kind of the managed service, one of the aspects you picked into Face AI. And I remember very closely, this is, twenty twenty. We both were chatting quite a bit, and, your team actually did a very thorough analysis.

We sent you a phone number to call, and a lot of your teams actually called that number. I remember you giving us feedback on the voice of the AI, like how human it is and things like that. Do you do you wanna, spend some time talking about other aspects beyond integration, managed services, specifically as as as the AI itself? What were options you had?

What did you like about interface and the meta selection process?

Right. You’re right. We did call it. And, you know, getting back to that thing of having a quality digital assistant.

Yeah. You know, everybody knows it’s a chatbot. They know it’s not a person. And so, you know, really being able to to have it sound more human human sounding, if that makes sense.

You know, when we called called the interface spot, first of all, kinda like, Bruce demonstrated earlier. You could ask multiple questions. That’s what members do when they call a agent. They ask multiple questions. You could ask multiple case questions. It was almost like you were engaging, you know, with a human.

And and we really didn’t like the fact that, you know, we could take a male bot as opposed to a female bot. You guys had both.

I I think when we were looking at some of the other, vendors, they just had the female voice. But, you know, what we really liked, about the digital assistant from AI was that it it was even though we you know, it’s a bot. You know, it’s a digital assistant. It was like engaging with a person.

And we felt like that that would be be meaningful, and it would make our members more receptive to to using a digital assistant.

Got it. Got it. Any other, any other, you know, tips you wanna share for the vendor selection process, Caroline, or should we move forward for the implementation process?

I think that that is probably it. I I would say that, you know, for us also, you know, we talked about the integration, but also the willingness to work with other other vendors, because at the end of the day, when you’re looking at getting more automation, you’re going to have to have those transactional experiences.

So having a vendor that, has worked with multiple partners, is willing to work with other partners, and that, you know, quite a honestly, like, you guys have done is willing to do a a little bit of the leg work to to help you advance, I think, is really valuable.

Thank you. Thank you, Caroline.

So just moving forward with the implementation process. So, I I I still remember I was very closely involved as well. This was an exciting time. We we got the Yeah. The project started. So we we start the implementation process.

You know, there was you have an amazing team. You know, Cherry and and they were absolutely amazing in terms of leading the project.

So just curious, you know, if if you if you would like to take us through, you know, what was the implementation process like? You know, let’s start at the beginning, you know, from the kickoff call, and and whatnot. So would you would you mind sharing more details there?

Absolutely. Because it it was really a great process. The the first thing I’ll say is that you guys did a great job in leading the way with a very disciplined and organized implementation process. And, of course, since this was new to us, that was really important.

It, it first started with you all looking at the data and I thought that was so valuable because you didn’t make an assumption about what experiences, are, you know, what questions were being asked. You wanted to to look at our data. And and quite honestly, we we we didn’t have very sophisticated data, but you guys worked with us to understand, well, you know, what types of calls are you getting? What are the top five calls, the top ten calls? I think we probably started out with maybe fifteen or so experiences. But you all work with us with the data that we had, which was basically just kinda tagging phone calls, as to what we needed to work with and what we needed to start with.

Then you guys took us through a two day workshop, which I thought was so valuable because you you you asked questions. Again, you didn’t assume. You ask questions. You learned about our systems.

You asked about how our members responded, and and you really took time to understand that those frequent types of calls, the top calls, and and how we, as agents or how the agents would actually resolve the inquiries.

Then you guys took all of that and then came back with a, functional I think it was called a functional specification design document.

And and you all did all the work. You know? You you took everything. I mean, you you listen to us. You took everything. You put this all in a document.

And it had the copy. It had the verbiage. It had the flow of the experiences as you guys recommended it for each inquiry.

And and that made it it’s so easy for us. You know, if we had had to do all of that, we probably would still be implemented.

But what was really cool about that is that you guys did that work, but we had full control on the customization. And I do remember when we were working with your team, we we drafted several iterations because, my contact center VP, she can kinda be a little bit picky sometimes about how things are saying. You guys did all of those revisions, and then we when we finally got to a a good document that we could all be happy with, then you get you did all the program. You did all the programming in in, good fashion.

And then we did our user acceptance testing, you know, which was on a customized number, and we were able to then move into production. It was really, really a great process. And you guys kept us accountable too. You know, if we dragged our feet a little bit, you all kept us accountable. And, I like that as well. But it was, it was really a great implementation process. Very organized.

