FinTechTalk: Are You Actually Improving? The Benchmark Reality for Dispute Leaders

Speaker A: Foreign.

Speaker B: Welcome to FinTech Talk. I'm your host, Charles Orton Jones. The highlight of any financial year, if you're in dispute resolution, is the publication of the Quavo Report, the State of Dispute Management. This is the industry's deepest dive into all the KPI's benchmarks and numbers that the industry relies on. It provides a guide for banks, fintechs, credit unions and any other organization involved in debit payments. Credit payments, and gives, uh, them an ability to track their performance. Well, the 2026 report publication is imminent, so we've got the authors of the report onto FinTechTalk just to pull out the highlights and talk us through the numbers the industry is going to be talking about after the publication of this report. So without further ado, let me welcome from, uh, Quavo we have, um, Ike Sullenberger. Ike, if you want to turn your camera on. Ike is customer experience leader, Fraud and disputes at Quavo. Uh, for those of you who know, Quavo is a data focused financial technology company with deep expertise in fraud management. Ike, great to have you with us.

Speaker A: Yeah, great to be here.

Speaker B: And assisting with the report was Oriemma Roundtables. So we welcome Jaime. Pass. Jaime is Director of Service Delivery at Oriema. Jaime, great to have you with us.

Speaker C: Great to be here.

Speaker B: Uh, well, let's just start with what the report is. I mean, Ike, just give us an overview of this report.

Speaker A: Yeah, absolutely. Um, so this report is actually something that we started doing last year, um, and this will be kind of our second iteration of it. But because of the nature of our business, we sit on data from a number of financial institutions, um, kind of spread across segments. So we've got credit unions and fintechs and enterprise banks and community banks. Um, and with all of that data, we've been able to collect a really good baseline within kind of our network of how they're performing both operationally and at the chargeback process. Um, and then partnering with Auriama, we can compare all of those numbers across the platforms they're using with us against what we see industry wide. So it gives a really good insight into how some of the best performers are performing and where maybe opportunities might be kind of spread across different segments.

Speaker B: That's great, Jaime. I mean, ARIMA is a leading provider of peer benchmarking. Just tell us about why the industry values this report so highly. Why do people need numbers?

Speaker C: Yeah, you need to be able to, um, uh, compare yourself with the industry. Uh, as Ike was saying, we will segregate or segmentate those um, metrics that we have in place. Um, and you want to be able to, you want to be able to compare yourself with the industry, right? To see how you're doing, where you're doing, where you stack, uh, with the other institutions, whether it's in the same peer group or different peer groups.

Speaker B: That's great. I can tell you as someone who's read the report, it is absolutely stacked with numbers. When I'm pulling out stuff like how many days it should take to resolve a dispute, what automation should be handling, what losses are considered reasonable, the amount of disputed dollars that can be recaptured in a well run organization. There's just so much information in there. Uh, so Ike, perhaps you could just kick us off really. So when looking at chargeback performance today, what numbers actually tell you something useful and what numbers are getting more attention than they deserve?

Speaker A: Yeah, that's a good question. Um, there's a lot of data around chargebacks and a lot of different steps in kind of the workflow for a chargeback. Um, you know, when I look at chargeback data, I think really the most important number is, is often the simplest. You know, what's the actual amount of dollars that you're recovering versus the amount that's disputed. It kind of helps to tell a true story of the value of your chargeback process, um, your ability to avoid losses and really your ability to recover on behalf of the cardholder. Um, when I think about it, where we might be looking at things that are getting attention that they don't necessarily deserve, I think more often than any metric, um, the chargeback win rate or chargeback win percentage tends to get a lot of attention. In reality, to me, that doesn't really encompass the full story. When I hear, you know, an organization cite a 90% win rate at chargeback, what that tells me is that they're spending probably a little bit too much time submitting only cases that they consider a sure win. Um, what we've seen, you know, in years past is that in the chargeback space, quantity over quality can often be a little bit more rewarding. If you can identify really any chargeback rights, your financial outcomes generally will be better than pursuing cases, um, that really only target the ones that you're confident you'll win.

Speaker B: That's great. And Jaime, for you, what numbers would you be directing people to? What are the highlights of the report?

