Name: Fin & Plaid: The AI-First Blueprint for Fintech Support
Uploaded: 2026-08-13T18:51:13.624Z
Duration: 32 min 28 s
Description: Fin & Plaid: The AI-First Blueprint for Fintech Support

Transcript for "Fin & Plaid: The AI-First Blueprint for Fintech Support": Hello. Hi, everyone. We'll give it one minute, and then we'll kick off. Emily, you have a pro headshot. I need to get, an updated headshot. I know. Right? Oh, yeah. My shirt says customers, customers, customers on it, so maybe I should send you one of those. Yeah. That would be amazing. That's I would love it. Very on brand. Yeah. Matching, headshots. You know? Cool. Well, I am to answer Alex's question, I'm joining from a little bit south of Boston, Massachusetts, so near the beach. And I am in San Francisco. I guess the whole city is kinda close to a beach. Not an not a nice beach, but That's true. We have warmer weather right now, I'd say, than SF. But. Not warm. Not warm today. cool. Alright. Well, I think we should just kick it off. I wanted to welcome everyone to the AI First Blueprint for Fintech Support webinar. I am Emily Harr, chief customer officer at Plaid. And I wanted to start by just sharing a little bit about what and who Plaid is, and then I'll turn it over to Junan. So what is Plaid? Plaid is really building the infrastructure for AI to work well in financial services. And we act as that trust layer that drives consumer adoption, and we, are connecting more than one in two US banked adults to over 9,000 apps. And Plaid also provides AI ready financial data through our more than 12,000 financial institutions and over 1,000,000 daily connections. So we'll talk about that today, and I'll turn it over to Junan for an intro, to Finn. Hey, everybody. My name is Janan Pang, and so I lead our solutions and customer success, functions here at Fin. So a bit about what Fin is, our mission is to make perfect customer experiences, the new standard for all businesses. And so, Fin is a solution, on the market that delivers for a modern CX team. And so we've got a world class AI agent that is called Fin, as well as the same name as the company, and a natively integrated, human help desk. And so together, we have 30,000 customers and 10,000 of them are using Fin as their AI agent for CX. And so, you know, this this webinar was really inspired by kind of the partnership of Plaid and Fin kind of working together. And so maybe, you know, as we get started, I wanna I wanna kinda kick off there and start here. So Platt and Fin, we actually share hundreds of mutual customers. And so when we started working together, we knew there was something, that, you know, our our customers cared a lot about and could gain a lot and get a lot of value from as if, you know, when we, you know, built an integration. And so, as as Emily mentioned, you know, Plaid has built this trusted, financial connectivity. And then Fin, has trusted experiences, and can bring that directly into a customer conversation. And so if you combine those two things together, you know, the goal here is like, how do you make support feel like like a very seamless extension of the product? And so when we started out to kinda partner together, you know, where we focused was, you know, I think the highest friction step in fintech, which is bank connectivity and kinda like the onboarding process. And so if you think about, you know, when a user's bank link fails, or, you know, a balance doesn't show, there's a lot of anxiety there. And so they will immediately open up a support ticket, support conversation. And then that goes in and requires a human to, you know, diagnose the error, troubleshoot that and help them through that. It also just blocks the, the value that like a fintech can provide if you don't get that set up right up front. So, with that, let's take a quick look at what this integration does. And so here you can see Fin. This is the support conversation. The user is saying they wanna link their bank account. And so Fin is actually going to generate a one time unique, connector into Plaid. And then this is Plaid's hosted link experience. So this is where Plaid helps you securely connect your bank, and be able to achieve your goal here all within the Fins support flow. And so then once you finish that, it's gonna take you back to Fin and then seamlessly, just right in the same experience, it will be connected. That's it. Plain and simple. Very, very quick. Great customer experience. Alright. So let's, let's get back to the webinar. We wanted so as Emily and I were talking, we wanted to kinda take this this webinar a little bit, in a different direction than, you know, what you might expect from a typical webinar. So, we we both have very similar roles. We talk to tons of customers and are kind of in this, like, you know, in the middle of, like, helping our customers kind of evolve and transform. And so so we wanted to take a a moment to actually just, like, spend some time to just chat and and to talk through kind of our experiences, what we are seeing. And so today, we're gonna keep it almost no slides, actually, no slides, and really just kind of talk through what we're experiencing. So there's three major areas that we'll cover. The first area is just velocity. And so how does AI help customers get answers and resolve their issues faster? The second thing is then security and why trust, permission data, governance, that's all so essential for AI in financial services. And then the third area is accuracy. So, you know, how trusted financial data really enables AI to deliver, you know, reliable, contextual, and just like actionable support, for customers. So with that, Emily, I'll I'll start off with maybe, like, first