Thank you, Caroline. And and there were a a a few more tracks beyond, the the specific, product implementation, you know, related to marketing, you know, related to branding. Like, for example, would you mind shedding some light on, how you prepare, to educate, the members, before the launch? And, you know, you you guys did a amazing job creating a landing page on the website and including Nick in the newsletter. So would you mind, telling us a little bit more about, that process of member education and and branding and picking the name, voice versus female? So there there are couple of tracks beyond the implementation.

Would you mind sharing some details on it?

Yes. Well, I I mentioned, you know, when we listen to the voices, we we we liked we liked the male voice better. We we thought that would be different too. We we had, an IVR, you know, like like most everybody has, and it had a female voice, but we thought the male voice would be different.

It it was around Christmas time. I I really do think that’s why, you know, that probably, you know, it’s kind of spurred Nick in some people’s minds. But my marketing folks say it it it Nick is not based on anything. We just thought it was a real cool name. Maybe it becomes, you know, Nick neighborhood.

Neighborhood. That’s what I thought.

They they picked Nick and and and and and gave that, digital assistant, a personality with with Nick.

We did it as you mentioned, Shrinivas. We created a a custom landing page, you know, with some visuals so you can kinda visualize visualize different mix.

We kinda change that out. And, we, educated our members with, newsletters.

In fact, we still have a, monthly article about Nick, so they can understand, you know, what new experiences Nick may have and, you know, what he can do. But, we promoted Nick in our newsletter, job members and then with several several emails.

But the probably the the greatest thing was, you know, having them just call and and be able to experience Nick.

And we did a lot of that in the contact center, as well.

Just kinda giving him that personality.

Got it. Got it. And I really appreciate for sharing more details on the member education preparation you you did and whatnot.

So, you know, so moving on to the next question, you know, with respect to the implementation, what was internally the preparation had to happen? Like, for example, resource allocation and from different teams. Would you if if you don’t mind, you know, talk more about different kinds of, teams that needed to get involved and and how much time they had to dedicate on such a project. And if you had to hire anyone new at all, like, with specific kind of skill set, like a some sort of an AI engineer or some sort of a new headcount to manage AI, or to manage this project. So, would you mind starting with, like, the different teams involved and the effort, and then there was any new hire or any new specialized skills had to build during the implementation?

Oh, sure.

So the specific teams that were, intricately involved in launching Nick was, of course, our contact center because those are the the folks that, you know, are going to be interfacing with members that interface with Nick potentially, and they give us the, the input as to how we need to to build Nick. So our contact center, our marketing team, because they promote obviously, Nick and then communications.

And then, our IT folks specifically for, our phone technician. And then for some of the integrations that we have with our core core system, we had a few, IT people involved for that.

And then, of course, compliance.

And and the great thing about, interface is that our compliance folks didn’t have, much issue. You know, your your clients people always, seem to have some issues, but, fortunately, because of the forward, like, forward thinking, and, forward thinking of Interface, we didn’t have, very many of any issues. I really don’t don’t recall any issues, to to be honest. All the proper verifications were taking place on the transactional, side. But, anyway, those those were the teams, that were involved with, implementing implementing Nick.

Okay.

As far as, you know, having to hire other people, or additional people, that is the great thing about interface. Again, because of the the managed service, module that, you know, was a part of of, our implementation.

We did not have to hire anybody, and we still we’ve we’ve been up a year, and we still haven’t hired anyone yet. My contact center VP Sherry, is one of the the main stakeholders.

Then my phone technician, from the IT side and then myself from an administrative standpoint.

As we, move toward our eventual goal of automation and doing more with AI, you know, we may hire some staff, but I will say that, you know, we haven’t had to do that yet. We’ve been able to make everything work. And and that was different from some other, partners that or some other vendors that we looked at because, we would have probably had to hire someone because of not having that managed service aspect, that interface that you all offer, Trinovox.

Got it. Thank you, Caroline. And, specifically, one of the aspects that get people nervous about is, you know, core integration. Right? So, and how compatible is that and and, you know, how easy or how much involvement they need to have on their end and whatnot.

So you’re on Episys. You’re you know, a lot of our prospects are also evaluating such a product, you know, about five hundred plus now. Many of them are on Episys as well. So, would you mind sharing some, details on how hard for, interface team to, plug into Episys core?

Was there any, like, significant investment in time and effort required from your team to do that?