Speaker C: Yeah, the, uh, recapture rate, um, I believe will be the one that people want to talk about it because it goes exactly with what Ike is saying. Right? Uh, it is answering, are you getting your money back? Ah, can you how much money you're getting back? Anything. It captures the real outcome. That's just, you know, uh, the activity or speed like Ike mentioned. You know, you'll see a lot of, a lot of data there. You know, you'll see, um, you know, automation rates. You see, uh, you know, disputes per fte, you see a lot of different reports. But if you look at the uh, uh, recovery dollars, that's probably the one, the one metric that will tell you, you know, if you're, if you are getting the money back, um, on your, from your disputes, that's great.

Speaker B: I don't know whether we can just give people some numbers. I mean, I'm looking at here that, um, there's quite a variety in organizational performance that I'm seeing. It's possible to resolve claims three times faster. Some organizations, you know, have a 90% user, uh, productivity increase. Um, I don't know whether you have the numbers to hand, but just tell us about the variety in performance in organizations.

Speaker A: Yeah, I mean, it varies heavily and a lot of it comes down to, you know, your procedural process and really what you value out of. Out of the process. So when we look at chargeback specifically, um, you know, everybody wants to recover funds. That's kind of the nature of the business. Right. Um, but how we do that also impacts, you know, depending on your solution, it can impact how long it takes to resolve a case. It can impact the time it takes for your account holder to get a credit. Um, so, so it varies a lot, but it varies in a good way because it's dependent on how you value your business and what you want your members to experience. Um, as kind of a general statement. Right. You can usually expect a credit union that's member centric to focus on resolutions first, and they'll likely resolve cases a little bit quicker than you would maybe see an enterprise bank doing. So. Um, but it's a wide swing and it all just depends on how you operate your business.

Speaker B: Yeah, it is a wide swing. I've just got the numbers here. Uh, resolution timelines. The average institution takes nearly 40 days to close a claim, whereas Quavo's most agile clients get there in 12. I mean, that's a three times difference. I mean, Jaime, isn't that really what we're drawing from this report, maybe is the most important headline is that the ability to improve performance is absolutely huge. And if you've got the right numbers, you know what to aim for.

Speaker C: 100%. Uh, uh, Charles, looking at those numbers yes. You want to make sure you have an effective team, uh, working your disputes. This could be automation, for example. Uh, no one better than Quavo to do that by. How are you automating, you know, uh, are you making actually good decisions on your automation instead of bad decisions? So, so things like that is where you have to take a look at, um, making sure that you know, um, you're, you're being effective the way you're working your, your disputes. But at the end of the day it is about recovering dollars. Right. You know, are you, are you tightening those out of credit rules? Are you uh, having the uh, the correct intake, decision, intake, you know, questions for your clients? Uh, that's what' to allow you to go ahead and be able to be good at processing those disputes.

Speaker B: Yeah, I love that you mentioned automation. And again, just looking at the report, we're looking at, uh, how it's possible to eliminate 90% of manual work. And I can see the data here just showing that some clients see a 96% increase in disputes handled per full time employee each month. So it's possible to release a double the workload with the right tech. And let's just talk about therefore what companies can do. Um, I mean, uh, Ike, for companies that know they need to improve, you can't just hire more people. Where should they be investing? What is the report telling us about the best areas to invest?

Speaker A: Um, I think, I mean really it sounds simple, but it's automation. Like it's automation of tasks, it's automation of, of chargeback processes, um, leveraging APIs. You know, most of the large associations have APIs available. Um, and really when we look at chargeback as a process, a large amount of time is spent just going through chargeback. Right. Determinations, making sure that you have rights to send a case off. Um, if you can automate that, you're going to remove an enormous operational burden. Um, the cost of having someone validate chargeback rights and then swivel chair and manually input a case into an association system can really quickly exceed the value of the chargeback that you're sending. Um, there's always a balance between what it costs to do the work and when it's worth pursuing. Um, automation can really shift that balance in your favor by handling some of the heavy lifting.