topic, first question. You sit at the intersection of thousands of businesses that build on top of Plaid. You know, what is the conversation that you keep having with customers about support and and customer experience? Let's start there. Yeah. Good question. I would say the one conversation that we keep having is just generally that customer expectations have really changed. You know, if you look back ten years ago, customers, of course, wanted digital experiences and fewer clicks and less friction. But now they expect, like, instant personalized support that helps them achieve their goals immediately. And they don't necessarily separate your support experience from your product. So if they can't connect an account, or complete a payment, that's the product experience that they're going to remember. And so in order to meet these new kind of customer expectations, you have to think about AI as the enabler, not the strategy in and of itself. Better customer experiences is the strategy, and I preach this a lot on my own team. Like, we should measure AI success by the customer outcomes that it creates, not that people are using AI or that we're, you know, spending thousands of dollars on our tokens. And so to really improve the consumer experience in financial services, you have to ensure that your AI efforts operate on a very secure, permissioned, like, trusted financial dataset. And you talked about this a little earlier. Like, consumers are, they're trusting us with their financial lives. It's it's emotional, and it's urgent for them. And we can really use AI in in this partnership too to help support their needs. Yeah. I love that. Like, you know, it's it remember what we're all doing all of this for. It is not just to use AI for AI's sake. Let's not let's not lose the plot there. So I think that's that's really helpful. You you talked about kind of the evolution. Like, how do you think about evolution and like how things have evolved from like the old school clunky click all the buttons chat bots to kind of like AI agents that you're seeing kind of like like Fin like. powering customer support. How do you think about Totally. I don't know about? you, but it's, like, very obvious when companies are using the clunky chatbots versus the AI stuff. But, generally, at Plaid, you know, we believe that the the future of AI isn't about just chatbots answering more and more questions. It, again, is really about helping customers accomplish their goals faster without having to go back and forth with chatbots. You you'll create a better experience when AI has that trusted financial context to move from just that back and forth conversation to actual resolution with, like, little to no friction. If we think back to kind of the, demo that we showed earlier, that's exactly what's happening within the Fin app. Yeah. I feel like this is gonna be a theme throughout the conversation, which is just, like, don't forget the outcome you're trying to get you you get the customer to. Okay. So, given the context that, you know, giving giving AI context is, you know, so important. But, like, there is a journey that organizations do need to go on in order to kind of, like, do this right. And you kinda mentioned, like, it's very obvious when there's a company that is, like, still using kind of the old clunky thing, and hasn't gone through that evolution yet. But it isn't it isn't, it's not just like deploy the software. There is like evolution that does need to happen. Let's actually take a moment. I wanna do a poll for the for everybody that's attending. So where do you feel like your organization sits on this curve today? And so like what we kept it very simple, are you AI primary, meaning AI handles the majority of customer interactions, you've gone through that full evolution transformation. Middle of the curve, AI handles the easy stuff, humans still handle at least half, maybe a lot of it, or you're just getting started, the majority of interactions are still handled by a human, we'd love to see. Alright, so the votes are coming in. We'll get some we'll get some lead changes in the in the results. It's very, across. yeah. Yeah. Yeah. So, yeah, it it feels like it's almost completely balanced across, you know, all three categories. Emily, any any takes on that? You know, I I think a lot I think even Plaid and probably Finn, like, we're all sort of, like, in the middle of our journey where everything's changing so quickly. And so it feels like while we have some things in place, it's always changing. So maybe we go back to getting started, and then maybe we're AI primary. It probably flips a lot, but I feel like. sometimes, in I think that reflects how I feel. You know, every every day you feel like you have it figured out, something changes and you you kinda have to Yeah. continue to iterate. It it does feel like, you know, the middle of the curve, if you're if you're sitting there, like, you know, the biggest thing is to, like, not let yourself plateau there and continue pushing. And then getting started is, like, you gotta get started somewhere. So, like, I think this is this is great to see. Okay. So, like, on that note, Emily, what do you think about, like, what is stopping organizations from, like, moving from whether you're sitting and getting started and get trying to get to middle of the curve or middle of the curve to AI primary. Like, what is stopping organizations from moving into that next stage? Is it is it. trust? Is data access? Like, what what are you seeing? Yeah. Good question. I talk to a lot of, like, support and customer success leaders, and I feel like there are kind of three themes of what's stopping them. One is their trust