You know, would you mind share shedding some light on that, Carol?

Absolutely. And and the answer to that is absolutely not. If if you have emphasis, then you have, SIM exchange, and that is what is used to to plug into interface. So, we had used that same platform with the smart teller system that we had, and so simply just had to plug into that.

And our people along with the interface team, are familiar with that. So, you know, no additional coding or anything like that. That and and that made it really easy. That that was really a win too because we knew that functionality. We knew, what that entailed. That’s the same really, really made it easy.

Got it. Was there any additional cost, from any other vendors that, you know, creating as you’re listening to this conversation should factor in beyond, interface AI cost? Like, was there any additional cost from the core vendor, or your telephone vendor that you had to factor in, into the overall project?

No, Shunabhas. It wasn’t because for our for our phone system, since what we did, we just simply, put interface on top of our phone system. It was a simple routing.

Mhmm.

There wasn’t any additional cost from our phone vendor to do that. We didn’t need integration with our phone vendor because interface is just sitting on top of that. So, that we did not have any additional calls from any of our partners, in the interface implementation.

Wonderful. Thank thanks a lot. And, Caroline, are you guys, I I forgot about this. Are you hosted for Episys or on prem?

We are hosted. We’re on the EASE platform. K. And so, we are hosted if, you know, some of the folks are are out there with that. Is still no additional calls. Just plug in to SIM Exchange.

Yeah. Yeah. And and, of course, if you’re hosted, it kind of make made things a lot easier for you because we already have a VPN tunnel with ease. But if you’re on prem, typically, other credit unions will listen to this.

You need to establish a VPN tunnel, beyond the same exchange. So something to keep in mind. Still pretty straightforward. And thank you, Calin, for sharing that.

So, and and was there any sort of preparation, in terms of the mindset of the team and the rest of the management who are getting involved in implementation launch of such a project, for example?

Would you have to educate your staff?

You know, this is not we’re not trying to replace any positions or, like, how did you present, the the the the, you know, the per the the kind of the concerns they may have is AI here to replace the jobs or, you know, and and in terms of can you elaborate on that if if you don’t mind the the topic personally?

Absolutely. Because we have several sets of stakeholders that we had to communicate with a little bit differently. Yeah. Of course, the executive team, you know, everyone has probably, or everyone had used some type of digital assistant with other businesses, so they were well familiar. And, we all saw from the executive level, the viability. So, honestly, there was not much, education that I had to do in that arena other than to present the, you know, the the advantages and and the features of of each vendor because everyone was was kind of acclimated and aware of that.

Our managers, were a little bit more, unaware.

We we had to kinda talk about what AI was, kinda compare some of the AI that they may be using in their personal lives.

But then talk about how, you know, it could help us with consistent, consistently managing our staff, being able to have better call volume, and being able to handle all of that. And so that was part of the education we did to bring them on board.

Of course, when we’re talking with staff, just did you mention, you know, of course, the first thing that, you know, people think of, well, am I gonna lose my job? And so we we did, you know, tell staff and really kind of present to them how, no. Nobody’s gonna lose their job. What this allows us to do is to be able to give more time to our members.

And so if interface at the digital assistant, Nick, takes care of these routine inquiries, then you can spend more time with the members. And, you know, agents in the contact center, that’s that’s really a value to them. They want to take care of their members. And I think that probably was what spoke to them the most is that, you know, I’ll be able to have more time with with the members.

You know, I won’t be a stress because, you know, some of those calls that I might get, they’re gonna go to Nick, and that really spoke to them. Some of them had had experience with digital assistants and some positive, some not so positive. But, you know, they they it makes sense to them that we’ll be able to help our members, and I think they were confident that we were choosing a a good quality digital assistant and would help our members. The the last audience was our board members.

And, sometimes the board members are the ones you really have to sell, but I have to tell you, I didn’t have to do too much with my board members. Again, you know, I introduced the concept of AI, what it was, you know, how they’re using it in their personal life. And then we, you know, transitioned over to okay. This is how AI is going to help us, in a better member experience and being able to help, you know, board members, get their answers very quickly and then to help us manage staff, you know, when we, you know, can’t get staff as quickly as as we need to.

So, a few different nuances, but I I think the the overall overarching goal was the same. You know, this is gonna help us with our members, and, you know, we’re going to do it by having a quality digital assistant.