Speaker B: And uh, Jaime, again, where are you going to recommend people look at? I mean automation is a big subject. There are other areas that can be improved. What caught your eye? You know, people reading this report, where are they going to start to think about Investing to improve their performance.

Speaker C: Correct? Yeah. Um, great question, Charles. I feel the first thing they have to look is their net impact on disputes. You get your recovery, you have your cost. At the end of the day you want to recover more, you want to decrease the cost at your institution. Right. So the way I look at things, when the members of our roundtables come in and discuss things, it's like, what is everybody doing? Um, to make sure that we're on the, on the positive net impact. Right on that. Um, you know, if, if we're on the positive, everything's looking good. If we're in the negative, okay, so what are we doing wrong here? So now there, it'll give you the ability to work with your, with your dispute teams, work with your fraud team, work with your operational team and make sure they're, they're, they're aligned, uh, towards the common goal. Right. Because at the end of the day what they want to do is that is um, uh, you know, reduce your cost, operational cost, while improving your recovery dollars. You know, everything else for me, you know, is yes, a lot of data. Yeah, you're, you're showing the effectiveness of your dispute team, but at the end of the day, the way you're going to be, you know, judge, uh, or you know, ranked is based on, on your, are you reducing your operational cost and the automation will do that for you.

Speaker B: So those are great things to be looking out for. Um, Ike, perhaps you could talk us through again. There's this huge differential between the best performing institutions and the laggards. What do the best performing institutions do differently to the worst performing institutions?

Speaker A: Yeah, um, I think when we look at productivity and just the ability to get through more cases, um, you know, it's automation, but if you look at it a little bit more granularly, um, we're really focus, firing users. Right. So you have institutions where they have a large back office and the tasks that they're working are the tasks that you want a human working. Right. They're doing final investigations, they're doing initial investigations. What they're not doing is going into the association's system and pulling out extra transaction data. They're not entering in case data, they're not firing off a pre arbitration like all of those really heavy lift but repeatable processes get automated. Um, and that's what allows us to really focus fire when we have a human in front of a screen and hit the important things and cycle through cases in a quicker fashion.

Speaker B: Perfect. And Jaime, again, you've been up close with These organizations. What do you notice the difference between the best performers and the worst performers? Is it maybe the philosophical difference that some are investing in automation, that some have their eye on the right benchmarks? Um, just elaborate for us in your view, what makes the difference between the top and the bottom.

Speaker C: Yeah, another great question, Charles. Um, you know, during my roundtables, you know, uh, I hear, I hear a lot of, a lot of rates, you know, left and right and this and that. What I, what I see from the best organizations is that they optimize the net financial impact, right? They make decisions based on expected financial outcome. Not, not so much on uh, uh, not so much on volume and things like that. Right? It's dollars is dollars. You know, adjust your strategies. People are adjusting their strategies based, uh, on dollars. You know, where, where are we seeing the losses? Where are we seeing instead of so much the volume? Right, because the volume could be like I said, you know, you know, you pick and choose, uh, where you want to work the easy, low hanger fruits and things like that. But if you, if you focus on where the dollars are being spent or losses, that's where you want to go and strategize what's great.

Speaker B: Because it really um, breaks the results down by financial organization, segment. I mean there's this fintech, credit union, bank industry and processor and you see some quite stratified results between them. Again and again we see fintechs performing really well on almost all metrics. And again it's that philosophical that they're dealing with. They've got modern systems, they have a psychology that invests in automation compared to, as we know, banks and some credit unions can be a little bit old fashioned. Is that what it comes down to?

Speaker A: Um, in general, yes. Uh, you put it pretty well. I think fintechs are kind of at the cutting edge of what we can do with technology and where we can leverage it to optimize operations not only in the fraud and dispute space, but probably across all of their, their organization. Um, I would, I would imagine that if we looked at data across different segments of their industry, you would probably see a similar pattern. Um, but it really does boil down to fintechs having kind of the urge and the desire to invest heavily in technology. And what we often see is they tend to have a smaller operating group. Um, you know, where you might have had an enterprise bank, where you would expect to have 150, you know, back office investigators of fintechs don't have those numbers. And it's because they're leveraging Technology to assist in that volume.