in AI data and in agents, which we're talking a lot about today. Two is their risk tolerance to experiment and try new things, especially as AI is evolving every minute of every day. And I would say the third is the internal resources to build with those AI tools. You know, who's is it a specialized team? Is it just support team in and of itself? So it's really that trust, risk, experimentation, and, having resources to build. Yeah. What. do you I'm super aligned with that, and I think, like, I think what we've seen in in working with customers is it does start with, like, an executive or a leader or somebody that's championing it, having a vision, and then clearly communicating that across kind of the team or the organization. Because I think a lot of the things that you you talked about, they're actually, like, it kinda comes down to the teams behind, you know, the technology. Right? And so, they need to kinda believe and want to change and drive that transformation. Otherwise, it ends up just being like a bolted on, like I'll just kind of stick it at the end, Yeah. type solution. And that will kind of force you into a plateauing if you don't think through the organizational structures, the processes, the data and the systems that need to be connected in for, you know, an AI agent to, like, truly thrive and deliver a good experience. Yeah. It's. So speaking. of transformation and evolution, customer expectations, you talked about this in the very beginning about, like, how they, like, it it the speed like, they they're expecting more now. How. have you seen things change, on the customer side? Yeah. It's exact exactly that. Right? Expecting more. We're seeing an increased need for support to be immediate, to be accurate, to be twenty four seven. And, you know, we continue to see AI evolve so quickly. But, again, it's not just about, like, answering questions. The real value now in using AI is to really help customers actually complete their financial tasks and to resolve issues and accomplish what they came to do very quickly and without friction. So, yeah, that's kinda what I think. Immediate, accurate, no friction, and helping them actually resolve the issue. Yeah. Get to that outcome. Exactly. The, I wanna share a little bit about, like, you know, how we help customers, think through that challenge. Because I think, like, if you're let's say you're aligned to okay. We gotta go solve the customers out get to help our customers get to that outcome. You know, the first question becomes, okay. Well, like, well, how do I do that? I have tens of thousands of support issues, and they're all not shaped the same. And so, like, what does that journey look like? And so one, you know, framework or model that we help customers through is what we call the conversation taxonomy. And it's basically, you know, if you take all of your support volume and you start to break it down into categories, you start to figure out, you know, the the different ways that you need to support, those customers and the ways you you kinda solve for that. And so you'll have a bucket of, support conversations that are very like informational. It's kinda like FAQ style things. Like I didn't read the docs, so I just, like, tell me the answer kind of thing. And, Yep. then you'll have, like, the next tranche of data that is, you know, what we call personalized. Meaning, like, in order for for you to, like, actually get that customer to the outcome they're looking for, you actually need to know a little bit about their information, their, you know, their account, things like that. And then the last, the third bucket is then like you actually need to take action on that user, for that user. And so, helping them through kind of the next step to get to their outcome. And then the last kind of like maybe most hairy bucket is like troubleshooting. And so, you know, if you think of things in that like informational personalized action and then troubleshooting, you can then start to think about and align like what are the systems, what are the knowledge, what's the content, what are the processes that support answering and getting to those outcomes? And so, you know, things like reconnecting an account, understanding a transaction, you know, resolving a failed payment, those are all things that, like, they fit into, you know, those four categories and so, we want to be able to kinda help solve those kind of things. So, I think the the role of human support, you know, if you you start to kinda solve those those four categories, the role of human support is actually starting to evolve as well. And so one of the things that we've seen at Fin is, like, we don't we don't see AI replacing people. We actually we see AI actually. We see Fin handling a lot of, you know, the things, so that way humans can go spend their time doing something higher leverage, higher value, more complex building relationships, be more proactive I think that's like the holy grail of customer experience right is like how do you be more proactive right? And so I think, like, when we think about velocity in, like, this first bucket is it's like, it's how do you reduce customer effort and get them to the outcome, that they're looking for. Yeah. Totally. I really like what you said about, conversation taxonomy. I I've haven't heard that before. And then I totally agree on the human support. I've actually seen many of our team members, like, really enjoy, having some of their work repeatable work done by AI so that they can build new things with AI. It's been really exciting to have them contribute to reducing customer effort for sure. Yeah. Yeah. I mean, it unlocks new skills, new roles. I was talking to a customer