Got it. Got it. I mean, picking a a quality digital assistant is is benefits and, you know, it’s a win win. Right? So better member experience more automation.

It it just is a win win in every, way. It it, you know, leads to better savings and justify the cost of ownership of such a a product too. So you talked about how AI learned. So kind of moving post implementation.

The the AI is live now and, you know, as members are using.

You know, let’s start with was there, you know, any, feedback from the members right after the launch with with that? Because it’s a big change. Some of the credit unions kind of think about, hey. You know what? Let me launch it after hours or let let me only launch it to few number of calls twenty four seven, like, you know, five percent, ten percent of the calls before I roll it out, to all the calls.

But in your case, we launched it to twenty four seven, all the members.

It’s a it’s a big release, from from day one. So just curious, was there any negative member feedback, right after the launch? And, would you mind sharing some details on that?

No. I don’t mind at all. You’re right. We’ve launched it to everybody. We we like to do things big here at at Neighborhood Credit Union.

And, you know, surprisingly, you know, we we we we didn’t get any feedback.

Initially, no feedback.

And so, you know, we thought, okay. Well, the members that, you know, will use Nick, they’re getting help. The members that, you know, prefer to talk to a person, they’re they’re getting to a person.

So, you know, I just did a little bit of informal, you know, asking members that I knew that had used Nick. And and and it was positive. They said, oh, yeah. I like Nick.

But I did receive one complaint. I one of my long time members, he called me up and he said, Caroline, you know, this this Nick, you know, it’s gonna cause all the members to close their accounts. I think I’m gonna have to close my account that this is just not good. It’s not good.

And and so, you know, I I talked with him and I said, well, just just try Nick. I said, you know, Nick is, you know, he’s available after hours. He’s gonna help you with all your balances. I know you like to get your transaction history.

He didn’t really like to use digital banking. I said, just just try it, just try him. And I said, I I I don’t think that, you know, that the member’s gonna close their accounts. I said, this is a way to have twenty four seven access, to your accounts, as much as you need and enter other inquiries that you might have.

So, you know, he, gradually consented.

And then I think probably about a month later, he called me back, and he said, well, Caroline, you know, I did try Nick. And and I think Nick’s pretty good. And and obviously, he he didn’t close his account, and we didn’t have vast closures of accounts. But it was interesting to me because he was so resistant at first. And then, you know, it it he did turn around his behavior because he saw the convenience.

And, you know, he became a a fan of dicks, you know? So, yeah, you’ll have some members that, you know, won’t they don’t want to use the digital assistant, and those people will call your agents and that’s fine. But I really do think that the people that use Nick, and experience, you know, how, convenient he is and how accessible he is. You know, they’re they love it.

Alright. Thank you, Colin. Real quick. Wanna, jump to kind of talking about the impact of Nick so far.

Right? So and and and Nick so far has automated about forty percent of the cost consistently, every single month. You know, and it took us a while to get there because you were one of our first customers going live. So the AI has to learn quite a bit and things like that.

So, but in the recent past, we’re able to get this far very quickly. And thanks to your team and having Nick and AI learn, through through the experience in the early days.

So real quick, wanted to, bring up a snapshot of the benefit kind of the, return on investment and the, impact of Nick that has been so far. It’s been about, four hundred and thirty days, Caroline, that we launched Nick, and time Oh my god.

Yeah. Time flies really fast. And, about ten thousands ten ten thousand hours of, staff time saved with Nick answering the calls, that leads to about about two hundred thousand costs Nick has completely handled itself.

And, the the the savings gained is measured in terms of, you know, zoomed cost per call.

You know, typically on an industry average, it’s just usually about ten dollars per call multiplied by the number of calls AI itself handles. So, but you, you know, you are able to also achieve efficiency, especially with your challenge with, not being able to hire fast enough when there’s a turnover in the call center, which happens a lot. And through that, you you’re also reduced their overall head count. Right? So it’s about by four, I believe. Callan, you want to say something?

We we we were. When we implemented Nick, we were able to attrition because we we didn’t want anybody to lose their job. But through attrition, we were able to, eliminate, four, four staff members.

And, you know, we’ve been able to maintain service levels since then. So we we we, did all of that, I think, toward the end of twenty twenty one. So this was actually not even recently, we were able to do that by the end of twenty twenty one was to reduce that, forehead count, from our contact center.

And that that was that was substantial. That was substantial.