Speaker B: Just to give you a very tricky question. How can these numbers be misinterpreted? We began this conversation. You said, watch out for stuff like a win rate. It's not always as revealing as you think it is. Now, I'm looking at a number here of recaptured dispute dollars. To me, that feels like a great measurement. I'm looking at, um, banks recovering 86% of dispute dollars. Um, industry down at 74%. But just give us a warning. How can some of these numbers be misinterpreted?

Speaker A: Yeah, I think recapture is a good example. Um, recapture is kind of a collective term that Quavo uses, and what it represents is the inverse of your loss. So it's not necessarily a term for just recovery, but it's recovery plus merchant credits, plus your denials. It's any dispute that came in the door that you had the potential to take the loss that you didn't because you recaptured it either through traditional chargeback or, you know, um, a merchant credit, or you were able to deny the case. Um, it's really easy to look at a high percentage recap.

Speaker B: Sure.

Speaker A: But without understanding what built that, um, it can be misconstrued. Um, you know, denial is kind of the hot and heavy word in this industry. You want to be cautious. Anytime you're denying a claim, make sure you have your ducks in a row. Um, what we can see, you know, on our platform and how people are operating today is not only are they recovering more, but they're able to deny more. Because we really build out an end to end process from intake to investigation, to where they're able to catch some of the bad actors early on and successfully recapture funds through that denial path.

Speaker C: That's great, Charles. Uh, uh, sorry for jumping in. I would say, and I don't know if you agree with me, Ike, or not, but this recapture the speed dollar is probably the number one metric in here or the one that catches the eye the most. Uh, this is what's going to, uh, tell you if you're being effective or not. That's my personal belief.

Speaker B: All right, yeah, keep going, Jaime. Because you said you hear people's qualitative concerns, um, when you hear organizations talking, what really concerns them? What's the. From an emotional level, what are they really worried about?

Speaker C: That's a big one, Charles. Um, being in so many different roundtables, I lead the, uh, chargeback disputes. I leaked the car fraud. I leaked the fintech fraud. Uh, the number one issue that I'm seeing today, of course, is first party fraud, uh, by, by far. Right. It's very hard to prove. Very hard to prove. It's very, uh, you know, hard to, to recover dollars, um, and a lot of over crediting, you know, as soon as the client calls the bank, hey, I, I was a victim. The bank is crediting the client instead of going through the whole, uh, dispute process. So it is, it's very hard to get a hold of over crediting in exchange of customer experience. Right. They want to, you want to make sure the clients are happy, satisfied, but they're losing money. They're, they're, they're not recovering that amount. Uh, so first party fraud will be the biggest challenge. The second, I would say is not getting the right data. So some clients do not trust their own data and they don't have any benchmarking ability. Right. Um, so if you're able to benchmark yourself with your peers, in this case with this report, you have the industry processor, credit union, fintech, you want to know where you stack against your peers, uh, for the same type of metric. So some of those clients don't have the, uh, visibility, uh, of benchmarking. That is definitely a must. I believe that is something that leaders should be focused on as well.

Speaker B: That's so valuable. We have a lovely question from the audience and if anyone watching wants to ask a question, uh, to Ike and Jaime, please, um, pose it because we love asking them difficult questions. We have from Clive Anderton. Maybe I could have a shot at this one. How do we measure the true operational effectiveness of recapturing dispute dollars when high recovery rates can be offset by soaring manual investigation costs and customer churn?

Speaker A: Yeah, yeah, I mean, it's a good question, I think. Um, you know, operational effectiveness is a constantly swaying metric. You can look at simple things like, you know, tasks per user per hour. Right. But that doesn't really paint the full picture. Um, so a lot of it is around time spent and where you're, where you're dedicating. You know, human eyes. I think we look at it in a pretty simplistic fashion, which is if it's repeatable and the values are similar, automate it. Um, you can measure operational effectiveness much quicker if you only have users looking at two things, which is invest like initial investigation and then, you know, final investigation. If everything else coming in the door is automated because it's repeatable and you can put logic behind it, you have a really clean path to understand how much you're truly spending on your opex versus what your outcomes are. When it comes to a recovery.