the other day where they actually have started to shift some of their support team into, almost more like engineering style roles where they're actually, like, solving bugs now, which is just such a cool unlock, for for CX organizations. Yeah. It's awesome. So, Emily, like, maybe shifting gears, like, where do you see kind of, like, the biggest opportunity? So we we talked about, like, the experience itself. Like, in financial service, where where's the biggest opportunity, to improve the customer experience? Yeah. I would say the biggest opportunity is to put that AI powered support and all the content and all the insights, like, upfront so that the customers may not even have to file that ticket. I don't know about you or anybody else on this call, but I never wanna have to file a support ticket if I don't have to. Who's with me on that? And so you really wanna just meet the customers, like, where they are at the moment that they need help. And so the more we can empower our customers to self serve, the better. But when they do need to to file that ticket, you wanna make sure that that experience is is pretty seamless. The other thing I'll say in terms of opportunity is you wanna make sure that your your product team is actually getting the feedback and insights from the AI and from support so that they can fix bugs. But more importantly, they can build the best possible product. So, again, customers don't have to file tickets, but the product works better and better through a lot of the data we're delivering to them. Yeah. Absolutely. And I know a lot of the, the majority of the audience today is is, you know, financials fintech, building in financial services. And so we started off the conversation with, you know, the account connection, you know, challenge. I wanna do another poll. What percentage of your inbound support volume is actually is really, like, some form of is my account connected correctly? Let's see from the audience how much of that comes from this. Again, quite a mix. The diverse audience. Yeah. Totally. Well what's interesting is it feels like there's actually kind of like two peaks here there's a group that's below 30% and then there's another big group that's above 50%. Oh, actually, no. Okay. The the 30 to 50 is catching up. Yeah. Interesting. I'd say regardless, like, Plaid and our a lot of our customers experience a fairly high volume of connection issues. And this is, like, what I think is one of the biggest opportunities and and why we built this partnership with Fin. But, again, this, like, this only works if AI has the right context, which in this case is obviously the trusted financial data from Plaid. And so once the AI has that data and really understands what's happening with a customer's financial accounts, it can deliver much more accurate, relevant, and personalized experiences to help that customer solve their goals. And you mentioned some of these connection issues earlier, but a lot of the, you know, opportunities are around those big challenges like connecting a bank account or reconnecting with expired credentials or verifying an account or, like, understanding why a payment or transfer didn't didn't work correctly. And as, you know, we talked about earlier as well, like, personal finance can be really emotional and really urgent. Like, you need access to your money. You wanna protect your money. You might need to share your money with your family overseas or whatever it may be. So a lot of these kind of account connection issues are are really urgent and the reason why customers want immediate accurate answers. But more than that, they want the resolution. And they also you you talked a lot about context earlier. They also don't wanna have to explain, like, their situation over and over again. They want AI or you to understand their issues the first time, and that's where that kind of context and accuracy comes in to build that trust. Yeah. And beyond that, you know, I think that instead of kind of, like, waiting for the customers to file a ticket and all of that, the the almost better use of AI is to really identify those issues beforehand and, notify the customers and help them resolve it before it even becomes a support case. That's where you're gonna get, like, the true win win for both the customer and and your business. So yeah. Yeah. How you go? I mean, it goes back to your point on, like, you know, sometimes oh, yeah. Like, the like, maybe the the best support is no support, but it's also, like, an extension of the product. And so it kinda goes both ways in in terms of, like, you're striving towards making sure your customers get value. And just to drive home kind of the, like, the size of the impact and, you know, how Plaid is providing the right data to actually make this happen. We saw a customer that integrated Plaid into Fin and saw a 30 percentage point increase in the resolution rate and that, you know, they're probably in the like above 50%, volume kind of, answer, but like, it's just like that type of, you know, that one use case to drive that kind of impact is just wild to me. And so it just it just shows how impactful something like that is. It also just shows, like, how much how important it is to, like, really understand the shape of your, you know, support volumes. That way you can, you know, meet your customers where they're at. Yeah. Totally. That's impressive. And we're just getting started. You know? So. just a few data points already, but, on the right track. Absolutely. Okay. So let's let's move on. So financial services, I think it's kind of like feels maybe even cliche to say this, but it's just financial services companies, they it just demands such a higher