Thanks a lot, Callan. So those that’s kind of the, end of the session here, letting you know about the impact, after Nick as well and the whole journey from, the beginning of finding a vendor I mean, thinking about the problem to solve, finding a vendor, going through the implementation, the support, Neighbourhood got from Interface on the post launch, and the impact eventually they gained. So really quick, Calin, if you don’t mind, let’s open it up for questions. We got five more minutes, for for the session here.

So I’m just gonna pull up, the q and a tab here, look for some of the questions that came in.

So let’s go from the top.

Give me one second.

So were you able to adjust the contact center hours in AI? And if so, how long before you did mind or adjust your hours? Also, you have been able to define what kind of impact AI has, had on the NPS. The first part of the question, Caroline, you already answered, which is you lost AI for twenty four seven on day one.

Right? So and and the second part of the question is new, which is I think you probably answered it. What kind of impact AI had on your NPS?

Well, we we don’t, measure NPS specifically, just, you know, member satisfaction.

So we don’t we don’t really look at that score. But, when we had I will say this. We had a a member survey where they did do the NPS, at the end of last year, and, our MPS actually went up, just to own our overall member survey.

Got it. Thank you. Moving to the next question here.

Were you able to sorry. That’s the same one. Would you be sending the copy of this recording? Absolutely. Yes.

What percentage of staff were able to downsize with their addition of AI? So you already answered the challenge, so we’ll move to the next question.

So give me one second.

Does Neighbor, Credit Union have live agents twenty four seven to handle calls that Nick cannot handle? So is the question is, do you have staff, twenty even after hours, to handle questions Nick cannot handle?

We do not, but we do have some, agreements with some of our vendors where they have some staff on hand for, instance, our card processor. They have staff for specific card, related, inquiries where if Nick can’t answer that, then we transfer to our, card processor vendor, but we do not have any staff on hand. If it’s something that cannot be transferred to our, partners, then, we take care of that member, the next day.

Got it. So we we got couple of minutes, probably a few more questions here. And, so this is about the learning of, Nick, Callan. So the the question is, is the machine learning continuous for Nick outside of scripted q and a developed during implementation?

So, you know, there are two aspects of learning, and and this question kind of, we gotta split it out, kind of talk about. One is, do do your team has to train AI with specific ways how your members are asking questions?

Like, for example, what’s my balance? Could be asking, like, hundred different ways. Like, how much money I have, how much, how much, change I got. You know, could be asking, ten different ways.

Is your team today investing any time telling AI and machine learning saying, hey. Here are the different ways my, members are asking these questions. That’s one part of the question. The second part of the question is when AI gets to learn new questions, is that automated, or is there a manual intervention as in you pick the questions and the interface AI team trains it?

So would you mind shedding light on both aspects of learning, and the first being different ways of asking questions? Is neighborhood doing anything with that?

We we actually aren’t. What we find is that, you know, that the AI can pick up the different ways that the question is being asked. So, you know, we’re really not, trying to train our members. Ask it like this, ask it like that because AI can can pick up can can pick up the keywords.

And and and it can make the distinction. And as it learns, it can learn that, you know, the different ways that members are asking for different, inquiries and different experiences.

As far as new things that, you know, that are coming, you know, we we do work with, Interface. You know, if there is an experience that we need to build, a lot of times Interface will pick that up before we say something.

But but we work, to build those those new experiences if we have to, you know, do some, additional sweeping.

Yeah. Some context, the reason this question is coming, Kalyn, because so the vendors out there, you actually get a tool. You have to train every single way members can ask questions to AI. So Oh, really?

Oh my gosh. Yes. So there is a number of vendors where creating this. Unfortunately, many of them move to interface now that they actually have to train, invest time, have team training every single way members are asking questions.

In your case, you’re not doing that at all, so that’s just We don’t have to do that.

Got it. I think we’re about time. So, thank you, for everyone for the participation.

And, Kalyn and Ginny, it it means a lot for us for both of you being here.

And thank you for sharing your, journey with us, Callan. I I really appreciate it. Thank you.

Thank you. It’s been a pleasure. And thank you, Jenny. And, to the audience out there, thank you all for attending. It’s been great to share.

I think we’re off the live. Yeah. So, Kellen, appreciate, spending time. I think we’re offline. Right, Vivek?

I think that’s I still see that maybe Yeah.

That’s the see not We have nine more attendees.

Okay. I think we could end the bridge and yeah. Yes. Thank you.

Thank you. Thank you.