Speaker B: Jaime, if I just tweak the question for you. The purpose of this report is to give organizations numbers that they would, under other circumstances, have no access to. Do you feel a lot of banks and fintechs and credit unions maybe operate in the dark, that the answer to this question may be is a lot of organizations would struggle to actually measure their effectiveness.

Speaker C: Correct, Correct. A lot of organizations don't have the ability to get this type of data that Kuaw was presenting here or that if you're a member of a realma, you're able to get from benchmarking. Uh, we have, I don't know how many 50 plus, 60 plus different metrics that you can get your hands on, uh, in our benchmarking. And you can get more by combining some of those metrics together. Right. So yes, a lot of the leaders are playing, uh, with, uh, with no visibility of what, what was out there. So they don't know if they're actually being effective or not. Kind of what you, I think you asked the question earlier. You know, some of the, some of the metrics can be misconceptions. Right. Uh, like the win rate. Like the win rate. Just to give you an example, you know, if I hear, I, I hear sometimes clients coming back to me saying, hey, I have a 50, 60, 70% win rate. Wow, that sounds amazing. Let's go ahead and do a little bit more investigation on that. Yeah, they're not giving you the real story. They're only representing the ones that they know they're going to win, but they're actually leaving, omitting the ones that they can't win. So it's not really 50, 60%. So it can definitely be, uh, I mean, you have to keep a good eye, uh, what data, what benchmarkings are out there in order for you to, you know, have visibility of what others are doing as well.

Speaker B: That's great. And just looking at the performance differences, I'm looking at the report at the combined losses, uh, the difference. Quavo likes to just ring fence the best in class and compare their activities to other market segments. And I can see that when it comes to losses, there's an eight fold improvement from the best in class to the, uh, processors who are struggling with, um, best in class losing only 3% compared to 24% of the processors. The performance difference is absolutely massive. Um, we've got a lovely interesting question from the audience. This is a very technical question from Elizabeth Leon. I get this. Um, she has a question about card testing. How can we effectively combat this type of fraud? Specifically, do you recommend implementing 3D Secure Access across all transactions, or are there better alternatives? I don't know the answer to this. What do you recommend, Ike?

Speaker A: Yeah, that's a good question. Um, and the landscape today merits it. Um, you know, a few years ago I maybe would have said yes. I think it's shifted enough now that, you know, you really have to be really agile in how you implement fraud strategy. Um, and it changes day to day. You kind of have to jump in on that. I think, at least for me, what I would say combats first party fraud best is introduction of friction. Um, and I know no one wants to do that. Right. Like it creates sometimes a negative experience for a cardholder that's truly just trying to transact, but the volume of which you're encountering, you kind of have to, um, 3D secure is a way to do that. You know, two way SMS is another strong combative method that really isn't a huge friction point. But it very easily gives you a quick and clear line of sight into being able to deny a claim when that comes back around from a first party fraudster. Um, there's a lot, uh, that's just changing constantly. Um, that's a really good question that's making me think harder than I wanted to.

Speaker B: I think you've handled it well. Yes, Jaime.

Speaker C: Go ahead, Charles. Uh, uh, something to keep in mind. And I believe this was, uh, Elizabeth, right, that sent. The question is do. I'm with Ike a couple years ago, three years ago, will blindly say yes. Now I'll probably have to sit down and think about it a little bit. But if you do decide to move forward with the 3Ds, make sure the implementation is the one you want. I'm seeing a lot of clients coming in, hey, I implemented three ds, but I don't have the ability to do an otp. So clients are just getting burned with fraud. Right. You know, these are ECI 5. If anybody knows here, ECI 5 secure transactions. Uh, if you're not able to OTP, those transactions are going through, uh, your, you will not be able to recover those funds. So make sure that you look at the way you're implementing it. And if they're not allowing you to do OTP in the first phase, they may tell you, hey, let's do it in second, third, fourth phase. Uh, absolutely not. Uh, either you go all in or you don't go in at all.

Speaker B: Love it. Great question again from the audience, uh, from AFI Summers. Um, she says you Mentioned not trusting data. Can you expand on why leaders don't always trust their data? How can data analysts overcome that attitude? Ike, leaders are not trusted?