standard when it comes to security and trust. And so we're gonna do our last poll for the day. How often is trust and security the blocker to AI adoption? Is it not very often, very often, or just constantly? I think you're right on the constantly. Yeah. And I don't I don't know if there's, like, a there's probably it's hard to differentiate between very often and constantly, but either way, it's it's a lot. Yeah. Yeah. I mean, this one was actually a pretty clear one. So, like, like, constantly, I would say dominated that. Emily, like, how does Plaid think about kind of balancing innovation with trust and security? Yeah. I I don't even know if I would, like, call it a balance. You know, at Plaid, we we believe that trust, like, isn't a feature of AI or of innovation, but it's, like, the foundation. You can't actually do anything without it, And AI can really only deliver meaningful value in financial services when it is built on secure, permissioned, and trusted financial data. And as we were talking about, like, consumers are trusting us with some of their most sensitive and urgent financial information. And that responsibility in and of itself really shapes how we think about every innovation including with AI. And as you mentioned, like, it's it's just different in financial services. It's not the same as AI support in other industries. You know, you're not, you're not using, you know, financial data to help reset someone's password, or, like, return a pair of shoes, both of which, you know, I I do often. But we're actually helping people manage their financial lives where accuracy, security, privacy, and speed are, like, absolutely critical. And trust is what really accelerates that. And I would say, like, the question is is always whether organizations can trust AI in these highly regulated environments. And the answer to that is when that AI is built on a secure infrastructure and has that reliable financial context, organizations can certainly move from experimenting to more being more confident in what they're deploying on the AI front. So. It's kinda like, maybe it's like a false trade off, but it is, like, to your point, it's a nonnegotiable, from, like, a a trust and security perspective. You just kinda have to find the right. infrastructure. to support it. Yeah. I totally agree. Alright. So we've talked about velocity. We've talked about security. We've talked about kind of accuracy and trust. Emily, share why does this partnership matter and, you know, where does it what does it say about kind of where financial services is headed kind of into the future? Yeah. I'm like Michael Carpenter. I love Plaid and I love Finn. So, this partnership matters because we, you know, we we both believe that the future of financial services just isn't about connecting financial accounts. It's really about helping people accomplish their financial goals with confidence. It's it's one of the reasons, honestly, I joined Plaid is because I really do believe in that mission of, you know, empowering our consumers and people like you and me to really have control and confidence about our financial lives. So, you know, that's where we're focused at Plaid on consumers connecting securely, and managing their financial lives. And and Fin helps to really extend that experience by helping these customers resolve these financial issues or issues with these apps more naturally within the conversation. So, again, we're just working together to to really meet them where they are and reduce that friction across their across their journey. Yeah. And, you know, I think financial services, just like AI, is is evolving every, you know, minute of every day, and we're really entering our next chapter here. Financial connectivity, you know, itself really transformed how consumers access financial products. But now with AI, it's really transforming how consumers, like, experience those products. And, again, it's hard to separate the two. Your support experience is the product experience. And so, you know, you have to combine all of this. You know, you need the trusted financial data with those kind of intelligent personalized interactions, and AI is what allows us to do that. And Plaid provides that trusted financial data to make those experiences not just faster, but more accurate and reliable. So to bring it back, you know, to the to the beginning of all of this, it's really about those three things that you talked about at the beginning of the webinar. It's about velocity so customers can accomplish their goals faster. It's about security so that every interaction is is built on trust. And then it's about accuracy so that AI is delivering those personalized, reliable, and, you know, actionable support experiences. So that's really it for me. I'll, I'll let you close this out. Yeah. Actually, I I shared the the the customer that had the 30 percentage point resolution rate increase. And so I don't wanna share just a cherry picked, metric. So over the past kind of few weeks, we've seen over 73,000 Cloud backed data connector calls in FIN, and it's actually delivering a resolution rate lift of over 11, percentage points. So we are excited to just continue to build on this integration and just to continue to, deliver more value for all of our mutual customers. And so if you're interested in getting the thin and plaid integration set up, we will drop a link into the chat now, and so that way you can get more info and we will also send that link out via email if you've registered through the webinar. But that, that wraps it up for us today. Thank you all so much, for joining us, and, thank you, Emily, for the for the great chat. Yeah. Thanks, everyone. Bye. See you.