Speaker A: Yeah, I think, uh, that's a really good question. And it's something that we see pretty often. There's probably two things that occur the most when we have either a new client or a prospect coming over. Um, number one is that they don't, they don't even have the data. Like, they're not collecting it or they don't know what it is or what it means. And the second one is they are collecting the data, but they don't have a very good way of translating that into true insight or operational value. Um, you know, I think, you know, it was called out really well. How can data analysts overcome that attitude? Data analysts have to be able to, like, picture a story. They have to, they have to write it out so that you can see what it means to an operator. An operator is going to look at a bunch of numbers on an Excel sheet and not. Not know what to do with it. You have to, you have to tell a story with that data so that you understand what's actionable and what can maybe be disregarded. Um, but it is a common thing across the industry. And because there's so much data out there now, and we're doing a lot with it, um, I anticipate that will slowly kind of erode. But, yeah, you know, a mom and pop credit union in Michigan may not really have the resource or really the history to understand the volume of transaction data coming into their organization day to day.

Speaker B: Love it. Hi, mate. I see you nodding away there. I mean, I love this idea of people not trusting the data. Is that because they're skeptical of the data or they think, um, that it doesn't necessarily apply to their organization? You know, every bank.

Speaker C: No, no, they don't know. They don't know how to translate it. Charles. That's exactly this. It's exactly it. So they, they have the Excel spreadsheet, but what does that mean? What, what is that telling me? Or what is that telling you? Uh, that's, that's, that's where the problem is. So it's not so much that they don't trust. It is one, they don't know how to do it if they're able to get to the data. Uh, that's a whole nother thing. Right. Uh, so that's that. So I was, I was kind of smiling a little bit because that's exactly, that's exactly it. You know, we have clients. Oh, we have this data. We have this data, but they don't know what to do with it. They don't know how to translate into insights.

Speaker B: We have another question, Douglas hall, that just juxtaposes onto this. What type of data would make it easier to trust? I mean, Ike, do we need more data? I mean, the amount of data in this report is phenomenal. I mean, my feeling is that people really ought to have the numbers there. But what do you sense? Are there some numbers that would make it easier for people to develop a trust?

Speaker A: Um, if it's data that you can back up and see, I think is the biggest thing. Right. It's really easy to put data on a piece of paper and show you numbers in a visual. Um, but if I give you a report that I can tie back to transactions that, you know, occurred, that you have historical disputes on, that you can run back the story against, makes it a little bit easier to trust. You can understand where the trend is happening. I mean, disputes are cyclic. We see this all the time. It's seasonal. Right. You can always anticipate a January, February spike after the holidays. Um, if I can show you data to support what were the highest transaction types that got hit during that spike, and what could you do to implement a strategy to avoid it? It's a little bit easier to trust something like that than me blindly throwing an Excel at you with a bunch of numbers on it. Um, it's really around again, like, you know, telling a story with the data that you have and tying it to something tangible.

Speaker B: I suppose it's. Yeah. Leading on from that. It's about being able to talk about the data and have those conversations. I mean, Jaime, how much do you think it helps organizations to be able to just have a dialogue, not just about their own number, but about industry benchmarks and just develop a habit of having a conversation around numbers which will then build trust.

Speaker C: Yeah. So what's going to help organizations, Charles? So, first of all, from the data itself, again, kind of what I mentioned, uh, earlier, you know, you get your recapture data, you get your cost. Ah. If you look at the numbers, if you're in the net and the net negative impact, you know, you can go in a room with your fraud teams, you get in with your dispute team, you go in with your operational statement and get aligned. Right. So front leaders, you know, when you get those three teams aligned, looking at what kind of, uh, performance you're seeing, that's what's going to get the best out of your. Out of Your, uh, you know, dispute teams.

Speaker B: That's great.

Speaker C: So strategize and go from there.

Speaker B: Uh, we have a question from the audience which again touches on one of the most important issues of the entire report, which is where would AI automation make the biggest impact to dispute operations? I mean, huge topic, Ike, automation. Where should it be focusing?

Speaker A: Everywhere. Everywhere. But I think if you want to, if you want to segment it a bit, uh, realistically, one of the larger things that you still have a human sitting in front of is representance. Um, they're unpredictable. You know, it could be seven pages, it could be 50 pages. You've got to scroll through it, you've got to pick out information, you've got to compare it to system of record. It's a lot. Um, the process takes a long time to just manage one of them. Generally, uh, I think you would get the largest lift operationally by implementing AI there, um, being able to scrape data off a representment, summarize it, and then tell you, here's what the representment says, here's what your system of record says. And you can very easily checkbox yes, no, yes, no, yes, no, and maybe reduce a process that goes from, you know, 30 to 45 minutes down to 15. And that's a huge cost save if you replicate that across every presentment in your door.

Speaker B: And uh, just when it comes to. Oh, sorry, Jaime, but go ahead.

Speaker C: No, just one more thing really quick. You know, I always, um, what I tell, you know, our clients and what they tell me is that, yeah, they want more, more, more AI, more AI, more. But make sure you're tuning those out because, uh, you don't want a, a, a, a automation or AI and making the wrong decisions. Right. That's just going to derail everything downstream. So it's not so much about more and more and more, it's about tuning it and making sure that you're getting the right decisioning and, and go from there. So, you know, automation is, is one of the key, but is good automation as well. Right? Uh, that's where it's good automation.

Speaker B: Yeah. Well, we've got a question from Shelley Summers in the audience, which I kind of like, which is more on AI. How do we prevent this accelerated front end from creating a massive bottleneck in downstream manual investigation and evidence gathering? That's a very sensitive and well posed question, Ike. How do we stop this? You know, AI in one department can overwhelm a back end department.

Speaker A: Um, well, two things. Use AI in both departments and capture your evidence on the front end. Um, you know, enrich transactions as they come in the door, get all of the transaction data up front. I would venture a guess that most organizations are still swivel chairing between systems to capture all the evidence that they need when they do an investigation. How's it all in one place? Get everything in one spot, Review it all in a concise manner that's, that's systematic. Um, but the reality of it is wherever you leverage AI on the back end, you should counterpart it on the front or at the very least automate some of that evidence collection so that it's sitting in one one system of record that every investigator has easy access to.

Speaker B: Um, hi Major. Do you have any thoughts on this AI being implemented in one part of the organization rather than another?

Speaker C: No, I mean, I think I'd said it best, I'd said it the right way. That's how I would do it as well.

Speaker B: Makes perfect sense. Just give us a clue. The report talks about the last year's report versus this year's report. It looks as though the numbers are kind of stable. Is that a fair interpretation, Ike? There's some gains here or there, but a lot of the last year, this year shows similar performance. There's no disaster. There's plenty of room for improvement though.

Speaker A: Yeah, it's relatively consistent. And I think that speaks to two things. Um, number one, it speaks to how these organizations are able to manage increases in volume, so scalability. And then number two, although you know, most people don't feel like we're winning this battle against first party fraud, at the very least we're keeping it at bay. Right. Um, year over year we're maintaining it. Our strategies are improving. There's more and more technology coming out that we can leverage. Um, their incremental wins, but they're wins nonetheless I think is the way I would look at that.

Speaker B: Incremental.

Speaker C: I would agree with that, Charles, as well as the first party fraud. You know, they're tools that we have. I mean we're getting better and better and better tools, but the ones that are frequently being used are not, are not catching that first party fraud. And it is a massive issue in the industry. So I think, um, uh, you know, that, that, that will probably be the one thing that may be increasing from last year's. Um, but, but I mean it's something to keep an eye on.

Speaker B: I thought I'd even push a little harder than that. Incremental improvement. When I look at the report, I see the difference between uh, the performance in best and worst. Absolutely huge. I'm um, looking at, um, for example, days to chargeback in Quavo. Best in class, um, two days compared to some banks, 12 days. These are big, big differences. Ike, just tell us, when people read this report, should they be thinking about making incremental improvement, 5, 10%, or is there the opportunity to really find a 3x4x5x improvement in performance depends on your appetite.

Speaker A: Um, I think both are valuable attempts. You can really go either direction. Um, I say automation is a very high level solution to a lot of this. There's different parts of the process you can automate. To make things simpler, a big leap would be chargeback rights determination, being able to automate that really quickly. Um, that's a large lift off your operational team. Another one that folks don't often consider but probably should is like provisional credits. At what point is a provisional credit really even worth having a user do an initial investigation? $5 worth it because you're probably paying the user more than that to just sit there and investigate it. Like having a threshold at which you're, you're comfortable and you accept the risk to say like we're, we're just going to issue on this one and then we'll follow up later on down the line. That's going to create those little incremental wins where you can move from, you know, eight days down to four.

Speaker B: And Jaime, tell me, so what are we thinking, Jaime? Incremental or step change? And it must surely depend on the organization people work for. But the gap between best and worst in this report is pretty startling, isn't it?

Speaker C: Oh yeah, absolutely. And it also, I mean it shows, it shows who has their, you know, who's going towards the right step. Right. Who's being effective with the data that you're showing. Absolutely. There's a big gap and everything we talked about here, you know, uh, are you comparing yourself with your recapture recovery, your combined losses by dollar. Uh, how good is your intake and decisioning? Uh, how do you reprioritize your work queue? How do you. I um, mean my kind of what Ike was saying, you know, where do you drive the threshold? You know, I will cut all low value, uh, activity. I mean it will definitely assist us with the, with the speed and the activity. But at the end of the day I go back to, you know, the most important ones are, or the recapture rate and that combined losses by dollar amount. Right. What's your cost? You want to make sure that at the end of the day the leaders want, like I said, uh, uh, increase the Number of uh, recapture rate or recover amounts and less cost.

Speaker B: And a final question guys, we've covered the report brilliantly is just again to emphasize how should organizations be using this? There's this ton of numbers here. Are these numbers intended for company benchmarking? They should be looking at best in class and thinking, yeah, we can be like that. What advice do you have for organizations who read this report on how they should get the most out of it?

Speaker A: Um, I think to what Jaime said earlier, number one is use it as a comparison to yourself. Understand where you're performing today and understand if that fits into the segment of your business. Like if you're, you know, if you are a fintech, are you aligned with how these, these folks are performing? If you're not, you know, where, where do you potentially have some opportunity and then secondary to that, you know, when you look at the best in class performance, I'm not necessarily saying don't target that as a goal, but understand what it takes to get there. Like dig into, to what it means to your business, your values, your operations to get to that point. Um, and then you can kind of pick and choose where you want to go as far as what you want to invest in from a technology perspective, where you want to leverage AI, where you want to look at automation, um, and kind of pick and choose your goal set from there. Again, incremental wins. But it's easy to see all of this and not absorb it. Look at where you sit first and understand where you want to be.

Speaker B: Terrific. And Jaime, just to finish us off, what advice do you have for people?

Speaker C: I mean what I've noticed is uh, having the most people or having the best, uh, or the best tools does not actually tell you that you're going to be one of the best out there. My suggestion, my feedback is hey, if you do see that you're not uh, you know, in, in the middle of this report, if you're, if you're not doing as good, get outside help, right? Get, get, get some, you know, the people out there that knows what they're doing to help you, uh, get your disputes in control to automate your, your, your processes and, and things like that. Right? You know, there are people out there that do this for, for a living. Uh, you can definitely go ahead and uh, and you know, get that help if you need to.

Speaker B: Fabuloso. Well, it's a blockbuster report. I um, understand it's available on June 2nd for full download, is that correct?

Speaker A: Correct. Available June 2nd for full download. And I believe we shared it here for those attending, so you should have access to it as well.

Speaker B: The link is there. So highly recommended. It really does contain every number that's relevant in the, um, in the disputes field. So it doesn't matter which organization you work for. It really is worth a deep dive. Guys, thank you so much for talking us through the report and highlighting the most important areas. That's hugely appreciated. Ike. Thank you so much.

Speaker A: Yeah, thank you.

Speaker B: And Jaime too, thank you for joining us.

Speaker C: Thank you so much. Thank, uh, you for inviting, uh, me.

Speaker B: Thanks for coming and for everyone watching. Thank you for your time from me and the production team here at fintech Talk. Thank you. See you next time. Mhm. It.